I  COMMISSION OF THE
   EUROPEAN
   COMMUNITIES
                           WORLD HEALTH
                           ORGANIZATION
    UNITED STATES
   ENVIRONMENTAL
PROTECTION AGENCY
                INTERNATIONAL  SYMPOSIUM


                         PROCEEDINGS


                        Recent Advances
             in the Assessment of the Health Effects
                   of Environmental Pollution
                             Volume III




                       !
      .-.
                                                  ^.••v 15 _




            _ __- /.--








   Paris. 24 to 28 June 1974

-------
LEGAL NOTICE

The Commission of the European Communities
and its departments decline all
responsibility with regard to the use of
the information contained herein.

-------
                           TAGUNGSBERICHTE
                      INTERNATIONALES SYMPOSIUM

              NEUESTE ERKENNTNISSE  IN DER BEURTEILUNG
        DER GESUNDHEITLICHEN FOLGEN  DER UMWELTVERSCHMUTZUNG
                             PROCEEDINGS
                       INTERNATIONAL SYMPOSIUM

                  RECENT ADVANCES  IN THE ASSESSMENT
         OF THE HEALTH EFFECTS  OF  ENVIRONMENTAL  POLLUTION
                                ACTES
                       SYMPOSIUM  INTERNATIONAL

               PROGRES REGENTS DANS  L*EVALUATION  DES
       EFFETS DE  LA  POLLUTION DE  L'ENVIRONNEMENT  SUR LA SANTE
                                ATT I
                       SIMPOSIO INTERNAZIONALE

                 RECENT I  PROGRESS I NELLA VALUTA!IONE
    DEGLI EFFETTI  DELL' INQUINAMENTO  DELL' A.'.BIENTE  SULLA SALUTE
                               VORSLAG
                      INTERNATIONAL SYMPOSIUM

            RECENTE  VORDERINGENBIJ  DE VASTSTELLING  VAN
     DE GEVOLGEN  VAN MILIEUVERONTREINIGING VOOR DE  GEZONDHEID
                  Paris (France)* June  24-28, 1974
                          organized jointly by

               CEC - Commission of the European Communities
    Directorate General for Social Affairs - Health Protection Directorate

            EPA - United States Environmental Protection Agency

                                 and

                    WHO - World Health Organization
          Published by the Commission of the European Communities
irectorate General Scientific and Technical Information and Information Management
                           Luxembourg, 1975

-------
                                    LXV
ORGANISATIONSAUSSCHUSS / ORGANIZING COMMITTEE / COMITC  ORGANISATEUR

             COMITATO ORGANIZZATIVO / ORGANISEREND COMITE


         Vorsitzend / Chairmen / Presidents / Presidenti / Voorsitters:

                         B.H.  DIETERICH  (W.H.O.)
                          S. GREENFIELD  (E .P A .)
                            P.  RECHT (C.E.C.)


        Stellvertretende  Vorsizende / Vice  Chairmen / Vice-presidents /
                      Viaepresidenti / Vice-Voorzitters:

                            D.  EARTH (E.P.A.)
                         M. CARPENTIER (C.E.C.)
                           J.  KUMPF  (W.H.O.)


           Generalsekretar / Secretary General / Secretaire general /
                  Segretario generate / Algemeen Searetaris:

                           J.  SHEETS (C.E.C.)


           Wissenschaftliohe Sekretariat /  Scientific Secretariat /
            Secretariat  soientifique / Segretariato saientifico  /
                        Wetenechappelijk Secretariaat:

                           A. BERLIN (C.E.C.)
                            R.  ENGEL (E.P.A.)
                           V.B. VOUK (W.H.O.)


        Aaaistentin / Assistant  / Assietante / Assietente / Asaistent:

                         G. TREU-RICCO1  (C.E.C.)


              Wissenschaftlicke  Beratcr /  Scientific advisors /
             Conseillera scientifiques  / Consiglieri scientific^ /
                        Wetenechappelijke Adviseurs:

            Ph.  BOURDEAU  (C.E.C.),  G.C. BUTLER  (Canada)
         M. COLOMBINI  (Italy), G.  DEAN  (Republic  of Ireland),
              J.F. PINKLEA (U.S.A.), L.  FRIBERG  (Sweden),
              H.E. GRIFFIN (U.S.A.), W.W.  HOLLAND  (U.K.),
               M. KEY  (U.S.A.), A.  LAFONTAINE  (Belgium),
           D.  RALL (U.S.A.),  H.-W.  SCHLIPKOTER  (F.R. Germany),
           R.  SENAULT  (France), J.  SPAANDER  (Netherlands),
               L. TEPPER  (U.S.A.),  N.K.  WEAVER  (U.S.A.)

-------
                         LXVII

                   TAGUNGSBERICHTE
                INTERNATIONALES  SYMPOSIUM
         NEUESTE ERKENNTNISSE  IN DER  BEURTEILUNG
   DER  GESUNDHEITLICHEN  FOLGEN DER  UMWELTVERSCHMUTZUNG


                         Band  I
                    ERO'FFNUNGSSITZUNG
                POPULATIONSUNTERSUCHUNGEN
   UNTERSUCHUNG  DER  WIRKUNGEN  AUF DEN MENSCHEN(Panel)
              TOXIKOLOGISCHE UNTERSUCHUNGEN
              EXPOSITIONSMESSUNGEN  (Panel)

                         Band  II
       UNTERSUCHUNG  DER  WIRKUNGEN AUF DEN MENSCHEN
                  TIERUNTERSUCHUNGEN
                      STOFFWECHSEL
                  WECHSELBEZIEHUNGEN
                 EXPOSITIONSINDIKATOREN

                        Band III
                    GEWEBSMESSUNGEN
        ERFORDERNISSE IM HINBLICK AUF MESSUNGEN
                    UMWELTMESSUNGEN
                 EXPOSITIONSMESSUNGEN
                        MODELLE
                 ZUSATZLICHE BERICHTE

                        Band IV
  DIE WISSENSCHAFTLICHEN DATENGRUNDLAGEN, DIE FUR DIE
ENTSCHEIDUNG UBER DEN GESUNDHEITSSCHUTZ  BENOTIGT WERDEN
       (Aussprache im  rahmen des  gesamtausschusses)
               EXPERIMENTELLE FORSCHUNG
    ERFORDERNISSE IM HINBLICH AUF MESSUNGEN (Panel)
                 GEWEBSMESSUNGEN (Panel)
      UNTERSUCHUNG DER GESUNDHEITLICHEN WIRKUNGEN
                     SCHLUSSITZUNG
                    TEILNEHMERLISTE

-------
                        LXVIII

                    PROCEEDINGS
              INTERNATIONAL SYMPOSIUM
   RECENT ADVANCES  IN THE ASSESSMENT OF THE HEALTH
        EFFECTS OF ENVIRONMENTAL POLLUTION

                     Volume I
                  OPENING SESSION
                POPULATION STUDIES
           HUMAN EFFECTS STUDIES (Panel)
               TOXICOLOGICAL STUDIES
            EXPOSURE MONITORING (Panel)

                     Volume II
               HUMAN EFFECTS STUDIES
                  ANIMAL STUDIES
                    METABOLISM
                   INTERACTIONS
              INDICATORS OF EXPOSURE

                    Volume III
                TISSUE MEASUREMENTS
                 MONITORING NEEDS
            ENVIRONMENTAL MEASUREMENTS
                EXPOSURE MONITORING
                      MODELS
               SUPPLEMENTARY PAPERS

                     Volume IV
THE SCIENTIFIC DATA BASE REQUIRED FOR DECISIONS TO
  PROTECT HUMAN HEALTH   (Plenary discussion group)
            EXPERIMENTAL INVESTIGATIONS
             MONITORING NEEDS  (Panel)
            TISSUE MEASUREMENTS  (Panel)
              HEALTH EFFECTS STUDIES
                  CLOSING SESSION
               LIST OF PARTICIPANTS

-------
                            LXIX

                           ACTES
                  SYMPOSIUM INTERNATIONAL
           PROGRES RECENTS DANS l/EVALUATION DES
  EFFETS DE LA POLLUTION DE L'ENVIRONNEMENT SUR LA SANTE

                         Volume I
                    SEANCE D'OUVERTURE
                 ETUDES SUR LA POPULATION
           ETUDES DES EFFETS SUR L*HOMME  (Panel)
                   ETUDES TOXICOLOGIOUES
               MESURE DE L'EXPOSITION  (Panel)

                         Volume II
               ETUDES DES EFFETS SUR L'HOMME
                  ETUDES SUR LES ANIMAUX
                        METABOLISME
                       INTERACTIONS
                 INDICATEURS D'EXPOSITION

                        Volume III
         MESURES RELATIVES AUX TISSUS BIOLOGIQUES
        BESOINS EN MATIERE DE MESURE DE L'EXPOSITION
               MESURES DANS L'ENVIRONNEMENT
                  MESURE DE L'EXPOSITION
                          MODELES
                 RAPPORTS SUPPLEMENTAIRES

                         Volume IV
          LES DONNEES SCIENTIFIQUES REOUISES  A
LA PRISE DE DECISION POUR LA PROTECTION DE LA SANTE HUMAINE
        (Groupe de discussion en assemblfie plfiniere)
                 RECHERCHES EXPERIMENTALES
    BESOINS EN MATICRE DE MESURE DE L'EXPOSITION  (Panel)
      MESURES RELATIVES AUX TISSUS BIOLORIQUES  (Panel)
              ETUDES DES EFFETS SUR LA SANTE
                     SEANCE DE CLOTURE
                  LISTE DES PARTICIPANTS

-------
                            LXX

                           ATT I
                  SIMPOSIO INTERNAZIONALE
            RECENTI PROGRESSI  NELLA VALUTAZIONE
DEGLI EFFETTI DELL'INOUINAMENTO DELL'AMBIENTE SULLA SALUTE

                         Volume I
                      SESSIONE  INAUGURALE
                  STUDI SULLA  POPOLAZIONE
           STUDI DEGLI EFFETTI SULL'UOMO (Panel)
                  STUDI TOSSICOLOGICI
              MISURA DELL'ESPOSIZIONE  (Panel)

                         Volume II
               STUDI DEGLI EFFETTI SULL'UOMO
                    STUDI SUGLI ANIMALI
                        METABOLISMO
                        INTERAZIONE
                 INDICATORI DI ESPOSIZIONE

                        Volume III
               MISURE NEI TESSUTI BIOLOGICI
      NECESSITAX RELATIVE ALLA MISURA DELL'ESPOSIZIONE
                     MISURE AMBIENTALI
                  MISURA DELL'ESPOSIZIONE
                          MODELLI
                  RAPPORTI SUPPLEMENTARI

                         Volume IV
         I DATI  SCIENTIFICI DI BASE RICHIESTI PER
 PRENDERE DECISIONI  AL FINE DI PROTEGGERE LA SALUTE UMANA
                (Gruppo di discussione plenaria)
                   INDAGINE SPERIMENTALE
NECESSITA' RELATIVE ALLA MISURA DELL'ESPOSIZIONE  (panel)
           MISURE NEI  TESSUTI  BIOLOGICI (Panel)
             STUDI  DEGLI  EFFETTI  SULLA SALUTE
                     SESSIONE FINALE
                  ELENCO DEI  PARTECIPANTI

-------
                             LXXI

                            VORSLAG
                   INTERNATIONAL SYMPOSIUM
            RECENTE VORDERINGEN BIJ DE VASTSTELLING
 VAN DE GEVOLGEN VAN MILIEUVERONTREINIGING VOOR DE GEZONDHEID

                             Band I
                        OPENINGSZITTING
                       BEVOLKINGSSTUDIES
        ONDERZOEKINGEN NAAR EFFECTEN BIJ DE MENS (Panel)
                 TOXICOLOGISCHE ONDERZOEKINGEN
                   METING VAN EXPOSIE  (Panel)

                             Band II
           ONDERZOEKINGEN NAAR EFFECTEN BIJ DE MENS
                   ONDERZOEKINGEN BIJ DIEREN
                          METABOLISMS
                          INTERACT IES
                    INDICATOREN VAN EXPOSIE

                           Band III
                METINGEN VAN BIOLOGISCH WEEFSEL
   EISEN VOOR HET TOT STAND BRENGEN VAN TOEZICHT OP EXPOSIE
                    METINGEN IN HET MILIEU
                     METINGEN VAN EXPOSIE
                           MODELLEN
                   SUPPLEMENTAIRE DOCUMENTEN

                            Band IV
  WETENSCHAPPELIJKE GEGEVENS NOODZAKELIJK VOOR HET NEMEN VAN
  BESLISSINGEN TER BESCHERMING VAN DE GEZONDHEID VAN DE MENS
                   (Pienaire discussiegroep)
                    EXDERIMENTEEL ONDERZOFK
EISEN VOOR HET TOT STAND BRENGEN VAN TOEZICHT OP EXPOSIE  (Panel)
            METINGEN VAN BIOLOGISCH WEEFSEL (Panel)
        ONDERZOEKINGEN NAAR GEVOLGEN VOOR DE GEZONDHEID
                          SLOTZITTING
                        DEELNEMERSLIJST

-------
INHALTSVERZEICHNIS




     CONTENTS





TABLE DES MATIERdS





      INDICE





   INHOUDSOPGAVE

-------
                                   LXXV
GEWEBSMESSUNGEN
TISSUE  MEASUREMENTS
MESURES RELATIVES  AUX TISSUS BIOLOGIQUES
MISURE  NEI TESSUTI BIOLOGICI
METING  VAN BIOLOGISCH WEEFSEL

    Evaluation en microscopic optique du degr£ de I'empoussierage
    pulmonaire asbestosique  par microfiltration de I1expectoration
    et  du sue  gastrique	1189
    J.  BIGNON, B.  HEM, G.  BONNAUD, P.  SEBASTIEN,
    G.  MONCHAUX

    Etude quantitative en tnicroscopie optique £lectronique  de
    I'empoussierage pulmonaire et pleural chez vingt sujets ...   1199
    A.  FONDIMARE,  J. DESBORDES,  J. TAYOT,  J. PERROTEV

    Determination of cadmium in blood using a U.OO mm paper
    disc-graphite furnace technique - Application to
    industry	1207
    A.A. CERNIK,  M.H.P.  SAYERS

    The incidence of elevated blood lead values in an asymptomatic
    pediatric  population residing in a major American industrial
    state	1223
    P.R. FINE, D.D. DOBIN

    Mercury in the placenta and  foetal membranes as an  indication
    of  low mercury exposure  	  1233
    A.  WANNAG, J.  SKJAERASEN


ERFORDERNISSE IM HINBLICK  AUF MESSUNGEN
MONITORING NEEDS
BESOINS EN MATIERE DE  MESURE DE  I_'EXPOS1TION
NECESSITAX RELATIVE ALLA  MISURA  DELL'ESPOSIZIONE
EISEN VOOR HET TOT STAND BRENGEN VAN TOEZICHT  OP  EXPOSIE

    Establishing safe ambient  air quality  levels  for eighteen
    hazardous pollutants 	   1241
    S.L. WILCOX,  E.L.  KEITZ, L.J.  DUNCAN

    Method for the  identification of  environmental noise levels
    requisite to protect public  health and welfare 	  1249
    S.L. YANIV,  H. VON GIERKE

    Relationship between acute respiratory illness and  air
    pollution levels in an industrial city	1263
    D.  LEVY,  M.  GENT,  M. NEWHOUSE

    Environmental  contamination  by automotive lead 	  1277
    H.W. EDWARDS

-------
                                   LXXVI
    Petrols additivated with isotopically differentiated lead.
    Proposal of an  experiment to estimate the incidence of traffic
    on the environment pollution by lead - first experimental
    results	1287
    P.  GARIBALDI, S.FACCHETTI, A.  QUAGLIARDI,  G.  VANINI
    P.P.  GADDO,  M. DE  BORTOLI, P.  GAGLIONE


UMWELTMESSUNGEN
ENVIRONMENTAL MEASUREMENTS
MESURES DANS  L;ENVIRONNEMENT
MI SURE  AMBIENTALI
METINGEN IN MET  MILIEU

    Application du  compteur COULTER a I1evaluation de la mesure de la
    pollution atmospherique par la mesure des particules solidep en
    suspension dans 1'atmosphere 	   1303
    J.  RUMMER, C. ALLART-DEMUL,  D. HEBBELINCK,
    H.  BASTENIER

    Ultraspurenanalyse organischer Schadstoffe in der Luft mittels
    der sogenannten Thermogradienttechnik  	   1317
    J.  ANGERER,  A. HAAG,  G. LEHNERT

    Improved methods of sampling and analysis of airborne hazardous
    particulate pollutants 	   1331
    R.C.  LAO,  R.S. THOMAS, L. DUBOIS,  J.L.  MONKMAN

    Surface area and adsorption properties of dust particles .  .  .   1341
    W.  KARCHER,  P. CLAUDE, R. DEPAUS

    An  improved technique for monitoring particulates in the atmosphere
    to  check air quality standards   	   1351
    J.E.  EVENDIJK, A.  A.  ROOSKEN,   N.  NIEUWSTRATEN

    Analytical studies on the speciation of ambient levels of
    mercury in air	1363
    R.S.  BRAMAN,  D.L.  JOHNSON

    New techniques  for the measurement of air pollutants  .  .  ...  1371
    J.R.  McNESBY

    Specific hydrocarbon monitoring at low concentrations ....  1385
    R.  PERRY,  J.D. TWIBELL

    A system for the rapid identification of toxic organic pollutants
    in water	     1399
    W.T.  DONALDSON, M.H.  CARTER,  J.M.  McGUIRE

    Vergleichende Schwermetallanalytik mit Hilfe der Polarographie
    und  Atomabsorption - Moglichkeiten und Grenzen der
    Reproduzierbarkeit   	  1407
    W.  EGELS,  G.   CR6SSMANN,  H.  RETHFELD,  D. SEIFERT

-------
                                   LXXVII
     Spurennachweis schadlicher Elemente in  Luft, Wasser und
     biologischen Medien mit der Rontgenfluoreszenzanalyse ....   1417
     L. BEITZ, J. HAASE, N.  WEICHERT


EXPOSITIONSMESSUNGEN
EXPOSURE MONITORING
MESURE DE  L EXPOSITION
MISURA DELL ESPOSIZIONE
METING VAN  EXPOS IE

     Etudes preparatoire en vue de 1'etablissement d'un programme
     coordonne de surveillance de la contamination de 1'homme par
     les produits chimiques   	   1433
     E. FOURNIER

     Methodologie pour  1'evaluation de 1'exposition de 1'nomine
     resultant de la contamination des aliments par les metaux
     lourds.  Etude de  quelques cas concrets   	   1441
     R. BITTEL, A. GARNIER,  G.  LACOURLY

     Mercury  pathways to man and "in vivo" content of the population
     of the Mt. Amiata  area	1451
     P. CAGNETTI, L.  CIGNA ROSSI,  C.F. CLEMENTE,
     G.P.  SANTARONI

     Accumulation and release of petroleum hydrocarbons "by Edible
     Marine Animals  	  1461
     J.W.  ANDERSON,  J.M. NEFF


MODELLE
MODELS
MODELES
MODELLI
MODELLEN

     Nonlinear models in multivariate analysis applied in exposure
     assessment	1473
     V. STEPANEK

     A laboratory model system for the study of the accumulation
     of chemicals in the aquatic food chain, exemplified on
     a-Hexachlorocyclohexane   	  1470
     J.H.  CANTON, P.A. GREVE,  W. STOOFF

    Measures  of association of some air  pollutants, natural
     ionizing radiation and cigarette smoking with mortality
     rates	1491
     R.C.  SCHWING, G.C.  McDONALD

-------
                                  LXXVIII
     Body burden of  pollutants:  compartmental  stochastic models  .  .   1505
     A.H.  MARCUS

     Analisi di regressione dei  valori di piombo nel sangue
     sull'esposizione atmosferica  	   1511
     L.  FAVRETTO,  L. FAVRETTO GABRIELLI

     Sur un modele epidemiologique, stochastique et compartimental
     simulant 1'apparition et 1'aggravation de la surdite ....    1519
     M.T.  HO,  E.  QUINOT

     Evaluation de la charge corporelle en mercure par simulation
     numerique    	1531
     R.  MAGNAVAL,  R. BATTI,  A. BOUVILLE
                       ZUSATZLICHE  BERICHTE
                       SUPPLEMENTARY PAPERS
                     RAPPORTS SUPPLEMENTAIRES
                      RAPPORTI SUPPLEMENTARI
                     SUPPLEMENTAIRE DOCUMENTEN
UNTERSUCHUNG DER WIRKUNGEN AUF DEN MENSCHEN
HUMAN  EFFECTS  STUDIES
ETUDES DES EFFETS SUR L'HOMME
STUDI  DEGLI EFFETTI SULl'UOMO
ONDERZOEKINGEN NAAR EFFECTEN BIJ  DE MENS

    The use of Canadian medical data in defining health effects
    of the environment on man	1543
    C.A.R. DENNIS

    The relationship between lead and hyperactivity  	   1549
    O. DAVID

    Untersuchungen an 12.000 Patienten iiber den Einfluss der
    Luftverschmutzung und des  Arbeitsplatzes auf die Haufigkeit der
    chronischen  Bronchitis   	    1557
    U. SMIDT,  G. WORTH

    "Eco-genetics": Host variability in health effects  of environmental
    Agents	1563
    G.S.  OMENN

    The effects  of particulate air pollution on expectation of life .1569
    T.J.  THOMAS, D.W. MALONE

    Illness due  to environmental pollutants in a physician's
    practice	    1575
    G.L.  WALDBOTT

-------
                                   LXXIX
     Assessment of the health effects caused by chronic exposure
     to solvents	    1579
     P.U.  CAPURRO

     Incidence de certaines maladies chez la population de deux
     centres fortement industrialises, comparativement a trois
     localites temoins  	   1583
     M. DELEANU, N.W.  GHELBERG,  I.  LENGHEL,  N. MOLDOVAN

     Bone marrow changes  in asbestosis   	  1589
     A.Z.  EL-SEWEFY,  H.  SHAHEEN, A. SHAMS EL-DEEN


TIERUNTERSUCHUNGEN
ANIMAL STUDIES
ETUDES SUR LES  ANIMAUX
STUDI  SUGLI  ANIMALI
ONDERZOEKINGEN  BIJ DIEREN

     Behavious of rats and the threshold limit  value of tri-
     chloroethylene   	   1601
     A.P.  SILVERMAN,  H.  WILLIAMS

     Mutagenitats-Vorteste auf Carcinogene Wirkungen vom
     Umweltchemikalien  	 1607
     H. MARQUARDT

     Les cha^hes trophodynamiques marines experimentales:
     methode d1etude  des effets pathologiques consecutifs  aux
     pollutions chimiques   	  1613
     M. AUBERT,  B.  DONNIER

     Versuche zur  medizinischen Bedeutung des Gleichfeld-  Oder
     Faraday-effektes; bautechnische Folgerungen   	   1617
     J.R.  MOSE

     Methodology of experimental and epidemiological research of
     different environmental physiopathological  factors   .  .  .  .   1625
     M. STUPFEL

     Utilisation de cultures de tissus humains  pour la recherche
     d'une  eventuelle synergic entre les effets toxiques de deux
     solvants usuels:  le toluene et le benzene   	 1631
     D. GRADISKI, J.L.  MAGADUR

     Exposure of laboratory  animals to urban  air pollution:
     physical plant,  experimental difficulties, observed effects . 1637
     A.L.  REEVES, T.F.  MOONEY jr.,  R.G. SMITH

     Effets synergiques des  polluants particulaires et d1infections
     experimentales tuberculeuse ou ornithosique   	  1645
     J.C.  MARTIN, L. LE BOUFFANT, H.  DANIEL

-------
                                    LXXX
     Les effets des polluants sur 1'epuration pulmonaire   .  .  .  .   1651
     L.  LE BOUFFANT, H. DANIEL, J.C.  MARTIN,  J.P. HENIN

     Exemple de recherche dans une atmohphere d'une substance
     chimique cause d1effets  pathologiques   	      1661
     A.  CICOLELLA,  D. GRADISKI, J.L.  MAGADUR,  B.  MONCELON


UMWELTMESSUNGEN
ENVIRONMENTAL MEASUREMENTS
MESURES DANS  L'ENVIRONNEMENT
MISURE  AMBIENTALI
METINGEN IN HET MILIEU

     Aspects of pollution in three  urban centre situations in
     Great Britain	1669
     R.G.  DERWENT, H. N.M. STEWART,  J.S.S.  REAY

     Etude de la contamination de fourrages et legumes par le plomb
     et les metaux lourds	1675
     E.  DELCARTE,  R. IMPENS,  R. KIRCHMANN,  P.  NANGNIOT,
     E.  FAGNIART

     Effects of secondary lead smelters on air, soil and  vegetation
     quality, and on local residents in Toronto  	  1685
     T.M.  ROBERTS,  T.C. HUTCHINSON, R.E.  JERVIS,  J. PACIGA
     A.  CHATTOPADHYAY, J.  VAN LOON,  R. CRUZ

     Sui vantaggi presentati  dall'uso della fornace a grafite come
     nuova tecnica di assorbimento atomico nell'analisi dell1
     inquinamento dell'ambiente di lavoro   	   1693
     D.  CASULA, G.  DEVOTO

     Levels of some trace elements in baby and adult diets in
     Italy	   1697
     M.  ALLEGRINI,  E. LANZOLA,  C. ROGGI,  G.  GIANANI,
     G.C.  SABATINI

                           32      35
     The use of radioisotopes  P and   3 in studies on behaviour
     of pesticides in the soil  	   1711
     Z.  SZPERLINSKI

     Die Belastung der Nahrung mit den Schwermetallen Blei, Cadmium,
     Quecksilber, Zinn und Arsen in Qsterreich   	   1719
     H.  WOIDICH, W. PFANNHAUSER

     Asbestos measurements in the California Environment  ....   1729
     J.J.  WESOLOWSKI, R.  STANLEY, G.  SMITH,  L. CARPENTER,
     A.  BERNER

-------
                               LXXXI
EXPOSITIONSMESSUNGEN
EXPOSURE MONITORING
MESURE DE !_'EXPOSITION
MISURA DELL'ESPOSIZIONE
METING VAN EXPOSIE

     Siting considerations for air  quality monitors as applied to
     epidemiological  studies   	   1741
     P.R. HARRISON

     Evaluation of a  rapid index of environmental  pollution by tobacco
     smoke and study  of the composition of the main-stream and side-
     stream cigarette smoke 	   1749
     G. SCASSELLATI- SFORZOLINI,A.  SAVING

     Beeinflussung des Kohlenmonoxidspiegels durch Luftungstechnische
     Anlagen	    1757
     H. RUDEN,  H. LANGER,  K. BOTZENHART

     Recapitulation of exposures around a point source of air
     pollution	1763
     M.B. HERTZ

     Recapitulation of metropolitan air pollution  exposures arising
     from mobile and  stationary sources   	   1769
     D.O. HINTON

     The problem of passive smoking in motor cars	1773
     H.P. HARKE

     Anaesthetic vapour pollution in  operating theatre atmospheres   1779
     R.S. BARRATT,  R.  BELCHER, W.I. STEPHEN,  J.M.  THOMPSON

     The contribution of organic lead compounds to total lead
     levels in urban  atmospheres   	  1783
     R.M. HARRISON, R. PERRY,  D.H.  SLATER

     Enquete portant  sur la consoramation familiale de pesticides dans
     les families habitant dans de  grands ensembles de la region
     parisienne jnetholologie )	1789
     C. PIVA,  M.L.  EFTHYMIOU,  E.  FOURNIER


MODELLE
MODELS
MODELES
MODELLI
MODELLEN

     Applicazione dell'analisi spettrale allo studio della
     dispersione degli inquinanti atmosferici  nell'area di
     Venezia	1797
     L. IOVENITTI,  M.  MENENTI, F.  FERRARI-BRAVO,
     I  GIOVANNINI,  A.  LEVY

-------
                               LXXXII
     Studio e valutazione degli effetti dell1inquinamento
     ambientale sulla salute dell'uomo attraverso simulazione
     con modelli bio-ingegneristico-chimici   	   1807
     M.  DEL BORGHI,  G. FUMAROLA,  R.  CARDI, G.  FERRAIOLO

     A statistical analysis  of the air pollution data  in connection
     with the meteorological conditions  	   1815
     D.  YORDANOV


EXPOSITIONS INDIKATOREN
INDICATORS OF  EXPOSURE
INDICATEURS  D  EXPOSITION
INDICATORI DI  ESPOSIZIONE
INDICATOREN  VON  EXPOSIE

     II contenuto di piombo  nei capelli come indice  di esposizione
     al tossico nell'ambiente   	   1823
     S.  CACCURI,  L.  PECORA,  A. ROSSI

     Application of lead mobilization for exposure evaluation  .  .   1829
     L.  GRAOVAC-LEPOSAVIC, D.  DJURIC,  S. MIL1C,
     V.  VALJAREVIC

     Beziehungen zwischen Raumluftkonzentration, Blutspiegel
     und Metabolitenausscheidung aromatischer Losungsmittel  .  .  .   1841
     D.  SZADKOWSKI


GEWEBSMESSUNGEN
TISSUE  MEASUREMENTS
MESURES RELATIVES  AUX TISSUS  BIOLOGIQUES
MISURE  NEI TESSUTI  BIOLOGICI
METING  VAN BIOLOGISCH WEEFSEL

     How useful are the bulk of environmental data?	1849
     J.  VAN LOON

     Auswirkungen der Blei-Uraweltkontamination auf Lebensmittel von
     Schlachttieren; ein Beitrag zur Probenahme, und Beurteilung.  .  1853
     W.  KREUZER,  P.  WISSMATH,  B.  SANSONI, W.  KRACKE,
     F.  DIETL

-------
                       GEWEBSMESSUNGEN



                     TISSUE MEASUREMENTS



          MESURES RELATIVES AUX TISSUS BIOLOGIQUES



                MISURE NEI TESSUTI BIOLOGICI



               METINGEN VAN BIOLOGISCH WEEFSEL




                         (Continued)
Voraitsender - Cnairman - President - Preeidente - Voorzitter






                    L. ROSIVAL (C.S.S.R.)

-------
                              1189
          EVALUATION EN MICROSCOPIE OPTIQUE DU DEGRE
          DE L'EMPOUSSIERAGE PULMONAIRE ASBESTOSIQUE
  PAR MICROFILTRATION DE L'EXPECTORATION ET DU SUC GASTRIQUE

     J, BIGNON+, B, HEM+, G, BONNAUD++, P, SEBASTIEN** ET
                         G, MONCHAUX++

+  Clinique Pneumo-Phtisiologique, Hopital Laennec, Universit§
   Paris V, France
++ Direction de I1Action Sanitaire et Sociale, Paris, France
RESUME

     L'identification et le denombrement dee corps ferrugineux
(asbestoaiques) et dee fibres nuea (d'aebeete) ont ete realises
par microfiltration de I'expectoration et du sue gaatrique.
Troia groupee de aujeta ont ete etudiea:  groups It aujeta ex-
poaea d. un fort empouaaierage aabeatoaiquei  groupe II, aujeta
ayant travaillS «J la transformation de produita d. baae d'aabeete
et groupe III, eujeta expoaea d. un empouaaierage de type indue-
triel comportant plua ou moina de t'aabeste.

     La numeration dea oorpa ferrugineux a r&vele qu'il existait
une correlation etroite entre le nombre de corps ferrugineux
trouvea dane  I'expectoration et le nombre de oea mSmee corpe
trouves dans  le poumon.   II eemble qu'au-deaaue de la concen-
tration intraparenchymateuae de 1000 corps ferrugineux par cm
de poumont on retrouve dea corps ferrugineux dane l'expeotorationt
tandis que dane le sue gastriquef le reflet de  I'empouaeierage
pulmonaire aabeatoaique n'eat mia en evidence qu'd, un aeuil plua
haut d'environ 10 000 corps ferrugineux par am  de poumon.   La
recherche de  corpe ferrugineux dans I 'expectoration apparait
done comme la methode la plus sensible pour appreoier le niveau
de I'empoueaierage pulmonaire par dea particulee fibreuses.

-------
                             1190
     L'analyse de I*Evolution dans 1e temps de 1'elimination
respective des corps ferrugineux et des fibres nues apres la
cessation de lfexposition professionnelle a permis une etude
dynamique de I'epuration <5 long terme des particules fibreuses:
les sujets eliminent d'abord les fibres nues les plus courtes,
puis apres des annees d'evolution, le pourcentage de corps fer-
rugineux augmente pour atteindre un taux de 80%, 10 ans apres
la cessation de I*exposition.
ABSTRACT

     Ferruginous (asbestos) bodies and naked asbestos fibres
were identified and counted by microfiltration of the sputum
and gastric secretions.   Three groups of subjects were studied:
Group I, subjects heavily exposed to asbestos dust:  Group II,
subjects who had been occupied in the manufacture of asbestos-
based products;  and Group III, subjects exposed to industrial-
type dust containing greater or lesser amounts of asbestos.

     Counting of the ferruginous bodies showed a close correla-
tion between the number of these bodies in the sputum and the
number found in the lung.   It appears that if the concentration
                                                               2
of ferruginous bodies within the parenchyma exceeds 1000 per cm
of Tung, these same bodies are found in the sputum, but that
they are found in the gastric secretions only when the concen-
tration in the lung exceeds the much higher threshold concentra-
                                        •j
tion of 10,000 ferruginous bodies per cm  of lung.   Thus, det-
ermination of ferruginous bodies in the sputum appears to be
the most sensitive method for assessing the level of fibrous
duet particles in the  lungs.

     A comparative analysis of the elimination of ferruginous
bodies and naked fibres in the course of time after occupational
exposure had ceased permitted a dynamic study of the long-term
clearance of fibrous particles.   The subjects first eliminate
the shortest naked fibres.   Then, after the condition has taken
its course for some years, the percentage of ferruginous bodies
increases, rising to 80% ten years after the end of exposure.

-------
                                   1191
1. INTRODUCTION
       Dans des publications  precedentes  (1,2), nous avons montre que le
                                                     ft
denombrement des corps  ferrugineux (CF) apres microfiltration de I'expecto-
ration constituait une methode d'investigation c Unique tres inte>essante,
permettant d'e"valuer, du vivant du ma lade, et sans avoir recours 6 la biopsie
pulmonaire, la concentration intra-pulmonaire en CF.
       Le  present travail a pour but d'etablir  des donnees de reference con-
cernant les relations qui existent entre le  nombre de CF dans I'expectoration
ou dans le  produit de tubage gastrique et la concentration intra-pulmonaire,
compte tenu de ('importance de ('exposition au risque asbestosique  et de
('elimination des  fibres apres  la cessation de ('exposition.

2. MATERIEL ET  METHODES
       L'6tude a port£ sur 193 sujets.
       L'interrogatoire professionne I a permis d'isoler 56 sujets temoins non
exposes a  I'asbeste et 137 sujets dont on pouvait supposer une exposition au
risque asbestosique au  cours de leur vie professionnelle.  Le  classement  des
sujets en  3 groupes,  en fonction du travail qu'ils ex6cutaient,  a permis
d'obtenir une evaluation grossiere de Pintensit6 de leur exposition :
groupe I  : 21  cas, a"ge moyen : 52 ans . "Manipulation de I'asbeste a I'etat
brut"  (broyage,  filage, tissage, projection,  manutention, etc...) ;
groupe II : 47 cas, Oge moyen : 53 'JTI.-,  "Transformation  de  produits & base
d'amiante" (isolation, chauffage, reparation freins d'automobiles,  etc...)  ;
groupe III : 69 ans ;  dge moyen : 56 ans.  "Exposition 5 un empoussierage
Industrie! comportant plus ou  mains de I'amiante".
       Si chez tous  les sujets des 3 groupes  precedents ont  ete precises :
nature et dur6e de la profession et delai depuis la  cessation de ('exposition
asbestosique. Par contre, la dens!t£ de I'empoussierage et (a nature des
fibres inhalees n'ont pu §tre obtenues qu'approximativement dans cette  etude
retrospective .
       Le denombrement des  fibres d'asbeste a et6 realist apres digestion
chimique par ITiyochlorite de  sodium et microfiltration sur 3 types de  preleve-
ments selon une  m§thode precedemment dScrite (2) : expectoration matinale

-------
                                  1192
chez tous les sujets (1 93 sujets),  liquide gastrique pr6lev6 a jeun avant le
lever (37 sujets) et parenchyma pulmondire apres fixation formolee en  insuffla-
tion (52 sujets). Les fibres gainees ou corps  ferrugineux (CF) et les fibres
nues ont 6te compiles au  microscope optique en contraste de phase  (x 400).
Le nombre de particules fibreuses a €t€ exprim£ : pour ('expectoration et
le produit de tubage gastrique  : nombre total par prr'levement  ; pour le
poumon : nombre de CF par cm3 de parenchyme pulmonaire fix6.
3. RESULT ATS
        1) Relation entre le nombre de corps ferrugineux rencontr6s dans
        I'expectoration ou le liquide de tubage gastrique et  le type d'expo-
        sition professionne lie.
1°) Epuration des CF dans  {'expectoration :
        Chez  les 56 sujets  temoins non exposes professionnellement 6 I'asbes-
te, il n'y avait  pas de CF dans ('expectoration.
        Chez  les 137 sujets pour lesquels il  existait  la notion d'une possible
exposition professionnelle a I'asbeste, la frequence des cas avec expectoration
positive eta it respectivement pour les groupes 1,  II et III de  95 %, 60% et
25 %. Cette relation apparait encore plus nettement lorsqu'on tient  compte
du nombre de CF par expectoration : le nombre moyen de CF trouv£s etait
de 100 pour les  Sujets du groupe I, 10 pour  les sujets du groupe II, 1  pour
les sujets du groupe HI  (Fig. 1).
2°) Resultats compare's de I'e'puration des  corps ferrugineux dans I'expectora-
tion et dans le produit de tubage gastrique (T.G.),:
        Chez  37 sujets, ont 6t6 compare's  les resultats de la recherche des
CF dans I'expectoration et dans le T.G. Chez 26 de ces sujets,  la recherche
etait negative dans  les  deux produits.  Pour  les 11 patients avec CF dans  ('ex-
pectoration, le  nombre de CF trouves dans le  T.G.  6tait toujours nettement
inferteur a celui trouv£ dans I'expectoration.
        II apparait done que la recherche des CF dans I'expectoration est
plus sensible que dans le T.G. pour apprecier le niveau d'6limination des
part icu les fibreuses. L'etude des s£cr€tions  gastriques etant a reserver aux
sujets sans expectoration.                  '

-------
                                1193
       2) Correlation avec I'empoussierage pulmonaire
       Dans 52 cas Studies, le nombre de  CF/cm3 de parenchyme pulmonai
                                                   ft
re (3) a £t6  compare a celui trouve dans I1 expectoration (Fig.  2)
JOOO

5OO
3OO
1OO
SO
IO
5
2
1
A

A
t
+ U
£
A
A
A

Nombre maximal de C F
dans 1'expectoration


*

j
+ •*
**
•••



A GT
G»
X O]|
•5- m

Ixx
XXX
XXX

                                     io6
                                     10
                                     io1
                                             t
                                               POUMON          4 Gr
                                             * NOMB8E DE Cf/CC   . Gi
                                             «.     *          w /^	
                                                              x C
                                                                Temolns T
                                                        J.
            Figure 1                           Figure 2
Fjgure 1  :   Correlation entre le nombre de CF rencontres dans ('expectora-
            tion et le type d'exposttion professionnelie.
            Pour  chaque cas, ce nombre est repr6sent6 en utilisant une
            £chelle logarithmique  ( -*- indique la moyenne du  nombre
            de CF dans chaque groupe).

Figure 2  ;   Correlation entre le nombre de CF rencontres par  cm3 de paren-
            chyme pulmonaire et le type d'exposition professionne He.
            Pour  chaque cas/ ce nombre est represent^ en utilisant une
            echelle logarithmique ( 4-  indique la moyenne du nombre de
            CF dans chaque  groupe).

       Chez les 20 t£moins de cette $6rie, sans exposition professionne Me
et avec expectoration negative, le  nombre de CF trouv6 par cm3 de
parenchyme pulmonaire variait entre 2 et 150 (moyenne - 50).
       Chez les 32 sujets avec risque asbestosique certain ou  probable,
les re" su I tats variaient en fonction de I'intensite du risque.
       Ainsi, pour le groupe  I,  la moyenne de 100 CF trouv£s dans I'expec-
toration correspondaient a une concentration pulmonaire moyenne d'environ
100 000 CF/cm3  ; pour le groupe II7 la moyenne de 10 CF de I'expectora-
tion correspondait  5 une contentration pulmonaire moyenne d'environ
10 000 CF/cm3 ; pour le groupe  III,  la moyenne de 1 CF dans  I'expectora-
tion correspondait  a une concentration pulmonaire d'environ 1000 CF/cm3.

-------
                                  1194
       D'une maniere genera le,  II semble done qu'il existe une relation
entre le nombre de CF trouvS dans ['expectoration et  la concentration en CF
Les requitals sont  exprimes sur la Fig. 3,  qui montre que = 1000 CF/cm3 de
parenchyme pulmonalre represente une valeur seoil (!es resultats sont repre-
sented suivant une double 6chelle  logarithmique).
    Expectoration
               f f
               i* r
    300
   -30
    -3
                                                           •*	
                                                            *   tt
                                               Figure 4
                 d* cr /«
         Figure 3
 Figure 3  :  Correlation entre le nombre de CF dans I'expectoration et  le
            nombre de CF rencontres dans les pr^levements de parenchyme
            pulmonaire ; les requitals sont representfis suivant une double
            echelle  logarithmique.
 Figure 4  :  Variation du rapport des fibres gainees ou CF par rapport au
            nombre total des CF + fibres nues en fonction du delai de
            cessaHon.du risque asbestosique.

        3) Etude de l'6purotion des porticules fibreuses ft long terme
        Dans 15 observations ou fibres nues et enrobees ont et6 d^nombrees
 a  la fois dans I 'expectoration et dans le parenchyme pulmonaire, le pourcen-
 tage des fibres enrobees (CF) et des fibres nues variait suivant  le delai  de
 cessation de ('exposition. Ainsi, dans les ann£es suivant la cessation de
 ('exposition, ce sont les fibres nues, surtout courtes (4) qui dominant dans
 I'expectoration. Par  la suite, on constate que le pourcenrage de CF, corres-
 pondent d I 'enrobe me nt de fibres longues augmente rapidement avec le

-------
                                 1195
temps pendant une p6riode de 1 0 ans environ, pour ensuite se stabiliser a
environ 80 % et rester a ce niveau a peu pres constant.
                                                    »
       II est cependant interessant d'observer que sur les  cas etudies,  nous
n'avons identlfi^  le chr/sotile qu'une seu le fois dans les pre lavements de
parenchyme  pulmonaire . Ces  r^sultats concordent avec ceux observes par
d'autres auteurs (5,6,7) qui constatent une predominance  des vari6t6s d'as-
bestes amphiboles dans  le tissu pulmonaire. Le probleme de la penetration
alv6olaire des fibres de chrysotile  et de la cin^tique de leur 6puration reste
done non r^sotu.
        Bien que pr6 liminaires, ces rgsultats  indiquent les modal!t£s  de  l'£pu-
ration  pulmonaire a long terme  des particules fibreuses chez I'homme.  Jusqu1
a ce four, les etudes analytiques r6alis£es, n'ont pu l'€tre que  chez I'animal
(8,9,10). Or la connaissance des modes depuration chez  I'homme apparait
pourtant fondamentale  (11), puisque de la dur€e de retention intra-pulmonai-
re des fibres d'asbeste depend ,°n effet pathogene  plus ou moins marqu6 (5),
notamment en ce  qui concerne  faction cance>igene.
                           BIBLIOGRAPHIE
 1  -  BIGNON,  J.7 et coll., Nile Presse  Med, 1973,  2,  1697-1700.
 2 - BIGNON, J. et coll., Biological effects of ingested  particles,
     Durham 1973,  (sous presse).
 3 -  BIGNON,  J.  et coll., Environ. Res.,  1970, J, 430-442.
 4 -  BEATTIE, J. et KNOX, J.F.  Inhaled particles and vapours, C . N.
     Daviesed., Pergamon Press, Oxford, 1961,  pp. 419-441.
 5 -  POOLEY, F.D.Proc. working group to review the Biological effects
     of asbestos, 1972,  Lyon, -
-------
                            1196
 9 - rVANS, J.C. et coll. Environ. Res., I 973, 6, 180-20' .

 10 -TIMBRELL, V. et SKIDMORE, J.W. Inhaled particles III, vol 1 , W.H.
    Walton Ed.,  Gresham Press, 1971,  pp. 49-56.

 11 -LE BOUFFANT, L. et coll.  Lille Med.,  1972, 1_7, 1091-1101.



                         DISCUSSION


COOPER (U.S.A.)

     Avez-vous  eu 1'occasion d'utiliser cette technique pour
fitudier le cas  d'ouvriers nouvellement embauches afin  de de-
terminer si cette technique pourrait constituer un systdme  de
detection et de surveillance de la surexposition?  II  est bien
connu que des ouvriers subissant  une forte dose d1 exposition
peuvent continuer a  expectorer des corps ferrugineux encore
pendant plusieurs annSes apres la cessation de 1'exposition,
mais cette mSthode pourrait etre  appliquSe aux nouveaux ouvriei
au moins pour la recherche des amphiboles.


BIGNON (France)

     Jusqu'a present, nous n'avons pas utilis£ cette technique
pour la surveillance d'ouvriers nouvellement embauch^a, mais
nous sommes persuades que 1'examen de I1expectoration  represent
un test tres sensible de surveillance de tels sujets.   On pour-
rait par exemple proposer d'interrorapre 1'exposition des qu1
apparaissent des corps ferrugineux dans I1expectoration.


DUPUIS (France)

     Pensez-vous que I1utilisation du microscope optique permet
d1identifier les fibres de petites dimensions qui sont vraisem-
blablement  les seuls susceptibles d'induire deS me'sotheliomesi


BIGNON (France)

     MalgrS I1utilisation de  techniques microscopiques elabor^<
(1), comme 1'observation en contraste de phase/  ou  la  techniqu<
des liqueurs d'indice, il est evident que les fibres de taille
inffirieure a 3  microns peuvent fichapper a 1'examen  en  microscoi
optique.

     Le probldme se  pose tout particulierement pour  la deter-
mination des fibrilles e'ISmentaires de chrysotile dont le dia-
metre de 30O A  les rend pratiquement  inaccessible a  la micro-
scopie optique, les  fibres que 1'on peut observer 6tant alors

-------
                              1197
des associations  de  fibrilles elementaires.   II faut utiliser
le microscope  electronique pour resoudre ces fibrilles.

     Neanmoins,  la microscopic optique cons'titue une mSthode
excellente  pour  apprecier le degre d'empoussierage d'une region
tissulaire,  ceci  grace aux methodes de preparation qui sont
simples et  tres  reproductibles.  II convient ensuite d1utiliser
la microscopie Electronique pour etablir une granulom£trie des
particules  et  realiser une identification plus precise par
microdiffraction.

(1)  BARTOSIEWICZ, L.: Improved techniques of identification and
determination  of  airborne asbestos. Amer. Ind. Hyg. Assoc. J.,
June 1973.
GIBBS (Canada)

     La methode que vous avez utilisee pour 1'identification du
type de fibre present dans les poumons des ouvriers vous permet-
elle de determiner de quel type d'amphibole il s'agissait?


BIGNON (France)

     Pour ce travail, 1'identification des fibres a uniquement
ete realised par microdiffraction electronique.  Si par cette
methode,  il est tres facile de distinguer amphibole et chrysotile,
la distinction entre les differents types d'amphibole est, a
notre sens, impossible d'apres le seul examen du clichS de
microdiffraction.

     Pour identifier precisement le type d1amphiboles, il faut
utiliser plusieurs -microm6thodes physiques:

     - Optique cristalline: par cette methode, on mesure
particulidrement 1'indice de refraction et le signe de 1'al-
longement.
     - Microanalyse chimique elementaire: les fibres d1amphiboles
peuvent avoir une taille suffisante pour qu'avec le MEB eguipe
d'un systeme dispersif, on accede a la chimie elementaire d'une
facon semi-quantitative.

     Mais ces methodes sont longues et couteuses et n'ont pas
    utilisees dans cette t lauete.

-------
                              1199
    ETUDE QUANTITATIVE EN MICROSCOPIE OPTIQUE ELECTRONIQUE
  DE L'EMPOUSSIERAGE  PULMONAIRE ET PLEURAL CHEZ VINGT SUJETS

      A, FONDIMARE, J,  DESBORDES,  J,  TAYOT ET J, PERROTEY

Laboratoire d'Anatomie Pathologique,  Centre Hospitaller du Havre
France
RESUME

     De nombreux  travaux menSs tant en France qu'a I'Etranger,
il ressort  que  la plupart des adultes vivant en milieu urbaint
ont des corps asbestosiques ou ferrugineux dans leur parenchyme
pulmonaire:  nous avons  fait la mSme constation ehez les sujets
vivant dans I'agglomeration Havraise.

     Or,  en mati£re  d'asbestose professionnelle, la plupart des
auteurs attribuent un pouvoir pathog&ne aux fibres d'amiante
restees nues, bien plus  qu'aux corps asbestosiques,

     Dans ces conditions., it nous a paru interessant de comparer
en microsoopie  optique et electronique la teneur en fibres d'a"-
miante du  tissu  pulmonaire et pleural de sujets soumis  les uns a
une exposition  massive^  les autres a un risque faible, bref ou
apparamment nul.
ABSTRACT

     Numerous studies in France and in other countries have  shown
that most adults living in an urban environment have asbestos  or
ferruginous bodies in the pulmonary parenchyma.   We have found
the same in subjects living in the area of Le Havre.

-------
                              1200
     With regard to occupational asbestos-is,  most authors consi-
der that asbestos fibres which have remained  naked are much more
pathogenic than asbestos bodies.

     In these circumstances^ we thought it would be of interest
to compare by optical and electron microscopy the asbestos fibre
content of the lung and pleural tissue of some subjects who had
undergone massive exposure and others who had been exposed to
slight or short-tern risk, or to no apparent  risk.

-------
                             1201
Materiel   et  Methodes
     Le present travail porte sur 2O maladee:
 - 12 ont ete soumis a un empoussierage professionnel massif
   par 1'amiante
    8 a un risque faible, bref ou apparamraent nul.
     Nous comparons les resultats a ceux d'une etude anteri^ure
en microscopie optique deja publiee portant sur:
 - 52 sujats  non selectionnes
 - 12 porteurs de mesotheliomes
 - 11 asbestoses majeures.
     Dans chacune de nos 2O observations, nous avons detruit 4
gr de tissu pulmonaire ou pleural formole par digestion alcaline,
Le produit obtenu est lave et ramene par centrifugation a un
culot de  1 cm .  Une goutte de 20 ml est prelevee pour etude
au microscope electronique Sopelem 75, soit directement  (empous-
sierage  faible) soit apres une nouvelle dilution  (empoussierage
eleve)   Le coulot initial est filtre sur une membrane Millipore
a 3 microns qui permet apres clarification, la numeration au
microscope optique des corps asbestosiques.
     Par  cette methode il est possible de comparer la teneur en
corps asbestosiques d'une part, et en micro-fibres d'amiante de
dimension infra-microscopique d'autre part, d'une masse standard
de tissu  fixe.
     En  raison de 1'existence de plusieurs temps de centrifu-
gation decantation, il est possible que nous sous-estimions la
teneur en micro-fibres, par centre une surevaluation nous parait
peu probable.

Resultats
     Nous rappellerons brievement les donnees de notre etude
quantitative prealable ou:
     96%  de nos sujets ont des corps asbestosiques dans leur
parenchyme pulmonaire.

-------
                             1202
     Nous avons ventilfi nos rSsultats en quatre groupes selon le
nombre de corps asbestosiques (C.A.)  contenus dans 4 gr de pourno;
 - 1 moins de 10  C.A.
 - 2  de  102  a 1O3  C.A.
 - 3  de  1O3  a 1O4  C.A.
 - 4 plus  de 1O4 C.A.
     Tous les sujets non selectionn^s figurent dans les groupes
1 et 2.  Les asbestoses professionnelles av£rees se situent
dans le groupe 4.  Les observations de mfisotheliome font tran-
sition et se classent dans le groupe 3.
     Ces donnees sont comparables a celles de la litterature.
     L'gtude couplfie en microscopie 61ectronique nous a permis
dans un premier temps de classer nos observations selon ces
groupes:
     En microscopie optique:
 - 13 observations sont du groupe 4
 -  3        "      sont du groupe 3
 -  4        "      sont du groupe 1 et 2.
      En raicroscopie ilectronique:
      Nous observons des fibres nues dans tous nos echanti 116ns
de parenchyme pulmonaire ou de plaque pleurale.
      Pour les empoussiSrages de type 4 en microscopie optique la
concentration pulmonaire en fibres infra-microscopiques atteint
des chiffres de 1'ordre de 106 a 107.
      Nous reproduisons ici un tableau portant sur 6 de  ces
observations oQ la repartition des fibres est assez homogene
sur plusieurs Schantilions pour autoriser une Evaluation  quanti-
tative rapportSe au volume initial dissous.

-------
                             1203
Observations
n°s
30. 606
A. 481
A. 691
A. 1384 b.
A. 384
A. 1154
C.A.
Optique
11. 104
6,9.105
1,5.104
4,7.105
1,1.10s
2,2.105
Fibres nues
electronique
12. 106
3. 107
3. 105
1,4.107
1,6.107
3,1.107
          les  observations de type 1 a 2 les concentrations en
micro-fibres sont beaucoup plus faibles.  L'on observe parfois
moins  de  10 micro-fibres pour une grille entiere.
    Dans les  limites de notre methode, le nombre des micro-
fibrilles depasse done exceptionnellement des chiffres de 1'ordre
de 10  .   Bien  que nous raisonnions sur un petit nombre d'obser-
vations,  il semble done exister un hiatus quantitatif conside-
rable  entre les  empoussierages pulmonaires de type professionnel
et ceux de rencontre tant en microscopie optique qu1electronique
surtout.
    Dans laplevre ou plutot dans les plaques pleurales les
concentrations en corps absestosiques visibles en microscopie
                                                    2
optique sont  tres faibles depassant rarement 10O  (1O ).
    En microscopie electronique 1'on retrouve constamment des
microfibres d'amiante: elles sont en general plus petites que
dans le poumon homologue.
     Certaincs  observations nous montrent d'importants pheno-
menes  de  concentration locale pleurale pour des empoussierages
pulmonaires faibles.
    Ex  : 75  C.A. dans 4 gr de tissu pleural hyalin
          69  C.A. dans 4 gr tissu pulmonaire homologue.

-------
                              1204
      Dans  ce  travail  nous  ne  retrouvons pas au niveau de  la
plevre  l'§cart  qui  existe  dans  le poumon entre les empoussieragei
de  type professionnel et de type occasionnel.

Discussion
      Bien  que notre enquete repose sur un nombre limite
d1observations,  il  apparalt cependant que les empoussierages
pulmonaires massifs des  asbestoses professionnelles corresponded
a des concentrations  parenchymateuses en fibres d'amiante sans
commune mesure  avec celles relevant d1expositions de recontre.
      Dans  le  premier  cas,  tous  nos  sujets pr^sentaient clini -
queraent et  radiologiquement d'intenses lesions de fibrose  pul-
monaire.   Dans  la seconde  Eventuality nous n'avons not£ au
maximum que de  discretes lesions de fibrose histologique.
      Au niveau  de la pievre par contre,  si  les  concentrations
tissulaires sont beaucoup  plus  faibles, 1'on ne retrouve  pas
un  tel  hiatus sur le  plan  quantitatif.  Cette constatation nous
semble  devoir etre  rapprochfie de la grande frequence avec
laquelle nous observons  des plaques pleurales hyalines cal-
cififies ou non  a 1'autopsie de  sujets faiblement exposes  d' une
part et de I1existence de  mesotheliomes chez le meme type de
malades d'autre part.
      D'un  point de  vue qualitatif, comme beaucoup d'auteurs ,
Heppelston et Ashcroft en  particulier, nous avons surtout iden-
tifig chez nos  sujets des  fibres du type amphibole et plus
rarement du chrysotile en  longues fibres ou Scheveaux touffus.
Nous n'avons  pas eu I1occasion  d1observer comme Selikoff  et
Langer  ont pu le faire chez les habitants de New York, de
nombreuses micro-fibrilles de chrysotile.

      Les enquetes de  risque nenees chez nos naiades ou dans  leur
entourage  professionnel  nous  laissent a penser ou'actucllemcnt
1'empoussierage pulnonaire par  1'aniante dans 1'agglomeration
havraise releve plutot d'une  exposition professionnellc faiblo  o.
occasionnelle,  dont la frequence va d'ailleurs augmentant, que  d

-------
                             1205
problems  de  pollution generale.   11 est possible que cette
situation  evolue  dans  les annees  a venir.
                                             »
     L1amplification des etudes quantitatives pour difficiles

qu'elles soient nous parait constituer une voie interessante

pour appr^cier  la signification de ces empoussierages faibles

en les  comparant  aux asbestoses massives.


References


1.   BROUET G., BIGNON, G. , BONNAUD, G., GONI J. , CHRETIEN J. ,
     PARIENTE R., :  Calcifications pleurales associees a un
     empoussierage asbestosique detect^ par 1'examen de micro
     prelevements au microscope polarisant. Journal Fran9ais
     de Medecine  et de Chirurgie thoraciques 2_1 (2) 181-195,1967

2.   ASHCROFT T., HEPPLESTON A.G., : The optical and electron
     microscopic  determination of pulmonary asbestos fibre
     concentration and its relation to the human pathological
     reaction.  Journal of clinical Pathology 2Jj 224 - 234,1973

3.   FONDIMARE A., TAYOT J., DESBORDES J. , ERNOULT J.L.,
     PERROTEY J., :  Etude syst£matique de 1'empoussierage
     pulmonaire par 1'amiante dans la region Havraise. Archives
     d'Anatomie Pathologique 21.  (3) 273 - 278,1973

4.   HAGERSTRAND  I., SEIFERT B.,: Asbestos bodies and pleural
     plaques  in human lungs at necropsy.  Acta Pathologica et
     Hicrobiologica  Scandinavica Section A Pathology  (81)
     457 - 460,1973.

5.   MEORMAN  L.,  : Asbestos bodies and pleural plaques in a
     finnish  series of autopsy cases.Acta Pathologica  et
     Microbiologica Scandinavica, Section A = Pathology Supplement
     181 7 -  107,1966.

6.   POOLEY F.D.  : Electron microscope characteristics of inhaled
     chrysotile asbestos fibre,British Journal of Industrial

     Medicine 29  146-153 1972.

-------
                             1206
                          DISCUSSION


BRUCH (Rgpubligue f6d6rale d'Allemagne)

     Avez-vous determine le rapport entre les particules d'amian
(PA) et les fibres nues dans le poumon pour les 4 groupes d1 ex-
perience?  En experimentation animale (avec P.P.  Holt)  la for-
mation de PA par deterioration de macrophages alv£olaires com-
binee a d'autres causes telles que le tabagisme a ete arretSe.
Le rfisultat fut que les lesions pulmonaires etaient d'autant
plus graves que I1 on dSnombrait moins de PA.  II se pourrait don
que, chez des malades  (mSsotheiiome et/ou fibrose)  le rapport
PA - fibres nues soit abaissS.  Nous partons de 1'hypothSse que
la formation de PA est un mecanisme d'epuration du corps.  Une
disposition individuelle differente du corps pourrait expliquer
que certaines personnes soient atteintes alors qu'elles n'ont
subi qu'une exposition relativement peu importante.


FONDIHARE (France)

     Nous avons etudie le rapport corps asbestosique/fibres nues
chez les sujets trSs einpoussier€s seulement.  Le nombre des corp
asbestosiques augmente en fonction du temps et varie avec le typ
d'amiante inhale.
GIBBS  (Canada)

     II nous arrive frequemment d'examiner le parenchyme pul-
monaire de sujets atteints de m£sotheiiome pour identifier le
type de fibre present  (par exemple amosite, crocidolite, tre-
molite, chrysolite, etc.).  Etant donne qu'il existe actuel-
lement des mfithodes permettant I1analyse microchimique des
differentes fibres, avez-vous I1intention d'utiliser ces mSthode
pour 1'obtention d'une telle information?


FONDIMARE  (France)

     Nous distinguons par diffraction d1 electrons le chrysotile
des amphiboles.  Nous ne disposons pas actuellement des mSthodes
de  microanalyse.

-------
                              1207
   DETERMINATION OF CADMIUM  IN BLOOD USING A 4,0 MM PAPER
     DISC - GRAPHITE FURNACE TECHNIQUE,   APPLICATION TO
                           INDUSTRY

              A, A, CERNIK AND M,  H, P,  SAVERS
Central Reference Laboratory, Department  of Employment, London,
United Kingdom
ABSTRACT

    A paper disc  flameless AAS method is described for the det-
ermination of  Cd in  blood,  enabling difficulties in sample pre-
paration  to be  minimized*    By control of the ashing step, the
matrix can be  removed  without loss of Cd.   Problems with the
fast signal response during atomization can be met by spectral
band width and  temperature  control.   At the 106 picogram level
(5,3 yg/JOO ml  blood)  the  P.S.D.  uas 0.06.

    Results in three  industrial situations are reported.   The
method should  facilitate further investigation of its application
to industry using  capillary or venous blood.

-------
                                  1208
1.   Introduction
     The increasing interest in environmental pollutants has focussed
recent attention on cadmium (Cd),  This metal is widely used in industry,
but knowledge regarding its absorption and excretion is incomplete, thus
rendering difficult the interpretation of industrial exposure using bio-
logical methods.  Measured by a polarographic technique, Cd in the urine
of the general population is less than 5 H«A.  In the industrially
exposed, greater quantities may be found.  Urinary excretion of Cd is
commonly used as an index of absorption, and the finding of 25 fig/1 is
considered evidence of undue exposure.  This does not, however, reflect
the total body burden or the concentration in vulnerable organs, such as
the kidney or the lung.
     Absorbed Cd is conveyed in the bloodstream, either cell or plasma
bound, throughout the body.  Correlation between the urine (UCd) and blood
cadmium (BCd) is poor and indeed neither measurement is regarded as a good
index of absorption by some authorities.  Moreover, the UCd does not neces-
sarily reflect recent exposure.  Tissue biopsy has been advocated by Ul-
lucci and Hwang (1) in view of the problems of blood and urinary findings.
     From the practical viewpoint, however, for the detection of evidence
of undue absorption of Cd in industrial workers, a simple test is needed,
acceptable to workers and management, which can be related to exposure in
order to alert the Industrial Hygienist to advise on control of environ-
ment before  concentration in the critical organs can occur, resulting, for
example, in  tubular nephritis with associated proteinuria.
     This paper describes recent developments in the use of atomic absorp-
tion spectrometry  (AAS) for the estimation of BCd which can be readily
adopted  for  the study of venous or capillary specimens collected from
workers  in  the  factory.
     The tantalum  sampling boat achieved a relative detection limit for Cd
of O.2 ng/ml~  using a 0.5 ml sample  of blood, but 16 hours were needed
for oxidation  (Hauser et al  (2)).  Ediger and Coleman (3) ignited 10 ml of
blood in a  Delves  sampling cup near the  flame and obtained results compar-
able  to  those  using an  extraction/graphite  furnace technique.  There may
be loss  of Cd  using this method of oxidation since the vapour pressure
 temperature (VFT)  of Cd is little higher than that required to burn blood.
Other methods  using the Delves cup after pre-treatment of the  sample have
 shown that recoveries,  reproducibility and  sensitivity are satisfactory
 (Joselow and Bogden CO and  Lieberman (5)).  The latter also reported that

-------
                                  1209
 the method overcame interference effects of different matrices.  Ullucci
 and Hwang (1) favoured solvent extraction of the Cd prior to determination
 on a tantalum ribbon, but calibrated the blood figures «rith aqueous stan-
 dards.
     There are few published references to flameless methods.  Dolinsek
 et al (6) determined Cd in blood using the graphite furnace without pre-
 treatraent of the sample which was dried, ashed and atomized in situ.  This
 direct approach to analysis of metals has the advantage of minimising
 problems due to contamination.  The graphite furnace technique enables Cd
 to be determined on capillary or venous samples directly without pre-
 treatment and with adequate sensitivity.
 2.   Majgrials and Methods
     2.1  Collection of Specimens
     Blood  Venous or capillary blood samples were used as indicated in
 the tables and text collected on the lines described by Cernik and Sayers
 (?).  Capillary blood was spotted onto Whatman's No. k filter paper.
     Urine  Casual specimens of urine were collected from workers in the
 factory.  Corrections were not made for specific gravity, which varied
 widely.
     2.2  Apparatus
     A Varian Techtron AA5 spectrometer was used with a CR63 graphite fur-
 nace workhead.  The graphite cups, 'f.5 mm in diameter, were supported by
 two side electrodes with the correct end radius.  Other features were as
 described by Cernik (8) except for the following:
 Cadmium lamp current taiA..  Spectral band  width 0.33nm.   Wavelength 228.8nm.
     Workhead Power Controls  For ashing,  12 seconds at 300°C (-10).  For
 atomization, 2.0 seconds at 600°C (-10).
     2.3  Method
     All blood specimens whether venous or capillary examined by AAS were
 done by the punched disc technique using k.O mm discs after previous spot-
 ting in the case of venous blood, as described elsewhere (Cernik (8)).
 Non-atomic absorption that appeared in the Cd position was controlled to
 0.01 absorbance by the use of a hydrogen lamp.  A non-interfering second
 peak was ignored (Fig. 1).
     Working Standards  Cadmium chloride (CdCl2) solution was mixed with
 blood samples to give a final added concentration of 0.0, 2.0, 'f.O, 6.0,
8.0 and  10.0 ug Cd/100 ml  blood.
    Ashing  of Matrix   Precise  control of  the  voltage  applied  over  a given

-------
                 1Omm
                                  1210
                              ASHING  5.6-12MCS.
                              ATOMIZATION 3.2-2.0s«cs.
                                             (soot)
                       \
                                          10mm
                          4.0mm Cd discs
                                   NAB O.7scc later than Cd signal
                                                      MAI
                                                       NAD
                            V.
                  Cd LAMP
H2 LAMP
   Fig. 1.   Non-Atomic signal found in the  cadmium position when
             4.0 mm blood  discs are atomized at 600°C.  (NA  = non-
             atomic)
time constant is important for the ashing of the matrix because of the
volatility of Cd.  Previous work with lead using the same technique had
shown that, with the voltage required to achieve 300°C for 12 seconds,
complete removal of all the molecular absorption from the matrix resulted.
The same ashing system was necessary for Cd because the matrix was identi-
cal, but, unlike lead, there was only approximately 100°C difference
between the ashing temperature and the VPT of Cd.  The possibility that Cd
might vapourize at lower temperatures than its VPT was considered, and to
check this, discs of blood punched from one sample were ashed at increas-
ing voltage settings, and the Cd content determined (Table I).

-------
                                  1211


                               TABLE  I
            LOSS OF CADMIUM BY INCREASING THE TEMPERATURE
            _ AT THE MATRIX ASHING STAGE _ •_

     Ash Control               Temperature               % Loss
       Setting                  °C  -10                  Cadmium


        5.6)                      500                    Nil

        5.7)                      3^0                    Nil
        5.8)  at                   370                     6.0
            >  12
        5.9)  seconds.             390                     10

        6.1)                       ^25                     ^5
                               TABLE II

             ATOMISATION TEMPERATURES AND RESIDUAL CADMIUM
                              CSBW 0.33nm)	

Atomize Control
Settine (All           Temperature           Cadmium         Residual Cd
    2.0 Sees)            °C I  10°           Peak Height       Peak Height
      2.0                  ^50                   32               8.0

      2.5                  1*90                   35               7.0

      3.0                  515                   35               3.0

      3.2                  600                   39               1.0

      3.5                  625                   ^3               1.0



                                 TABLE III
        PRECISION OF REPLICATE  ANALYSES USING *t.O mm
        DISCS  AT THREF DIFFERENT CONCENTRATIONS OF
                         BLOOD  CADMIUM
K Cd/100 ml
0.8
2.0
8.2
Standard Deviation
0.07
0.11
0.35
Relative
Standard Deviation
0.088
0.055
0.0^3
    No.  of replications  at  each concentration = 11

-------
 1212
TABLE IV
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Mean
S.D.
COMPARISON
4.0 mt
Venous
0.4
0.4
0.5
0.5
0.9
1.4
2.2
2.8
4.8
5.0
5.7
6.5
6.8
7.8
3.26
2.73
OF MICROMETHOD
UK Cd per
i P.D.
Capillary
0.6
0.6
1.2
1.1
1.4
1.4
1.9
5.4
5.5
4.8
5.0
7.4
7.2
7.7
3.66
2.73
AND CATHODE RAY POLAROiRAPHY
100 ml BLOOD
Pplarggraphy
Venous
^•MA^^H^^^
0.5
0.9
0.8
0.8
1.6
1.0
3.0
3.5
8.1
4.9
1.9
5.4
5.5
6.0
3.14
2.45

-------
                                1213
            ABSORBANCE
                 0.4r
                 0.3
                0.2
                       5.6-12
                       3.2-2.0   SBW 099nmB
                        (600°C)
                                      Control bkx>d value
                                       (=4.0-4.5pgCd)
            jjg Cd/100ml.
8
10
12
    Fig. 2.  Effect of  spectral band width and volatilization
             temperature  on  the linearity of calibration.

     Atomization and SBW Effect.   Using  a  relatively high atomiza-
tion temperature (1200°C), calibration  graphs  obtained from the  range
0-iO  g Cd/100  ml blood curved  irrespective  of SBW  (0.53  to U.99 nm).
Reproducibility was also erratic.    Reduction  of  the  Volatilization
temperature  to  approximately 600°C produced  linear  analytical  graphs
only when  the  "BW was  0.3? nm.    (Fig.  2.)   To establish the  optomum
lowest temperature setting for  the best vapourisation of  the Cd,  blood
discs from a sample were atomized at  increasing temperatures Cueing
SBW of 0.33  nm) and the residual Cd  checked  (Table  II).
3»   Results
     3.1 Venous and capillary blood specimens and urine samples were col-

-------
                                   12U
lected from workers engaged in two industrial processes using Cd for pig-
ment manufacture and smelting respectively.   The  first  is a dusty process
where ingestion is the chief mode of absorption,  whereas fume is the prin-
cipal hazard in the second.  Specimens were  also  collected from workers
where there was little Cd risk and from non-exposed  controls.
     3.2  Precision and Accuracy
     Three different naturally occurring Cd  concentrations in blood were
determined and the results from eleven replications  are shown in Table III,
To evaluate the punched disc (PD) results, we compared  both the capillary
and venous samples against results obtained  using our standard method of
cathode ray polarography in 14 cases (Table  IV).   The correlation between
the venous results done by FD and polarography (POL) was 0.8l and between
the capillary results and POL was 0.&5.  The corresponding correlation
                                 TABLE  Y

                       BLOOD CADMIUM IN THE UN EXPOS ED
                           (»g  per 100 al BLOOD)
  1.
  2.
  3.
  «u
  5.
  6.
  7.
  8.
  9.
  10.
  •  Smokers

(Micromethod)
Venous Capillary
0.2
0.2
0.2
0.8
0.5
0.2
0.2
0.8
0.3
0.6
0.2
0.2
0.2
0.6 •
0.5
0.3
0.2
0.6 *
0.2
0.6 *

-------
         TABLE  VI
LEVELS OF BLOOD CADMIUM FOUND

Number of
Comparisons
( Venous
Mean (
( Capillary
Standard < VenOUS
Deviation( ^^y
Range of < Venous
Samples ( Capillflry
"Unexposed"
Lab. Staff
10
o.to
0.36
0.25
0.19
0.2 - 0.8
0.2 - 0.6
Naval Dockyard
(Silver Brazing)
33
0.37
0.51
0.19
0.3^
0.2 - 0.9
0.2 - l.k
Pigment Factory
(Duet Exposure)
Ik
2.66
2.7^
2.37
1.81
0.5 - 10.3
1.1 - 8.5
Smelting Factory-1
(Fume Exposure)
17
4.20
5.23
2.38
3.55
1.1 - 8.9
1.5 -13-3
Smelting Factory-2
(Fun* Exposure)
8
5.20
5.61
1.9^
1.89
2.2 - 7.8
1.9 - 7.7
                                                                                NS

-------
                                1216
between the 14 capillary and venous results was 0.96.
     3.3  Normal Values
Ten volunteers with no
lary
and venous specimens ei
V. No urines


were examined
FINDINGS
pg Cd per 100 ml Blood


1.
2.
3.
k.
5-
6.
7.
8.
9.
10.
11.
12.
13.
14.

Venous
0.5
1.0
1.5
1.7
1.7
1.8
1.8
2.0
2.0
2.8
3.0
3.3
3.8
10.3
P.P.
Capillary
1.7
1.1
1.4
2.6
1.3
2.6
3-0
2.4
2.4
2.8
2.6
2.2
3.8
8.5
                               TABLE  VII
                   FINDINGS IN COLOUR FACTORY WORKERS
                                                Mg Cd/litre
                                       Urinary Cd (not corrected for SG)

                                                    POL

                                                    N.D.
                                                    N.D.
                                                    N.D.
                                                    N.D.
                                                    4.6
                                                    N.D.
                                                   <3.0
                                                    N.D.

                                                    3.9
                                                    N.D.
                                                    N.D.
                                                    7.2
                                                   13.8
                                                   47.0

-------
                                  1217
     3»4  Workers  engaged  in  brazing (with no exposure for 5-8 weeks)
     33 workers  in a  Naval dockyard who had been engaged in pipe jointing
by silver brazing  with  Cd-containing alloy were examined.  None of these
men had been  exposed  for 5-8  weeks prior to examination.  .Results of BCd
done on venous and capillary  specimens are summarised in Table VI.  Al-
though the  low blood  findings confirmed the absence of recent exposure,
the urinary results showed quantities ranging from 2-83 (JgA in 9 of the
33 workers.   (It was  not detectable in the remaining 24.)  It is well
known that  urinary excretion  of Cd persists long after exposure to Cd has
ceased.

     3.5  Workers  in  a  Pigment Factory exposed to Cd Oxide Dust
     The corresponding  results in Ik workers engaged in pigment manufac-
ture are given in  Table VII.   Environmental measurements had not been taken
recently in this factory.   (We also sampled managerial, clerical and lab-
oratory staff whose findings  were normal and are excluded from the table.)
     3.6  Workers  in  Two Cd Smelting Factories
     Results  in  25 workers in two factories exposed to Cd fume are also
summarised  in Table VT  and the actual findings in the second factory are
given in Table VIII.  Environmental measurements in this factory showed

                              TABLE  VIII
                   FINDINGS IN SMELTING FACTORY WORKERS
            ug Cd per 100 ml Blood                  ug Cd per litre Urine
                    P.P.                            (Not corrected for SG)
            Venous       Capillary                          POL

  1.          7.8           7.7                             35.0
  2.          4.8           5.5                             14.8
  3.          2.2           1.9                             12.6
  4.          2.8           5.4                              9.7
  5.          6.8           7.2                             75.0
  6.          5.7           5.0                             31.0
  7.          6.5           7.4                             22.0
  8.         5.0           4.8                             31.0

-------
                                  1218
seven samples out of ten to be abore the current TLV (0.1 mg Cd/M5).  (5
workers in the first factory with only intermittent exposure are excluded
from Table VI.)
4.   Statistical Analysis of Results
     ^.1  Venous against Capillary BCd
     There is good correlation (0.93) between the results of BCd done on
venous and capillary specimens (Fig.3).  All results are included and the
mean of the venous samples was 1.7*t (ig/100 ml (SD = 2.2?) and the mean of
the capillary samples was 2.02 pg/100 ml (SD = 2.6?).  Capillary results
tend to be higher in the exposed groups which could be regarded as an
advantage in the context for which the technique is recommended in that
the method "falls safe".
                 Capitoy
                  -More than MM
                  at this point
                      Approximate 9S%
                      teUranc. Hmtta
                                            Cap.-V09Van.+0-12
                                              /
                                                CORRELATION'O-93
   Fig. 3.
 0      1234BI7S910
A comparison of results obtained .for blood cadmium
using venous and capillary specimens. The regression
line is shown and  approximate 95$ tolerance limits.

-------
                                  1219
    k.2  Comparison of Blood and jJrinary  Cd
    Comparisons of BCd and UCd  show  that  if  Cd  is detected in the urine
then the mean BCd of the workers concerned is greater than the correspond-
                                                            *
ing mean in those with no Cd detectable in the urine.  There also appears
to be an upward trend in the mean  BCd as the  amount of UCd increases
(Table IX).  The range of the BCd  at  the different urinary concentrations
was, however, wide.
                                TABLE  DC

                  COMPARISON OF BLOOD  AND IWINS CADMIUM
Cd in
Urine
V8/l
Cd in
Blood
pg/lOOml

Not
-*'l
No.

47
Detectable
Mean.- S.D.

0.60 i 0.73
1
< 5 >5 i 1n
~i n ^ f
No. Mean. S.D. No. ! Mean. S.D. No. Mean S.D.
1 • 1
i 1
17 j 2.5 2.4 6 J3.23 2.26 I 22 3.69 j 3.00
5.   Discussion
     It is known  that concentrations of Cd in the blood do not necessarily
reflect the  body  burden,  but it would seem from the results reported here
that BCd may well prove a valuable index of the factory hazard (Table VI).
     Until recently results of Cd determinations in blood at the levels
found in unexposed workers have been conflicting.  Elizabeth Willden (9)
found approximately 2 Mg/10O ml in normal subjects, while earlier reports,
for example.  Butt et al (1O) have quoted much greater quantities.  Problems
of methodology may well account for these variations.  Cd quantities of
less than 1  ^e/100 ml quoted by Friberg et al (11) for unexposed workers
in Sweden, U.S.A.  and West Germany are of the same order as reported here.
     There is general agreement that despite fluctuations, BCd concentra-
tions in exposed  workers  are significantly higher than in unexposed
controls.  Friberg et al  (12)  and our experience certainly bear this out.
     Although the published data (see Friberg et al (13)) suggest that the
BCd nay not  prove a useful index for control of exposure in man, further
investigation is  clearly  needed.  The method described here may facilitate
this.

-------
                                   1220
                               References

1   ULLUCCI, A.P.,  HWANG, J.Y.,  "Determination of cadmium in biological
    materials  by  flame and  non-flame atomic absorption techniques".
    Applications  Laboratory,  Instrumentation Laboratory Inc., 115 Kartwell
    Avenue, Lexington, Mass.

2   HAUSER, T.R., HINNERS,  T.A., KENT, J.L., "Atomic absorption determina-
    tion  of cadmium and  lead  in  whole blood by a reagent-free method",
    Anal. Chem.,  Vol. Mt, 1819 (1972).

3   EDIGER, R.D., COLKMAN,  H.L., "Determination of cadmium in blood by a
    Delves cup technique",  At. Absorption Kewslett, Vol. 12, 3 (1973).

If   JOSKLOW, H.M.,  BOGDEN,  J.O., "Multi  element microanalysis by Delves
    cup A.A. spectrophotometry on chelate/solvent extracts", At. Absorp-
    tion  Newslett.  Vol.  11, 127  (1972).

5   LiraiiHMAN, K.W., "Determination of Cd in biological fluids by the Delves
    modification of A.A.S.",  Clin. Chim. Aeta. U6, 3 21? (1973).

6   DOLIN&EK,  F., STUPAR, J.t GLA2ER, L.t "Determination of Pb and Cd in
    whole blood by flameless  atomic absorption technique", **th Yugoslav
    Conference on Applied Spectroscopy,  (1972).

7   CERNIK,  A.A., SAYERS, M.H.P., "Determination of lead in capillary blood
    using a paper punched disc atomic absorption technique", Brit. J. Industr.
    (ted., 28,  392 (1971).

8    CERNIK,  A.A., "The determination of blood lead using a 4.0 mm paper
     punched disc carbon sampling cup technique", Brit. J. Industr. Med.,
     in the press (197*0.

9    WILLDEN,  Elizabeth G.,  "Drinary and whole blood cadmium concentrations
     in renal disease",  Ann. Clin. Biochem.. 10, 10? (1973).

10   BUTT, E.M.,  et al,  "Trace metal levels  in human serum and blood",
     Arch. Environ. Health,  8, 52 (196*0.

11   FRIBtfRG, L., PISCATOR,  M., NORDBKRG, G., "Cadmium  in the environment",
     p.50  C R C  Press,  Cleveland, Ohio, U.S.A.  (1971).

12   IBID, p.63.

13   IBID, p.65.
The work  presented here  has been  submitted for publication to the
British Journal of Industrial Medicine.

-------
                             1221
                          DISCUSSION

KJELLSTROM  (Sweden)

     We  have  studied in Sweden the cadmium level in blood and
urine of three newly employed workers in a Cd-Ni battery
factory  where the total cadmium level in suspended particulates
were about  50/mg/m3.  Blood and urine samples were tak^n each
week during the first two months and then each month during one
year. The  blood cadmium increased more or dess linearly during
the first four months of exposure to reach a steady state 3
times/ 4 times and 1O times higher then the pre-exposure level.

     During the first year no significant change in urinary
cadmium  occurred.  Thus for early cadmium exposure it seems that
blood cadmium after some months should be a better indicator of
exposure than urine.  For long time low-level exposure we believe
increased urinary cadmium is more a reflexion of tubular damage
then the exposure.


CERNIK (U.K.)

     In  our experience the blood Cd will rise above unexposed
levels if the environmental Cu levels are in excess of the thres-
hold limit values.  With oxide dust hazard the absorption tends
to be slower but nevertheless demonstrable.  Urinary Cd will not
be of much value in these circumstances since tissue uptake and
stabilization, will interfere with the Cd absorption - excretion
mechanism.


PISCATOR (Sweden)

     I want to add  to Dr. Kjellstrbm's statement that the workers
we examined were exposed to cadmium oxide dust.  Exposure to high
concentrations of cadmium fumes may cause sharper  rises in blood
levels of cadmium and may even cause increases in  urinary
excretion of cadmium.
CERNIK (U.K.)

     I agree.  It tends to be easier  to  determine  Cd  fume  absorp-
tion, because the blood Cd response  can  be  rapid  and  reach high
initial levels, especially if the  sample is obtained  during or
after the work shift.  The urinary Cd in these  workers  will rise
significantly and rapidly if  the Cd  concentration  in  the air is
prolonged.

-------
                              1223
       THE  INCIDENCE OF ELEVATED BLOOD LEAD VALUES IN AN
        ASYMPTOMATIC PEDIATRIC POPULATION RESIDING IN A
                MAJOR AMERICAN INDUSTRIAL STATE

                  P, R, FINE AND D, D, DOBIN
Master Health Services, Chicago, Illinois, USA
ABSTRACT

     In  the  nation's first state-wide investigation and survey
for lead poisoning in an asymptomatic population, blood speci-
mens were obtained from more than 7,700 children between 1 and
6 years  of age  who resided in 20 Illinois cities of intermediate
size (10,000 to 150,000).

     During  the first screening phase 6,151 children were sampled;
1,14? (18.6%) were found to have evidence of undue absorption of
lead (greater  than or equal to 40 ,ug/100 ml) while in 51 (0.8%)
of the children unequivocal instances of lead poisoning occurred
(greater than  or equal to 80,ug/200 ml), according to the guide-
lines of the Surgeon General of the United States of America.

     The percentage of children evidencing undue absorption of
lead in  each city varied from 9.6% to 31.3%, while the main blood
lead levels  of  the samples from the different cities varied from
19.8,ug/100  ml  to 32.9,ug/100 ml.

     Preliminary analysis of Phase Two Screening has revealed
extremely interesting results.   The percent of asymptomatic
children screened who demonstrated blood lead values in excess
of 40 mcg.%  ranged from 5.5% in one location to 13.5% in another
locale.    These figures are lower than the range reported during

-------
                               1224
the 1971 campaign  (9.6% - 31.3*),   Despite the apparent  reduction
in range values^ it is still alarming to note the  significant per-
centage of children, randomly screened, who demonstrated  elevated
blood lead levels.

     Perhaps a more meaningful and certainly a more  significant
finding which was  demonstrated during the second phase  was  the
fact that while the ran numbers of children (and therefore  the
relative percentage) with blood lead values equal  to or greater
than 40 mcg.% Has  somewhat lower in 1972t the mean blood  lead
values remained remarkably similar between the two years.    The
Phase One mean blood lead value for all children tested was det-
ermined to be 25.5 mcg.%.   Preliminary examination  of  the  1972
                                           *
data indicates a mean blood lead value of 27.32 mcg.%  (corrected
to reduce influence of fluctuating sample size).

     The findings  suggest that pediatric lead intoxication  is
not confined to urban slumst but does occur with alarming fre-
quency in smaller  communities.

-------
                                   1225


BACKGROUND

     Early in 1971, the Governor of Illinois instructed the Department of
Public Health to  initiate a "lead poisoning program".   Program officials
questioned the relevancy of a single  city approach and chose to expand
the project to include approximately  14 cities, each with a minimum
population  base  of 10,000 persons.   The marked variance in population
density in Illinois served as the primary reason for the approach, and
cities were selected which included a wide range of population densities.
The data obtained from the testing conducted in each city would help to
verify or disprove the hypothesis that population density and housing
serve as two basic variables directly associated with lead poisoning(7J-
A complete analysis of data from the  1971 program and a preliminary
analysis of the 1972 date follows.

MATERIALS AND METHODS

     Prior to 1971, the City of Chicago had the only lead poisoning
detection program in the State of Illinois,8 yet this program provided
services only to one-third of the children in the State.  The remainder
of the State's population is divided  nearly evenly between communities
of less than 10,000 people (including rural areas) and communities
between 10,000 and 150,000 population|9].

     A group of potential target cities from the 149 in Illinois with
populations between 10,000 and 150,000 was identified.  Initial selection
of cities was made on the basis of geographic location, racial composition,
population density and available medical facilities necessary for medical
follow-up.

     Table 1 lists the cities that were screened and describes the popula-
tion of each.  Two of the communities have less than the stated minimum
of 10,000 population.  They are, however, part of a contiguous urban area,
Cook County: hence, their population size is a reflection of more polit-
ical subdivision than of sociological smallness.

     Because earlier work has established a definite seasonal variation
of blood lead levels and of number of lead posoning cases, with the peak
occurrence of both during the summer months  [8,10.11J program  implementa-
tion was scheduled for the summer of 1971.

     Several workers have established that lead posoning cases are most
common  in children 1-3 years of  age  [Il,l2].ln order to  include any older
children that might possibly have elevated blood  lead  levels, the  Illinois
Department of Public Health Program was  focused on  children 1 to 6 years
of age.

     Blood specimens were collected with  sterile,  12 cc. disposable
syringes  fitted  with  1"  21 gauge-A bevel  disposable needles.  Aliauots
of blood were transferred  into lead-free  heparinized vacutainer tubes
 (for lead  and hemoglobin determinations),  and  into  non-heparinized  sterile
tubes  for  serological  testing.

      The  5.0 ml. of  heparinized  whole  blood  for  the lead determination
was  hemolyzed in distilled water,  chelated with  sodium diethyl dithio-
carbonate  at a pH  of  6.5, extracted with methyl-iso-butyl  ketone and
analyzed with an atomic  absorption spectrophotometer,  using a single  slot

-------
                                    1226
                                   Table 1


                       POPl'LATKK DMA Of THE 14 COKtlXITIES

                                  IN UHtCH

                    CHILDREN'*S M.OOD LEAD VALVES WERE DETERMINED
CITY
Aucor*
Springfield
Peorla
E. St. Loul*
Decatur
Joliet
Rock I.l.nd
Eaat Kollne
Harvey & Phoenix
E. Chgo. Hght*.
Chicago Hght*.
Robblna
Carbondale
Rockford
TOTAL
DATE
OF
SCREENING
IK 1971
7-20,21
8*3.4
8-15.16
8-17.18
S-J3.24
8-26.17
8-30
8-31
9-7
9-8
9-9
9-10
9-U.U
9-20,21,12

1970
TOTAL
POPULATION
74.182
«,753
126.963
69,996
90.397
80.378
50.166
20.812
38.242
5,000
40,900
9.641
22,816
147.370
868.636**
CHILDREN
ACES
1-6
8.817
9,203
13,043
9,187
13,572
8.441
5.091
1.419
4,612*
1,004*
5.181
1,520*
1,512
16,531
100,134
SHJMBER
OF
CHILDREN
SCREENED
449
670
387
376
793
383
285
298
226
172
537
103
164
1,208
6,151
I OF CITY'S
CHILDREN
AGES 1-6
SCREENED
5.09
7.28
2.97
4.09
5.84
4.54
5.60
12.32
4.90
17.13
10.36
6.78
17.46
7.31
6.14
       *  Population data available from the Bureau of the Cenaut varle* with the alt*
         of the city.  Centu* data for the** commit lei did not Include nmber* of
         children by yearly age grouping*.  Therefore, the rater* with aiterUka vere
         calculated from the ratio of under 5 year* population to th« 1-6 year popula-
         tion* of the el t let for which both nurfiera were available.
       ** 20.91 of the total population In middle *l*ed Illlnol* cltle* and 7.81 of the
         total 111 loo 1» population.

burner and a  wave  length  of 2170 Ao13  Standarlzation against controls
of 0,20,40,70,100,200 and 300 mcg.% was performed.   Pooled samples were
run  daily along with the  program specimens.   The standard deviation of
the  results of these pooled specimens  was  3.0 mcg.%£j.A blood  lead value
of 40 mcg.% or greater was considered  evidence of undue absorption of
lead,  and a blood lead value of 80 mcg.% or greater was considered an
unequivocal care of lead  poisoning [12].

RESULTS

      In 1971, six thousand one hundred and fifty one children  were
screened in  14 cities.   This figure  represents approximately 6.2% of the
total  children, in the  state, between  the  ages of 1 and 6 years.   Over
5% of the children in  the "at risk"  age group (1-6 years) were tested  in
10 of the 14  communities, and over  10% of children in the "at-risk" age
group were  screened  in  the remaining four.  A frequency distribution
analysis of blood lead  reported  between 0 mcg.% and 65+ mcg.% revealed
marked digital preference at values  ending in the integers zero and five.

-------
                                   1227
Table 2 shows the mean blood lead  levels  ranged  from a  low of 19.8 mcg.%
to a high of 32.9% with the mean blood  lead level  determined  to be 25.5 meg.
The percentage of children with blood lead  values  40 mcg.% ranged from
9.6% to 31.3%.  For all cities the number and percent of children with
elevated blood lead levels (40 mcg.% or greater) were found to be 1,147
and 18.6%.  It is interesting to report that more  children (4.4.5%) had
blood lead values between 20 and 39 mcg.%  than  between 1  and 19 ncg.%
(36.9%).  Subsequent grouping of 40-59  mcg.%,  60-79 mcg.% etc. showed
a decline in the percentage of children tested found to have  those blood

                             Table  2

        PERCENT OF CHILDREN TESTED HAVING BLOOD LEAD VALUES IN THE
        FOLLOWING 20 MCG.% RANGES AND THE MEAN BLOOD LEAD VALUE  IN
                          EACH COMMUNITY

CITY


Aurora
Springfield
Peorla
E. St. Louis
Decatur
Jollet
Rock Island
East Moline
Harvey & Phoenix
E. Chgo. Hgta.
Chicago Hgts.
Robblns
Carbondale
Rockford
TOTAL

0-19
Meg. 7.


37.9%
21.5
22.0
30.1
46.5
31.9
44.9
37.2
44.7
25.0
38.0
41.7
22.3
48.6
36.9%

20-39
Mcg.%


37.97.
48.4
46.8
45.2
41.2
43.9
34.0
51,3
38.9
61.0
42.6
45.6
61.0
41.8
44.57.

40-59
Mcg.%


16.9%
23iO
22.2
19.9
9.5
21.1
14.4
9.4
13.3
12.8
16.0
12.6
15.2
8.4
14.8%

60-79
Mcg.%


3.8%
6.1
7.5
4.3
2.0
2.3
5.6
2.0
2.2
0.6
3.0
0.0
1.1
1.1
3.1%

80+
Mcg.%


3.6%
1.0
1.6
0.5
0.8
0.8
1.1
0.0
0.9
0.6
0.4
0.0
0.4
0.2
0.8%
Mean
Blood
Lead
Value
Meg. 7.
28.2%
31.5
32.9
28.6
21.5
27.8
25.0
23.5
22.6
27.3
25.2
22.2
28.5
19.8
25.5%
 lead values  (Table  3).  The  mean  of the mean blood lead values was
 calculated in order to  reduce  the influence of the size of the samples
 in the different  communities.  This mean of the mean was 26.0 mcg.%.  A
 similar calculation for the percentage of children 40 mcg.% and over
 80 mcg.% showed averages of 19.2% and 0.9% respectively of the children
 tested had blood  lead values above those levels[l].Evaluation and assess-

-------
                                 1228




                                  Tlble 3




                     X OF CHILDREN HAVING BLOOD LEW VALVES

                        III THE FOLLOWIKC 20 MCC.l RANGES
CITY
Aurora
Springfield
Peoria
E. St. Louis
Decatur
Jollet
Rock Island
East Mollne
Harvey & Phoenix
E. Chicago Hgts.
Chicago Heights
Robblns
Car bond* le
Rockford
TOTAL
0-19
Hcg. X__
37.91
21.5
22.0
30.1
46.5
31.9
44.9
37.2
44.7
25.0
38.0
41.7
22.3
48.6
36.91
20 - 39
Meg. %
37.91
48.4
46.8
45.2
41.2
43.9
34.0
51.3
38.9
61.0
42.6
45.6
61.0
41.8
44.51
40 - 59
Mca. I
16.91
23.0
22.2
19.9
9.5
21.1
14.4
9.4
13.3
12.8
16.0
12.6
15.2
8.4
14.81
60-79
Mcjt. 7
3. 81
6.1
7.5
4.3
2.0
2.3
5.6
2.0
2.2.
0.6
3.0
0.0
1.1
1.1
3.11
80 +
We*. I
3.6X
1.0
1.6
0.5
0.8
0.8
1.1
0.0
0.9
0.6
0.4
0.0
0.4
0.2
0.81
ment of the 1972 data  is  currently being completed.   Preliminary analysis,
has revealed some  extremely interesting results.  The percent of children
screened who demonstrated blood  lead values in excess of 40 mcg.% ranged
from 5.5% to 13.5%.  These figures are lower than the range reported
during the 1971 campaign.  Despite the apparent reduction in range figures,
it is still alarming to note the significant percentage of children,
randomly screened, who demonstrated elevated blood lead levels.   Perhaps,
a more meaningful  and  certainly  a more significant finding which was
demonstrated during  the 1972 campaign was the fact that while the raw
numbers of children  (and therefore the relative percentage) with blood
lead values equal  to or greater  than 40 mcg.% was somewhat lower in 1972,
the mean blood lead  values remained remarkedly similar between the two
years.  Analysis  of  the 1972 data indicates a mean blood lead value of
27.32 mcg.%  (corrected to reduce influence of fluctuating sample size).

-------
                                   1229
DISCUSSION

     Over 81  percent of the children screened were found to have blood
lead values below 40 mcg.% cutoff point established by the U.S.P.H.S.
Approximately 18.6% of the children screened demonstrated blood lead levels
indicative of undue absorption of lead while 0.8% were classified as
unequivocal cases of lead poisoning according to the U.S.P.H.S. guide-
lines[l).It is interesting to note that if the distribution of blood
lead values is lumped by 20 mcg.% groupings, nearly one-half of the
children  (44.5%) demonstrated blood lead values in the 20-39 mcg.% range,
while only slightly more than a third (36.9%) showed values less than
20 mcg.%.  This may indicate  that previously accepted normal blood
lead values  may be in error.  Earlier workers have suggested that the
median blood lead level in urban populations ranges from 16 to 27 mcg$[8T]./j].
Since the children in this program were substantially from the most "at
risk" group, ages 1-6 and living in deteriorated housing, and since their
median blood lead value was 24.1, the values cited for the median blood
lead values for the general population may be excessive.

     A marked digital preference for "zero's" and "five's" in the
reporting of results has already been mentioned.  The five most
frequently appearing values were 20, 25, 30, 10 and 15 mcg.%.  50% of all
values reported in the range between 0 and 100 mcg.% ended in either a
zero  or  a 5, while such values represent only 20% of the possible
reportable values in that range.  Possibly, a sensitive digital recorder
attached  to the atomic absorption spectrophotometer unit would reduce
the digital preference[ 1], The authors wish to point out that there is
much discussion and controversy about the many methods which "may" be
used to determine^ lead levels in the body, recently, the urinary
aminolevulinic acid test described by Davis et al [16]seemed to hold most
promise =is a' quick, simple and inexpensive procedure.  However, Blanksma
et al[17jand Blumenthal [18] amply demonstrated that while the collection of
urine is  more convenient than the collection of blood, the reliability of the
ALA test  1s  open to question.  In fact, in a comparative analysis of the
two methods the correlation coefficient between the urinary ALA test and
the blood lead test was determined to be 0.11.  In that study, the percent
of children that would have been missed if the urinary test was substituted
for the blood test was 33%.  Those findings lead the authors to conclude
that urinary ALA is not a dependable screening test for childhood lead
poisoning.  Based upon these findings as well as an exhaustive review of
literature it is our opinion that the blood lead analysis, although
some-what more cumbersome, expensive and critical is the most reliable
laboratory and screening method available.

CONCLUSIONS
     The Surgeon General has estimated that upwards of 400,000 children
across the nation may possibly have elevated blood lead Ievels[l5] «If
the Surgeon General's estimate was based upon the previously accepted
hypothesis that elevated blood lead levels are confined to major metro-
politan areas, then it would seem that this estimation of the extent of
the problem must be revised upward to include the large numbers of child-
ren outside of major metropolitan areas also affected.  For as this study
has shown, elevated pediatric blood lead levels do exist in considerable
numbers in communities of  intermediate population.

-------
                                 1230
                          BIBLIOGRAPHY

 1.  Fine,  P.R.,  et.al.,  "Pediatrlc Blood Lead Levels:  A Study In
     14  Cities of Intermediate Population."  JAMA September 15,  1972.

 2.  Christian, J.J.:   Remarks, Pediatric Lead Poisoning - The Silent
     Menace:  Proceedings  of the First Annual Symposium, Chicago, June 6,
     1971.

 3.  Chisolm, J.J.,  Kaplan E.:  Lead Poisoning in Childhood - Compre-
     hensive  Management and Prevention.  J. Pediatr.  73:942-950, 1968.

 4.  Grlggs R.C.,  Sunshine I., Newill, V.A., Newton,  B.W., Buchanan S.,
     Rasch  C.A.:   Environmental Factors in Childhood  Lead Poisoning.
     JAMA 187:703-707,  1964.

 5.  Greenberg, M.,  Jacobziner, H., McLaughlin, M., Furst H.T.,  Pelliti-
     eri 0.:   A Study of  Pica in Relation to Lead Poisoning.
     Pediatrics 22:756-759, 1958.

 6.  Jacobziner,  H.:  Lead Poisoning in Childhood: Epidemiology, Mani-
     festations and  Prevention.  Clin. Pediatr. 5:277-286, 1966.

 7.  Fine,  P.R.,  Suhs,  R.H.:  Pediatric Lead Poisoning  in Illinois:
     Principles of Chemical Epidemiology, pp. 119-128.   A Study  Book
     published by the Environmental Protection Agency,  Chamblee, Georgia.
     1971.

 8.  Blanksma, L.A., Sachs H.K., Murray E.F., O'Connel  J-{M.: Incidence
     of  High  Blood Lead Levels in Chicago Children: Pediatrics 44:661-
     667, 1969.
                                     %
 9.  U.S. Bureau  of  Census, Census of Population:  1970, General Popula-
     tion Characteristics Final Report, PC (1)-B15.  Illinois U.S.
     Government Printing  Office, Washington, D.C., 1971.

10.  Chisolm  J.J., Hanson H.E.:  The Exposure of Children to Lead.
     Pediatrics 18:943-958. 1956.

11.  Ingalls  T.H., Tiboni A., Werrin H.:  Lead Poisoning in Philadelphia
     1955-1960.   Arch Environ Health 3:575-579, 1961.

12.  Medical  Aspects of Childhood Lead Poisoning, Statement approved by
     the Surgeon  General, HSMRA Reports 86:140-143, 1971.

13.  Berman E.:   lexicological Applications of Atomic Absorption Spectro-
     photometry,  in  Progress in Chemical Toxicology,  A. Stolman, ed.,
     Academic Press, 1969.

14.  Goldwater L.J., Hoover A.W.:  An International Study of "Normal"
     Levels of Lead  in  Blood and Urine.  Arch. Environmental Health
     15:60-63, 1967.

-------
                                  1231


BIBLIOGRAPHY (continued)
15.   Steinfeld, J.L.:   Public News Release, HEW Z94, U.S. Department
     of Health, Education and Welfare, November 8, 1970.

16.   Davis, J.R., Andelman, S.L.:  Urinary Delta-Aminolevulinic Acid
     (ALA) Levels in Lead Poisoning, Amer J Clin Path 53:956-962,  1970.

17.   Blanksma, Larry A., et al, :  Failure of Urinary Delta-Aminolevu-
     linic Acid Test to Detect  Pediatric Lead Poisoning.
     Amer J Clin Path 53:956-962, 1970.

18.   Blumenthal, S., et al.: A  comparison between two diagnostic tests
     for lead poisoning.  Amer  J. Pub Health. Vol 62 Ho  8:
     1060-1064.

-------
                               1233
      MERCURY IN THE PLACENTA AND FOETAL MEMBRANES AS AN
              INDICATION OF LOW MERCURY EXPOSURE

                A, WANNAG+ AND J, SKJAERASEN++

+  Institute of Occupational Health, Oslo, Norway
++ Department of Obstetrics and Gynecology, Rikshospitalet, Oslo,
   Norway
ABSTRACT

     In order to evaluate the foetal load of mercury when mothers
were exposed to elementary mercury, we have compared two groups
of women and their babies at birth.   One group consisted of
dentists, dentists' assistants and dental technicians.   The
control group were arbitrarily chosen among other women deliver-
ing at the same hospital.

     There were no differences between the groups in mercury con-
tent of erythrocytes or plasma in mothers or babies.   Amnionic
fluid also contained the same amount of mercury in both groups.

     The placenta, the chorion-allantois membrane and the amnion
membrane contained, however, significant more mercury in the ex-
posed group than in the control group.

     Thus the mercury content of the placenta, the chorion-allantois
membrane and the amnion membrane might give an indication of excess
mercury load to the foetus.   This even with exposure levels that
do not alter the erythrocyte or plasma content of mercury in the
mothers or in the babies.

-------
                                1234
1.     Introduction
      Placenta represents  a  barrier to inorganic mercuric ions.(Friberg
and Vostal  ID.)  Elementary mercury will within minutes be oxidized
to inorganic mercuric ion in  the  body, - however, it  crosses the
placenta while being in the elementary state.  ( Clarkson et al.tZ].)
In order to investigate if  the babies  of mothers occupationally
exposed to  low levels  of elementary  mercury  had an increased
mercury load,  one group  of dentists, dentists' assistants  and
dental technicians and one group of non exposed  women were studied.

 2.    Material and Methods
      The women  in the exposed group were working in different
 dental offices in  the Oslo region.  No measurements of their
 workplace exposure were  made, but from  other  studies (unpublished)
 the exposures are known  to be  around the TLV value (0.05 mg
 Hg/m3) or  lower.   The non exposed group was arbitrarily  chosen
 among other women delivering  at the same hospital.   At  birth
 samples of  the mother's  and the baby's  erythrocytes and  plasma,
 the amnionic  fluid, the  placenta,  the chorion-allantois and amnion
 membrane  were  taken for analysis  of total mercury content.   The
 samples were pyrolyzed and the mercury  determined with atomic
 absorption.    The Student's  t-test.was applied  to evaluate the
 differences between the groups, and p values 0,0500 and  less accepted
 as significant.
 3.    Results and  Discussion
       There  was  no difference  between  the groups in mercury
 content of  the amnionic fluid.   Neither were  there any differences
 between the groups in mercury content of erythrocytes and plasma
 in the mothers or in the babies (Table 1).  However, the blood
 content of mercury falls  rapidly after exposure  to elementary
 mercury,  and blood is unreliable as a biological parameter for
 inorganic  mercuric ion load in target  organs.
       The  mercury levels in erythrocytes and plasma are similar
 to those previously found by Tejning 13J  (8.6 ng  Hg/g)  in  Sweden,
 Suzuki et  al.[4)  in Japan,  however, found levels which were  2-3
 times higher (22.9 ng  Hg/g).    Both found that the erythrocytes  of
 the  babies  contained  30-50% more  mercury than the mothers'.
 while the  plasma of  the  babies contained  10-20%  less than the

-------
                                   1235
  mothers'.   The same tendencies appear in our  results (Table 1),
  but the differences are not statistically  significant  ( p  =  0.42).
        The  exposed group  has a higher mercury  load in the placenta
  and  in  the  foetal membranes than the control  group.   The values
  for the placenta are  24.47 ng Hg/g tissue and 12.00 ng Hg/g
  tissue,  respectively (p =  0.0013).  for the chorion-allantois mem-
  brane 21.61 ng  Hg/g and  11.47 ng Hg/g (p  »  0.0032)  and for the
  amnion membrane 14.54  ng Hg/g and 5.32 Hg/g (p =  0.0134)
  (Table  1).
                       Total mercury  content of tissues.

                        The mean value and standard  error of the  mean
                         are given in ng Hg/g tissue
Table I.
Mother erythrocytes
Mother plasma
Baby  erythrocytes
Baby  plasma
Amnionic fluid
Placenta
                   Control

                 7.96 ± 0.49
                 4.62 + 0.48
                 8.85 - 0. 74
                 4. 31 ± 0.57
                 3. 23 ± 0.39
                12.00 ± 0.93
                                                  Exposed
               Student's  t-
               test
Chorion-allantois membrane 11. 47 + 1,25
Amnion membrane            5.32 + 0,76
 8.84 +0.87   p = 0,3520
 4.63 ±0.68   p = 0,9841
10. 18 + 1.44   p = 0,5736
 4.06 1 0.66   p - 0,77^1
 2.33 i 0.49   p = 0,30^6
24.47 ±4.06   P = 0.0013
21.61 ± 3.38   P = 0.0032
14.54 + 4. 12   P - 0.0134
         The mercury content in the  placenta of both groups,  in both
   foetal membranes of the  exposed and in the chorion-allantois of the
   controls  is higher than in maternal  and foetal blood.  Thus both
   placenta  and  membranes  might act as a "sink"  for mercury.
   However, the importance  of  this "sink" function as  protection for
   the foetus against mercury is not known.   Neither is it known if
   the mercury  load will alter the function of these organs,  thus
   impairing the conditions  for  the foetus.   Garrett et al. C 5] and
                                                      203
   Wannag (unpublished) injecting pregnant rats  with    HgCl,  found
   mercury to accumulate in the placenta and in the foetal membranes

-------
                                1236
while little entered the foetus.   The accumulation was larger in
the chorion-allantois  than in  the amnion,  similar  to  humans.   This
in contrast to the greater binding capacity for mercury by the
amnion in vitro.   The in  vivo/in vitro difference  might be caused
by the better blood supply to the chorion-allantois which should
give this membrane a higher mercury exposure than the amnion.

4.    Conclusion
      Even with an exposure level  of elementary mercury which
do  not alter the mercury  content of blood in mothers or babies,
the placenta and the  foetal membranes get an increased  load of
mercury.   This reflects an increased  mercury  exposure to the
foetus,  and  the mercury levels  in  the  placenta and  foetal mem-
branes  might give an indication of the mercury load in foetal
target organs.   However, no  study of  the mercury  content in
foetal organs has  yet been undertaken.

      References
1     FRIBERG, L. ,  VOSTAL,  J. , "Mercury in the environment",
      US  Environmental Protection Agency (1971).

Z     CLARKSON,  T.W.  et  al. ,  "The  transport of  elemental
      mercury  into fetal  tissues",  Biol.  Neonate.  21,  pp.  239-44
      (1972).

3     TEJNING. S. ,  "Kvicksilverhalterna i  blodkroppar och i
      blodplasma hos "normala"  m0drar och deras  nyf0dda barn",
      Pep, of Occupational  Medicine.   Lasarettet i  Lund. Sweden,
      Report No.  680220 (l9b8)."

4     SUZUKI  T.  et al. ,  "Comparison  of mercury contents  in
      maternal blood,  umbilical  cord blood and placental tissues",
      Bull.  Environ.   Entam.  fc Toxicol. , 5.  pp.  502-8 (1971).
 5     GARRETT.  N.E. et al. ,  "Placental transmission of mercury
      to  the fetal  rat", Toxicol. appl. Pharmacol.. 22, pp.  649-54
      (1972).

-------
                             1237
                         DISCUSSION
ROELS (Belgium)
1.    Concerning the blood in the placenta, what pretreatment
did you apply to remove it?

2.    What is the contribution of blood-mercury to the values
you measured in the placenta's?


WANNAG  (Norway)

1.    No specific: the tissue pieces were squeezed to get as
much as possible of the blood out, they were washed in deionized
water.

2.    As we don't know the exact amount of blood - we naturally
don't know the amount of Hg which comes from the blood.  However,
the Hg blood valves (normal group: erythrccytea 7.96 ng HG/g
and plasma 4,62 ng Hg/g) are lower than that of the placenta
(12,OO ng Hg/g).  Thus the "contamination" of blood in the
placenta in this case lowers the calculated result for Hg in
the placenta tissue.


CERNIK  (U.K.)

      What volumes of blood was required for the analysis of
mercury in the unexposed control group.  Also which method of
analysis was used?  Flameless AAS with reduction?


WANNAG  (Norway)

      Around 1 g, both of erythrocytes and plasma was pyrolyzed.

      The Hg from the pyrolyzed tissue was collected in per-
manganate, from which it was driven  of by  stanous chloride and
analyzed in AAS.  However, I am sorry that I can not give the
technical details of the AAS analyses as the analytical side
of  this study was taken care of by our department of chemistry.


ROSIVAL (Czechoslovakia)

      How  large was in your  study  the exposed and control group
of  women?

-------
                             1238
WANNAG (Norway)

      Exposed group 19 women, nonexposed group 26 women.
                  •

BERLIN (C.E.C)

      What was the range of mercury values observed in the
various tissues and how were these values distributed?


WANNAG INorway)

      I am sorry I cannot give the range here accurately, as I
do not have all the data with me.  However,  from the standard
errors of the means found in table 1, it can be seen that the
range is quite large and naturally most so for the exposed
group.

      Ho analysis of the distribution of the values has been
made.


LAUWERYS  (Belgium)

      Do you have any information about the  chemical forms of
mercury in the placenta of both  groups?

      I suppose they could be quite different.  In the control
group mercury is probably mostly under the organic form  (methyl-
mercury e.g.); in the group exposed to mercury vapor, a certain
proportion of the accumulated mercury could  be inorganic mercury,


WANNAG (Norway)
                                                            ~»
      No, only total mercury was analyzed.   However, we hope
soon to be able to analyze the different mercury compounds
separately.  The different compounds are in  fact what really
interests us, as the different mercury compounds have different
toxicokinetics and toxicodynamics.  I feel the time is overdue
to talk about "Mercury" meaning  total mercury, what we should
talk about is the specific mercury compounds like - elementary
mercury - inorganic mercury - methyl mercury, and so forth and
keeping these distinctly appart.

-------
           ERFORDERNISSE IM HINBLICK AUF MESSUNGEN
                      MONITORING NEEDS
        BESOINS EN MATIERE DE MESURE DE [.'EXPOSITION
      NECESSITA' RELATIVE ALLA MISURA DELL'ESPOSIZIONE
  EISEN VOOR MET TOT STAND BRENGEN VAN TOEZICHT OP EXPOSIE
                        (Continued)
Vorsitzender - Chairman - President - Presidente - Voorzitter

                   J.  BOUQU1AUX (Belgique)

-------
                             1241
   ESTABLISHING SAFE AMBIENT AIR QUALITY LEVELS FOR EIGHTEEN
                     HAZARDOUS POLLUTANTS

     SANDRA L, WILCOX, EDWIN L, KEITZ AND LINDA J, DUNCAN
The Mitre Corporation, McLean, Virginia, USA
ABSTRACT

     The central objective of this study was to provide data for
the development of safe ambient air quality levels for eighteen
hazardous pollutants, including thirteen metals, asbestos, boron,
chlorine, fluorides and selenium.   This proposed safe level for
each pollutant was then compared with the magnitude and perva-,
siveness of that pollutant to provide insight for planning and
developing hazardous pollutant control technology.   An additional
value of these proposed safe concentrations is  to provide input
to the current U.S. Government effort to establish official safe
levels of pollutants in the air.

     The principal effort  in gathering the data for developing
safe levels consisted of an in-depth investigation conducted
through personal contact with twenty three recognized  experts in
the epidemiology of air pollution  from the medical and public
health professions.   The  investigations were designed to provide
the most current integrated opinion on this subject.   An analysis
and discussion of  the results are  presented.    In addition, a
comparison  is  shown  between these  results  and ambient  air quality
standards which have been  established by several nations and
various  states in  the  U.S.

     A  comparison  among  the  tonnage of pollutant  emitted, the
population  in  areas  adjacent  to  the major  emitter, and the  safe

-------
                             1242
ambient air quality level was made in order to indicate priorities
which should be established for research and development of control'
technology for these pollutants.

     This paper presents the results of two small parts of a
much larger study performed for the U.S. Environmental Protection
Agency,   Obviously the data presented here must be brief and in
summary form.   Information on all the other aspects of the study
is available from the authors.

-------
                                    1243
1.  Introduction
In order to establish safe ambient air quality (AAQ) levels, it is
necessary to determine a measure of toxicity to humans for the pollutants
under investigation.  These relative toxicity values can then be used to
define safe rates of inhalation of the elements and compounds and, thus,
to estimate safe AAQ levels.  If toxicological information were complete
for the concentrations generally found in the ambient air, and if this
information included data on the effects of these concentrations on the
entire population, a safe AAQ level for each pollutant could be set with
confidence.  Unfortunately, not only are the required data unavailable,
but necessary analytical techniques are beyond present capabilities.
Ideally, the final setting of a safe or tolerable AAQ level for each
pollutant would result from a thorough understanding of the varied effects
on humans at all stages of development as they are exposed to different
combinations of pollutants, the synergistic effect of pollutants, the
changing effects of pollutants as they are chemically altered with
changing weather, and the reinforcement of an airborne pollutant's effects
when the same material is ingested.  However, the determination of
estimates of safe concentrations of pollutants in the ambient air demands  .
an extension of known data into a realm of great uncertainty.
2.  Presumed Safe Ambient Air Quality Levels
Experience, intuition, analysis and information must all be used to arrive
at best possible estimates of levels of pollution that will not prove
harmful to man.  This has been done in the course of this study.  Existing
relevant data have been assimilated from literature and personal interview,
and are presented in Table I for each of the eighteen hazardous pollutants
studied.
The information presented includes first, maximum concentrations measured
in the ambient air, second, the average of expert estimates on the 24 hour
maximum safe ambient air quality level, third, the average of existing
ambient standards, and finally, threshold limit values.
The estimates of presumed safe levels shown in the table represent the
average opinion of the following six experts in the field of air pollution
and human health who suggested numerical estimates:
    1.  Dr. H. A. Schroeder, Dartmouth Medical School
    2.  Dr. H. E. Christensen, National Inst.for Occupational Safety

-------
                                    1244
health.
    3.  Dr. H.B. Elkins, Commonwealth of Massachusetts, Division of
        Occupational Hygiene
    4.  Dr. B. Ferris, Harvard School of Public Health
    5.  Mr. G. Boylen, Industrial Hygiene Office, Massachusetts Institute
        of Technology
    6.  Dr. G. Gordon, Department of Chemistry, University of Maryland
    These experts in the field of epidemiology of air pollution and
toxicity took experimental results, statistical data, and subjective
philosophies of life into consideration to arrive at  meaningful, safe AAQ
levels.  Not all of the estimates for each pollutant are the same; not all
of the experts have identical bases for their opinions.  However, the
averages shown should provide a good basis for further investigation.
3.  Population Affected
Table II presents data showing the total annual emissions for each of the
hazardous pollutants, the major emitters of these pollutants and the
annual emissions from these sources.  Also shown are the number of cities
where these major emitters are known or estimated to be and the adjacent
population.  The annual emissions shown are for the U.S.A. in 1968.  The
adjacent population figures were calculated by assuming that all persons
in the city where an emitter was located were potentially exposed.  How-
ever, an exception was made for very large cities or metropolitan complexes
where an upper limit of one million persons per location was used.  These
figures are intended only to give gross indications of the population
affected and should not be considered immutable.
It is quite clear that the largest segments of the population are affected
by emissions from both coal and oil burning utility plants and residential
fuel oil combustion.  However, it does not follow automatically that these
emitters deserve first priority for future research and control.  The rela-
tive toxicity of the emissions must also be taken into account by compar-
ing the average safe AAQ level with typical AAQ values found in the areas
surrounding the major emitters.  Unfortunately, sufficient data does not
exist  to clearly establish existing AAQ values, and it is possible only
to indicate the approach which should be taken.  For example, the safe
                                                 3
level  for  tin is shown in Table I to be 13.7 ug/m  and a typical AAQ value
                                           3
for tin  in a steelmaking city is 0.01 ug/m .  Thus the safe level is
three  orders of magnitude greater than the observed level.  In a similar

-------
                             1245
                            TABLE I

Ambient Air Quality Values  for Potentially Hazardous Pollutants
                          3
                    ( jjg/m  except as noted)
Pollutant
Arsenic
Asbestos
Barium
Beryllium
Boron
Cadmium
Chlorine
Chromium
Copper
Fluorides
Lead
Manganese
Mercury
Nickel
Selenium
Tin
Vanadium
Zinc
Urban/Non-Urban
Values
Maximum
Observed
— —
— — —
1.50/0.00
.002/.00019

.37/.043

.107. 270
1.20/.25
1.89/.16
16.00/.20
1.44/.08
	
.42/.76
	
.10/.01
1.30/.024
2. SO/. 77
Average
Experts
24hr Safe
Level
5.9
.05
fiber/ml
77.5
.003
50
1.2
30
4.6
10
47
4.7
11.4
0.8
3.7
5.4
13.7
6.8
34
Average* of
Existing
Ambient
Standards
3
— — —
	
.01

— 	
30
1.5
—
8.1
1.0
10
0.3
—
—
—
2
~
TLV •!• 100
Workweek
Time-weighted
5 arsenic and cpds.
2 arsine
.05 fiber/ml longer
than 5(j.
5
.02
100 B203 & BBr3
30 BF3
1 oxide fume
2 sol. salts & dust
30 C12; 3 C102
70 HCl; 4 GIF-
1 Cr03;5 sol. salts
10 metallic and
insoluble salts
1 fume; 10 dust
20 HF;25F; 2 FZ
1.5 lead ar senate
2 others
50
.1 alkyls;.5 others
.07 nickel carbonyl
10 others
4 selenium fluoride
2 others
1 organic; 20 inorg.
.5 fume; 5 dust
V2°5
50 ZnO; 10 ZnCl2

-------
                                   1246
                                   Table II
      Population Affected by Major Emitters  of Hazardous Pollutants
Pollutant
Arsenic
Asbestos
Barium
Beryllium
Boron
Cadmium
Chlorine
Chromium
Copper
Fluorides
Lead
Manganese
Mercury
Nickel
Selenium
Tin
Vanadium
Zinc
Total
Emissions
(£opg/yr)
9268
6261
10,826
144
9520
2305
78,200
12,006
13,463
163,140
9,000
18,993
793
6,233
855
356
18,873
150,515
Tons/yr
Major Emitters Emitted
agric applications 2925
primary Zn roasting 1390
mining 5610
Ba chemical mfg 4400
milling & handling 2700
utility coal boilers 101
utility coal boilers 3055
B chemical mfg 2400
iron blast furnaces 1000
liquefaction & 43000*
handling
utility coal boilers 6425
Cr alloy production 4200*
primary Cu converters 3729
utility coal boilers 28,384
steel open hearth 25,400
secondary Pb smelters 2,000
gray iron foundary 1,400
ferro-alloy 3,669
electric furnaces
application of paints 215
utility coal boilers 173
residential fuel oil 2,435
utility oil boilers 1,441
utility coal boilers 413
glass manufacture 203
steel open hearth 260
residential fuel oil 8,330
utility oil boilers 4,930
steel open hearth 39,000
primary Zn roasting 31,818
Number of
Locations
WJ
estimate
17
9
11
7
325
325
54
45
58
325
16
19
325
64
56
1680
53
no
estimate
325
14429,250
122
32$
249
64
14429,250
122
64
17
Ad j acent
Population
(millions)
no estimate
0.5
negligible
1.1
0.8
at least 15
at least 15
no estimate
5.6
5.7
at least 15
1.3
0.6
at least 15
8.6
17
no estimate
no estimate
no estinate
at least 15
at least 40
at least 13
at least 15
no estimate
8.6
at least 40
at least 13
8.6
0.5
*Before controls; all others after existing controls,  if any.

-------
                               1247
way the safe level for beryllium is 0.005  (JLg/m while a typical AAQ
                                                     3
in a city with a coal burning utility plant is 0.0003  u^/m .  Here the
safe level is only one order of magnitude greater and if other factors
are assumed equal, the beryllium emitter should have priority over  the
tin emitter for future study and control.
Therefore, the establishment of priorities for future action should be
based on both the nuraber of people exposed and the relative toxicity of
each pollutant in the ambient air.
In this study it has been our objective to indicate in a gross way  the
average expert opinion on twenty-four hour safe levels.  We have also
given estimates of the pervasiveness of hazardous pollutant emissions in
the U. S. A.  The work is only beginning and we invite others to join
us in the study of this very important area of air pollution research
and development.
                         DISCUSSION
HOGGER (Switzerland)
      Did the  speaker take  particle size  into .account when evalu-
ating airborne particulates  and their deleterious effects on
man?   We know that with many substances  particle size  can deter-
mine  how far  the  effects are harmful.
KEITZ (U.S.A.)
      The toxicity of the  various pollutants relative  to their
particle size  was not within the preview of this study.  Since
it would be rather difficult to set ambient air quality stan-
dards based on particle size the Environmental Protection
Agency did not ask for this type of information.  We  agree that
such information is of great importance  in hazardous  pollutant
studies and such studies  should be undertaken.
DEN BOER  (Netherlands)
      You gave  us the averages from the  experts 24 h  safe levels,
How close were their levels together; what were the  maximun
and minimum  values?

-------
                              1248
KEITZ  (U.S.A.)

     The spread of this data varied greatly among the various
pollutants.  In the case of barium compounds it was an order of
magnitude.  For beryllium the spread was very narrow.  Informa-
tion on this subject is much more detailed in our complete study
which is available on request from the authors.


LEUNG  (U.S.A.)

     In obtaining the expert estimates, how the solicitation
was done?  Was it by personal interview, group interview, or by
questionnaire that each expert has to provide written answers
separately?  If questionnaire was used, please describe the
questionnaire briefly.  Was any statistical analysis made to
determine the group estimates dispersion, and to ascertain "the
trustworthiness" of these experts estimates?


KEITZ  (U.S.A.)

     The experts were interviewed singly by personal contact.
There was no  formal questionnaire. Infact out of a total of 23
experts who were  interviewed only 6 were willing to give nu-
merical opinions.

     The  others did not  desire  to state  numerical ambient air
qualities.  Due to the small  sample size statistical analysis
was not done.  The dispersion of estimates was large for some
cases  and  small for others.   Information on these details is
available  from the authors on request.   The "trustwortheness" of
the estimates resides solely  in the credibility and professional
reputation of each of the experts.  We feel that an excellent
group was interviewed.

-------
                               1249
        METHOD FOR THE IDENTIFICATION OF ENVIRONMENTAL
       NOISE LEVELS TO PROTECT PUBLIC HEALTH AND WELFARE

           SIMONE L, YANIV+ AND HENNING VON CIERKE**

+  Environmental Protection Agency, Washington D.C., USA
-H- U.S. Air Force, Dayton, Ohio, USA
ABSTRACT

     This paper deals with environmental noise exposure affecting
the public health and welfare and proposes a uniform, predictable,
and easily measured method for characterising the impact of en-
vironmental noise on public health and welfare independently of .
the noise sources.   The concept serves as the basis for the def-
inition of safe noise exposure levels.

-------
                                1250
1.   Int roduc t i on
     The purpose of this paper is to discuss a method for assessing
environmental noise and to identify levels which are protective of
public health and welfare.  The philosophy behind this work can be
summarized in three statements: (1) discussion of public health
makes sense only in terms of the total noise exposure sustained by
the public; (2) desirable levels of noise from each source, and the
duration of people's exposure to these levels, must be derived
from the desirable maximum noise from any source to which the
public may be exposed without undue effects on health; (3) identi-
fication of long-range noise goals requires that one method be used
to characterize the impact of environmental noise independently of
the noise sources which contribute to environmental noise.
     Implicit in the above, is the requirement that any given noise
exposure be predictable and verifiable through measurements by the
same yardstick as any other noise exposure so that the total
cumulative noise exposure of people be obtainable, irrespective of
                                                             1 2
where or under what conditions these exposures are sustained. '
     Thus, a tool emerges which may neglect some details in the
measurement  of noise for  simplicity and uniformity sake.  However,
so long as the agreed upon method is easily translatable to other
more detailed and hence complicated measures, details may be best
taken into account in procedures associated with emission standards.
This development represents a step forward in the noise field since
the lack of progress in noise abatement may have been due to the
absence of a simple, uniform, predictable and inexpensively moni-
tored scheme for assessing noise impact.
2.   Methodology for Specification of Environmental Noise
     The accumulated evidence of research on human response to
sound indicates that the magnitude of sound as a function of
frequency and time are the basic indicators of human response.  This
consideration, combined with the philosophy presented above, has
led EPA to choose the equivalent A-weighted sound level, LA, in
                                                          A
decibels as its basic measure for environmental noise.  Its basic
definition is:

-------
         = 10 loglrt     1        /         LA(t)/10
                                 1251
/
                                        10         dt          (1)
                      "^    x   **!              •
where t  - t  is the time interval over which the levels are eval-
       4-e    J.
uated.
     This equivalent sound level is the constant sound level which,
in a given situation and time period, would convey the same amount
of energy to the ear as does the actual time-varying noise pattern.
Thus, two sounds, one of which contains twice as much energy as the
other but lasts only half as long as the other, would convey the
same amount of energy to the ear and would be characterized by the
same sound level.
     This concept is not new and has been utilized to describe the
onset and progression of Noise Induced permanent Hearing Loss
(NIPTS).   Furthermore, there is evidence that something much like
                                                              4
it relates to the general adverse reaction of people to noise.
     Thus, EPA has evolved the Day-Night Sound Level, L., to
describe the noise climate in those environments in which individ-
uals are expected to be present for prolonged periods of time such
as residential dwellings.  Other schemes have been proposed such as
NEF, CNR, NPL, etc.; however, these schemes are much more compli-
cated, often more limited in application, difficult and expensive
to monitor, while not significantly increasing the accuracy of
predictions.
     The primary interval of time of interest in L .  is the 24-hour
period.  Community measurements have shown that during the night-
time, exterior background noises generally drop in level from day-
time values.  Also, the activities of most households decrea: e at
night, thus, lowering the internally generated noise levels as
shown in Figure 1.  As a result, intrusive noise events become
nore noticeable at night. These differences are accounted for by
applying a penalty to noise events occurring at night, that is, by
treating nighttime noises as if they were several decibels noisier
than they actually are.
     For any reasonable distribution of environmental noise levels,
a 2-period day and a 3-period day yields identical levels within a

-------
                                1252
           8Or
            70
to
g  60
£
o
£
o  5O
        UJ
        o:
            40
            3O
            2O
                                                 p.'
                          s;«.
                                           t
                                           *
                                                      .•
                                                    ' M
                                   i
                                      i J
             0
            MIDNITC
                          12
                         NOON
                       HOURS OF DAY
                                        15
                                                  21
                                24
                               MIDNITE
 Figure  1.
   Noise Inside Living Areas of 12 Homes - values of
   Hourly Equivalent Sound Level as a Function of Hour
   of Day.
few tenths of a decibel.  Therefore, in the development of L,,
the 24-hour day has been divided into a 2-period day with the day
extending from 7:OO a.m. to 10:OO p.m., and the night period extend-
ing from 1O:OO p.m. to 7:00 a.m. the next day.
     To actually arrive at a magnitude of the penalty, an examina-
tion was performed of the diurnal variation in the equivalent noise
levels during the day. L,, and during the night, L , as a function
                        u                         n
of L,.  Examination of a set of 63 community measurements spanning
environments ranging from the quiet wilderness to the noisiest areas
around airports and highways was utilized.  This set of data is
shown in Figure 2.  It can be seen in Figure 2 that in quiet
environments (L.  < 55) L. is the controlling element in determining
L,  because of the reduction in noise levels at night, usually
 on
around 1O dB, while in noisier environments (L,  > 65} the values of
L  are only 3O to 4 dB below daytime values.  Thus, a 1O dB penalty
was chosen which applies pressure towards an around-the-clock relief
in noise reduction.
             The actual equation for L.  is:
 Ldn = 10
          1
         24
(15) (10
                                 'd/10.
(2)
where L, and L  are the daytime" and nighttime equivalent levels
       a      n

-------
                                     1253
respectively.

     To assist  the reader, Figure 3 shows  typical values of L
for various  types of  environments found  in the U.S.A.  together
with corresponding subjective  evaluations.
                                                      dn
tf>
5 "*
S Z.
c
"rid
^
"• I
i

T: 3
V* ~
c *
J* O
a .-
> D
'o-'3'
Is
>- w
Q Z



20

IB

T6


14

12
10

3
&
4

2
0

I T^
0 '

—
..

~ •
Approx . •
Curve, 0 - °
_]_ ° 0 %

O ° \ • w
* *^»c«^)
°Q \ D
0 ° \ -°«
° ° « V
O o 0 • •
-
L_ 1 1 1 1 .,
9
	 1 	 1 	 i 	 1
1 1 1
& R-j-ol/ Wilderness ~
O Residential, Light Traffic
• Rciidenliat, Heavy Traffic ~
• Dene Residential
O C"t», Comrnereiol _
t> Railroad
' Airport
T Freeway ~

m~
a _
• V
B
T —
Q * _
	 \ 	 1 1*
                      50
                            53
                                  40
                                         45
                                               70
                            Doy-Nighl A««roge Sound Level, L , fdB)
                                                            80
                                                                  es
   Figure 2.  Comparison  of the Difference Between Day and Night
               Values of the Equivalent Sound  Level with the
               Day-Night Average Sound Level,  L. .
                                                   dn
          QUALITATIVE
         DESCRIPTIONS
             DAY-NIGHT
            SOUND LEVEL
             DECIBELS
              _90,.           OUTDOOR LOCATIONS
                       LOS ANGELES- 3rd FLOOR APARTMENT NEXT TO
                  	         FhEEWAY
                  "•^  LOS AMGELES- 3/4 MILE FROM TOUCH DOWN AT
                                        MAJOR AIRPORT
               CIT
               (DO
               ME"
  CITY NOISE
  (DOWNTOWN MAJOR
  METROPOLIS )
                            -80-
                VERY NOISY
                 NOISY URBAN
                            -70-
 LOS ANGELES- DOWNTOWN WITH SOME CON-
	   STRUCTION ACTIVITY
 HARLEM- 2nd FLOOR APARTMENT
                                    BOSTON-ROW MOUSING ON MAJOR AVENUE
                                    WATTS-B MILES FROM TOUCH DOWN
                                     .       AT MAJOR AIRPORT
                  MALL TOWN 8
                     OUIET  -
                    SUBURBAN
                                     NEWPORT- 3.5 MILES FROM TAKEOFF AT
                                                 SMALL AIRPORT
                                     LOS ANGELES—OLD RESIDENTIAL AREA
                                    FILLMORE- SMALL TOWN CUL- de-SAC
                       SAN DIEGO- WOODED RESIDENTIAL
                                    CALIFORNIA- TOMATO FIELD ON FARM
                           — 40—
 Figure 3.
Examples of Outdoor Day-Night Sound Level in dB
(re 20 micropascals) Measured at Various Locations
Throughout the U.S.A.

-------
                                1254
3.   Long Range Goals  for Environmental Noise Levels
     3.1  General Approach
     An ideal approach to establishing criteria for environmental
noise levels in terms  of their effects on the long-term public
health and welfare would be to start by establishing the cumulative
noise dose which an individual, or a population, may sustain over  a
lifetime without undue effect, irrespective of where or under what
situation this exposure is received.  This approach has been taken
whenever possible.  There are, however, many difficulties:
     (a)  The dose which an individual accumulates is a function of
his life style.  For example, exposure patterns of office workers,
housewives and school  children are quite different.  Furthermore,
 within each group the exposure varies widely as a function  of
 working, recreational  and sleeping habits.  Thus, two individuals
 working in the same office accumulate different total doses depend-
                                               *
 ing upon where they live, how they commute, and their TV habits.
 Examples of such variations are shown in Figure 4-6.
      (b)  There is little information on the statistical distribu-
 tion of actual doses  and the relationships to long-term health
 effects except for two selected situations:  (1) the long-term
 effects of occupational noise exposure on hearing; (2) the average
 long-term exposure to noise in residential areas leading to annoy-
 ance and complaints.
S
90
TO
60
5O
40
3O
MO
SUBURBAN 	 60
1 - - -*
' I
^j n
* §
Si »• >• i M
•3 153 a ^ "
J> a t*j a ui a ui
	 |_ : ] ' ' J 	 	 	 1 	
£*T * ' " NOON S '
0 S 6 9 .2 IS « »
MOOR Of OAT




•GMT
     Figure 4.
Typical Noise Exposure Pattern of a Pre-School.Child.

-------
                               1255
 OFFICE WORKERS
                                         FACTORY WORKER

-J
90

eo

TO

60

5O
40
so


*U6URBAfJ
1 IBCUIU — -• — —
URBAN - — 	





.


-
-
0.
UI
ui
1





r
i
J
H
n
i



j












''tq ( 24 J
72

n




r-i













Si - s
Q 









CE L.
0 <
£ U











L-



L



r -i~~ 	
CT

K Cn
1 I 5 S
9t a >- _i _i
I ^ O < Ul UJ
J * U UJ DC »-
1 I 1 1
9 12 J
NOON
9 12 15
HOUR OF DAY
6 9
18 21





....

O.
UJ
UJ
_J
c/i







12
MIDNIGHT
24
Typical Noise Exposui
Pattern of an Office Worker.
Pattern of a Factory Worker.
     3.2  Goals for Hearing Preservation
     To establish criteria for hearing, the hearing threshold of a
noise-exposed population has been compared to the hearing threshold
of a non-exposed population.   The analysis has been made in terms
of the most noise sensitive frequency {4OOO Hz) and the least amount
of N1PTS which is measurable (5 dB).  This approach represents a
departure from traditional hearing criteria which are,at least in
the U.S., predicated on averaged frequencies of 500, 10OO, and
20OO Hz and a fence of 25 dB.  These differences arise from EPA's
concern with the total exposure of man rather than occupational
exposure and problems related in occupational criteria developments
to compensation for hearing damage incurred during work.  The result
of these analyses are shown in Figure 7.  Inspection of Figure 7
reveals that virtually the entire population will be protected from
any significant hearing loss if the noise dose does not exceed
73 dB over an 8-hour exposure.
     Since most of the noise induced hearing loss data comes from
occupational exposures, corrections were necessary to generalize
the results to general environmental situations.

-------
                                  1256
          1  2
Figure '
                  10   20  30 40 M 60 70  80  90 95   98 99
                                   •JTILES
                                              998999  9999
                 Exposure  Level  and Hearing Level as a Function of
                 Population Percent Showing Merging of Different
                 N1PTS Curves with PHS Curves.
      (a)  Most environmental noise is not continuous.  Intermittent
 exposure is less damaging than continuous exposure; thus, the
 exposure level can be 5 dB higher than the dose identified.
      (b)  since most of the data base is based on an average work
 year of 25O days,  a -1.6 dB correction factor is applied to account
 for the total  year.
      (c)  To account for the fact that a day is made up of 24 hours
 rather  than 8  hours, a correction factor of -5 dB is also applied.
      Thus,  the yearly cumulative dose which will protect  the
 general  population  from  NIPTS  is:
          L    =73-5+  5-1.6= 71.4 ~ 70  dB
      3.3 Goals lor Environmental Noise with  Respect  to Activity
          Interference and Annoyance
      The underlying concept  in establishing goals for environmental
noise has also been statistical.  This is  not  to say  that  individ-
uals  have the  same  susceptibility; they do not.  Even groups  of
people may vary in  response depending on previous exposure, age,
socio-economic status, political cohesiveness  and other social
variables.   in the  aggregate, however, the average adverse response

-------
                                 1257
of groups  of people is predictable  and related to cumulative noise
exposure.   The response we use  is a combination of such factors as
speech interference, sleep interference,  desire for a tranquil
environment and the ability to  use  telephones,, radio, and TV
satisfactorily.
     In order to derive a dose-response relationship between noise
exposure and human response, the results  of the social surveys of
the last 2O years have been examined.   For the purpose of this
analysis,  and when possible, the noise levels in each survey has
been converted to L, .  The resulting  relations are shown in
Figure 8,  together with a relationship showing the actual number of
highly annoyed persons who actually lodge complaints.
                  Relative Importance of Aircraft As A
                  Factor in DisliXina Area or Wanting to
                      (Heathrow 1st Study) t>7. D-10. I>U, I>12 and D-13
          5
          O
          U
          Id
          O
          Ld
          O
          UJ
          a.
COMPLAINTS AND
 THREATS OF
LEGAL ACTION
              50    55     60     65     70     75     80
               OUTDOOR DAY /NIGHT SOUND LEVEL(Ldn) IN dB (RE 20 MICRO-
                                PASCALS)  °
   Figure  8.   Summary of Annoyance Survey and  Community Reaction
               Results.
     For special  uses such as schools  and offices where criteria
for speech communication is the primary  concern, a  similar averaging
process is used to estimate the potential response  of people as a
group, again  ignoring individual variations.   Research since the
192O's has quantitatively characterized  the masking effects of noise
on speech  perception.  The effects of masking  on speech perception

-------
                                  1258
 indoors are summarized  in Figure 9.  This analysis assumes the
 existence of a reverberant field in the room and an acoustic absorp-
 tion of 3OO sabins typical of living rooms and bedrooms.
                 100
                    45
                  STENX A-*EKKTH> SOWD LFVFL IN dB (re 20 micropascals)
                                    absorPtion ^ical of living rtxro
                           bedrooms and is valid for distances greater than
                        one meter.
   Figure 9.  Normal Voice Sentence Intelligibility as a Function
              Situation?    Back9round Sound Level in an indoor

      ?•*»  Goals  for Enviromental noise  in  Various Spaces and
           Land Uses
      The above considerations lead to the  specification for a
variety of spaces and land uses of numerical values of noise levels
which if not  exceeded,  would provide entirely acceptable acoustical
environaents  from a public health point  of  view.   These can be
summarized in terms of yearly equivalent sound levels as follows:
in residential areas, hospital and other areas where  people are
expected to be present for prolonged periods of time, an L,  level
of **
          55 dB  for indoor and outdoor areas respectively;  for areas
where speech is  of primary concern, an L   ,.<. of  ^5  indoors,  and for
those areas where  hearing protection is of primary concern,  an L  t->k\
of 70 dB irrespective of area.

-------
                               1259
                         References


1.   von Gierke, H., "Impact  Characterization of Noise Including
     Implications of Identifying  and Achieving Levels of
     Cumulative Noise Exposure,"  EPA, NTID 73.4* July 1973.


2.   "Information on Levels of Environmental Noise Requisite to
     Protect Public Health and Welfare with an Adequate Margin of
     Safety," EPA 55O/9-74-OO4, March 1974.


3.   Robinson, D.W. and Cook, J.P. , NPL Aero Report No. Ac 31,
     National Physical Laboratory, England, June 1968.


4.   Meister, F.J., "Der Einfluss der Einwirkdauer bei der
     Beschallvmg des Ohres,"  Larmbekampfung 1O,  (3/4) June/August
     1966.


5.   Johnson, D.L., "Prediction of NIPTS Due to Continuous Noise
     Exposure," Joint BPA/USAF Study, EPA 55O/9-73-OO1-B,
     July 1973.
                            DISCUSSION
 BERLIN  (Sweden)
       I  have difficulties to  accept your assumption that  there
 is a  simple relation between  A-weighted average noise levels
 and the degree of disturbance  of the population.  I think  the
 weight  of the evidence in the  literature is that the character
 and frequency of noise are at  least as important as the  level.
 What  neurophysiological basis  do you have for this assumption?

       I  also like to point out  that according to our experience
 and acceptance of your standard would lead us into difficulties,
 e.g.  noise from single aeroplanes, motorvehicules/ trains  etc.
 passing with a low frequency,  can be as disturbing for the
 sleep in a quite area than a continuous noise of much higher
 A-weighted    value.  We have  positive evidence for that from
 investigations going on in our laboratory.

-------
                              1260
YANIV (U.S.A.)

     I certainly would agree with you that the response of
human beings to sound depends strongly on the frequency of sound.
Two basic approaches can be utilized to account for these dif-
ferences:

     - one may divide the sound pressure spectrum into a series
of contiguous frequency bands by electrical filters so as to
display the distribution of sound energy over the frequency
range;

     - one may apply a weighting to the overall spectrum in
such a way that the sounds at various frequencies are weighted
in much the same way as the human ear hears them.

     The first approach is employed in detailed engineering
studies of noise sources.  However, for monitoring purposes it
is much too complicated.  The second approa'ch uses a special
electrical weighting network in the measurement system.  This
network weights the contributions of sounds of different fre-
quencies so that the response of the average human ear is sim-
ulated.  Each frequency of the noise then contributes to the
total reading an amount approximately proportional to the sub-
jective response associated with that frequency.  Measurement
of the overall noise with a sound  level meter incorporating
such  a weighting network yields a  single number such as A-
weighted sound level.  For monitoring purposes this represents
a marked simplification.

      The bulk of  the  literature  tends  to support  the position
that  the A-weighting  is  an  adequate way to deal with the  ear's
differing  sensitivity  to sounds of different  frequencies.   In-
deed,  for  many noises  the A-weighted sound  level  has been found
to correlate  as well with human response as more  complex measures
such  as  calculated  perceived noise level of the loudne.ss  level
derived  from  spectral  analysis.

     .(1) It is true that psychoacoustic research  indicates  that
for some noises a different frequency weighting which increases
the sensitivity to  the 10OO-4OOO Hz  region might  be more desir-
able.   (2) However, various forms  of this alternative are cur-
rently in  existence; none of these have yet progressed in accep-
tance to the  point  where a standard  exists thus limiting  its
applicability for monitoring purposes at this time even though
it might be more desirable.

     The second major characteristic of noise is  that it fluc-
tuates considerably over time.  Thus, one cannot  simply say
that the noise level at a given location or experienced by  a
person is  that many decibels.  To describe the noise completly
one must use a method for accounting for the time-varying signal.
A  number of methodologies for combining the noise from both      !
individuals events  and quasi-steady state sources into measures
of cumulative noise exposure has been developed: NEF, CNR, CNEL,
NNI, PNL etc.  Furthermore these  methodologies while differing
in technical details are in fact conceptually fairly similar
and correlate well with each other.  Also, the relationships

-------
                              1261


between cumulative exposure and community annoyance also cor-
relate well.   (3-4)  The equivalent sound level is formulated
in terms of the equivalent steady noise level in a stated per-
iod of time that would contain the same noise energy as the
time-varying noise during the time period.  Again the bulk of
the literature supports its use.  In Germany it has been used
to evaluate the impact of aircraft noise in the neighbourhood
of airports.   (5) In Austria it is used for evaluating the im-
pact of street traffic on dwellings (6) and in schoolrooms.  (7)

     Research experiments by Pearsons on the trade-off of level
and duration of noisy events  (8) and other investigations of
the trade off between the number of events and noise level in
aircraft flyovers also support this approach.(9)

     As to your  second comment regarding sleep, I appreciate
your concern which I share.  Nevertheless, as you know there
exists  a very limited amount of consistant data regarding en-
vironmental noise effects on sleep.  I do not know how one may
handle  this problem at this time except by limiting noise levels
at night.  Surely you cannot be implying that we raise back-
ground  noise  levels at night so as to  reduce  the effects of
intrusive  single events.  To my knowledge the value to  32 dB
that would result indoors at night from a level of 55 dB outdoors
is consistent with the limited amount of criteria for sleep.
However, as more data developes the method which we have proposed
will be refined and revised.
                           References

 1.   Bishop,  D.E.   "Judgements  of  the  Relative and Absolute
     Acceptability  of Aircraft  Noise"  J.A.S.A.  40, December
     1966,  p.  103

 2.   Kuyter,  K.D.   The Effects  of  Noise  on Man, Academic  Press
     NY:  1970

 3.   House  Noise -  Reduction Measurements for  Use in Studies  of
     Aircraft Flyover Noise.  Society  of Automotive Engineers,
     Inc  AIR  1081,  October 1971.

 4.   Bishop,  D.E. and Homonjeff, R.O.  Procedures for Developing
     Noise  Exposure Forecast Areas for Aircraft Flight  Operations,
     FAA  Reports DS - 67  - 10,  August  1967.

 5.   Burck, W.  Grutzmacher, M.  and Meister,  F.J. Muller E.A.
     Gottingen 1965.

 6.   Bruckmayer, F.  and Lang J. Disturbance of  Population  by
     Traffic  Noise.  Oesterreichische  Ingenieur - Zeitschrift,
     1967,  H.  8, 302 - 306

-------
                             1262
7.   Bruckmayer, F., and Lang J., Disturbance due to Traffic
     Noise in Schoolrooms.  Oesterrerichische Ingenieur -
     Zeitschrift, 11, (3) : 73-77, 1968

8.   Pearsons, K.S., The Effects of Duration and Background
     Noise Level of Perceived Noisiness, FAA - ADS 78, April
     1966.

9.   Galloway W.J. and Bishop D.E. Noise Exposure Forecasts :
     Evolution, Evaluation, Extensions and Land use Interpretation
     BBN, Inc. Report No. 1862 August 1970 also FAA - 70 -9 -.

-------
                             1263
    RELATIONSHIP BETWEEN ACUTE RESPIRATORY JLLNESS AND AIR
            POLLUTION LEVELS IN AN INDUSTRIAL CITY

               D, LEVY, M, GENT AND M, NEWHOUSE

Respiratory Unit, St. Joseph's Hospital and McMaster University
Medical Centre, Hamilton, Ontario, Canada
ABSTRACT

     A retrospective study relating hospital admissions for ac-
ute exacerbations among adults with chronic respiratory illness,
and among children with acute respiratory disease, to air pollu-
tion levels was carried out in a highly industrialized Canadian
city.

     Two thousand patient records were collated and daily,
weekly and monthly averages for admissions to hospital were re-
lated to a contemporary standard air pollution index  (A.P.I.)
which takes into account particulaie matter and S0g.   A corre-
lation coefficient of 0.53  (P <.0.10) was found for the city as
a whole, but the relationship was more marked where industry
was most concentrated.   Admission data from the  hospital clo-
sest to the industrialized  area showed a much higher  correlation
coefficient with A.P.I.  (0.80, P <^0.05).

     Thus there is a strong relationship between  A.P.I, and
hospital admissions due  to  respiratory illness in a susceptible
population.

     (Supported by National Health Grant No. 215-6225 Government
of Canada).

-------
                              1264
     During the past decade several epidemiological studies have
suggested a relationship between air pollution and morbidity or
mortality due to respiratory disease.  While there has been little
doubt of the association during the notorious disasters which
occurred in the Meuse Valley, London, England, and Denora,Pa.,
•the association of health effects and exposure to lower, apparently
acceptable levels of air pollution over long periods, remains a
matter of some contention.
     For this reason a  retrospective study was done, which in-
cluded all children and adults admitted to all of the hospitals in
a highly industrialized city, for an acute respiratory illness
or an acute exacerbation of a chronic respiratory illness during
a period of one year.   We were thus able to study a large patient
population that varied  in age, type of respiratory disease and
domicile with respect to industry.
     The aim of this study was
1. to establish correlations between hospital admissions for
   acute respiratory illness and an air pollution index,
2. to examine separate  effects of sulfur dioxide and particulates
   as well as other common pollutants, and
3  to study the effects of meteorological  factors .
     Hamilton, Ontario, Canada's largest steel  producing and one
of its most polluted cities,  lies at  the south-west tip of Lake
Ontario and has a metropolitan population  of  approximately
35O,OOO.   It  is unique  by virtue of  the  fact  that there is a  single
point source  of intense pollution -  the  steel factories and  that
it can be  readily divided  into  four main districts  each of which
has a biological monitoring  station,  otherwise  known  as a hospital
which,  for the most part,  serves  the  local district population.
He will refer to these hospitals as A,B,C, and D.  (figure 1).
Hospitals A and B lie below a 3OO foot plateau, 2.5 and 4-5 Km
 from industry,  while C and D are on the plateau 3.5 and 8 Km
 away (figure 2).   The  single air pollution monitoring station is
 situated in the centre of the city at (P).  For most of the
 year the prevailing wind is from the south-west.

-------
                              1265
        Air Pollution &
        Respiratory Illness
Figure 1  :  Location  of the smelter, air pollution measuring
            station and biological monitoring  stations  (hospitals)
            in  Hamilton.
         Air Pollution &
         Respiratory Illness

 Figure 2 :   Distance of biological monitoring stations  (hospitals)
             from steel factories in Hamilton.

-------
                            1266
     Patient  selection was made by reviewing the charts  of those
admitted for  acute respiratory illness  (figure 3).   After com-
paring the admitting and discharge diagnosis and evaluating the
relevant clinical data, only those suffering from acute  respir-
atory disease related to bronchitis,  bronchiolitis,  emphysema,
pneumonia and asthma were chosen for  this study.  Only those
patients whose histories revealed that  they had suffered exacer-
bations of symptoms of shortness of breath, wheezing and cough
within 5 days of admission were included.  From approximately
5,OOO charts  reviewed only 1,78O were found to meet all  of our
criteria and  these patients were then classified by age  and sex.
     Acute respiratory admissions were related to calculated
Air Pollution Index, as well as sulfur dioxide, particulates,
hydrocarbons, carbon monoxide, oxidants,  pollens, temperature
and humidity.
  AIR  POLLUTION  & RESPIRATORY ILLNESS
                       HOSPITAL
                     ADMISSIONS
                                            HYDROCARBONS
                                             TEMPERATURE
                                              RELATIVE HUMIDITY
                                        VIRAL ILLNESS
                            Figure 3

-------
                             1267
     Figure 4 shows the  formula used for calculating the API.
It is obtained empirically  using two components, sulfur dioxide
and the coefficient of haze (or floating particulates) as its
two variables.  When  the concentration of SO2 is .1 parts per
million and that of the  coefficient of haze is 1 unit, the index
is at our highest  "acceptable" level of 32.
 AIR POLLUTION  & RESPIRATORY ILLNESS
                       THE AIR POLLUTION INDEX
            IS OBTAINED FROM THE FOLLOWING EMPIRICAL FORMULA

                A.P.I.  =2.5 [13.9  C.O.H.  +  104.5 S02] °'8
                     C. 0. H. s Coefficient of Haze
                             (OR Particulate Matter)
                     S02   * Sulphur Dioxide

                 When C. 0. H.  • 1.0 and  S02 =  0.1 ppm
                          Then A. P. I. • 32
                             Figure  4
      Analyses of the air pollution  index and the number of
 admissions were done on a daily, weekly and monthly basis and
 these two variables were related graphically.  As previous
 studies had indicated that  there is a 2 - 3 day lag between days
 of high air pollution index and admissions to hospital, we
 attempted to allow for this delay by looking at average weekly
 pollutant figures and total weekly  admission.

-------
                                1268
      Admissions fell broadly into a younger age group  under lo
 and an older group over 40, 65% of the number being male.
      On graph 1 are plotted the total weekly admissions  to all
 the hospitals and the average weekly air pollution index along
 the ordinate and time in months from July 1970 - June  1971 along
 the abscissa.  This demonstrates that peaks and valleys  in the
 air pollution index are generally reflected in the number of
 patients admitted to hospital - the most marked relationship
 appearing in October and February while the last week  of
 December showed a striking disparity.  The correlation coefficient
 describing this relationship was 0.77, highly significant at the
 1% level.
     60-i
     50-
     4O-
           ADMISSKDNS	
   _
   -
     30-
     20-
     10-





\ ^
1 l\ /vl
\l " 1
1
'l!
A 1
/ * i'
/ » li
\
N / \ ' ^
\ ;\ / \ ! /\
Vv/ iji/^
']
A
CO 7 7
 P<.01
       A
      /
                       O    N    D   Jan.
crapn i Total Weekly Admissions vs. Average Weekly API. for Whole City.

-------
                              1269
     There are several additional  points  to  note here.   The air
pollution index was higher  on  the  average in 1971 than  in 1970
and this was true also of the  total  number of admissions for
respiratory illnesses.  The peaks  of both air pollution and
hospital admissions in October and February  were times  of most
frequent thermal inversions.   Lastly,  we  noted a marked drop in
admissions in the last week of December  in spite of the high air
pollution index.  This is probably explained by the fact that
this week encompasses both  Christmas and  New Years, a time when
people like to remain at home,  and this  hypothesis was  borne
out by a sharp drop in the  total admissions  for all causes to
each hospital during.that week.
     Having looked at the city as  a  whole, let us now examine
each of the four regions by assessing admissions to the hospital
serving that region.
     On graph 2 are plotted the relationship between the average
air pollution index and the total  number  of  admissions  per month
at each district hospital as before.  For hospitals A,B, and C,
a similar relationship was  found while no association was noted
at D.

      _eo
      u.
      *»
       50

           ~i	1—i	1	1	1—i	r—i	r

         JulyASONDJ FMAMJ  July A  S O N D J FMAMJ
                              Months
   Graph 2 :       Monthly Average AR+Total Monthly Admissions
                             vs. Time

-------
                              1270
     On graph 3 we see  the  relationship between total weekly
admissions to each of the four hospitals along the ordinate and
the average weekly air  pollution index along the abscissa.   We
noted that the correlation  coefficient measuring this relation-
ship was highest at hospital A with a value of O.71, O.57 at
hospital B and O.5O at  C, all measurements highly significant
at the 1% level.  The correlation coefficient at hospital D was
.OO7 and not significant.
30
25
20

15
10

5
D CC-57
P • f 1 1
• • *
• *
A
i
.
• •Is • ••
. I:'* •
• V • *
a

                                                  CC-71
                                                  P
-------
                              1271
      HO
      70
      eo
      50
      4d
      30
      «f
      10
B
               cc»32
                (X.05
     I
cc-77
 P<01

80
70
60
50
40
30
20
d
	
D cc=21
N/S


1
4
A
..;».
1 	 . 	 . . . 	 r-^

c

• •
*•:•
, •


                                                   P<-05
           •2  3  A .5 Sppm

       Graph 4
                            -2 3  A  5  J6ppm
                                802
          Total Monthly Admissions vs. SO^
          for  the  four biological monitoring
          stations (hospitals)
     Graph  5 shows  a  significant correlation in all three
hospitals between the number of admissions and the coefficient
of haze.  And again there was no correlation seen at hospital D.
     When we looked at the other pollutants, carbon monoxide,
oxidants and hydrocarbons, there was no correlation seen in  any
of the hospitals.   When we examined the effects of pollens,
temperature and  relative humidity on admissions it was noted
that while  there was  a moderate negative  correlation with
temperature, that is, more admissions on colder days, pollens
and humidity appeared to have no effect.

-------
                               1272
80
70
60
SO
40
30
20
10
B CC-.57
P<001
-
• • *
• • * .



A
• *
A 1
*
« •
•

                                                        CC=.48
70
60
5O
40
30
20
K)

D CC--XJ5
nre



. • • •

c
• <
•
• *


                                                      CC-.34
123456789
      OOH
            Grapn  5  :
                                      123456789
                                            COH
                                           >
                       for the four biological monitoring
                       stations  (hospitals)
Discussion
     Clearly  the most important finding  resulting from our study
is the marked association between acute  respiratory disease and
the Air Pollution Index and its components, SO2 and floating
particulates  for the whole city.  This relationship is seen to
be more marked the closer to industry patients live.
     Furthermore, when other common pollutants were assessed,
no association with respiratory illness  could be found, suggesting
that SO-  and  particulates are indeed most important in the
pathogenesis  of acute pulmonary disease  due to air pollution and
indicating  the value of the calculated Air Pollution  Index as it
is presently  determined.
     If the monthly average for the Air  Pollution Index is
plotted against admissions for the 12 months of the study, a
remarkably  linear relationship is obtained which allows

-------
                             1273
extrapolation to a "O" pollution level which  predicts a basic
pool of 60 admissions.  When the mean Air Pollution Index for
a month rises from 6 to 22, the "excess  admissions" increase
from 40 to approximately 12O - a 3-fold  change which represents
a more than 2-fold increase in the total admissions,  (graph 6)
 AIR POLLUTION & RESPIRATORY ILLNESS

.
E
&
£
1
g
<
3
K



Grap!
220-i
200-
180-
160-
140-
120-
100-
80-
60-
40-
20-
-

l 6

^
s*
9? •
s' **
5-
r'^
„'
'


246 8 101^1416 182022 2426 283b
Average API per ma
AVERAGE MONTHS API. vsTOtALMOmHY ADM.
     Quite a definite age distribution was noted in this  study.
It is generally accepted that young children and adults past
middle-age are more  susceptible to illness.  In this regard, our
study supports similar work reported recently from both London
and Chicago which showed that for a similar age distribution mild
respiratory illness became more serious in the face of high pol-
lution levels.  The high proportion of male admissions might
be accounted for by considering smoking habits and job exposure
but these factors as well as socioeconomic conditions were
outside the scope of our study.

-------
                              1274
     The marked increase in admissions for respiratory disease
 in October  and to a lesser extent February is  a  phenomenon well-
 known  to chest physicians.  Similar findings were  reported by
 Lawther et  al in London and this may relate to the unusually
 frequent thermal inversions at those times, accounting for
 higher pollution levels.  In the studies of Lawther et al and
Carnow, the reporting technique was inherently limited by the
 subjectivity of the responses.  We think that  by looking at
 hospital admissions in relation to an Air Pollution Index in
 this way, we have been able to introduce a measure of objec-
 tivity although we realize that we are merely  observing the
 tip of an  iceberg.
     In conclusion, this study has shown:
 1. A marked association between hospital admissions for acute
   respiratory illness and a calculated Air Pollution Index.
 2  This relationship also holds true for the  components of the
   Air Pollution Index, namely SO- and floating  particulate
   matter  as expressed in the coefficient  of haze.
 3  As  indicated by the data from district hospitals, it appears
   that the correlation is greater at hospitals  closer to
   industry.
 4  Surprisingly, none of the other pollutants studied showed
   any relationship to acute respiratory illness.
 5  While  a reduction in temperature appeared to  influence the
    incidence of acute respiratory illness, relative humidity
    did not.

-------
                               1275




                           DISCUSSION

STUPFEL  (France)

     During the time of collection of your data did you
observe acute infectious episodes like influenza which could
have been modified by the measured pollutant and meteorological
parameter?


LEVY (Canada)

     No.  Unfortunately, from an epidemiological point of
view, no viral endemies occurred during the study period.


GOLDSTEIN (U.S.A.)

     In point 3 of your conclusions in which the correlation
between air pollution levels and hospital admissions is
greater at hospitals closer to industry (e.g. hospital A)
are not these hospitals also closer to your single monitoring
station than is hospital D which is furthest away from the
monitoring station?


LEVY (Canada)

     Since the pollution monitoring station was in the
centre of the city, the city hospitals were, of course,
closer to the single pollution monitoring station than
hospital D.  Vlithin the city, however, hospital A (closest
to both industry and the pollution monitoring station) had
the highest correlation, while hospital B which was further
away than hospital C  (4.5 vs. 3.5 km) from industry (but
below the plateau) showed a greater correlation.

     Because of the prevailing wind  (away from the hospitals +
the monitoring station), the assumption of Lower Air Pollution
levels at hospital D is a reasonable one.  Hospital D was
thus chosen as a "control."


MAGE (Danmark)

     Considering the probability that the air pollution
levels surrounding the only monitoring station are not
perfectly correlated with the air pollution data you used
and the correlation decreases with distance, how can you
justify a study of this type when all your conclusions could
be changed by moving your station by one kilometer?

-------
                              1276
LEVY (Canada)

    While  it  is  highly unlikely that the data from the
single centrally  located pollution monitoring station was
completely  representative of air pollution in districts
represented by hospitals B and C, (and certainly was not
representative of D),  the city's geographical location (in
a bowl) , and prevailing wind, (from the southwest and lending
to carry pollutants  away from the population) , would lead
to an underestimate  of maximum downwind concentrations and
provides a  reasonable  "average*1 value (determined by spot
sampling throughout  the city.)  Thus, in our situation
moving the  pollution monitoring station one or even more
kilometers, while obviously producing an effect (higher
if it were  moved  N.E., lower if moved S.W.) would be unlikely
to change the  air quality data greatly within the city.
Even if one had a pollution monitoring station in each
district, the  same problem might arise.  In spite of this,
the fact remains  that  a very strong correlation exists
between this "average" air pollution value and admissions
for the city,  a correlation which becomes "weaker" in proportion
to the distance from the source of pollution.  Clearly the
more physical  topographic pollution information available
the better; however, as stated at the outset, in our
context, 4  biological  monitoring stations were available
and indeed, from  a health effects point of view, these
appeared to be quite appropriate.

    Fortunately,  nore physical monitoring stations are now
being installed while  at the same time industry is slowly
improving pollution  control.  Our data will hopefully
provide a baseline for future, similar, studies in .our
community to assess  the beneficial health effects of this
investment  in pollution control equipment.

-------
                              1277
        ENVIRONMENTAL CONTAMINATION BY AUTOMOTIVE LEAD

                       HARRY W, EDWARDS

Department of Mechanical Engineering, Colorado State University,
Fort Collins, Colorado, USA
ABSTRACT

     For the past four years3 an interdisciplinary team of re-
searchers at Colorado State  University has been studying environ-
mental contamination by automotive  lead.   Reported in this com-
munication are some of the major findings relative to monitoring
the atmospheric component of the environmental flow of automotive
lead.

     A sampling technique which permits  half-hour collection
periods has been developed using low  lead-blank porous graphite
electrodes as filter media.   Analysis by emission spectrography
simultaneously provides data for other trace  elements in atmos-
pheric particulates.

     Results of tests  in  which  various membrane filters were
backed by  liquid scrubbers containing nitric  acid indicate  that
these membrane filters allow a  significant portion of the  total
atmospheric  lead to  escape collection.    This deficiency may  not
be  solely  due  to passage  of  organic lead;  some  lead reaching
the scrubber is believed  to  be  in  the form of a  very finely
divided  partieulate.    The  implications  are  that many previously
reported atmospheric lead values may  be  low  and  that many  pre-
viously  computed  lead particle  MMEDs  may be  high.

      Field and wind-tunnel tests  confirm that urban  atmospheric
 lead levels  are profoundly influenced by city geometry.    Measure-

-------
                              1278
ment of Kr-85 concentration profiles in a scale model placed in
a wind tunnel indicate that street canyons formed by two and
three storey buildings* such as would be the case in a small
town or in a suburban complex* can inhibit ventilation substan-
tially.   Moreover* under moderate crosswind conditions pollutant
concentrations within the canyon are strikingly anisotropic* and
street-level lead concentrations along the upwind building face
are substantially higher than along the downwind face.   Lead
                                3
concentrations exceeding 10/ug/m  were measured in the street
                                                        3
canyon whil-e rooftop samplers indicated less than l,ug/m .

     The evaluation methodologies discussed are capable of pro-
viding considerable detail  concerning the environmental flow of
automotive lead and should  be useful for other trace substances.

-------
                                   1279
1.   Introduction
     The purpose of this contnunication is to report several  new methodolo-
gies for quantitatively describing the environmental flow of automotive
lead.  The extent and significance of environmental contamination by auto-
motive lead, both in terms of effects on human health and environmental
effects, have recently come under intensive investigation in a number of
countries where there is widespread use of lead alkyls as gasoline addi-
tives.   While there is general agreement that man has substantially lead-
contaminated some portions of his environment fay combustion  of leaded gaso-
lines, scientific data are presently insufficient to adequately define the
medical and environmental consequences {Commission of the European Commun-
ities [1], National Academy of Sciences [2]).  For the past  four years,
an interdisciplinary team of researchers at Colorado State University has
been studying the atmospheric component of dispersal of automotive lead in
the environment.  Summarized in this paper are recent findings concerning
the dispersal and monitoring of automotive lead in the environment.
2.   Atmospheric Sampling
     Of critical importance to the accurate assessment of human exposures
to airborne lead are effective sampling methodologies.  The  most commonly
employed sampling technique involves drawing air through a membrane filter
which is subsequently analyzed for lead.  Two difficulties are now recog-
nized.  The first is that high lead blanks for many commonly used filter
materials preclude short sampling times for lead levels found in typical
urban atmospheres.  Twenty-four hour sampling times are therefore commonly
employed to swamp out the lead blank of the filter.  These sampling times
are far too long to permit correlation of observed atmospheric lead levels
with short-term meteorological phenomena.  The second difficulty is that
a significant portion of the total atmospheric lead escapes  collection by
conventional filters (Purdue [3]).  The portion that passes  through the
filter consists in part of organic lead compounds that reach the atmos-
phere through incomplete gasoline combustion and through gasoline leakages
and spillages.  A second portion of the nonfilterable lead may consist of
very finely divided lead-containing particulates.  Analysis  of atmospheric
particulates indicates that, of the metals ordinarily found  in the atmos-
phere, only lead and vanadium are largely associated with submicron
particles (Lee [4]}.

-------
                                    1280
     A combined sampling-analysis method for the determination of trace
elements in atmospheric participates was developed by Skogerboe and Seeley
as a part of this project  (Skogerboe [5], Seeley [6]).  Porous graphite
spectroscopic electrodes are used as filter media; analysis for several
elements is carried out using an emission spectrographic method.  A major
advantage is that the low  lead blanks of the graphite electrodes permit
much shorter sampling times, e.g., one-half hour with a flow rate of 800-
1200 ml/min for atmospheric lead concentrations in the range of 1-10 pg/m3.
A second major finding of  this work is that the porous graphite electrodes
are more efficient for atmospheric lead than are a number of membrane fil-
ters.  Experiments consisted of drawing air through a filter train in
which various membrane filters were backed by the porous graphite elec-
trodes.  The tests demonstrate that the graphite filters collect about 5 %
more lead than the membrane filters,on the average.
       For other elements, including calcium, aluminum, and magnesium, the
fraction passing through the membrane filter is normally substantially
less, approximately a few  tenths of a per cent of the total mass of the
element collected.  A third advantage of the graphite electrode sampling
technique is the capability for simultaneous multielemental analysis.
     Follow-up studies (Skogerboe [5]}, in which membrane filters were
backed by activated charcoal traps followed by liquid scrubbers containing
nitric acid, indicate that membrane filter inefficiencies  juay not be sole-
 ly due to passage of organic lead.  Simultaneous atmospheric samples were
obtained with the above filter trains downwind from a traffic source.  The
fractions of the total atmospheric lead collected in the train components,
averaged for three determinations, are as follows:  0.45 p membrane filter,
72%; activated charcoal adsorber, 5£; nitric acid scrubber, 23%.  Lead
found in the charcoal trap is tentatively designated organic, and lead
found in the nitrio  acid  scrubber may  be ascribed to very fine lead-con-
taining partioulates which pass both the membrane filter and the charcoal
trap.  More  elaborate tests are under way in an attempt  to provide a clear
distinction between  the organic and particulate portions of .the nonfliter-
able lead.  Additional experiments,  in  which the charcoal traps were replaced
with the previously  described graphite electrode filters, confirm that some
lead which  passes  through the membrane filter is caught in the graphite cup.
These experiments  also indicate that membrane filter efficiency for atmos-
pheric lead decreases with increasing  distance from the source. For example,

-------
                                    1281
percentages of the total lead collected which were found on the membrane
filters at various distances from the readway are as follows : 92 % at 2 m,
68 % at 28 m, and 66 % at 66 m. Thu8| while use of membrane filter samplers
alone would not lead to major errors in computing atmospheric lead burdens
very near the source, atmospheric lead levels could be seriously under-
estimated at greater distances. If the smaller lead-containing particulates
also escape collection by most inertial impaotors used for particle sizing
and the commonly used back-up membrane filters, the result could be a
serious overestimation of lead particle sizes, particularly for samples
taken beyond the region immediately adjacent to a roadway. These and other
deficiencies of sampling techniques for atmospheric lead are under investi-
gation by R. K. Skogerboe of this project.
3.   Lead Dispersion in City Street Canyons
     Human exposures to airborne lead from traffic exhaust tend to be
greatest in city street canyons.  Poor ventilation of city street canyons
can have the effect in many cases of elevating pollutant concentrations
substantially above the levels that would be predicted on the basis of
standard diffusion equations.  There is presently no totally satisfactory
approach for mathematically modeling pollutant dispersion under these con-
ditions.  Physical modeling is presently the only effective technique for
simulating street-level dispersion of atmospheric pollutants released from
automotive exhaust in urban complexes.  By means of physical modeling, the
effects of local geometry of buildings, local topography, trees, and
meteorological factors can be determined.  Measurements are taken over a
scale model of an urban complex in a wind tunnel in which various atmos-
pheric boundary-layer conditions can be simulated (Cermak [7]).  Concen-
tration profiles of a radioactive tracer (Kr-85) released at "street-level"
in the scale model are measured at various points to reveal pollutant dis-
persion patterns, and flow visualization is accomplished by observing
movement of a smoke tracer.  Physical modeling can therefore be highly
useful for predicting the effects of a proposed new structure or traffic
route upon local pollutant dispersion.
     Physical modeling studies have been carried out within this project
by Cermak, Lombardi and Thompson (Cermak [8]) to determine whether major
lead concentration inhomogeneities could be expected in street canyons of
Fort Collins, a medium sized town (population approximately 50,000).  The
geometry selected for modeling consists of a section along the main street
which is lined by two and three-story buildings.  Daytime traffic density

-------
                                    1282
averages 900-1000 vehicles per hour.  The average lead content of gasoline
sold in this area is 2*s grams per gallon.  In the laboratory model, uniform
city blocks were constructed which measured 4Hx4HxH, where H is the model
building height, 5 cm.  Street widths are 1.5H in the present model.  To
simulate pollutant release from traffic on the four-lane street, the radio-
active tracer was released at ground level from two parallel line sources
in the street of the model.  Air flows in the wind tunnel were thermally
neutral.
     The physical modeling studies to date reveal striking concentration
inhomogeneities in the street canyons of the uniform geometry under moder-
ate crosswind conditions.  For pollutants released from two parallel
ground-level line sources under 2 m/sec crosswind conditions, concentra-
tions are higher along the upwind building face than along the downwind
building face.  On the upwind side, relative concentrations decrease from
approximately 5  at street level to approximately 2 at H.  On the downwind
side, relative concentrations decrease from approximately 1.2 at street
level to approximately 0.4 at H..
     Field measurements in downtown Fort Collins provide qualitative con-
firmation of the wind tunnel measurements, notwithstanding some important
differences between the present physical model and the actual street can-
yon geometry (Katen [9]).  The height of the two and three story buildings
forming the street canyon averages 9 m, and the actual street width is 42
m.  With a 2.3 m/sec wind from the northeast (the street runs north-south),
lead concentrations along the east side of the street canyon ranged frori
11.3 yg/m3 at street level to 4.0 yg/m3 at roof level.  On the west side
of the street, lead concentrations ranged from 0.9 yg/m3 at street level
to 1.3 vg/m3 at  roof level.  Vertical measurements along a light pole in
the center of the  street  (a  parking  area) ranged from 3.7 yg/m3 at street
level  to 1.4 yg/m3 at roof level.  Simultaneous measurement of atmospheric
CO levels at each  of these points suggests that vertical concentration
gradients are steeper for Pb than for CO.  CO levels are also higher along
the upwind building face  than along  the downwind building face.  Data from
two additional sampling points  above the  rooftops on each side of the
street are as follows:  east side, 0.4 yg/m3; west side, 0.9 yg/m3.
Correspondingly, above-rooftop  CO levels were higher on  the downwind side
under these  conditions.   Additional  measurements in nearby  residential
areas confirm that atmospheric  lead  concentrations decay rapidly with

-------
                                    1283
increasing distance from this source; lead concentrations 2-5 blocks away
were in the range of 0.1 -to 0.3 wg/m3.  A detailed physical model is pres-
ently under construction to permit more quantitative correlation with
field data.  Additional field studies are underway to.measure the nonfil-
terable portion of the atmospheric lead content at various points, both in
the street canyon and in the surrounding residential areas.
4.   Conclusions
     Use of porous graphite speotrographio electrodes for air filters over-
comes the high lead blank problem of many conventional membrane filters and
thus enables ambient lead levels to be monitored for relatively short time
periods. A second deficiency of many membrane filters is that in addition
to organic lead, very fine lead-containing partioulates may escape collect-
ion. These small partioulates may comprise a significant portion of tne
total atmospheric lead content. Reexamination of many previously reported
airborne lead and lead-particle size data may therefore be desirable.
     Air-lead levels along busy streets in smaller towns may not be great-
ly different than those in street canyons of large cities, notwithstanding
a large difference in total traffic volume.  Ventilation is profoundly
inhibited even in canyons formed by two and three story buildings.  Within
these city street canyons, air-lead values exceeding 10 pg/m3 were measured
while values in the adjacent residential area averaged a few tenths of a
ug/m3.  These results demonstrate that data obtained from rooftop samplers
alone may not adequately reflect actual lead exposures of persons living
and working in city street canyons.
5.   Acknowledgement
     This research was supported by the National Science Foundation through
the Research Applied to National Needs program on Environmental Systems and
Resources, Grant GI-34813X1.

-------
                                   1284
6.   References
     [1]  Commission of the European Communities,  Environmental  Health
          Aspects of Lead, International Symposium,  Proceedings, Amsterdam,
          October 2-6, 1972.
     [2]  National Academy of Sciences, Airborne Lead in Perspective,
          Washington, D. C., 1972.
     [3]  L. J. Purdue, R. E. Enrione, R. J. Thompson, and B.  A. Bonfield,
          Anal. Chem., 4£, 527 (1973).
     [4]  R. E. Lee, Jr., S. S. Goranson, R. E. Enrioni and G. B.  Morgan,
          Environ. Sci. Technol., 6, 1025 (1972).
     [5]  R. K. Skogerboe, "Analytical Chemistry," in Impact on Man of
          Environmental Contamination Caused by Lead, H. W. Edwards, ed.,
          Interim Report of the period July 1, 1972 to December 31, 1973,
          NSF Grant GI-34813X1, Colorado State University, Fort Collins,
          Colorado, 1974.
     [6]  J. L. Seeley and R. K. Skogerboe, Anal. Chem, 46, 415 (1974).
     [7]  J. E. Cermak, AIAA J., 9,  1746 (1971).
     [8]  J. E. Cermak, D. Lombard!, and R. Thompson, "Effects of City
          Geometry upon Contaminant  Dispersion in City Streets," in Impact
          on Han of Environmental Contamination Caused by Lead, H. W.
          Edwards, ed., Interim Report  for  the period July 1, 1972 to
          December 30,  1973, NSF Grant  GI-34813X1, Colorado State Univer-
          sity, Fort  Collins, Colorado,  1974.
     [9]  P. C. Katen and  E. R. Reiter,  "Atmospheric Transport," in Impact
          on Man of Environmental Contamination Caused by Lead, H. W.
          Edwards, ed.,  Interim Report  for  the period July 1, 1972 to
          December 30, 1973, NSF Grant  GI-34813X1, Colorado State Univer-
          sity,  Fort  Collins,  Colorado, 1974.

-------
                          1285
                       DISCUSSION
HARRISON (U.K.)
     I would like to point out that one of the published proced-
ures for determination of organic lead in street air involved the
passage of the air sample through a membrane filter, assumed to
collect particulate lead and then through a scrubber containing
iodine monochloride in hydrochloric acid solution.  In view of
the results presented by Dr. Edwards collection of lead particles
passing the filter by the I Cl acid scrubber is probable and this
may explain elevated results for levels of organic lead found
when using this method, as compared to results determined by
other methods.

     I have two questions:

1.   What experiments were performed to determine blank values
for the nitric acid scrubber for lead in reagents and leached
from the glass?

2.   What was the face velocity of particles at the surface of
the membrane filter?  Membrane filter efficiences have been
shown to be critically dependent upon this velocity.


EDWARDS  (U.S.A.)

1.   Lead blanks were measured, and the major_contribution is
the nitric acid reagent blank, about O.3 X 1O 9 grams of lead
per ml of HNO^ redistilled twice in quartz.  Lead leached from
the glass contributes negligibly on a relative basis.  Analyses
were carried out at lead levels about a factor of ten above the
scrubber blanks.

2.   Tests with this system were carried out over the face ve-
locity range of about 2 to 5O cm/sec.  In this range, no depen-
dence of collection efficiency upon face velocity was observed.
PERRY  (U.K.

1.   Have you  attempted  to  establish  the  particle  size  range of
the  lead compounds  that  passed  through  the  filter  paper?

2.   What is the  proportion of  this material  that  exists  as
organic lead?

-------
                               1286
EDWARDS (U.S.A.)

     No, but this work is underway.  One of the difficulties is
that in this particle size region, there is no totally satisfac-
tory analytical methodology.  A combination of complementary
approaches, including electron microscopy and fluorescence, is
presently favored.

2.   The data presently available do not provide an unambiguous
answer to your question.  Lead found in the carbon adsorbent
may not be only organic in nature, and some organic lead may be
eluted from the adsorbent and be caught in the scrubber.  At
this point, I would agree with use of the term "nonfilterable"
to describe lead passing the membrane filter.

-------
                              1287
   PETROLS ADDITIVATED WITH ISOTOPICALLY DIFFERENTIATED LEAD
      PROPOSAL OF AN EXPERIMENT TO ESTIMATE THE INCIDENCE
        OF TRAFFIC ON THE ENVIRONMENT POLLUTION BY LEAD
                 - FIRST EXPERIMENTAL RESULTS
       p, GARIBALDI'*', s, FACCHETTI\ A.
       G. VANINI, P, P, GADDO, M. DE BORTOLI
                       AND P. GAGLIONE1""

+     Snam Progetti - Petroliferous Products Research Laborato-
      ries, Milan, Italy
++    Joint Research Centre, Euratom, Ispra, Varese, Italy
+++   Societa Italiana Additivi per Carburanti S.p.A., Italy
++++  Institute of Hygiene, Turin University, Italy
+++++ Fiat S.p.A., Central Research Laboratories Management,
      Turin, Italy
ABSTRACT

     When correctly applied the technique used to determine the
isotopic composition of  lead can be used to monitor the contri-
bution of road traffic to  lead pollution in the environment,
especially in the atmosphere and in soil* vegetation, food and
the human body.

     Tests mill be carried out in at  least two regions of Italy,
using petrols containing lead with an isotopic structure which
is either identical to or different from that of the  lead they
currently contain (whether naturally  or from some other source).

     Analyses of  lead samples from major mines have shown that
Australian lead  (Broken  Bill Mine) has a markedly different
isotopic composition.

-------
                               1288
     We intend to use this type of lead to prepare anti-knock
additives /or use in petrols to be marketed in the regions under
study.   Appropriate sampling should then enable us to determine
how far lead pollution is due to exhaust emissions from oars.

     Analytical tests on the possible presence of other metals
in the atmosphere will provide useful information on the more
general problem of atmospheric particulates.   The areas selec-
ted for this study are Piedmont (Turin and district) and Sardinia
(Cagliari City and district);  the first was chosen for its high
traffic density and high level of industrializationf and the
second because it is exceptional in having a high lead content
in the soil, which could influence diet.   Both these regions
are well suited to market distribution of the required type of
petrol.

     The programme of tests is scheduled to  last for three
years beginning in spring of 1974.   Petrols containing addi-
tives of Australian lead will be distributed on the market for
a period of 18 months.

-------
                              1289
     Even to-day there still exists differences of opinion on the
importance of the different sources of lead pollution, especially
as regards motor-car exhaust.  In fact, some experts retain the
additive-containing petrols as the main source of pollution (1-3),
others are not of the same opinion  (4-7). Our*laboratory tests  (8)
have enabled us to estimate the amount of lead exhausted in the case
of European motor-cars in town traffic conditions.  5O-6O% of the
lead contained in the petrol is exhausted, the remainder is re-
tained by the various parts of the engine  (lubricating oil and
deposits) and by the exhaust system.  Other European Laboratories
(9-10) obtained the same results.
     The lead contained in the larger particles emitted by the
exhaust of the vehicles, falls to the ground at distances of
some hundred metres from the roads, whereas the smaller particles
are diffused into the atmosphere and are transported even great
distances due to the movement of the mass of air  (11).  The
interest in correlating the various sources of lead to  environmental
pollution is, therefore, evident.  Of extreme interest, more-
over, is an estimation of the amount of lead exhausted by vehicles
absorbed by the human body.  The technique of determining the
lead isotopic composition, appropriately used, is considered as
satisfactorily solving this problem.  In fact, several studies
(7, 12-22) made to discover the sources of the pollution by means
of isotopic analysis, led to the conclusion that such technique
can be used successfully providing that the isotopic relation is
known in the various sources, that it remains constant therein
and that it is sufficiently differentiated in each of them.  At
present, the mass spectrometry technique enables measurements to
be made on quantities of product of less than one microgram with
a precision of O.2-O.3/5 on the Pb-204 isotope and of O.l% on the
remaining natural Pb-206, Pb-207 and Pb-208 isotopes.
     An experiment should be carried out in Italy using lead
additive-containing petrol having a differentiated isotopic com-
position to be employed in large areas having different inten-
sities of traffic.  In fact, it is difficult to correlate the
lead used as an additive to the petrols to that present in the

-------
                              1290
environment, since the additives normally employed have different
and variable isotopic compositions  (12).  This experiment should
lead to:
     1) An estimation of the  contribution of the vehicles to  the
        lead pollution in  the atmosphere and ground in town and
        country areas.
     2) An estimation of the  diffusion  in the atmosphere at
        different heights  and at various distances from the
        currents of  traffic.
     3) A clarification of how  the  lead is transferred to the
        vegetation.
     4) An  estimation of how  much of  the absorption is inhaled
        by  man.
     The  experiment  can be outlined as  follows: the SocietS
 Italiana  Additivi per Carburanti  (SIAC) will use for  at least
 18 months only Australian  lead  (Broken  Hill mine) .  In fact,  our
 analyses  on metallic lead  samples obtained from the mines that
 normally  supply  Europe, have  shown  that the Australian Pb pos-
 sesses a  sufficiently differentiated  isotopic composition  (Table I) .
                           Table I
     ISOTOPIC RATIOS OP LEAD  SAMPLES  USUALLY SUPPLIED TO
                      THE ITALIAN MARKET
Origin
Australia (Brokei
Hill mine)
Sample arrived at
Bussi 4.11.72
Sample arrived at
Bussi 19.2.73
Sample arrived at
Bussi 21.9.73
South Africa
Cerro de Pasco,
Peru
Mexico
Canada
Southern Italy
Pb 206
Pb 204

15,7
15,7
15,7
17,8
18,0
15,9-19,5
18,2
17,4-18,3
Pb 207 |Pb 208
Pb 204

15,1
15,1
15,1
15,4
15,0
Pb 204

35,1
35,1
35,1
37,4
37,6
13, 2-16, ^32, 6-40, 6
15,7 38,6 t
15, 2-15, 9)37, 9-39, i
Pb 206
Pb 207

1,04
1,04
1,04
1,16
1,20
1,21
1,16
1,15
Pb 208
Pb 207

2,32
2,32
2,32
2,43
2,51
2,47-2,52
2,46
2,48

-------
                               1291
In certain areas, only petrol containing Australian lead will be
distributed.  Before this petrol will be placed on the market,
the lead isotopic composition will be determined in these areas
on all types of samples which will later be analysed.  Once the
                                              *
Australian lead additive-containing petrol is introduced on the
market, the same determinations will be repeated at intervals in
order to be able to follow the variations in the isotopic com-
position of the lead.  Once the production of anti-knock additives
having a base of Australian lead have been concluded, a new vari-
ation will be made in the isotopic composition of the lead in the
same types of samples.  During the experiment, a control will be
effected on the lead imported into Italy for any other use.
     It is clear from the foregoing that the experiment is both
vast and complex and necessary in order to have a significant
analysis that can solve such an important problem.
     Italy  is particularly suitable for such an experiment for
the following reasons:
-  its refining capacity exceeds its consumption and  therefore
   exports part of the petrol produced
-  a high percentage of the alkyl  lead  used  as  an additive is pro-
   duced by  only  one firm  (SIAC)
-  the  importation of  lead, no matter how  used,  is  subject to
   government authorisation;  it  is,  therefore,  possible  to con-
   trol the  isotopic composition of  all the  lead imported
-  there  is  a high  traffic intensity.
                            Table  II
            ITALIAN PETROLIFEROUS SITUATION IN 1972

Unrefined petrol processed by the refineries
that produce petrol
Petrol produced
Petrol consumed
tons x 1 00
110.329
t2.?07
10.400

-------
                               1292
     Table II gives the total quantity of unrefined petrol
processed in Italy, the amount of petrol produced and the quantity
of petrol consumed in the year 1972.
                         TABLE III
       ANNUAL CONSUMPTION OF LEAD IN ITALY IN 1971
                                            tons x 1000   %
METALLIC STATE
Semi-processed
Electric cables
Accumulators
Alloys

Total metallic state

CHEMICAL COMPONENTS

Oxides
Anti-knock additives
Other chemicals
Sundry

Total chemical components

Total lead consumed
148,0
 60,0
71 ,2
17,1
 7,3
 1,6
 2,8

28,8
208,0    100,0
      Table III relates to the consumption of lead during 1971
 divided according to use.
                          TABLE IV
                CIRCULATION IN ITALY IN 1972
Hotor-cars
Lorries
Buses

Total
Density  (n° inhabitants x vehicles)
    12.484.313
      995.591
       36.150

    13.516.054
          4.0

-------
                               1293
     Table IV indicates the circulation for the year 1972
divided between petrol driven vehicles and diesel driven vehicles.
     Since it is not possible, for obvious reasons, to effect
the experiment all over Italy, it was decided to choose the
most suitable regions.  Selection was made as "follows:
- possibility of the complete supply of petrol by the refineries
  located in the region or adjacent regions
- present vast use of alkyl lead produced by SIAC in these re-
  fineries so as to limit exchanges with other suppliers
- availability of a city having a high traffic intensity and a
  high industrial level.
     An appropriate study carried out showed that Piedmont
 (especially Turin and province) offered interesting prospects for
the work to be carried out.  In fact, the area satisfies the
above requirements and in Turin, for some time now, an efficient
network of centres for the recording of atmospheric conditions
 (Fiat centres)has been in operation.  Sardinia  (Cagliari and
province) was chosen as a comparison for the following reasons:
- excess of refined petrol
- SIAC produced alkyl lead  (>9O%)
- the region is completely isolated
- the high lead concentration in the ground offered the possibi-
  lity of being able to evidence both the contribution due to
  the traffic and that due to natural lead in the environment
  and in man.
     The above experiment should take about 3 years.  We have
 started an experimental investigation to determine the present
 level of lead pollution especially  for the determination of the
 lead isotopic composition present in the various sample environ-
ments .  This  investigation is at present limited to Piedmont
 (Turin and  province)  and to the following aspects:
 - atmosphere  involved
 - washout

-------
                              1294
- human body  (blood and urine)
Atmosphere involved
     Samples were taken in Turin and analyses were made of the
isotopic composition by means of mass spectrometry and of the
concentration by means of atomic absorption.  At Turin, and for
comparison at Ispra which can be considered as a semi-rural
area, samples were also taken by means of the Andersen particle
fractionator which enables the dust to be divided into fractions
of different granulometry.  On the dust thus fractioned, also
semiquantitative determinations of other metals are made in
addition to the isotopic and quantitative analyses of lead.
The  first results obtained are given in Table V.

                          Table  V
Isotopic rations in  the  lead  present in the atmosphere of Turin
Sample
No
1
2
3
4
5
6
7
8
9
Date
28.2.74
28.3.74
2.3.74
2.3.74
1.3.74
20.3.74
20.3.74
20.3.74
24.3.74
Place
C . Novara
P. Car due ci
P. Statuto
C. Novara
P. Carducci
P. Statuto
P . Carducci
C . Novara
P. Carducci
Pb Cone.
vg/ro
8.1
6.2
5.0
1.5
5.5
5.1
3.2
4.0
6.5
Pb 206
Pb 207
1.21
1.22
1.22
1.20
1.22
1.22
1.20
1.21
1.18
Pb 208
Pb 207
2.51
2.49
2.46
2.52
2.48
2.46
2.49
2.54
2.49
Pb 206
Pb 204
19.58
19.83
17.80
18.83
19.84
18.97
18.46
19.43
21.70
 Washout
      The experiment of radioactive fallout, that is, of the
 radionuclides produced and dispersed all over the world fol-
 lowing nuclear explosions, has demonstrated that the non-gaseous
 substances emitted into the atmosphere return in time and dif-
 ferent mechanisms on the surfaces of the world.  Generally, in
 the removal of the substances tied to the solid particles dis-

-------
                             1295
persed in the atmosphere, the action of the precipitations
greatly prevails on all the other causes.  Such removal occurs
through two main mechanisms: The first, called rainout  (or
snowout) consists in the incorporation of the particles by the
drops of rain in the formation of clouds; the second, called
washout consists in the elimination of the particles by the drops
that fall.  Since the greater part of these remarks in this field
concern the radionuclides of the fallout, released at a high
altitude, a lot of information is available on the global effect
of the mechanisms of rainout + washout but there is very little
information on the washout effect alont.  In view of the fact
that the lead, with the particular isotopic relations of this
experiment, will be released into the atmosphere at ground level,
it will, therefore, be possible to determine the entity of the
washout effect for lead by measuring the concentration  in air
and in rainwater.  This phenomenon has already been observed
in New York  (23, 24) for the global lead and other metals.  In
Turin and at  Ispra we have started to observe the quantity of
Pb globally deposited, in precipitation essence and with single
showers of rain; in paralles, daily determinations are  made of
the Pb  concentration in the atmosphere by collecting the dust
of some hundred cu.m of air on filters.
Human body
     The  final aim of the investigation  is the estimation of
the contribution made by the traffic to  the  lead absorption by
the human body.  As far as we know, there are only two  jobs
 (7, 22)  that  have faced this problem by means of isotopic ana-
lyses.   Due to the different metabolic models used, different
results were  obtained, especially on the  time necessary to
observe a variation of the  isotopic composition in the  blood,
once  that of  the environment has been modified.  Rabinowtz and
coll.  (7) determined  the variation of  the lead isotopic composition
                                                          204
in the  blood  caused by the  ingestion of  food marked with    Pb;
consequently, the human blood  in a relatively short time  (some
weeks)  is affected by the variation of  the  isotopic composition
of the  diet.  Although  limited to one  person, it was demonstrated

-------
                              1296
that the contribution of the diet is approximately 50%.
     The research proposed by us is complementary to this work
in that it refers, above all, to the contribution due to inha-
lation and will have, in addition, the advantage of having
available a practically unlimited number of subjects.
     Determinations of the isotopic composition and concentration
are being made on a series of blood and urine specimens taken by
the Institutes of Hygiene in the city and provinces of Turin,
in Milan and Ispra.
     The present value of the isotopic relations in the human
body and the reproducibility in time of such measurements will
also be determined.
                           Table VI
  Isotopic ratios  of  4 blood  samples  from Ispra  and Turin




sample
No
1
2
3
4
City
Ispra
Ispra
Turin
Turin
PbConc.
^g\iOOml
36
26
34
50
Pb206
Pb207
1.18
1.19
1.19
1.18
Pb208
Pb207
2.45
2.45
2.43
2.46
Pb206
Pb204
18.91
19.04
19.53
17.92
      In this way,  it is wished to ensure that possible variations
due  to the additive-containing petrol with Australian lead are
not  hidden by fluctuations to be attributed to other causes.
The  first results are given in Table VI.

References
1.    D. Bryce-Smith Chem. Brit.7, 54 (1971)
2.    D. Bryce-Smith Chem. Brit.7, (7), 284 (1971)
3.    J. Thinker New Scientist and Science Journal  2 Aug.  71,
      p. 323
4.    A.L. Mills Chem. Brit. 7  (4), 160  (1971).

-------
                               1297
 5.    N.K.  Weawer, Ind.  Med. Surg. 40 (9), 31 (1971)

 6.    H.M.  Kirton, New Scientist and Science Journal, 30 Sept.
      1971  p.  772

 7.    M.B.  Rabinowitz, G.W. Wetherill, J-D „ Kogple, Science, 182,
      725 (1973)

 8.    P.  Garibaldi, M. Zerra, N. Passarini, Rivista Combustibili
      27, (11, 12) 505.( 1973)

 9.    K.  Cambell, P.L. Dartnell, Institution of Mechanical Engi-
      neering C 123/71

10.    P.  Clayton, D.J. Ellis, P.L. Palmer, C.J. Potter, S.C. Wal-
      ling, The evaluation of a filter for the removal of lead from
      the exhausts of petrol engines. Warren Spring Laboratory,
      P.O.  pox 20, Gunnels wood Road, Stevenage, Hertfordshire
      SG 1 2BX

11.    M.  Murozumi, T.J.  Crow, C. Patterson, Geochim. Cosmochim.
      Acta 169, 577  (1969)

12.    S.  Facchetti, P. Garibaldi, Rivista Combustibili 27,  (10),
      474  (1973)

13    G.S.  Parkinson, W.M. Catchpole, J. Inst. of Petrol. 59  (566)
      59  (1973)

14.    S.  Facchetti, P. Garibaldi, The lead isotopic composition'
      determination for the  identification of environment pollution
      Intern. Symp. Environmental Health Aspects of Lead, Amsterdam
      (1972) .

15-    T.J. Chow, M. Johnstone, Science, 147, 502  (1965)

16.    W.V. Ault, R.C. Senechal, W.E. Erlebach, "The use of  lead
      Isotopic Composition  as and Environmental Tracer". Final
      reporter of project LH 11, LH  119 April 1968, Isotope Ltd,
      50 van Buren Avenue.  Westwood, New Jersey.

17.    W.V. Ault, R.G. Senechal, W.E. Erlebach, Environ. Sci.  Techn.
      4, 3055  (1970)

18.    T.  Chow, "Isotopic  Identification of Industrial Pollutant
      Lead" Second International Clean Air Congress, Washington
      1970.

19.   Holtzmann., J.  Env.  Science and Technology,  4, 313  (1970)

20.   M.B. Rabinowitz, G.W.  Wetherill, J. Env. Science and  Tech-
      nology, 6,  705  (1970)

21.   Nathans and Stopps,  Global Effects of Lead Containing
      Pollutants, 64th Annual Meeting of  Air Pollution Control
      Association, Atlantic City, N.J., 1971.

-------
                                1298
22.   W.J. Manton., Nature 244, 165  (1973)

23.   H.L. Volchok, USAEC Report - HASL 242 (1971)

24.   H.L. Volchok, L. Toolkel, M. Schonberg., USAEC Report HASL
      281 April 1974 p. 1-143
                           DISCUSSION
 BOUQUIAUX (Belgium)
      What time is required for the isotopic change in lead in
 gasoline additives to have a significant effect on the isotopic
 composition of lead in human blood?


 FACCHETTI (JRC - Ispra)

      I should like to refer to the experimental results of
 Wetherill et al, which although based on a small number of sub-
 jects, I consider to be more reliable than forecasts obtained
 from metabolic models.  With the use of a Pb 2O4 labelled diet
 a significant response was obtained after some two weeks.
 Assuming, still on the basis of the Wetherill study, that the
 incidence due to inhalation is approximately 3O-40% of the total
 blood lead,   a measurable change in the isotopic composition of
 blood lead ought to be observed after a few months since the
 gasolines will have lead additives which have an isotopic com-
 position differing by 1O-15% from that presently in use.

      As stated earlier, these gasolines are to be used in Italy
 for a period of at least two years.


 WETHERILL (U.S.A.)

 1.   Will this experiment change the isotopic composition of
 lead in gasoline over all of Italy of just in the two regions
 under study?

 2.   What will be the time required to change the isotopic
 composition of gasoline?

-------
                               1299
FACCHETTI  (JRC - Ispra)

1.   SIAC, who supply  75-8O% of  the Italian market, are only going
to use Australian lead.  At present they supply over 9O% of the
gasoline in the two regions in question  (Piedmont and Sardinia).
Agreements have been made  to the effect that  in these two regions
and neighbouring districts exclusively Australian lead will be
used, while the lead distributed by other  suppliers {^20-25% of
the total) will be used  in regions far from those under study.

2.   The time required in  the SIAC plants  for stoppage, clean-
up and restart is estimated to be around two  weeks.  The produc-
tion of a  sufficient quantity of lead alkyls  and distribution  to
the refineries should  take a further two months.  This would
give enough time to use  up old stocks.  Furthermore, the refine-
ries have  agreed not to  use any  additive  left over.  Thus,  from
technical  aspects, it  will take  altogether some  4 months before
the regions under  study  and their neighbouring districts are
actually being supplied  exclusively with gasolines  with Australian
lead additives.

-------
                      UMWELTMESSUNGEN
                ENVIRONMENTAL MEASUREMENTS
               MESURES DANS L'ENVIRCNNEMENT
                     MI SURE AMBIENTALI
                  METINGEN IN MET MILIEU
Vorsitzender - Chairman - President -  Presidents - Voorzitter
                 S. CERQUIGLINI-MONTERIOLO
                         (Italia)

-------
                              1303
        APPLICATION DU COMPTEUR COULTER A DEVALUATION
          DE LA MESURE DE LA POLLUTION ATMOSPHERIQUE
             PAR LA MESURE DES PARTICULES SOLIDES
                EN SUSPENSION DANS L'ATMOSPHERE

    J, KUMMER, C, ALLART-DEMUL, D, HEBBELINCK, H. HASTENIER

Ecole de Santfi Publique, University Libre de Bruxelles, Belgique
RESUME

     Une methode se basant SUP I'utilisation du compteur Coulter
est proposes pour mesurer 1e nombre de particules dans I'atmos-
phere en fonction de leur dimension.

     Des partioules atmospheriques sont prelevees sur un filtre
en nitrate de cellulose puis sont mises en suspension par dis-
solution du filtre dans un liquids organique rendu prealablement
oonduoteur.   Connaissant la dilution effeetu.ee et la quantite
d'air preleve on peut oalculer la eourbe de distribution des
partioules atmospheriques par le compteur Coulter.

     Le compteur Coulter est base sur le prinaipe suivant:  un
tube en verre ayant une ouverture ealibree est plonge dans un
liquide oonduoteur.   De part et d'autre de cette ouverture, se
trouve une electrode immergee.   On fait circuler le liquide aveo
les particules en travers de I'ouverture du tube;  chaque passage
de particule fait varier la resistance entre les deux electrodes
et produit une impulsion de tension dont I'amplitude est pro-
portionnelle au volume de la particule.   Moyennant des discri-
minations electroniques, on peut alors faire aisement le- comptage
suivant diverses dimensions et obtenir la eourbe de distribution
des particules en suspension.

-------
                              1304
     Nos recherahes ont ports sur  1 'efficacite de diffevents
solvants en vue de realise** un  solvant organique conducteur don-
nant les resultats les plus stables et les plus reproductibles.

     Nous avons effectue pap aette methods une etude comparative
de la pollution atmospherique entre un site rural et un site
urbain.

     Nous soulignons  la facility,  la  rapidite3 la reproductibi-
lite de cette methode, les dimensions des particules mesurees
pouvant aller de  200  a 0,6 ,u.
ABSTRACT

     A method based  on  the  use  of the  Coulter counter  is  proposed
to measure  the  number of particles present  in the  atmosphere  by
size.

     Atmospheric  particles  are  collected on a cellulose  nitrate
filter and  placed in suspension by dissolving the  filter  in an
organic liquid  which has first  been made conductive.   From the
"known degree of dilution achieved and  volume of aiT samples^  it
•is possible to  calculate the distribution curve for the  atmos-
pheric particles  by  means of the Coulter counter.

     The  Coulter  counter is based on the following principle:
a glass tube with a  calibrated aperture  is  placed  in a conduc-
ting liquid.    To each  side of this aperture there is  an  elect-
rode immersed  in  the liquid.   The liquid containing the  part-
icles  is  passed through the tube aperture;   each particle passing
through causes  the resistance between  the two electrodes  to vary3

-------
                              1305

producing a voltage pulse,  the amplitude of which is proportion-
al to the volume of the particle.   By means of electronic dis-
criminations it is then easy  to  count the suspended particles
according to size and plot  their distribution curve.

     Our experiments concentrated on testing the efficacity of
Various solvents with a view  to  finding a conductive organic
solvent which would yield the most  stable and most reproducible
results.

     Using  this method we carried out a comparative study of
air pollution in a rural and  an  urban location,

     We should  like to emphasize that this method is easy, fast
and reproducible, and can measure particles of  200  to  0.6 .u in
s^ze.

-------
                                   1306
 1 . Introduction .

Les probleraes poses par la determination de la distribution granulometri-
que des particules solides dans  1'air sont de deux ordres: d'une part le
prelevement de ces particules en suspension dans 1'air, et d'autre part le
comptage des particules p re levees et leur repartition en fonction de leur
dimension .
D'uns maniere gSnerale. le pr£levement des particules peut etre incomplet
soit pour les petites particules dans IB cas de 1'impinger et de la sedi-
mentation (1,2) soit pour les grosses particules dans le cas de la preci-
pitation thermique. Nous avons utilisg des membranes filtrantes dont la
dimension des pores est connue,  ceci permet de fixer avec certitude la 11-
mite infgrieure de la dimension  des particules prelevees.
Pour le comptage des particules  nous proposons la misa en suspension des
particules dans un liquide organique par dissolution du f litre et leur
comptage a 1'aide d'un compteur  Coulter ZB. ce compteur permet de compter
at de classer rapldement un grand nombre de particules. La dimension das
particules se situant dans le cas qui nous preoccupe entre 0,6/*- et 40/«-.

2. Description de la methode de  prelsvement et d* analyse.

2 .1 . _Pr616veinent_des_garticules .

L'air est aspire par une pompe a vide sur la membrane filtrante 3 1'aide
du dispositif decrit par la figure 1.
La pompe a vide (Becher VT.3) assure un debit nominal de 1'ordra de 3m3/h.
Ella est pourvue d'un vacuometre et d'une soupape de regulation de vide
qui permet un reglage continu du debit entre 0 et 3m3/h.
On utilise un support de f litre  Millipore en P.V.C. d'un diametre de 50mm.
Un Joint thorique assure une etancheit§ parfaite. Un capuchon de protec-
tion place au sommet du porte-filtre protege efficacement la membrane de
la pluie et de retombees di verses.
Les membranes filtrantes sont das membranes Sartorius type 11302 en nitra-
te de cellulose. El les sont caracterisees par une structure reguliere et
spongieuse avec des grandeurs de pores bien ddfinies. Nous avons utilise
des membranes de porosit€ 1.2/*-. Nous notons que pour la filtration des
gaz. le degr6 d'efficacit£ des membranes filtrantes est considerablement
augmente par des forces electrostatiques . C'est ainsi que la membrane uti-
Iis6e possede un diametre de pores de 1,2/*mais retient egalement des par-
ticules dont le diametre est de  1'ordre de 0.4/c.
La mesure du debit est ass urea par un rotametre (Rota L 63/2400) insure
entre la t&te de pr61evemant et  la pompe a vide. Ce debimetre permet 1' eva-
luation de debits compris entre  0 et 2,4m3/h.
Le compteur  Coulter permet  de  compter £lectroniquement les particules sui-
vant leur dimension.  Son  prlnclpe a d^ja  6te decrit  longuement dans la lit-
terature et  nous renvoyons  aux diff6rents articles s'y rapportant.
Rappelons toutefois que les particules doivent etre  mises en suspension
dans un liquide conducteur  (3,  4. 5,  7).
Le compteur  utllisd est le  models ZB;il permet de situer les seuils infe-
rieurs et superieurs  des  dimensions des particules,  ou de faire des comp-
tages du nombre des particules au-dessus  d'un diametre donne (nombre cu -
nulatif). Ce compteur a §te prealablement etalonnS a I1 aide de particultss
spheriques de latex de dimensions homeodisperses de  5,02/Aet

-------
           port*- Hltr»

                                                              tcocla  CoUUr
Figure 1:  Mode  de  prelevement et d'analyse des particules  solides  en suspension dans
           1'atmosphere  et  traitement du filtre pour le comptage  au compteur Coulter.

-------
                                   1308
 Les dimensions des particules compt£es sont exprim6es en diametre d'une
sphere de volume Equivalent, etant donne que la reponse du compteur est pro-
portionnelle au volume des particules. quells que soit leur forme.

2i3i_Traitement_du_filtre_en_vue_dy_cgmgtage_des_garticules.
Apres un prelevement de 24heures  (soit environ 35m3 d'air aspire), les par-
ticules sont detachees de la membrane par mise en solution de celle-ci
dans un solvant organique compos§ d'un m§lange 1:1 acetone-methanol. Ce sol-
vant a ete filtr£ sous pression sur une membrane en cellulose regenerSe de
0,2/^de poroslte. Ce solvant a et£ rendu, au prealable. conducteur par 1'ad-
dition de 3\ de chlorure de lithium. On dissout ggneralement un quart de
membrane dans 200cm3 de solution. Cette solution est directement prete
pour le comptage . La possibility de dissoudre le flitre permet d'ecarter
les Inconvenients de la mise en suspension des particules par  ultra-sons
(7). Le comptage des poussieres dSfini comma le nombre.de particules conte-
nu dana un volume determine de solutionest r6alis£ apres le passage de 0.5
ml de la suspension du travers du tube a orifice Coulter de 100/t.
L'utilisation d'un tel tube permet de compter des particules dont le diame-
tre se situe entre 1,&Aet 40AC.
La concentration particulalre des echantlllons pour les diverses dimensions
est evaluee a partir de 1'equation suivante:
              . Cn - np) Vd
Ipl
n
 no

 Vd
 Vm • volume du solvant sur  lequel  s'effectue  le comptage.
 Va - volume d'air aspir£.
3. ValiditS de la m§thode et resultats  obtenus.

La prlncipale difficult^ fut le choix d'un solvant organique compatible
d'une part avec le comptage  au  Coulter  c'est-a-dire conducteur et d'autre
part dissolvent parfaitement la membrane  afin  de mettre les particules en
suspension. Nous avons v6rifi6  la validit£ de  chacunes des Stapes du com-
portement du solvant  seul. de la mise en  solution d'un f litre vierge, la mi-
se en solution d'un flitre chargg,  la statistique du comptage, la reproduc-
tibilite dans 1'echantillonnage.
D'une maniere gen&rale, les  resultats sont exprimes en nombre  cuaulatif de
particules pour une dimension de particules donn£e. Les comptages se font
sur O.Scnr* de solution  (Vm).
Le tableau 1 donne pour deux dimensions de particules le nombre des parti-
cules restant apres que le solvant  ait  iti flitre.

Tableau 1.
                                                   (1)
              Vm- Va
• concentration des poussieres en nombre de particules par m3 d'air.
• nombre moyen de particules dans le solvant.
• nombre moyen de particules dans la solution d'Electrolyte ayant dis-
  sous'une membrane filtrante vierge (blanc).
* volume de solution dans lequel est dissous la membrane filtrante char-
Dimension des particules.
2.1 ju
10.6 /«•*-
n
60
2,4
deviattin standart.
12
0,4
Coef. de
variation.
0,18
 Lorsqu'on dissout 1/8 de flitre vlerge dans 9Dcm3 de solvant on obtient
 alors IBS taux de comptage suivants (moyenne sur 20 comptages):

-------
                                 1309
          too«o
          sooo
          1000
           100
            90
        8
         lit
         o
        w
        tS
         I
        2
        g
        u.
        I   -
        U
        s    »
              12         S      10      20        SO     100
                           MAMETM OE» MflTICULES EN MICRONS

Figure 2:  Representation graphique de comptages  au compteur
           Coulter pour
           •  solvant seul
           c  solvant avec 1/8 de filtre  dans 50 cm » (blanc)
           +  1/8  de filtre charge dissous  dans 50cm  de solvant,
           D  difference entre le filtre  charge et le blanc.

-------
                                   1310
Tableau 2.
Dimensions des
parti cules
2.1/*
10. 6/*
n
711
13
Deviation
Standart
138
4,5
Coefficient de
variation.
0,194
0,34
Cas resultats montrsnt qus la dissolution du filtre apporte malgre tout
des impuretes.
CBS resultats pourralent etre quelque peu ameliores par 1'utilisation d'a-
cetone pure mais d'une part celle-ci s'evapore rapidement ce qui donnera
une imprecision sur le volume at d'autre part on est confronte a das pro-
blemes techniques tal la deterioration rapids de certains elements de
1*apparel11age lors d'une utilisation intensive (gralssage des robinets,
tuyaux en P.V.C.).
Lorsqu'on effectue des comptages a differents niveaux electroniques [c'est-
a-dire differents diametres) sur un flitre charge, la reproductibilite
des comptages est donnee par le tableau 3: les calculs ont ate effectues
sur 50 comptages.

Tableau 3.
Dimension des
parti cules.
2.1/*
4.2/u.
22.4^*
n
12.310
71.9
5,4
Deviation
Standart .
266
6
2,5
Coefficient de
variation .
0.02
0,11
0.46
Ayant resolu les problems se  rapportant au comptage meme on peut a lors SB
poser la question de la reproductibilite de la fissure  lorsque differentes
parties du f litre sont utilisees  (1/8  de f litre  est analyst! cheque fois).
O'autres part, on a verifie  la valid!te des resultats  lorsque 1'on effectue
das prelevements slmultanes  a  1'aide de deux systemes  de prelevements pla-
ces au memo endroit.
Les resultats obtenus sont resumes  dans le tableau 4 ou 1'on a chaque fois
utilis6 4 x 1/6 de filtre dlssous dans 50cm3 de  solution et ensuite les
resultats ont fte exprimes en nombre de particules par  m3 en SB basant sur
la relatlon(l).Les calculs se  rapportent a des particules de2,l/X
Tableau 4.
               Prelevements  1
Echanti lions.
1
2
3
4
moyenne
nbre m3 d'air
aspire
nbre part. .
par m3
n
10750
11210
10491
10726
10797
s CV
245 0.02
223 0,02
179 0.017
204 0,02
339 0.03
38,189 m3
209.400 part./m3
n
11366
11281
10866
11121
11158
s
121
195
122
160
243
cv
0,01
0,017
0.01
0,014
0,02
39,205 m3

213.177
part./m3
PrSlevemsnts 2
 La  figure II donne un example de courbe de distribution des particules
 dans  le solvent pur,  dans le solvent plus un filtre vierge dissous.

-------
                          1311
         1 «tct«r
          Wss
          ueos
           HOC
        K
        <
        b
       M
        5
           i*e»
            $09
            It
                                Uc3 PAr.T
Figure J:  Exemples de representation graphique du
           nonibre cumulatif de particules par ar d'air
           en  fonction de leur diametre.
             x Bruxelles le 1 avril  1973
             • Bruxelles le 21 novembre 1973

-------
                                         1312
     ISM*
S
     SMM
                                                             r-
                                                             i
                                              i  i
                                              •  i
                                              i  •
                                            r-J
                       ..J
                            "
                              I!
                              L.J
r-i
                                         U
                                                     L
                    i
                    *
                    I.
                                         IS

                                          MAI 19T1
                    3O
           Figure ^t  Evolution du n ombre  cumulatif du noatbre de particules
                      de diametre superieur  a 4,2  par m  d'air au cours du
                      mois de mai 1973  pour  Bruxelles (- - -) et
                      Dourbes (	).

-------
                                   1313
ainsi qua pour un f iltre charge, le difference entre un filtre chargfi et
la blanc. Les comptages ont ete effectues sur 0,5cm3 de solution.
La figure III donne un exemple de courbe de distribution obtenue.  Enfin on a
a montre que 1'on peut appliquer sur les prelevemsnts atmosphe'rique cette
methods pour suivre 1'evolution de la pollution atmospherique en particules
pour une dimension de particules bien dfifinies.
La figure IV donne un exemple de mesure de 1'evolution du nombre total de
particules de 4,2jU pendant un mois et ce en comparant un site urbain (Brux-
elles) et un site rural (Dourbes).

4. Conclusion.
On peut dire que cette methods SB caracterise par sa simplicite, par sa ra-
pidity et par sa facilite de manipulation. Elle permet des comptages rapi-
des et statistiquament valables.
Ces differents criteres font qu'elle peut aisement etre adaptee sur une
grande echelle a une etude de la pollution atmospherique par les particules
solides.
 BIBLIOGRAPHIE.
 1.  Mr. JACOBSON, S.L.  TERRY,  D.A.  AM3ROSIA
    "Evaluation  of  some Parameters  Affecting the Collection and Analysis
     of Midget Impinger Samples"
    Am. Ind. Hyg.
    Assoc.  J.                          p. 442 (1970).

 2.  P.M.  RENSHAW, J.fl.  BACHMAN, J.  PIERCE
    " The Use of Midget Impinger  and  Membrane filtrer for determining
     Particle Counts  "
    Am. Ind. Hyg.
    Assoc.  J.                          p. 113 (1969).

 3.  R.R.  IRANI
    " Evaluation of particle Size Distribution Obtained from Electro-
     lytic Resistivity Changes ".
    Analytical Chemistry    vol.  32n°9        p.  1163 (1960).

 4.  Manuel  Coulter.

 5.  Coulter Counter Industrial Bibliography (1972)

 6.  K..R.  SCHRAG, M.  CORN.
    " Compansion of Particle Size Determined with  the Coulter Counter and
     bij optical Microscopy"
    Am. Ind. Hyg. Assoc. J.                    p .446     (197O).

 7.  F.G.  ANDERSON.  T.F. TOMB.  M.  JACOBSON.
    " Analysing  Midget  Impinger dust  Samples with  an Electronic Counter"
    (U.S. Dept of the  Interior, Bureau of Nines)
    Report  of Investigations 7105,  1968.

-------
                               13 U
6. H.R. FRIEDBERG,
   " Particle Size  Distribution of Air Borne Particles and air Pollution"
   (Crobangh Laboratories) Presented at the American Institute of Chemical
   Engineers 64th National Meeting, New Orleans, Louisana.
   riach 19th . 1969.

9. R.E.G. Rendall,  G.C.H. VAN SITTERT.
   " The Determination of the Size Distribution of nine dust Using the
    Coulter Counter".
   (Pneumoconiosis  Research Unit of the Souths African Medical Research
   Council. Johanesburg.) P.R.U. Report 10/70
   June 1Bth, 1970.
                          DISCUSSION
PICHE  (Canada)
     Quelles  sont les raisons qui vous  empechent d1installer
votre appareil  compteur sur le terrain  meme,  de fa9on permanente
ou temporaire (roulotte)  afin d'fiviter  les  difficult£s inh§rentes
a la m^tnode  que vous utilises qui est  sujette a toutes les
interventions du solvant dans lequel  il vous  faut §ventuellement
61uer les prglevements rapportfes au laboratoire?

RUMMER  (Belgique)

     Le compteur COULTER se prete mal 3. une utilisation sur le
terrain ou  en aufcomatique notamment a cause dlune possibility de
bouchage du tube d. orifice de la sonde  Coulter.

     Ensuite  le but de 1'Stude £tait  de faire des mesures de la
pollution atmosph£rique par les particules, cette £tude est
notamment a mettre en relation avec celle effectuee  par le Dr.
Dams,  et par consequent,nous n'Stions  pas  int§ress§s de disposer
d'appareils automatiques.

     II faut  aussi tenir compte du fait que le cout des instal-
lations automatiques placets en divers  sites  est nettement plus
61ev6 que si  1'on dispose d'une seule installation au laboratoire
qui analyse les Schantilions arrivant des diff€rents sites.
D'autre part, je tiens S signaler que nous  n'avons pas observe
de difference notable entre un 6chantillon  traits directement au
laboratoire imm6diatement aprSs le prSlSvement et un filtre du
meme prfil&vement qui a 6t6 au prfialable envoy§ par la poste.


DUPUIS  (France)

     Avez-vous  fait des  prfelevements  S  1'aide d'appareils de
prSlSvements  individuels?  Dans 1'affirmative, les r^sultats
obtenus concordent-ils avec les r§sultats th^oriques calculfes  ^
a partir des  comptages faits avec des appareils a postes fixes?

-------
                              1315
RUMMER  (Belgique)

     L'appareillage  de  prglevement  dont  on dispose  se prete
assez mal  a des  prelevements  individuels, quoi  aue  I1on  pourrait
envisager  une  adaptation a cet effet.

     Seuls des prelgvements a  des postes fixes  ont  ete effectu£s
avec cet appareillage.


MOREAU  (France)

     La technique proposSe par 1'auteur  recueille a la fois  les
particules de  nature "physiaue" et  les autres "biologiaues"
c'est a dire, dans les  tallies retenues:  bact^ries spores  de
champignons, grains  de  pollen. La technique mettant les filtres
en dissolution - et  utilisation des ultrasons - va  sfeparer en
£laments isoles  ces  particules biologiques .-  elle risque done
d'etre faussfie lors de la multiplication  de ces  microorganismes
 (fonction  des  conditions §cologiques  ...)

     En effet, les   celles bacteriennes  sont  rarement vehiculeeF
seules dans 1'air ma is  seraient absorbees (cf. re'centes journSes
ASPEC a 1'Institue Pasteur: biocontamination  en salles d'opera-
tion et milieu  hospitalier)    sur des  particules inertes>a
6,u  ou = a 2.u selon leur structure de paroi (c'est 3  dire
selon  leur reaction  de  Gram).   Quant  aux spores de  champignons,
selon les  cas, elles "voyagent" seules ou groupies  mais  il y a
des tr&s fortes fluctuations  journalidres (certaines spores
lib§rSes a 12  h  d'autres a 24  h etc ...) selon  que  1'air est
sec ou  humide, calme ou agitS, etc  ...

     Comme certains  de  ces organismes  sont pathogenes, que cer-
taines especes  fongiques sont  allergfenes  ...  on  ne peut guere
ne pas  en  tenir  compte  dans  une enquete  g§n€rale.


KUMMER  (Belgique)

1.   Etant donnS que nous mettons des  filtres sur lesquels ont
fite" faits  les  pr^ldvements en  solution,  nous ^cartons certains
problemes  tels que  la d^sagr^gation des  particules  corrane c'est
le cas  lors de 1'utilisation des ultra-sons.

 2.   Dans  cette  mesure  on tient compte de  la dimension globale
de  la particule  (meme s'il y  a des  particules qui restent
absorb^es  sur  celle-ci).  C'est d'ailleurs  cette dimension glo-
bale qui  est  importante lorsqu'on examine  les problSmes  de pene-
 tration des particules  dans  les voies respiratoires.

-------
                               1316
3.   Etant donnS que les prelevements se font a I'ext4rieur, les
quantity's de particules d'origine biologique sont certainement
negligeables a cote des particules d'origine minerals.


GOOTJES  (Pays-Bas)

1.   Quel est 1'interet pratique de la methode hors laboratoire?

2.   Comment definit-on la poussiere?  II est de peu d'int^ret
de parler des methodes si 1'on ne connait pas les substances qui
la composent.


KUMMER  (Belgique)

     L'interet pratique de cette §tude reside dans le fait qu'il
est certainement plus interessant de connaltre la distribution
granulomfitrique des particules en suspension dans 1'atmosphere
et par consequent de connaltre le nombre des particules sus-
ceptibles de p§n€trer dans les voies respiratoires, que d1avoir
une donn§e elementaire qui est le poids total des particules en
suspension dans 1'air.

     Pour ce qui est de la definition de la poussiere du point
de vue composition, cette etude est 3 mettre en relation avec
celle notamment effectuee par le Dr. Dams pour la composition
minerale des particules ou encore les mesures des composes
organiques propos£es par divers auteurs.

-------
                                1317
          ULTRASPURENANALYSE ORGANISCHER SCHADSTOFFE
             IN DER LUFT MITTELTS DER SOGENANNTEN
                     THERMGRADIENTTECHNIK

                J, ANGERER, A, HAAG, G, LEHNERT

Zentralinstitut  fur Arbeitsmedizin der Gesundheitsbehorde der
Freien und Hansestadt Hamburg, BRD
KURZFASSUNG

     Es vird eine Methode vorgestellt, die einen qualitativen
und quantitativen Nachueis organischer Luftverunreinigungen im
Ultraspurenbereich erlaubt. Ihr {jkotoxikologischer Wert wird
ei-nmal davin gesehen, dass Schadstoffe bereits in Konzentra-
tionen nachgewiesen uerden kdnnen, die isoliert betvaahtet
vielfaah wohl als atoxisch gelten diirfen, in ihrer Summe durch
Additions- oder Potenzierungseffekte abev umweltmedizinisah
relevant werden. Periodisch zuf Qualittitskontrolle der Atmosphare
eingesetzt ergeben sieh hier ausserdem Ansatzpunkte fur QkoprS.-
ventive Massnahmen, indem neu immittierte Schadstoffe friihseitig
erkannt und vechtzeitig entspreahende Gegenmassnahmen eingeleitet
werden konnen.
ABSTRACT

     A method  is presented  which  enables  organic  air pollution
in  the ultra-trace  range  to be  detected qualitatively and  quanti-
tatively.  One  aspect  of its value for  the study of  environmental
pollution  is that pollutants can  be  detected  in concentrations
which, taken in isolation,  may  Dell  often be  considered  atoxic
but uhiah,  in  combination,  have a bearing on  environmental health

-------
                              1318
as a result of their cumulative or potentiating effects. Another
is that, when used for periodic atmospheric quality cheeks, this
method provides information on which environmental protection
measures can be based, since it enables newly immitted pollutants
to be detected early and appropriate counter-measures to be taken
in time.

-------
                             1319
Im vergangenen Jahrzehnt 1st auf Teilgebieten der Okotoxi-
kologie ein groBer Wissensfundus erarbeitet worden. Andere
Sektoren, insbesondere die atmospharische Verunreinigung
mit organischen Schadstoffen blieben bis heute dagegen
weitgehend unerforscht. Wenngleich davon ausgegangen werden
darf, dafl jede einzelne dieser Substanzen fiir sich unseren
Lebensraum meist nur in subtoxischen Konzentrationen konta-
minieren, so gewinnen sie doch dadurch an umweltmedizini-
scher Bedeutung, daB sich nach neueren Erkenntnissen zahl-
reiche Agentien in ihrer biologischen Wirkung zu addieren
oder zu potenzieren vermogen. Einer unter diesen Aspekten
notwendig werdenden routinemafiigen Erfassung der auch in
Ultraspurenkonzentrationen vorhandenen Luftverunreinigungen
war bisher aber durch die Nachweisgrenzen der analytischen
Methoden eine deutliche Grenze gesetzt.

Art und Vielzahl der nachzuweisenden organischen Stoffe
lassen zur qualitativen wle quantitativen Identifizierung
der Komponenten einen Einsatz der Kombination Gaschromato-
graphie-Massenspektrometrie als Methode der Wahl erschei-
nen. Die zuvor genannten Griinde machen jedoch eine vor-
hergehende Anreicherung der Komponenten unijmganglich, auch
dann, wenn statt des Massenspektrometers relativ empfind-
lichere Detektoren verwendet werden. Zur Anreicherung der
Komponenten verwenden wir ein sogenanntes Thermogradient-
rohr, wie es  zuerst von Kaiser  [l] vorgeschlagen wurde.
Dabei werden 'ca. 2 1. der Luftprobe durch das tiber  eine
Doppelwandung geklihlte, mit einem Sorptionsmaterial ge-
fullte Thermogradientrohr gesaugt. AnschlieBend werden die
retinierten Komponenten unter Erwarmung des Thermogradient-
rohres mit dem Tragergasstrom auf die gekuhlte Trennsaule
eines Gaschromatographen uberfuhrt.  Unter Anwendung eines
linearen Temperaturprogramms wird der Ofenraum des Gas-
chromatographen von -150 °C auf +200 °C aufgeheizt. Die
gaschromatographisch getrennten Komponenten werden dann,

-------
                                  1320
                       Lufiprab*
           DrucUufl
Abb. 1.  Schematische Darstellung des apparativen Aufbaues bei der
         1 hermogradient- Gaschromatographie-Massenspektrometrie-
         Analyse von Luftproben.

                                    n
                                            -
                                                   -
 Abb  2.  Verwendung von 10*  Dexsil  GC  5OO  auf  Chroraosorb G/AW-DNCS,
          80 - 100 mesh als Sorptionsmittel im  Thermogradientrohr.
          1 Mcthanol; 2 Aceton;  3 Pentadien;  4  Hexan;  5 Benzol;
          6 Toluol; ? - 9  o,  m,  p-Xylol;  10 - 11 Triraethylbenzole.

-------
                             1321
wie von uns andernorts  [2] beschrieben, wahlweise mit einem
FID bzw. mit dem'BID eines Massenspektrometers nachgewiesen.
Die qualitative Identifizierung  der Komponenten erfolgt
durch die mit einem Galvano-m-Schreiber gleichzeitig aufge-
nommenen Massenspektren.  Den Aufbau der gesamten von uns
verwendeten Geratekombination veranschaulicht schematisch
Abb. 1.

Mit dieser Meflanordnung konnten  wir in  "normaler" Raumluft
aus der Vielzahl der tatsachlich vorhandenen Fraktionen
bisher etwa 30 organische Komponenten ahnlicher Konzen-
trationen abtrennen und identifizieren  (Abb. 2 und  Tab. 1).
Ahnlich giinstige Resultate wurden inzwischen mit einer im
Prinzip unserem Verfahren entsprechenden  Geratekombination
von einer anderen westdeutschen  Arbeitsgruppe  [3] mitge-
teilt.

Im Falle der Abb. 2 diente als Fiillung  des Thermogradient-
rohres Dexil 300 GC auf Chromosorb. Verwendet man statt-
dessen Porapak Q als Sorptionsmaterial,  so liefert  die
gleiche Raumluftprobe  ein vollig anderes  Chromatogramm.
Wie Abb. 3  zeigt, werden  mit diesem Sorptionsmaterial
wesentlich  weniger Komponenten retiniert. Andererseits
werden im unteren Bereich des Temperaturprogramms im Gegen-
satz zu Dexil Methan,  A'thylen und A than zuriickgehalten.
Nach Erprobung einer Reihe anderer Saulenfiillmaterialien
konnen wir  zusammenfassend sagen, dafi  die Probennahme mit
dem Thermogradientrohr recht spezifisch erfolgen kann,
wobei das Ausmafl der Spezifitat  von der Art des verwendeten
Sorptionsmaterials abhangt.  Hieraus ergibt sich, daO  in
keinem Fall alle in der Luft befindlichen Komponenten  in
einem Analysengang zu  erfassen sind.  Dies ist  im Hinblick
auf eine selektivere Detektion der Komponenten wegen  ihrer
auBerordentlichen Vielzahl durchaus wiinschenswert.  Fur  den
Analytiker  bedeutet diese Tatsache jedoch, daG  je nach
der erwunschten  Information  das  Sorptionsmaterial dem  je-
weiligen Analysenproblem in  optimaler Weise angepaflt

-------
                           1322
                    TABELLE I
Massenspektrometrisch identifizierte Komponenten
Methan
Methanol
Dichlormethan
Methylchlorid
Freon 22 (CF
Freon 12 (CF2C12)
A than
Athanol
Athylen
Aceton
TrichlorSthylen
Tetrachlorathylen
Fluor-1, 1-DichlorSthan
Propan
Butan
Buten
Butin
Isobutanol
Diathyiather
Pentan •
Penten
Pentadien
Rexan
Benzol
Trichlorbenzol
Heptan
Toluol
Ok tan
p-Xylol
m-Xylol
o-Xylol
Trimethylbenzole
MethylSthylbenzol
Chlorbenzol
Dichlorbenzol

-------
                                    1323
                          1  3
Abb.  1>.   Verwendung von Porapak  Q als Sorptionsmittel  im Thermo-
          gradientrohr. 1 Methan;  2 Athylen; J> Athan.
                                      -i r
                      Kuhlunq
                                        Hetttutt
                    •=11	
                      I HetBluf I
                                        4	 Sorptionsrkchtung

                                        —<> Elutlonsrlchlung

Abb. *f.   Scheraatische Darstellung der chromatographischen Vorgange
          auf  dem Thermogradientrohr bei der Probenanreicherung (1)
          sowie  bei der Probenelution ohne Umkehr der  Betriebs-
          richtung (2) und mit  Umkehr der Betriebsrichtung (3) im
          Gradient.

-------
                            1324
werden muB.

Hatten uns unsere Untersuchungen gezeigt,  daB die  Fullung
des Thermogradientrohres  den fur die  qualitative Analyse
wichtigsten Faktor  dieser Untersuchungsmethode darstellt,
so schien  es uns zumindest von theoretischem Interesse,
die Vorgange im Thermogradientrohr wahrend der Probennahme
und besenders wahrend der Probenaufgabe  zu untersuchen.
¥ie Abb. 4 verdeutlicht,  besteht das  Thermogradientrohr
aus zwei konzentrischen Rohren,  wobei das  innere Rohr
einer klassischen gepackten gaschromatographischen Trenn-
saule entspricht. Wahrend der Probenahme wird nun  der
auSere  Mantel  im Gegenstrom mit dem Dampf  von fliissigem
Stickstoff gekuhlt, bei der anschliefienden Probenaufgabe
wird  dieser Mantel  - ebenfalls im Gradienten - mit HeiB-
luft  auf  220 °C erwarmt.  Dabei werden die  retinierten
Substanzen mit einem TrSgergasstrom auf die geklihlte
gaschromatographische Saule eluiert.  Erfolgt die Elution
in der  Ansaugrichtung (Abb. 4 Mitte), so wird eine bereits
wahrend der Probennahme erreichte gaschromatographische
Vortrennung der Komponenten verstarkt. Eluiert man jedoch
entgegen der Ansaugrichtung (Abb. 3 unter) so wird diese
Vortrennung wieder  aufgehoben. Derartige Vorgange  auf der
Trennsaule konnten  wir dadurch dokumentieren, daB  wir das
Thermogradientrohr  wahrend der Elution direkt mit  den je-
weiligen Detektoren koppelten. Uber Einzelheiten dieser
Versuche haben wir  an anderer Stelle berichtet [4]. Zur
Verdeutlichung der  Vorgange auf dem Thermogradientrohr
 sei hier -lediglich  das Ergebnis einer Elution in Sorptions-
richtung (Abb. 5) und entgegen der Sorptionsrichtung
 (Abb. 6)  demonstriert. Man erkennt deutlich, daB die im
 ersteren Fall  eingetretene Vortrennung bei Umkehr  der
Elutionsrichtung wieder aufgehoben wird. Dieses Faktum
 ist insofern von Bedeutung, als die Vortrennung der Kompo-
 nennten auf der ThermogradientsSule prinzipiell eine noch
 spezifischere  Aufgabe ermQglichen sollte,  indem z. B. eine

-------
                                   1325
                   1M
                        180
                             120
                                   90
                                        •0
                                             30     0
                                              Z>lt[*«e]
Abb. 5-  Elutionschroraatogramm durch direkte Kopplung des Thermo-
         gradientrohres  an  den Detektor (FID). Die Elution erfolgt
         ohne Umkehr der Betriebsrichtung.  Stationare Phase: Dexsil
         GC JOO. 1  Pentadien;  2 Benzol; J> Toluol, Tetrachlorathylen;
         k Xylole;  5 Trimethylbenzole;  6 Wasser.
                      WO
12O
90
BO
30     0

  Z«lt [MC]
Abb.  6.   Elutionschromatogramm durch direkte Kopplung des Thermo-
          gradientrohres an den Detektor (FID). Die Elution erfolgt
          mit  Umkehr der Betriebsrichtung. Stationare Phase: Dexsil 500
          GC.

-------
                             1326
vorgetrennte Substanzgruppe durch Umleitung  des  Tragergas-
stromes allein auf die gaschromatographische Trennsaule  auf-
gebracht werden kbnnte. Auf diese Weise  lieBe sich z.  B.
das im EID-Chromatogramm  storende Wasser abtrennen.  Inwie-
weit sich die auf dem Thermogradientrohr festgestellte
Vortrennung auf das endgiiltige  Trennergebnis auswirkt,
kann an dieser Stelle noch nicht diskutiert  werden.  Dies-
beztigliche Uhtersuchungen werden von uns momentan durchge-
fUhrt.

Zusammenfassend und ausblickend ist damit festzustellen:
Bei der von uns verwendeten Technik erfolgt  die  Proben-
nahme mehr Oder minder  spezifisch.  Je grb'Ber die Spezifitat
der Probennahme und -aufgabe,  desto geriAger wird bei
gleichem apparativem Aufwand  jedoch der Informationsumfang.
Gleichzeitig wird aber  die gewiinschte Selektivitat des
Nachwe is systems erhSht.  In Zukxinft  wird allerdings die
Erhohung der Trennleistung durch Verwendung  von Trenn-
kapillaren bzv. die Anwendung von  selektiveren Detektoren,
wie der Massenspektroskopie und besonders der Massen-
fragmentographie, eine  unspezifischere Probennahme und
damit eine ErhShung des Informationsumfanges zulassen.
Dadurch wird einmal die Moglichkeit gegeben, eine Bestands-
aufnahme der oft  in Ultraspurenkonzentrationen vorliegenden
Umweltschadstoffe durchzufiihren.  Zum anderen erlaubt der
periodische Einsatz der Methode wegen der ihr eigenen
niedrigen Nachweisgrenzen die fruhzeitige Erkennung neu
in die Atmospha're gelangender Noxen. Sie bietet einen
echten Ansatzpunkt fUr  okopraventive MaBnahmen,indem beim
ersten Auftreten neuer  Schadstoffe  GegenmaBnahmen zu einem
Zeitpunkt  getroffen werden konnen,  zu dem mit einem noch
verhaltnismSBig geringen Aufwand  gravierendere Umweltge-
fahren durch die  spezifische  Noxe  verhindert werden konnen.

-------
                              1327
  LITERATUH

  1. Kaiser,  R.,  Europaische Tagung fur biochemische und  in-
    strumentelle Analysen. Seminar "Gaschromatographic",
    MUnchen,  25.-28.4.1972

  2. Angerer,  J., Haag, A., Szadkowski, D. und Lehnert, G.,
    (1973),  Z.Klin.Chem.Klin.Biochem., 133-134

  3. Bergert,  K.-H.,  Betz, V. und Pruggmayer, D.,  (1974),
    Chromatographia, 3f 115-121

  4. Angerer,  J.  und Haag, A., (1974), Z.Klin.Chem.Klin.
    Biochem., im Druck
                        DISKUSSION
HARKE  (Bundesrepublik Deutschland)

      Ich mochte Sie fragen, ob die vorgestellte Methode
geeignet ist

a) ouantitative Bestirtunungen durchzufvihren
b) Aerosole aus leichtf liichtigen Substanzen in Raumluft zu
   bestimmen.
ANGERER   (Bundesrepublik Deutschland)

a) Zu Ihrer ersten Frage mochte ich bemerken, dass es durchaus
   moglich ist, mit dieser Methode quantitative Bestimmungen
   durchzufuhren. Dies ist ein Problem der Eichung. Es gibt
   dafur prinzipiell folgende Moglichkeiten:

   - Bekannte Konzentrationen, moglicherweise Gemische inter-
     essierender  stoffe werden in gleicher Weise wie die zu
     bestimmenden Luftproben aufgearbeitet und analysiert.
     Dadurch lasst sich eine Eichkurve fiir diese Stoffe erstel-
     len;

   - Die zu analysierende Luftprobe wird vor Ihrer Anreicherung
     aus dem TG-Rohr in einem definierten Volumen gesammelt.
     Dazu gibt man eine bestimmte Menge eines gleichsam als
     innerer Standart dienenden anderen Stoffes. Aufgrund der
     spezifischen Empfindlichkeiten der  verwendeten Detektoren
     gegeniiber diesen Stoffen lassen sich nun die Konzentra-
     tionen der einzelnen Stoffe berechnen;

   - Die von R. Kaiser vorgeschlagene Moglichkeit einer "abso-
     luten  Eichung", die ebenfalls auf den spezifischen Detek-
     torempfindlichkeiten beruht, lasst ebenfalls eine Konzen-
     trationsbestimmung der einzelnen Stoffe zu.

-------
                              1328
b) Zu Ihrer zweiten Frage mochte  ich sagen, dass wir noch
   keine Erfahrung mit Aerosolen  sairaneln konnten.


PERRY   (United Kingdom)

      Wenn das von Ihnen beschriebene Verfahren dazu benutzt
wird, einen spezifischen organischen Stoff  in der Atmosphare
zu messen,wie wollen Sie das System eichen?


ANGERER  (Bundesrepublik Deutschland)

      Zu  Ihrer Frage ist festzustellen, dass dieses Verfahren
dazu dienen soil, nicht nur  eine  sondern gleichzeitig eine Viel-
zahl von  organischen Spurenkomponenten spezifisch in der At-
mosphare  zu messen. Was das Problem der Eichung betrifft,
darf ich  Sie auf meine diesbezuglichen Ausserungen Herrn
Dr. Harke gegeniiber verweisen.


PFANNHAUSER  (Austria)

      Hie haben  Sie versucht,  die mogliche  Storung durch Was-
ser auszuschalten?

ANGERER  (Bundesrepublik Deutschland)

      Bei der Verwendung eines Flammenionisationsdetektors
 stellt  das  aus dem TG-Rohr  kondensierte Wasser  kein  analy-
 tisches Problem  dar, da der FID  bekanntlich Wasser nicht an-
 zeigt.  Dies zeigt im ubrigen  auch Abb.  I.

      Allerdings muss  bei der  Verwendung   eines Wasserspektro-
meters  eine Trockenstrecke  vor das TG-Rohr  geschaltet  werden,
 da anderenfalls  das  fur die qualitative Analyse notwendige  EID-
Leitchromatogramm empfindlich  gestort werden wiirde.


 BRAHAM  (U.S.A.)


       Flussiger  Stickstoff  kann die leichten Kohlenwasserstof-
 fe quantitativ ausfrieren.  Die Methode bietet  ausgezeichnete
 Moglichkeiten.


 ANGERER  (Bundesrepublik  Deutschland)

       Zunachst darf ich mich fur Ihre freundliche  Bemerkung
 bedanken, da auch wir von den Moglichkeiten dieser Methode
 uberzeugt sind.  Ich darf Sie jedoch darauf hinweisen,  dass  nicht
 flussiger Stickstoff,  sondern dessen Dampf bei dieser  Analy-
 senmethode verwendet wird.  Ausserdem handelt es sich bei  der  An-
reicherung  der Substanzen aus dem TG-Rohr urn einen gaschroma-

-------
                             1329
tographischen Vorgang, nicht primar um ein Ausfrieren der Kom-
ponenten.

      Ich darf jedoch an dieser Stelle sagen, dasr wir augen-
blicklich an einer Methode arbeiten, bei der die Komponenten
tatsachlich mit flussigem Stickstoff ausgefroren werden. Diese
Methode wird eine Erganzung der geschilderten Methode darstel-
len.

-------
                              1331
         IMPROVED METHODS OF SAMPLING AND ANALYSIS OF
           AIRBORNE HAZARDOUS PARTICULATE POLLUTANTS

     R. C, LAO, R, S, THOMAS/ L, DUBOIS AND J, L, MONKMAN

Chemistry Division,  (Air) Technology Development Branch, Air
Pollution Control Directorate, Department of the Environment
Ottawa, Canada
ABSTRACT

     Studies based on classical air filtration sampling have
neglected to consider losses due  to the volatility of  the com-
pounds sampled.   A class of hazardous air pollutants  among
them PAEt arsenic oxides, selenium oxide  show vapor pressures
which raise serious doubts  about  the validity of air quality
data for this  class of  compounds.

-------
                                1332
     The collection, identification and measurement of airborne



particulate matter continues to be one of the primary goals of



environmental assessment surveys.   A great variety of inorganic



and organic pollutants, some posing a hazard to human health, are



collected, ordinarily, by air filtration and analyzed by a variety



of techniques.   Most attention has been focussed upon those



compounds whose published biological activity has suggested



carcinogenicity or toxicity at ambient air levels.  Notable among



these "hazardous" pollutants are the PAH and compounds of arsenic,



selenium and mercury.



     Classical methods for the determination of compounds of arsenic,



selenium and mercury are being upgraded with respect to sensitivity



and  accuracy by procedures which Include concentration and chelation



steps and highly specific ultraviolet photometry.   For the separa-



tion and determination of the various PAH, a number of analytical



schemes have been published.   Characteristic of  these analytical



procedures  are combinations of extraction, chrooatography,



spectroscopy using  ultraviolet absorption or optical fluorescence.  '



Most recently, the  technique which uses gas chromatography, mass



spectrometry and  automatic data  processing as the total instrument,



has been successfully applied  to PAH analysis.^'^



      While the analytical techniques for pollutant assessment have



 continually undergone evaluation and improvement, few changes or



 investigations have been made  with respect to the methodology of  the



 collection of the pollutants  from  the  ambient air.   Traditionally,



 airborne particulates have  been  sampled by drawing a known volume



 of contaminated  air through filters made of  glass fibre, cellulose



 or organic membranes.  The  particulate matter,  entrained on  the

-------
                              1333
filter, has then been assayed.   The fact that simple air filtration

sampling has been so universally applied as a standard procedure for

such a long period of time, without any attempts at re-evaluation,

has serious implications.   It must be realized that a number of

particulate pollutants, having high vapor pressures, will either

pass through the filter medium without being trapped or they may be

evaporated from the collected particulates with the continuing

passage of the air stream.

     A few studies Of the volatility of some PAH have been made in

the past and some vapor pressure data have been published.  However,

these studies were cursory in nature and the accuracy of the results

is suspect.   Recently the "equilibrium vapor pressures" of some

PAH were determined. *    The compounds investigated were benzo(a)-

anthracene (BaA), benzo(a)pyrene (BaP), benzo(e)pyrene (BeP),

benzo(k)fluoranthene (BkF), benzo(g»h,i)perylene (BghiP) and

coronene.   These PAH were chosen, by the International Association

for Research on Cancer, as the PAH which should be analyzed in any

program of monitoring the PAH content of air.   The compounds in
                                                 ft       I
question exhibit vapor pressures in the range 10~  to 10"4

atmospheres over a range of 70 to 150°C.    The equivalent vapor

pressures at the ambient temperatures can be estimated by the least

squares fitted equation,
                            (atm)
      The  concentrations  of  PAH,  as  found in  air, are usually report-

 ed  in micrograms  per  1000 cubic  metres.  The ranges of published

 concentrations  of PAH in ambient air  at 25°Cl are given below, with

-------
                                1334
the corresponding equilibrium vapor concentration (EVC) calculated



from the equilibrium vapor pressure (EVP):
compound
pyrene
BaA
BaP
BeP
BkF
BghiP
coronene
pg/1000 m3
1.3-35
0.1-22
0.2-39
1.0-26
1.0-20
2 -^6
0 -20
eve
74,000
1,050
85
85
16
1.6
0.1
 It is clear that, except for coronene and possibly BghiP, vapor



 losses on sampling may be appreciable.   There are two effects



 which can reduce losses below those represented by the EVCs of the



 pure substances.   Firstly, the EVCs may not be reached due to a



 slow rate of vaporization;  secondly, the pollutant may be adsorbed



 on an active particulate substrate and its EVC, in this state, may



 be significantly less than for the pure state.   The kinetic effect



 of slow vaporization is considered to be small and has its effect



 mainly on the  rate of loss of material already collected on the



 filters.   On  the other hand, even when the EVCs are reduced by



 adsorption effects, the EVCs of the pure substance give a relative



 measure of the volatility and corresponding inefficiency of collect-



 ion.   However, since adsorption is not immediately effective ,

-------
                               1335
vapor concentrations of PAH and compounds of arsenic, selenium and

mercury encountered during emission measurements are probably of

the order of the EVCs regardless of the presence of soot or other

particulates.

     Failure to collect a concentration equal to the vapor concentra-

tion and subsequent sublimation from the filter are sources of error

in high volume sampling.   A measurement, based on particulate

sampling, should be considered uncertain by an amount equal to the

EVC unless reduction of the EVC by the effects of adsorption can be

established.

     A series of experiments have been designed in this laboratory

to assess the influence of temperature, air flow rate, vapor press-

ure and adsorption effects upon the collection efficiency of the air

filtration sampling technique  for particulate contaminants which

have been found in air.   The  first system investigated was arsenic

oxide.   In  this experiment, the  filtration sampler  was connected

to a Greenburg-Smith impinger, containing sodium hydroxide solution,

in order to  collect gaseous Ae^Og which escaped from the filter

medium.   Details of the  apparatus procedures and results are  given

in a recent  paper.7    The equilibrium vapor pressure (EVP) of  As^Og

was calculated  from


                        log10P     6067

                        	 =  -  	  * 9,905
                           atm        T


The  date quoted or  calculated  from a generally  accepted reference

handbook  can generate serious errors7  due in part  to the  fact that

arsenic oxide does  not exist  in  the  gaseous state  as A^O  but as

-------
                                1336
         o 60
         §
         1
         o
         fe
                           4     6     8    10     12
                               AT °C
    Fig. 1.   Calculated filter collection efficiency from EVP

     Figure 1 illustrates the filter collection efficiency as
calculated from the EVP and represents the minimum theoretical
efficiency for the temperature differential, AT, between the tempera-
ture of a pure sample of A82(H and the filter.   The experimental
results in Figure 2, obtained by dividing arsenic trioxlde on the
filter by the sum of arsenic trioxide on filter plus arsenic trioxide
in the absorbing solution, indicate that for the same  AT, the EVC
for AcgjO^ Is lower than that calculated from theory.   Due to
variations in the uncontrolled ambient temperatures over the long
experimental period, the data points deviate from the normal straight

-------
                                   1337
                                468
                                    AT °C
10    12
      Fig. 2.  Experimental filter collection efficiency from
               EVP for As20 .
line relationship.  However, in spite of these deviations, the tem-
perature dependent relationship of collection efficiency holds.
Therefore, the air filtration sampling technique for particulate
arsenic compounds is unsatisfactory when the temperature differential
between filter and ambient air is small.  The same is true for other
airborne particulates which have an appreciable EVC.
     Work is under way to design a very precisely temperature controlled
apparatus, to assess the adsorption of pollutants or particulates and
to investigate the evaporation of such pollutants while trapped on fil-
ter media.  It is hoped that such investigations may lead to quanti-
tative improvements in the sampling of important pollutants, whose
fundamental chemical and physical properties ensure anomalies in
sampling.

-------
                                 1338
References:
1.   Hoffmann, D., Wynder, E.I., in Air Pollution. 2nd ed., Vol. 2,
     ppl87-247i Academic Press, New York, USA (1968).
2.   Sawicki, E., Analysis of Airborne Particulate Hydrocarbons.
     National Cancer Institute Monograph, No. 9, p. 201,
     Washington, D.C., USA (1962).
3.   Lao, R.C., Monkman, J.L., Pottle, R.F., "Mass Spectroscopic
     Identification and Measurement of Polycyclic Aromatic Hydro-
     carbons in Air Pollutants".  Proceedings of International
     Symposium on Identification and Measurement of Environmental
     Pollutants, pp 144-146.  National Research Council, Ottawa,
     Canada  (1971).
4.   Lao, R.C., Thomas, B.S., OJa, H., Dubois, L.  "Application of
     a QC-HS-computer system to the Analysis of the PAH Content of
     Airborne  Pollutants", Analytical Chemistry. 45,  908 (1973).
5.   Murray, J.L., Pottie, R.F., Pupp, C.,  Lao, E.G.  "The Vapor
     Pressure  and Enthalpies of Sublimation of Polycyclic Aromatic
     Hydrocarbons", Canadian__J_._ of Chemistry. 52, pp  557-563  (1974).
6.   Murray, J.L., Pottie, R.F., Pupp, C.,  Lao, P.C.  "Equilibrium
     Vapor Concentration  of Some PAH, AS 0  and SeO   and the
     Collection Efficiencies of these Air -Pollutants", International
     Journal of Atmospheric Environment, in press (1974).
 7.   Lao,  R.C.,  Thomas, R.S., Teichman, T., Dubois, L. "Efficiency
     of Collection of  Arsenic Oxide in High Volume Sampling",
     Science..._q_f_the_ Total ^yironment. 2, pp283 (1974).
 8.   Handbook of Chemistry and  Physics, Chemical Rubber Company,
     Clevelend,  U.S.A.  Annual  publication.

-------
                             1339
                         DISCUSSION

RONDIA (Belgium)

     The work of Commins and of our laboratory have shown al-
ready sometime ago the volatility of polycylic hydrocarbons with
relatively low boiling point.  This work has recently been re-
sumed and shows that the organic material present in air escapes
normal sampling using high volume samplers such as the Staplex.
A large number of substances with a boiling point below 35O C
are retained on the filter as a function of the air temperature
and the length of sampling.

     This could invalidate, for example, the results from the
national network of the U.S.A. as regards soluble benzene.
Similarly, in our countries the difference of BaP concentration
in air between winter and summer has been unanimously attributed
to domestic heating.  Present experience shows that other factors,
such as sampling method, are involved, and that industrial sour-
ces  (particularly in Liege where this study was carried out)
also undoubtedly play an important role.


LAO  (Canada)

     I agree with you.


BUSH  (U.S.A.)

     Did you have any difficulty separating BaP from BeP on
your gas-chromatographic columns?

LAO  (Canada)

     There  is no difficult  if only BeP  is separated from BaP.
However,  if a sample has more than 7O or 80 compounds, BeP will
not  separate  from BaP under normal gas-chromatographic conditions
 (e.g. use only  one  column).  In our experiment when BaP and BeP
are  in a same fraction and BaP is determining quantitativly by
fluorescence  spectroscopy the amount of BeP is then obtained by
subtraction.


RUMMER  (Belgium)  *•

     Would  not  the  addition of an  activated carbon  filter  enable
the  polycyclic  aromatic  hydrocarbons  to be  collected more  effec-
tively?

-------
                             1340
LAO (Canada)

     We have not tested this method.  Perhaps it will be dif-
ficult to recover all PAH entrained on an activated carbon fil-
ter.
SHERWOOD  (U.K.)

     The problem of evaporation of particles  from air sample
filter papers can also occur when sampling mineral oil mists.
Although vapour can be collected by a silica  gel trap placed
behind the filter, this does not distinguish  between ambient
vapour concentration and vapour released  from particles on the
filter.
LAO  (Canada)

      Yes,  if  the total  vapour  concentration  does  not  reach  the
"equivalent vapour concentration"  at  the  ambient  temperature,
particles  collected  on  the  filter  will  evaporate.

-------
                              1341
   SURFACE AREA AMD ADSORPTION PROPERTIES OP DUST PARTICLES

               W, KARCHER, P, GLAUDE, R, DEPAUS

Euratom, CCR Petten, the Netherlands
ABSTRACT

     The surface area and adsorption capacity for benzene and
naphtalene has been determined for two natural dust specimens
which were obtained by filtration or impaction in an urban
(Milan) and rural  (Ispra/Varese) area respectively.   To dis-
tinguish between the adsorption properties of the bulk samples
and the respirable size fraction, the same measurements have
been repeated for  the size fraction with a particle diameter
C 5 .urn.

     The results indicate only marginally hi'gher specific sur-
face areas for the small size fraction than for the bulk speci-
mens.   Also,, the  amount of  hydrocarbons which can be adsorbed
at normal pollution levels  (^lmg/m ) is so low «l,ug/g) that
pollutant removal  by a pure  adsorption process on dust or aero-
sol particles, as  it has been discussed repeatedly in the
literature, seems  rather unlikely.

-------
                          1342
1.    Introduction



      Repeatedly it has been suggested in the literature



as a result of adsorption measurements at very high



partial pressures of pollutants (> 1  g/m ), that certain



atmospheric pollutants may be removed from the atmosphere



by an adsorption process on airborne particulate matter (1,2)



Though a recent theoretical study has shovn that rather



high surface areas would be required for an appreciable



rate of removal by an adsorption process (3), no definite



conclusions can be drawn from the available literature



data, since no adsorption measurements had been performed



at concentrations corresponding -with natural pollution



levels. To fill this gap, some adsorption measurements



have been carried out with benzene and naphtalene in the



concentration range of 0.3 - 50 mg/m  (average atmospheric



concentrations 0.05 - 1 mg/m ) on dust particles which



had been precipitated from the atmosphere by impaction



or filtration. To this effect, the break-through curves



of benzene or naphtalene in a stream of carrier gas have



been measured at room temperature on two dust specimens



which had been obtained in an urban (Milan) and rural



(Ispra) area respectively. The surface area for the total



dust samples and for the respirable site fraction (< 5m)



has been measured with the conventional BET-method (U).






2.    Experiments



      For  the adsorption and surface  area measurements,



two dust  specimens  have been selected, which had been



obtained  by  filtration  or  impactibn in an urban area (Milan)

-------
                             1343
  and   rural area (ispra) respectively.  The  results  of  a
  qualitative chemical analysis of the  two specimens
  are  presented in table I. It is evident from these
  figures, that the urban dust sample contains much  more
  carbon, but considerably less inorganic material
    iOp, Al 0,, CaO) than the specimen which was  obtained
  from a rural site.
                       Table I

  Chemical Analysis of Dust Specimens


Constituent              Urban           Rural Sample
                             (weight percent)

    C                   > 1*0                 16,2
    H                     1,7                 2,25
    N                     1,9                 1,33
                         12,9                 32
   A1203                   3,8                 5,7
   Fe203                   5,6                 8,6
   CaO                     7                   1»»
   MgO                     0,8
   Zn                      1,3                 1,25
   TiO2                    0,U                 0,7
   Pb02                    0,9
   KiO                     0,3                 0,U
   (The trace metals  are  shown  arbitrarily as oxides)

-------
                              1344
          The adsorption experiments have been performed

    in  a fixed bed  arrangement  under flow conditions with

    a dust specimen of 0,2 - 0,3  g  at room temperature

    (after degassing at room temperature)* using calibrated

    gas mixtures air/benzene or helium/benzene at a flov rate

    of  2.65 ml/min. The hydrocarbon concentration in the

    exit stream has been measured by gas-chromatography with

    a flame ionisation detector.  From the resulting break-

    through curves  the break-through time has been derived

    graphically and the corresponding dynamic adsorption

    capacity calculated accordingly.



                        Table  II


    Surface Area of Rural and  Urban Dust Specimens
    Sample
 Rural  (orig.)
(orig. after
 extraction
 vith CH)
5,3
              degassed at
                          25°C
                              300°C
surface
area
(m2/g)
2,2
2,6
f. 5-3,0
3,9
3,1
micro
porosity
(cm3/g)


1.9.10"3


(m2/g)
10,8
6,3
11-12
11,9
8,8
(cm3/g)


9,*.10-3


U.6.10"3
11,8
1 .10
                                       -2

-------
                         1345
      The BET-surface area has been determined volumetrically


by argon adsorption at -196°C after the specimen had


been degassed at room temperature or 300°C under vacuum.


In some instances, the micropore size distribution


curves have been calculated from the complete argon


adsorption-desorption isotherms by means of the BJH-method (5)


The micropore volume measured in this way is indicated in


table II.




3.    Results


      The results of the surface area measurements are


given in table II. It can be seen that the surface area

                                                2
for both dust specimens varies between 2 and k m /g,


with the respirable size fractions with a particle


diameter < 5 ym showing only slightly higher surface


areas than the original dust samples. After degassing


at 300°C, 3-1* times higher surface areas are obtained,


indicating a release of some adsorbed or constituent


species under degassing at 300 C.


      As it is to be expected from these relatively


low specific surface areas, the evaluation of the


break-through curves indicates a very low dynamic


adsorption capacity of the dust specimens of benzene


concentrations which are comparable with atmospheric


conditions. For instance, at a benzene partial pressure


of  1 mg/ra  and under flow conditions  (13,5 cm/min)

                                          _2
an adsorption capacity of approximately  10   and


0.3 pg/g has been derived from the  break-through curves


for the urban dust  specimens for c/c  - 0.01 and

-------
                             1346
   0.8 respectively (C  and CQ denote the benzene  concen-



   tration at the in- and outlet of the dust bed).  Even



   at a dust level of 1 mg/m  no measurable concentration



   effect due to benzene adsorption on dust particles



   vould be evident under these circumstances,  since per



   unit volume of air less than 10~ % of the original



   partial pressure of benzene vould be adsorbed on airborne



   particles.



         The dynamic adsorption capacity of the urban  dust



   specimen for benzene vapour has been plotted in  fig.  1



   for C /C  ratios between 0.01 and 0.8 as a function of
        x  o


   benzene concentration in air. The adsorption isotherms
                                                   *0%Co
Figure 1.  Dynamic adsorption capacity of Milan dust ae a function

          of Benzene partial pressure.

-------
                              1347
obtained  in  this  way  resemble closely a Langmuir-type
adsorption isotherm so that the amount of benzene
adsorbed  (W  )  for any given benzene partial pressure
           €L                                •
(C) may be calculated with good accuracy by means of
the equation
                             K  K  C
                    w
                     ft
       The  break-through curves of naphtalene obtained
on the urban dust specimen in the concentration range
0.5  -  50 mg C1oHft/m  are comparable to those of benzene
However, at comparable partial pressures, naphtalene
is retained about 10 to 100 times more effectively
than benzene (see fig. 2). But even for  naphtalene,
less than  O.IJto of the concentration  in  the vapour
phase  would be in the adsorbed state  at  a dust  and
hvdrocarbon level of 1 mg/m  .
2.5,
                                                          10% C0
                                                         • *0% Co
                                                         . 20% Co

                                                         . 1% Co
                           100
                 ,0             20             30             »  """
Fig.  2.  Dynamic adsorption capacity of Milan dust in presence of
        naphthalene.

-------
                          1348
      From these results it must be concluded that at



normal atmospheric dust levels removal of gaseous



pollutants due to adsorption alone on airborne particles



seems highly unlikely. This estimate does not take into



account adsorption phenomena by soil and plant surfaces,



where much higher surface areas are available. Also,



chemical reactions on catalytically active dust



particles may very veil contribute significantly to



removal or transfer processes of specific gaseous



pollutants in air.
      Acknowledgments








      The dust samples and analytical data have been



obtained by courtesy of Messrs. Geiss, Versino and



Zamorani from Euratom OCR Ispra, Chemistry Division.

-------
                              1349
                          REFERENCES

1. LIBERTI, A.,
   J. Pure Appl. Chem., 24, 631  (1970).

2. DEVITO, F.G., RANKE, F., PETRONIO, B.M.
   Staub-Reinh. Luft, 32, 103  (1972).

3. JUDEIKIS, H.S., SIEGEL, S.,
   Atmosph. Environm., 7, 619  (1973).

4. BRUNAUER, S., EMMET, P.H., TELLER, E.,
   J. Am. Chem. Soc., 6O, 309  (1938).

5. BARRETT, E.P., JOYNER, L.G., HALENDA, P.P.,
   J. Am. Chem. soc., 73, 373  (1951).
                          DISCUSSION
HARRISON  (U.K.)

     I am referring to your chemical analysis data in Table I.
Various metals are listed as their oxides.  I would like to
know whether any evidence was obtained to the effect that these
metals were in fact present in air as their oxides.  The pres-
ence of Pb in urban particulates as PbO_  (O.9% of total weight)
would be particularly surprising.
KARCHER  (C.C.R. - Petten)

     The dust specimens have been provided and analysed by
Euratom, CCR Ispra.  As far as I know the inorganic components
have been analysed by means of neutron activation analysis,
polarography and X-ray diffraction.  The metals have been pre-
sented arbitarily as oxides and have not been analysed with
regard to chemical composition since in view of the large number
and low concentrations of trace metals present in the samples
the task of identifying the chemical nature of the numerous
constituents would have been very difficult indeed.

-------
                              1351
       AN IMPROVED TECHNIQUE FOR MONITORING PARTICULATES
       IN THE ATMOSPHERE TO CHECK AIR QUALITY STANDARDS

        J, E, EVENDIJK, A, A, ROOSKEN, N, NIEUWSTRATEN

Central Environmental Management Service Rijnmond, Schiedam,
The Netherlands
ABSTRACT

     Standards for allowable concentrations of toxic substances
in ambient air often appear to be incompatible with available
measuring methods and apparatus.   In the US there is a tendency
to combine air quality standards with an analytical method.   On
the other hand the Dutch proposed standard for SO  is coupled
with the particulate-concentration, which should be expressed in
units of standard smoke according to the OECD specification of
1964.   When an air quality standard does not indicate a measu-
ring method, the method should be adjusted analytically or tech-
nically to that standard.

     The council of Rotterdam declared an air quality standard
for particulates in 1970,  when south of Hook of Holland an iron-
ore terminal started  to operate.   This standard was clearly
inspired by  the US standard proposal and the American method of
using high volume samplers.   In  the given regulations was laid
down, that by exceeding the standard, activities of the iron-ore
terminal should be stopped.   This implied continuous monitoring.
Because there was no  commercially available apparatus that ful-
filled  the needed requirements, an adequate instrument was dev-
eloped, to meet the following requirements:

-------
                              1352
2.    The apparatus must be operating continuously.
2.    Contamination of the filter's by the sampler-motor must be
     avoided.
3.    Periods between servicing must be as long as possible.
4.    The instrument should have a constant flow devise.
S.    Inserting new filters in the filter holder must be done in
     the laboratory to prevent unnecessary damages of the fil-
     ter,
S.    Linear air velocity and amount of sucked air must be com-
     parable with that of commercial apparatus,

     A description of the instrument is given and some monitoring
results are discussed.

-------
                                 1353
 1.
      Introduction
      After  the  second world war the port of Rotterdam has expanded
 enormously.  Because  of its favourable geographical situation at the
 mouth of  the river Rhine  a large number of refineries and petrochem-
 ical  industries has  settled on sites west of the city of Rotterdam.
 The reclamation of about  24 square kilometers of land from the sea
 in front  of the original  coastal line offered new possibilities to
 industries,  dependant on  deep water for the supply of raw materials.
 This  area,  called  Maasvlakte,  can be reached by ships up to 240.000
 tons  (figure 1).  The possible establishment of steel works in 1970,
                                             combined with coal and
                                             ore handling facilities,
                                             made the problem of
                                             contamination by
                                             paririculates acute.  The
                                             fear for suspended
                                             particulate  matter was
                                             dictated by  the known
                                             synergistic  effects with
                                             sulphur dioxide (Amdur
                                             [ 1 ]) and the catalytic
                                             action of iron and other
                                             metals upon  the oxidat-
                                             ion of sulphur dioxide
 (Bracewell [2]). Because  of this  reason together with infrastruc-
 ;ural problems,  the  original  plan for the steelworks has been with-
 drawn. However, newcoming coal-  and ore handling companies were
 faced with an air  quality standard for particulate matter, that  had
 not to be exceeded in  neighbouring living areas.  This standard re-
 lates to measurements  with a  sampling time  of 24 hours over a period
 of one year  and contained the  following limits  [3,  4] '
     50 percentile   (median value)             80
Location of samplingsite and ore handling facilities.
     75 percentile                            120  ug/nr
     98 percentile                            240  ug/m  .
The standard has been inspired by the use of  sampling  with  commerci-
ally available high volume samplers, but neglected the limited  relia
bility and serviceability of these samplers.  In the first year  the

-------
                                 1354
monitoring has "been done continuously, because it was not clear
whether at random sampling provided a fair estimate of the 98 percen-
tile. Selected data of the  first year of monitoring show that the
estimated 98 percentile value was inaccurate. For this reason monito-
ring had  to "be continued on the same continuous base and the high
volume  samplers had to be adapted to this purpose.
2.   Design of an improved  High Volume Sampler
     The  main objections against the available high volume samplers
are:
     1. The flow rate measurement by means of a rotameter device is
        not accurate. The rotameter measures only a small proportion
        of the total flow through the sampler.
     2. The total amount of sampled air after a sampling period of
        24 hours is determined by a flow rate measurement at the be-
        ginning and the end of the sampling period. Usually a linear
        decrease in flow rate is assumed. However, the decrease in
        flow rate is related to the filter loading and - at least  in
        the Rijnmond area - a relative strong decrease in flow rate
        can be expected between 8 and 10 a.m.
     3. In the high volume  samplers known to us, collector motors  up
        to 20.000 r.p.m. are used. It appeared, that they are not
        suitable for continuous use because of their limited life
        time.  Fast rotation wears down the motor quickly. Collector
        wear  frees copper,  that is monitored eventually. The
        abrasion of  the  carbon-brushes originates carbon, that pos-
        sibly Influences  the filter  loading. The life of the carbon-
        brushes varied in extreme cases between one day up to one
        month.
      4. Due to the  fact that the  sampling site and laboratory were
         relatively at a distance,  the  filters were changed by people
         working in the vicinity of  the  site. Their initial
          enthusiasm slackened after some  time.
      5.  Filter changing and transportation  of  the  filters to the
          laboratory often caused damage  of filters.
      The disappointing experiences  with the available high volume
 samplers and the need for automation were  the  main reasons  for

-------
                         1355
                                                          byt>»t* with vpindl* vain
                                                            s-ervorootor
                                                          C*ntrifiiB»l pump
      Fig.  2.   Front-view of  the  high volume sampler
            housing
         tiller holder
          ball valve
         cam ml tube
 programmed camshaft


             orific*
         servomotor
            recorder
   pressure  regulator
       power  supply
Fig.  3.   Side-view  of the high  volume  sampler

-------
                                 1356
developing a new sampler with  the  following requirements:
     1. The apparatus must "be  operating continuously.
     2. Contamination of the filters "by the sampler-motor must be
        avoided.
     3- Periods "between servicing  must be as long as possible.
     4. The instrument should  have a constant flow device to keep
        the mass by unit of time of monitored air constant.
     5- Inserting new filters  in the filterholder must be done in
        the laboratory, just to prevent unnecessary damages of the
        filter.
     6. The specifications of  the  developed apparatus should be the
        same as for other high volume samplers, especially with res-
        pect to the lineair velocity of sampled air (jutze [5]i
        Sholtes (6]).
     7. Preference should be given to commercially available parts.
     Regarding the first three requirements, the choice of the pump
is essential. The system should meet the need of unattended running
for a year, sucking a sufficient and constant air flow and hardly
any emission of carbon particles,  copper or other metals from the
pump should take place. Motors of  the current collector type are
less suitable. A single or three-phase AC motor with a two stage
compressor can meet the requirements.
     A typical example is a centrifugal pump of 1,5 hp, 2600 r.p.m.
displacing 5 m/min when running unloaded and operating down to
0,8 ata. A disadvantage of this pump is, that the volume of the
transported air cannot be regulated by voltage control: variations
of more than 10 volts result in a  sharp decrease of the volume of
passed air. Therefore the air  flow is controlled by using a bypass
regulator  (figure 2, 5). A P.Y.C.  spindle valve, with 20 turns from
open to close, inner diameter  40 mm, is used as a bypass, because of
the availability and the easy  fitting qualities with the existing
tubing. The valve is regulated by  a  servomotor, making one rev/min.
The bypass only works effectively  in the last 6 turns of the
spindle-valve, so it takes less than 6 minutes before a maximum
disregulation has been adjusted completely. Two micro switches are
safe-guarding the valve by deactivating the servomotor at extreme
settings of the valve.

-------
                                 1357
     Measuring air flow rate through the filter seemed to "be optimal
by means of an orifice. Variations of the differential pressure over
this orifice should be translated into an electrical signal. One of
the possibilities consists of using a flowmeter parallel to the
                                                 »
orifice and catching the movements of the floater by two lightbeams
and fotocells. A disadvantage of this construction is that a vari-
ation in a small part of the total air stream is regulating the
bypass system. Furtheron fast flow changes disarrange the system.
However, a static pressure regulator, used in a differential pres-
sure sensing mode, offered a.  useful  alternative. It causes an
electrical signal, becoming larger the more the differential pres-
sure is drifted from a preset value. This signal controles the
servomotor. Deviations of less  than 5 f° from the equilibrium are
corrected, but the effect of the decreasing absolute pressure is
neglected, when the resistance  of the filter grows with increasing
loading. By neglecting these effects an error of maximum 10% is
introduced. In the actual monitoring situations, however, the error
is less than 5 %•
     The mentioned regulation of the flow is effective under chosen
conditions up to 300 ug/m  particulate matter as a daily mean value.
This means a filter loading of  150 mg. Underestimation of the
particulate concentration occurs, when more than 300 ug/m  as a
daily mean value is present. With this pressure regulation system
the second requirement has been met. Moreover the differential pres-
sure over the orifice is recorded.
     Inserting new filters should be kept to a minimum; that means
that the sampling site is visited once a week.
     Figure 2 shows that 8 filterholders are connected via 8 ball-
valves to a central tube. The ball-valves are opened and closed by
a programmed camshaft, controlling micro switches. The switching
procedure takes about  30 seconds. To prevent overload of the pump,
one valve is gradually closed and at the same time the next valve
is slowly opened.
     The construction  of the filter holder  is shown in figure 4« To
obtain a favourable effect on the homogeneous distribution of
particulates on the filter surface, the shape of  the filter is
circular. Glass fiber  filter is preferred because of its easy hand-

-------
                                  1358
       • cassette
                                             DEza
       b adapter
 c face plate

 g gasket

 d support plate



e fine-mesh screen
e coarse.rnesh screen
g gasket

f connectkM plate

g gasket
                                                   k funnel
                                                   It mounting plate
          Fig. *t.  Assembled and disassembled  filter-holder.
     properties. The maximum diameter of circular glass  fiber filters
obtainable commercially,  is about 15 cms. Due to possible damaging
circumstances, it is preferable that the filter is  stored in a cas-
sette.  In this way it is  possible that only the laboratory personnel
handles the filter itself.  This implies that the filter  holder con-
sists  of a cassette (a) and an adapter (b) which are easily connected
and disconnected (figure  4). The cassette is built  from  a face plate
(c), a support plate  (d)  containing stainless-steel screens as a
support for the filter  (e) and a connection plate  (f) that is con-
nected with the mounting plate (h) at th& funnel  (k). Gaskets (g)

-------
                               1359
number of data
average
50 percentile
75 percentile
98 percentile
stand, deviation
                            Table I
                      SUMMARY OP ALL DATA
                     APRIL 1971/MAECH 1972   APRIL 1972/MARCH 1973
         285
          95
          77
         1U
         270
          69
298
 73
 65
 91
201
 56
                             Table II
            AVERAGE CONCENTRATION PER DIRECTION OF WIND
WIND DIRECTION

      N
      NO
      0
      ZO
      Z
      ZW
      W
      NW
APRIL 1971/MARCH 1972   APRIL 1972/MARCH 1973
          54
          86
         104
         100
          85
         122
          71
          34
 42
 63
111
108
 91
 97
 47
 41

-------
                                 1360

                            Table III

              AVERAGE CONCENTRATION VERSUS WINDSPEED

 CLASS OF
WINDSPEED            APRIL  1971/MARCH 1972   APRIL 1972/MARCH 1973
 (M/SEC)
  0-2                       123                     106
  2-3                        98                      73
  3-4                        82                      69
  4-5                        82                      64
  5-6                        77                      66
  6-7                        80                      52
  7-8                        89                      61
  8-9                       173                      73
  9 -10                                                83
                                                      166
       The measurements with the described high volume sampler
 started on January 197^•  From this time a sharp decrease in the
 copper content of the particulate matter was observed.

            PERIOD                           Mean copper content
       January - Harch     1973.                      0,52.
       April   - June      1973.                      0,56.
       July    - September 1973.                      0,65.
       October - December  1973.                      0,37.
       January - March     197^.                      0,03.
       This is an indication that copper contamination from the
 motor does not take place.

-------
                                  1361
secure a complete tightness. The cassette is made of P.V.C., the
funnel, which is commercially available, is made of vulcathene. The
dimensions of the ball-valves and filter holders limit the amount of
air drawn through the filter to about 25 nT/hr. *° reduce time con-
suming calculation in the laboratory, the apparatus is adjusted to
a flow of 500 ar/day.
     On determining the linear velocity of air at the entry
clearance between the edge of the roof and the main housing, the
advice of the U.S. Intersociety Committee of 1,13 m /min of air
sampled through a clearance of 645 cm  has "been followed (Sholtes
[6]). In this apparatus 0,35 m /min is sampled through a clearance
         2
of 200 cm . The entry surface has been positioned horizontally to
avoid catching particles generally larger than 100 \i (jutze [5]).
The whole apparatus has a housing of aluminium, to prevent corrosion
at the coastal site.
3.   Results
     Some data of two years sampling with conventional high volume
samplers are tabulated. Table I shows that for more than 75 % of the
days data are available. The overall average of the particulate con-
centration is nearly 25 % lower in the second year.
     During the first year the standard is exceeded because more
than 2 % of the data are higher than 240 ng/m . The lower mean level
in the second year is partly explained by a different frequency dis-
tribution of the winddirection and the lower level of particulates
at western and southwestern winddirection (table II). The remarkable
influence of windspeed on the particulate concentration is shown in
table IH.Both at low windspeed (0 - 2 m/sec) and at high windspeed
(more than 8 m/sec) a maximum level is reached.

-------
                            1362
                        References

AMDUR, M.O., CORN, M. ,  "Irritant Potency of  Zinc Ammonium
Sulfate of Different  Particle  Sizes", Amer.  Ind. Hyg. Jlssoc
    326 (1963).
BRACEWELL, J.M., GALL, D. ,  "The  catalytic oxidation of  sulfur
dioxide in solution at concentrations occurring  in  fog
droplets", Proceedings Symposium on the Physico-Chemical  Trans-
formation of Sulfur Compounds  in the Atmosphere  and the
formation of Acid  Smogs,  O.E.C.D., Mainz, Germany June  196?.

Collection 1970, number 465, H.B. nr. 48/1   ,  "Immission
standard for suspended particulate matter of the 7.O.M. Letter
of the Mayor to the Community-coi
December 1970", (text in  Dutch).
of the Mayor to the Community-council of Rotterdam, 4   of
 Collection 1971. nr.  61,  V.G.  nr.  3156,  O.W. nr.  27959-
 "Answer to questions  of mrs. F.  van Eeezik  and L.M.  Bliek
 about  the  Ertsoverslagbedrijf  Euxopoort  C.V." (text  in Dutch).

 JUTZE, 6.A.,  FOSTER,  K.E.,  "Recommanded  standard  method  for
 atmospheric sampling of fine particulate matter by  filter
 media  - high volume sampler",  J. Air Poll. Ass.. 17. 17-25 (196?).

 SHOLTES, R.S. e.a., "Tentative method of analysis for suspended
 particulate matter in the atmosphere (high  volume method)",
 Health Laboratory  Science,  J_,  279-286 (1970).

-------
                               1363
        ANALYTICAL STUDIES ON THE SPECIATION 9F AMBIENT
                   LEVELS OF NERCURY IN AIR

                  R, S, BRAMAN, D, L, JOHNSON
Department of Chemistry/ University of South Florida, Tampa,
Florida, USA
ABSTRACT

     An analysis method has been developed'for the determination
of several forms of mercury in air at concentrations doan to 0.1
to 1 nanogram per cubic meter.   Particulate mercury, elemental
mercuryf methyImercury(II) compounds, volatile inorganic mercuric
compounds and dime thyImercury are collected from air samples onto
previously blanked specific absorption tubes.   These are ana-
lyzed by thermal desorption of the mercury compounds and detec-
tion of the mercury by emission spectroscope in an electrical
discharge in helium carrier gas.   The analysis system has a
limit of detection near 0.03 ng m~  for elemental mercury, di-
me thy Imercury and particulate mercury.   The specific absorption
tubes for methyImercury(II) compounds and volatile inorganic
                                                         _ 7
mercuric compounds have limits of detection near 0.3 ng m
The analysis of ambient air for the several forms of mercury has
been carried out over  a sampling grid in the Tampa, Florida area.
Diurnal variations, short  term fluctuations, mercury forms in
homes and buildings, and mercury evolution from lakes have also
been studied.

     Current results indicate that elemental mercury is the ma-
jor form of mercury in air both in buildings and outside.   Sig-
nificant amounts of mercuric chloride-type compounds are also
found.   High values for mercuric chloride have occasionally

-------
                              1364
been obtained.   Mercury concentrations inside places of human
habitation are far greater than outside.   Inorganic mercury is
rapidly converted to methyl-mercury compounds and elemental mer-
cury in the soil.   Mercury in air above the Sargasso Sea was
found to be generally lower in total mercury -than over land areas
Because of the biological transformations, studies on the types
of volatile mercury compounds present in air are needed.

-------
                                    1365
1.  Introduction
     Mercury is perhaps the most widely distributed heavy metal pollutant
due largely to the volatility of elemental mercury.  Methylmercury halides,
dimethylmercury and mercuric chloride are also known to be volatile.
     Inorganic mercury has been  shown  to be  converted»into the more vola-
tile methyl-mercury forms through  biological activity by  Imura et al  [1]
and by others.
     Because  of the lack of  suitable analytical methods,  until now no in-
formation has been available on  the types of mercury compounds and their
concentrations in the environment.  Particulate mercury in air can be de-
termined by activation analysis  of filter pads and elemental mercury alone
                                                                  — O
by direct atomic absorption  if present above approximately 15 ng m~ .  The
combination of emission spectroscopy with the use  of sequential specific
absorption  tubes has provided the  needed method for specific detection and
analysis of the volatile mercury forms, and  particulate,  anywhere in the
environment.
2.   Sequential Specific Absorption Tube Sampling  Svseem
     A sampling system was designed for the  selective absorption of the
several suspected environmental  forms  of mercury.  Amalgamation was first
studied.  Silvered glass beads and gold coated glass beads were found to
rapidly absorb elemental mercury.  Dimethylmercury was quantitatively ab-
sorbed only on the gold beads thus providing a basis for  separation of ele-
mental mercury from dimethylmercury.
     After  considerable subsequent work sample tube packings were found
which would specifically absorb  methylmercury chloride and mercuric chlo-
ride.  A tube containing silicone  coated (3% SE-30) on 45/60 mesh Chromo-
sorb-W, HC1,  vapor treated,  was  found  to retain mercuric  chloride while
passing methylmercury(II) chloride, elemental mercury and dimethyl-mercury.
Methylmercury(II) bromide or iodide or acetate or  other volatile forms are
likely converted to methylmercury(II)  chloride on  the HCl-treated Chromo-
sorb-U tube.
     A non-siliconized Chromosorb-W column initially treated with NaOH to
0.15 milliequivalents per gram of  column packing was found to quantitative-
ly absorb methylmercury(II)  chloride while quantitatively passing elemental
mercury and dimethylmercury.  Methylmercury(II) chloride  is likely convert-
ed to the less volatile hydroxide  by the alkaline  column.

-------
                                    1366
     A  glass wool  filter, Gellman Instrument Co.  Type A, was  used as
a particulate filter to complete  the stack of  absorption tubes.  By
assembling the absorption tubes in the following order and  pumping
air  through them it was possible  to separate mercury from a single
air  sample into five forms.   Sample volumes from 0.025 to 0.2OO m
were taken.
     Glass wool filter preheated to blank   removes particulate  only
      Siliconized Chromosorb-W, HC1 vapor    removes HgC^ type compounds
         treated                             and particulate which passes
                                            the glass wool filters
      NaOH treated Chromosorb-W              removes CH3HgCl type compounds
      Silvered glass beads                   removes elemental mercury
      Gold coated glass beads                removes (CH3>2Hg
      The efficiency of the individual tubes and the assembled stack was
 tested using elemental mercury, HgCl,, CH^HgCl and (CH-j^Hg in operation
 under a variety of temperatures and conditions of humidity.  Methylmercuty
 (II) chloride was 83 - 100Z retained on the NaOH-Chromosorb-W tube at tem-
 peratures as high as 33°C and 85Z relative humidity.  Mercuric chloride was
 90 - 100Z retained on the siliconized Chromosorb-W column up to 33°C and
 85Z relative humidity.  Both compounds were tested for one and two-hour
 pumping periods.  Dimethylmercury and elemental mercury were 100Z absorbed
 on the gold and silver coated beads respectively under all conditions
 tested.
      Careful blanking of all sample tubes was necessary to achieve good
 lower Units of detection for analysis.
 3.   Analysis System,
      After sampling,  the stack  of tubes were separated and analyzed Indi-
 vidually for trapped mercury.   Mercury was driven off from each tube by
 heating and passed  through  a quartz discharge chamber by means of He
 carrier gas.  A direct current  discharge in the chamber was used to excite
 the 253.65 nm Hg emission  line. A conventional type recording spectrometer
 system described previously was used  [2].  The gold and silver sections
 were heated directly into  the discharge and de-amalgamated to give sharp
 mercury peaks when heated  to 300 - 400°C.  Chromosorb-W tubes were first
 heated into blanked gold coated bead  transfer tubes to eliminate some
 volatalixed interferences  from air and to sharpen mercury peak signals.

-------
                                    1367
The scanning capability of the system permitted verification of the pres-
ence of mercury.  Calibration was accomplished by using mercury vapor.
Lower limits of detection were found to be near  01 nanograms.
3.  . Limitations
     Since no other current methods exist for the detection of  specific
mercury compounds in the 1-20 ng/m3 range it was not possible to directly
compare this method to another one.  The identity of raethylmercury(II)
compounds or mercuric chloride compounds sampled from the environment de-
pends upon comparison to known compounds (i.e. HgC^, CH-jHgCl,  and
(CHj^Hg).  Compounds sampled from the environment did contain mercury as
determined by scanning the mercury emission line during analyses.
     Methylmercury(II) type compounds and dimethylmercury far above normal
concentrations were detected in air above soil treated with mercuric chlo-
ride.
     H2S and S02 were found to overload the Chromosorb-W (HC1)  column if
2-hour pumping times were employed but were not otherwise an interference.
     For the most accurate work, very careful handling of the absorption
tubes is necessary to avoid contamination.
     Mercury in laboratory air is a definite contamination problem.
4.   Applications
     The method has been used to study mercury in air in houses, aboard a
ship, in school buildings and in a sampling grid in the Tampa,  Florida, USA,
area.
     Tables I and II give data indicating some of our finding.   Air over
the Sargasso Sea (from shipboard analyses) contained only small amounts of
elemental mercury.  The larger fraction we believe was from aerosolization
of sea water and is likely HgCl2 type compounds.  The total mercury in air
found at sea compares reasonably well with the amount of mercury in the sea
water evaporated into air.  Sea water is on the order of 5 - 10 ppb mercury,
                                                                       3
or 5 - 10 ng/liter.  If evaporated, one liter of sea water occupies 1 m .
     Houses and buildings are obviously the greatest source of air mercury
exposure to the general population.  Amounts of mercury in buildings ranged
from 10 to 100 times higher than mercury in ambient out-of-door air.

-------
                      TABLE I  Mercury in Air in Building! and Aboard Ship (R/V Trident)

                                                                    3C
         Sample

Lecture Hall*
(Before claaa)

Same
Total only

Lecture Hall*
(After claea)

Same
Total only

House
   Bathroom
   Bedroom
   Kitchen
   Living room

Ship (totals only)
   Laboratory
   Engine Room
   Paint Locker

Faculty Office (Several)
   (total only)
   a                                         3
    Out of doors total ambient Kg, 3-6 ng/m

    Out of doors total ambient Hg, average 7.5 ng/m

   Uncertainties in data are ±0.3 ng/m  or ± 5%
———---—----------------———--—--- ng/m — '•• 	 • 	
Hg particulate HgCl2 MHC HE* DMM
2.9 4.2 12 310 0
-
0 12 25 500 12
.

43 4,4 200 0
0 10 37 0
12 0 88 0
2.7 23 110 0



Total
367
351
549
557
U)
247 £
47
100
136
21
51
62
110 - 140

-------
                                       TABLE II  Mercury in Ambient Air
            Sample

Tampa Area (same day)
   Suburban Location

   Bayside

   Urban Downtown

Trapped air over soil
   treated with HgCl2
   (I day after)

Trapped air over untreated soil
   (1 day after)

Sargasso Sea
   (underway)

Sargasso Sea
   (on station)

Sargasso Seaa

Sargasso Sea
   (underway - near waterline)

   (underway - top of deck)

   (on station)

Hg particulate
2
3
12
-

0.1
3.1

HgCl2
22
10
0
20
-
10
0
	 ng/m — -
MMC
14
54
3
20
-
0.2
0.7

Sal
4
6
4
400
-
0.9
1.1

DMM
0
3
1
60
-
0
0

Total
42
76
20
500
3-6
11
4.7
                         0.89

                   3
7.5, 12.0, 8.1 ng/m   Totals only

 4.0, 11, 2.4 ng/m3  Totals only

   4.0, 0.6 ng/m   Totals only
                                                           VO
   aSouth of Bermuda

-------
                                1370
    Acknowledgements

    This work was supported by the National Science Foundation Grant

No.  Gl-3479AX, Research Applied to National Needs Program.

    The following individuals assisted in the development  of the tech-

nique:  A. D. Shumaker, J. L. Bricker, M. Ammons and C. C.  Foreback.

    Literature

     [1]  N. Imura, E. S. S. Pan, K. N. J. Kim and T. K. T. Ukita, Science,

    172. 1248-1249 (1971).

     [2]  R. S. Braman and A. Dynako, Anal. Chem., 40, 95-106 (1968).



                          DISCUSSION
HARRISON  (U.K.)

     During  air  sampling, other materials  may be trapped in the
sampling  tubes.   Do these materials cause  any matrix effects  in
the mercury  discharge chamber analysis?


BRAMAN  (U.S.A.)

     Hydrocarbons and other organic compounds can produce a
background emission (as well as the predictable band emission
systems)  but they become a problem only  if present in exception-
ally high concentrations - such as direct  analysis of engine
exhausts.  Even  then, most organic materials are driven off of
the tubes prior  to detection of mercury.   No problems of this
type arose in our environmental sampling.


BYRNE  (Yugoslavia)

a)   The  late Dr. Williston, in his pioneering measurements of
mercury in air in California, observed a peak at sunrise which
he attributed to commencement of photosynthesis.  Have you
observed  a similar effect?

b)   Have you investigated, or do you have any information about
the nature of "particulate mercury."


BRAMAN  (U.S.A.)

a)   During  diurnal variation studies we have observed a delayed
peak concentration several hours after sunrise.

b)   We have no  information on the chemical composition of our
J£f;J  ?  *te  me^curv ^action.  It is not any of the volatile
chemical  forms of mercury.

-------
                              1371
     NEW TECHNIQUES FOR THE MEASUREMENT OF MR POLLUTANTS

                       JAMES R, MCNESBY
National Bureau of Standards, Washington DC, USA
ABSTRACT

     Measurement research in air pollution at the National Bu-
reau of Standards includes that applicable to ambient air, stack
emissions and automotive emissions.   Some of the measurement
techniques developed for source emission application have been
refined and made sufficiently sensitive to measure pollutant
concentrations as low as background levels.   The fluorescence
monitor for sulfur dioxide developed by Dr, Okabe and his col-
leagues has been applied commercially to the problem of stack
monitoring.   More recently the sensitivity of the technique has
been improved Very substantially so that ambient air sulfur di-
oxide concentrations can be measured at least down to 1 part in
  n
10  .   This apparatus will be described and its capabilities de-
lineated.   The technique of laser magnetic resonance for the
           4
measurement of NO at concentrations typical of stack emission
and auto exhaust has similarly been extended to include ambient
concentrations.   In addition,, progress has been made in  the
direction of simultaneous measurement of NO- by the same  tech-
nique.   The most recent data obtained by  the laser magnetic re-
sonance technique will be presented.   A third measurement tech-
nique developed at the National Bureau of  Standards* that of
dual angle, forward laser scattering for the real time measure-
ment of particle size distribution has now been shown to  be cap-
able of revealing morphological and chemical information.   The
development of pyrolysis as an alternative to permeation  tubes
and standard gas cylinders will be described.   The principle of

-------
                                1372
the pyrolytic technique is that a thermally stable molecule which
may be stored indefinitely is subjected to thermal decomposition
producing a non-storable molecular species plus an exactly equal
number of storable, readily measured molecular species.   The
pyrolytic technique has the inherent capability of tying together
virtually all gaseous pollutant molecules in a singlet self-
consistent measurement and calibration scheme.

     Finallyf an absolute coulometrie technique for the determi-
nation of parts per million concentrations of NO in nitrogen will
be described.   The technique has been used to generate confir-
matory evidence for establishing the integrity of NO in nitrogen
Standard Reference Materials.

-------
                              1373
1.   Introduction
     Air pollution measurement science is no exception to
the axiom that however good things are, they can be improved.
A corollary is that however bad things are .they can be made
worse.  One job we have set out to do at the National Bureau
of Standards is to improve our ability to measure ambient
air pollutants and to exert every effort to avoid inventing
a new, inferior measurement method.  As is true with most
research programs, we are not always successful and we have
sometimes succeeded in inventing a square wheel.  My intent
here is to discuss only those aspects of our program we
consider to have had a successful outcome.  Because the
National Bureau of Standards houses many physical measurement
specialists, it was natural for us to favor the
application of the more sophisticated methods of modern physics
and physical chemistry in our methods-development program.
Our attention has been confined for the most part to
those air pollutants designated as such by the United States
Environmental Protection Agency — SO , NO  , CO, 0 , hydro-
carbons and particulates.  I will be describing only the
NBS efforts.
2.   Fluorescence

     Until the advent of flame photometry  sulfur dioxide
was measured by wet chemistry.  Flame photometry was highly
sensitive and fast, but was unable to distinguish SO  from H  S
and other sulfur  containing compounds.   Its specificity problem
was solved by incorporating gas chromatography.  Unfortunately,
this entailed sacrificing speed of response in the process.

-------
                              1374
     The sulfur dioxide molecule had been under  study  in
the NBS laboratory of Dr. Okabe with the objective  of
measuring the strength of 0-S bond, when it was  rediscovered
that there is a very strong fluorescence excited by  radiation
in the region of 214 nm.  In the course of the study of this
fluorescence the radiative lifetime of the excited  state  that
produces the fluorescence was found to be a few  nanoseconds.
The significance of this .finding is that an excited  S0?
molecule has about one chance in ten of radiating its  energy
in the form of light before it suffers even a single
debilitating collison with an air molecule.  It was  confirmed
experimentally, that 10-15 percent of the excited states  of
SO. radiate light (fluoresce) at atmospheric pressure.  The
basis for the design of a radically new instrument  for
measuring sulfur dioxide had been discovered.  Once  the
basic discovery was made, it was possible to design  an
extremely simple apparatus of a design shown in  figure 1.
                      CU FILTER
    Zn LAMP
                                     MONEL
                                   HONEYCOMB
                                   PMflJBE
                                 HZ!
                                  DC POWER SUPPLY
S02
PERMEATION TUBE
                                                  -_«__ CLEAN
                                                       AIR
                     AMPLIFIER fj
                             LJ
                           RECORDER OR  SCALER
Figure 1.  Schematic Diagram of Sulfur Dioxide Fluorescence
           Apparatus.

-------
                              1375
A chlorine  filter  transmits  90 percent  of  the Zn (214 nm) line
and  the Cd  (229  nm)  line  while removing the emission lines in
the  light source from  270 to 390  nm.   The  color filter
transmits the  S0?  fluorescence and  rejects the 214 nm and 229 nm
lines of the light sources Zn and Cd,  respectively.
The  performance  of this apparatus on  samples of air  containing
sulfur dioxide at  concentrations  determined by means of
certified NBS  sulfur dioxide permeation tubes is shown in
Figures 2 and  3  and  shows a  linear  signal  from 0.2 to 200 ppm
with the Zn lamp and to considerably  higher concentrations
using the Cd lamp.   Recent results  indicate that this linearity
is maintained  at  least  down to 10  parts  per billion.
Figure 2.
Figure 3.
            SO,CONCENTRATION, PPM
Response of Sulfur Dioxide Fluorescence Monitor
at Low Concentrations.
             1*0-
            r:
                              ZriLomp
                  T
                     400
              r
 I
MO
I
I
I
1*00
I
 I
ton
                       I
                      IZOO
                  SOf, PPM
Response of Sulfur Dioxide Fluorescence Monitor
at High Concentrations.

-------
                              1376
     An extension of the fluorescence technique using  laser
excitation is being explored.  A potassium pentaborate  crystal
is being used to double a tunable dye laser down  to  217 nm.
Early results  already show an ability to measure  S02  by  laser
induced fluorescence at 1 ppb essentially instantaneously.   Remote
sensing of stack emissions is one objective of  this  research.
However, the main purpose of the research is  to measure ambient
S09 instantaneously at 1 ppb.
     It has been found that the 214 nm Zn line also  excites
                                              24-       2
the fluorescence of NO in the gamma system, A I 	^- X IT.
The radiative lifetime of the A2I+ state of NO is  2.2  x 10~
seconds and is readily quenched by oxygen but not  by nitrogen.
Therefore, NO in N_ can be measured by fluorescence  down  to
a. few ppm using a 25-second integration time.  The number of
counts has been found to be proportional to the concentration
of NO in nitrogen from one ppm to a few hundred parts  per
million.  The method is currently being explored  as  a  means  to
monitor routinely concentrations of NO in N2  in the  NBS
automotive exhaust emission Standard Reference Material.

3.   Zeeman Spectroscopy
      Two  closely related techniques have been used at  the
National  Bureau of Standards to measure paramagnetic molecules.
Laser Magnetic Resonance-, first used in Dr. Evenson's  laboratory
to  study  rotational Zeeman spectra of 0_ has  now  been  extended
to  the free  radicals NO, N02, OH, CH, H02, and  HCO.  The  method
is  based  on  the property of  a paramagnetic molecule  that  its
rotational energy  levels are split into  (2J + 1)  sublevels  when
a  magnetic field is applied.  A particular wavelength  of  a
water vapor  laser  (78.4,  79.1 or  118.6 urn) is incident on a
cell containing  the paramagnetic  molecule which has  a
rotational  transition  sufficiently close  to  the  laser  wave-
length that  absorption can  be made to occur by  application
of a magnetic field  to the  cell.   In  the  case of  the OH
radical,  the J =  5/2-^—J  =  3/2  rotational  transition is in
near coincidence with  the  laser  radiation.   Four
 transitions  in which  AM -  0,  are  the  allowed  transitions

-------
                               1377
in which the laser  electric field is parallel to the applied
magnetic field.  Abosprtion is measured by ac modulation  of
the magnetic field  and  phase sensitive detection of the
                                                            g
laser signal.  The  technique is capable, of measuring 2 x  10
                  3
molecules of OH/cra  .   The kinetics of reactions of OH with
CO and with NO and  NO.  have been studied using this technique.
Zeeman-Vibrational  Spectroscopy has been employed by Maki,
Kaldor, and Olson in  a  similar way to the measurement of  NO.
                    2
In this case the X  II3/2  state of NO has a near coincidence in
                                                      -1
its J = 5/2-«-3/2,  V  -  1-^—0 transition at 1884.32 cm" " with a
CO laser line   (v = 9-^-8)P(13)   at 1884.37 cm"1.  As shown
in figure 4, very modest  magnetic fields are able to
             (O
             Z
             UJ
                  M-f-f
                   M
                                            (a)
                                            (b)
4  H(kGauss)

 ^calculated
    AM*0
   transitions
                                         (d)
   calculated
    AM = *|
   transitions
 Figure  4.   Response of Zeeman Vibrational  Spectroscopy
            Apparatus to Nitric Oxide.

-------
                              1378
bring the vibrational transitions of NO into  coincidence with
the CO laser transition.  In a collaborative  effort with
Professor A. Cabana, Sherbrooke University, the v, band of  the
N0_ spectrum is being measured in high resolution and  is
being analyzed at NBS.  It  is expected that this will  result
eventually in an apparatus  capable of simultaneous measurement
of NO and NO. by Zeeman Spectroscopy.

4.  Stark Spectroscopy
     A Stark cell has been  built in Deslattes'  laboratory
and will shortly address  itself to the question of whether
advantage can be taken of the Stark  effect, modulation of  the
electric field and  phase  sensitive detection  of absorption
by such pollutant molecules as formaldehyde,  hydrogen
sulfide, ozone, etc.

5.  Piezoelectric Sensors

     The change in  the fundamental frequency  of a piezoelectric
crystal is  proportional to  the increase of  mass per  unit  area
when a pollutant is deposited on the  face  of  the  crystal.
Advantage has been  taken  by Scheide  of  this property  of
piezoelectric crystals by evaporating  a thin  film of  gold
onto the crystal and measuring the increase in  mass  resulting
from the interaction of mercury with  gold.  The detection  limit
     -12
is  10    grams and  concentrations  of  mercury  in air  are
readily measured at one part in  10

6.  Coulometric Measurement of NO
     In order to obtain confirmatory evidence for the
composition of NO in N_ Standard Reference Materials in the
parts per million range,  a  method independent of gravimetry  is
required.  Coulometry represents such an independent method
which has been developed  by Marinenko.  The chemical scheme
employed in the analysis  is as follows:

-------
                              1379
     1.  2NO  +  5H      Catalyst^
                   2
         2N02 +  7H2     Catalyst^
                                        *
     2.  NH3  +   H20  • - *- (NH4OH)£T* NH* +  OH~

The electrochemical measurement scheme is based on the anodic
reaction

        3.  2H20   - *" °2  +  4H+   +  4e'

The final stoichometric reaction is

     4.  H+   +   OH~  - »- H0
i.e. neutralization of ammonium hydroxide with elec trogenerated
sulfuric acid.
     All samples in the 100% to 0.1% NO concentration range were
analyzed by injection of a known volume of gas (25 cc) into the
H_ stream and titration of the absorbed ammonia.  Below 0.1%
NO, gas samples were continuously injected at a known rate and
the flux balance method of analysis was employed.  The results
of analysis of known mixtures are summarized graphically in
Figure 5.
     The data obtained for NO-N2 mixtures, covering  six orders
of magnitude  concentration range, confirm absence of  any
systematic measurement error  (bias) .   Work  is proceeding
in cooperation with the gravimetric preparation  group in
an effort  to  resolve  some observed  descrepancies .
7.  Particle  Size  by  Light Scattering
     An  instrument has been  designed,  built,  calibrated and
tested  in  Gravatt's laboratory which  sizes particulates by  the
intensity  ratio  technique.   The basic  concept  involves  the
simultaneous  measurement  of  the intensity of light  scattered
by  a single  particle  at two  small scattering angles.  The
ratio  of  the  two intensities  is a direct measure of  the size
and is  fairly independent of the  index of  refraction of  the
particle.   The  optical system consists of  an argon  ion  laser

-------
   •t
  I
  •••

  ••'
   I

   n
« w
3 Ml
O Ml
e H-
3 n
  H-
3! (»
H- B
X O
r» X
C
H O
(t) Ml
00
  90
O ID
MI n
  o
z <
H- m
n H.
H ••<
H-
r> o
  M»
O
D. rt
n  H
   H-
H- O
D
   O
a «
H- H-
rr a.
H  n
o
ao  ^
(B  rt
3  O
•  a
  no


  105



100


   95


   90
                                             O
0.0001
0001
                                                  0.01
                                      O
                                      o
                                      o
                                                                                             o
                                                                          0




-------
                              1381
source and an annular iris-fiber optic scattered light
detector.  The maximum count rate of  the system is  20,000
particles/sec and the diameter range  covered is 0.2 ym  to  4 ym.
For the scattering angles of 10°and 5° this technique can  size
particles in the 0.2 ym to 4 ym range with an error no  greater
than ±15% for all possible materials  in air.  In addition,
optimum angular ratios and apertures  have been determined
for a wide range of sizes.
     The conversion of the analog ratio light scattering
instrument to a digital instrument with computer data
acquisition and analysis has been completed.
     A very simple method for distinguishing between optically
absorbing particles, e.g., carbon and metals, and non-absorbing
particles was found from theoretical  studies.  This involves a
measurement of the absolute scattered intensity over the
angular range 30° to 70°,
     A preliminary theoretical study  of the depolarized light
scattering by long rods has been performed.  It appears that
a measurement of the ratio I (depolarized)/I polarized would
serve as a way of detecting asbestos  particles.
     Instrumentation for field use has been developed by
Greenough for the Environmental Protection Agency to determine
the size of particles in the 0.1 to 5 micrometer diameter
range.

8.   Calibration Standards by Thermal  Decomposition
     The aim of this project is to investigate the possibility
of  generating dilute, known quantities of pollutant gases
through the thermal decomposition of  appropriate "parent"
compounds.  Such an in situ procedure provides a simple
solution to the difficult problem of  preserving the integrity
of  dilute samples of pollutant gases  prepared in the usual
static manner.  Furthermore, through  the proper selection
of  the parent molecule, such a procedure may be easily calibrated
That is, for the reaction A (parent)-^B (pollutant) + C, where
C is a more stable and easily calibrated species, then a
determination of the concentration of C will fix the value for
B.   Finally, by controlling the concentration of the "parent"

-------
                              1382
molecule and/or the extent of decomposition, any concentration
of pollutant can be readily produced.  Overall, the present
procedure calls for a minimal amount of experimental skill
and is thus ideal for incorporation into test instruments and
routine field usage.
     The experimental procedure is to use as the parent
compound a molecule which decomposes into the desired components
via a molecular mechanism.  It is particularly  important
that there is no possibility of free radical induced
decomposition since such systems may not be reproducible.
The  selected compound is then pyrolyzed in a gold tube
reactor and the requirements are that only  the  expected
products be produced and in the proper stoichiometric
proportions .
     A  demonstration of  the feasibility of  this method as a
means of generating formaldehyde, acetaldehyde  and acrolein  has
been completed  in Wing Tsang's laboratory.  The relevant
reactions  are
     H2OC-CH2CH2OH    - »- isobutene +  HCHO

     H2C-CH-CH2CH(OH) CH3  - »- propylene +  CH
 and  H2          /  °  - ^-ethylene + H-C-CHCHO  (acrolein)
            2— H2
 The key points to  be  noted  are  the  equality  (on  a  mole  basis)
 of the olefins and aldephydes  formed,  the  great  variation
 available in the concentration  of pollutant  In the gas  delivered
 to the detector and the wide range  of  temperatures possible.
 It is  expected that the generation  of  organic  acids,  alcohols,
 ketones and hydrogen  halides will proceed  with equal  facility.
      Two substances that appear to  behave  in the required
 manner for generation of S02 and NO are shown in the
 following reactions.

      C1U   -  SO,     - ^-   cyclopropane  +  SO.
                I                    I
            -  CH               propylene

-------
                               1383
and
     CH  CH-CH CH-	»-2CH,CHO  +  2NO
         II
         0  0
         NO NO
     It should be noted that the trimethylene sulfur dioxide
system has characteristics which make it ideal for present
applications.  For example, at 0°C it is a solid with
a vapor pressure of the order of 0.03 Pa (0.2 microns).  Thus,
with a carrier gas at 1 atm and with complete reactions  (>800°7.)
one obtains S02 at the .3 ppm level.  At 70°C it is a liquid
and its vapor pressure is such that with 1 atm of the carrier
gas one obtains ~300 ppm samples.
     I have given you a few examples of what we at NBS regard
as new techniques in measuring air pollutants.  Obviously
these examples represent only a small part of the whole  of
new-techniques research.  Nevertheless, they represent our
best efforts and we believe they are not insignificant in
the total picture.
                           DISCUSSION
CERQUIGLINI  (Italv)
      Calibration  of  monitoring  systems  in automatic  networks
with  transmission of data to  a  central  data collection  oper-
ative unit,  made  by  current dynamic  methods,  is  quite expensive.
      Do you  think that  in the future the  thermal decomposition
method of  calibration could be  adopted  to this use?

McNESBY  (U.S.A.)
      Yes.  This  is exactly the point  of  developing the thermal
reactor.   The same reactor can  be  used  to generate known  concen-
trations of  a number of pollutants.

-------
                                1385
       SPECIFIC HYDROCARBON MONITORING AT'LOW CONCENTRATIONS

                    R, PERRY AND J, D, TWIBELL

Public Health Engineering, Imperial College, London, United Kingdom
  ABSTRACT

       The extensive use of hydrocarbons and the wide variation in
  their physiological effects necessitates a more detailed study
  of the hydrocarbon composition of the atmosphere.   Total hydro-
  carbon measurements3 the method used extensively at present,
  give no indication of the levels of individual potentially dan-
  gerous compounds present in the urban atmosphere.

       One of the major problems involved in studying the medical
  effects of individual hydrocarbon air pollutants is that of sam-
  pling f at street air concentrations^ in such a way as to render
  subsequent separation and analysis possible.   Jfj for instance,
  adsorption tubes containing standard chromatographic materials
  are used at sub-ambient temperaturesf difficulties arise with
  water condensation.   In addition the amounts of volatile mat-
  erials collected are limited by the relatively low retention
  volumes involved.   These problems of sampling and analysis can
  be overcome by use of the "Timed Elution Procedure".   Here the
  hydrocarbons are sampled quantitatively at ambient temperatures
  by adsorption onto Chromosorb 102.   The sample tubes are then
  heated to 130°C and a proportion of the collected material is
  eluted via a sampling valve onto the main chromatographic column
  for analysis.   For a given sampling tube, the proportion of
  each species eluted in this timed-interval is a linear function
  of the total amount collected allowing both quantitative analysis
  andt if necessary, GC/MS identification of the species present.

-------
                             1386
     This technique allows, for example, street air benzene
levels to be readily assessed.   It is also important in that
it is within the Cfi ~ C7,. hydrocarbon range that petrol and
diesel exhaust emissions differ most markedly.   These hydro-
carbons can be readily sampled and separated by this method and
their relative contributions to the street air hydrocarbon levels
may be measured.   Very low levels of alkylbenzenes, indans and
tetralinSj which contribute to exhaust odour can also be esti-
mated.

-------
                                   1387
    In the  U.K., where the tendency towards photochemical smog formation 15
not very great,  the  main problem caused by vehicle exhaust emissions is that
of odour nuisance  with its possible health hazard.   For this reason con-
siderable effort has been directed towards identifying the  individual com-
pounds present in  exhaust emissions, with particular emphasis being given
to diesel exhausts.
    No single component  is responsible for exhaust odour and complex
synergisms undoubtedly occur.    One group of exhaust components, the un-
saturated hydrocarbons,  arising from the original fuel or formed as pyro-
lysis products in the combustion process,  are  undoubtedly  responsible for a
considerable part of the odour.   Other  possible  odourants include oxy-
genates, and compounds of sulphur and nitrogen.   The  former  occur at levels
at  least an order of magnitude lower than the  aromatics  (A.D.  Little,L1J)
but may possibly contribute to exhaust odour  to an  equivalent  extent.   The
low molecular weight aldehydes such as formaldehyde,  acrolein and croto-
naldehyde, once suspected of being prominent  odourants,  are no longer con-
sidered to be important  (Vogh. [2.1).   At  the  odour threshold of diesel ex-
haust in air, (a dilution factor of about two  hundred) levels of these
species are typically only a tenth of their  odour thresholds  (Table 1) (Vogh
 [2].).
      TABLE 1.          Odour  thresholds of  possible exhaust odourents
                    (after Leonardos, Kendall and Barnard  8 , Gerarde  3 )
  Chemical                                           Odour threshold  ppm
  Acetaldehyde                                                 0*21
  Acetone                                                     100.00
  Acrolein                                                     °*21
  Benzene                                                      4'68
  Cyclohexane                                                 300.00
  p-Cresol                                                     0-001
  Dimethylsulphide                                            0.001
  Formaldehyde                                                  i-00
  Phenol
  Styrene
  Toluene                                                       2.1^
  o-Xylene                                                     °*1?
  p-Xylene                                                     °'1+7

-------
                                    1388
    Organic sulphur compounds have been suggested as possible exhaust
odourants due to their very low odour threshold levels in air (approx 0.001
ppm Table 1.).   Even if the total exhaust hydrocarbon fraction consisted
of entirely unchanged fuel, then individual sulphur species would only just
exceed their threshold odour levels in the diluted exhaust gas.   However
this is far from being the case in that the most reactive species present
in the fuels and presumably the odourous sulphur compounds fall into this
category, are readily oxidised or otherwise modified.   Further it has been
 suggested   that with modern fuels of low sulphur content, the final com-
bustion product sulphur dioxide contributes little to the exhaust odour
 (Vogh[2].).   Nitrogen dioxide is thought to be the only nitrogen compound
 that contributes strongly to the exhaust odour of a diesel engine, but only
 in the idling mode (Vogh[2l.).
    The  spectrum of hydrocarbons  emitted from a source can be related to the
 type of  odour produced.   Thus  the major differences in exhaust odour note
 between  petrol, jet aircraft and  diesel emissions broadly parallel the fuel
 composition,  which in turn determines the reactive hydrocarbon and oxygenate
  emissions  (Table 2.).   For petrol exhausts the main aromatic species are


                                  TABLE; 2

               The Relative Contributions of Various Structure
                 Types to the Total Exhaust Aromatic Fraction
                                               Emission  Source
    Structure Type                     Petrol   Aircraft (JP*0   Diesel
    Cg - C9 Benzenoid*                    69*
    o-Xylene + remaining Benzenoid       29*
    Indans.                              1-2           5.           9-
    Naphthalenes                         0.7           5.           9.


   •Benzene, toluene, m, p-xylenes, ethylbenzene, ethyltoluenee.
    Engines run at fast idle, under no-load conditions.

-------
                                    1389
simple alkylbenzenes, which with the exception of o-xylene are sweet in
odour.   Other materials of increased complexity arising from increased
ring substitution or chain branching cause the odour to become more acrid
(Gerarde[3].)•   These are more prevalent in jet aircraft and diesel ex-
hausts.   In addition indans and other mixed cyclic species also have
kerosine-like odours.   Thus where procedures are available for the coll-
ection of hydrocarbons from street air, the contribution of aromatic
species to the 'oily-kerosine'  odour note can readily be assessed.
     The most abundant hydrocarbon species found in street air, apart from
the simple alkanes, are the early members of the aromatic series.   Toluene
is present to the greatest extent in the street air of the U.K. and North
America (Perry and Twibell [V|., Pilar and Graydon [5J.), where it occurs
at levels slightly over twice those of benzene.   Total levels of C0H ^
                                                                   o 10
isomers roughly equal the toluene level.   The situation is apparently
different in Switzerland where benzene appears to be the most abundant
(Grob & Grob I 6].).  This may reflect differences in petrol composition,
petrol exhaust being the major contributor of these species.
     The basic C^-C   hydrocarbon composition of the urban atmosphere re-
flects the composition and densities of the traffic in the street, and
could be used as an indication of the odour nuisance and potential health
hazards.
Sample Collection.   Procedure for Hydrocarbon Monitoring
     Modern porous polymeric adsorbants have large retention volumes for
volatile organic species at ambient temperature.   Water is not readily
retained and the sample, following collection, may be stored for several
weeks.
     Desorption of this sample prior to analysis can be a problem. Although
solvent extraction may be used for this purpose it results in considerable
sample dilution in addition to the loss of the most volatile materials.
     Furthermore, if gas chromatography linked to mass spectrometry is to
be used for identification, the limiting source pressure precludes the
identification of species eluted with, or close to, the main solvent peak.
A Timed Elution procedure was therefore devised to overcome these problems
(Perry and Twibell Ik} , [7] .).
     The sample is collected in short lengths of 6mm O.D. stainless steel
tubes packed with 0.3g of Chromosorb 102.   The tubes are returned to the

-------
                                    1390
laboratory and connected into the gas sampling valve system depicted in
Figure 1, which is then  heated to 130 C.   At this temperature equilibrium
distributions are established for each species between the adsorbant and
the gas present in the void spaces.   The relative amount of a specific
hydrocarbon in the vapour phase, at a given temperature, thus depends upon
the volatility of the species as well as upon the space available.   This
in turn is determined by the tightness of the adsorbant packing and the
volume of any associated pipework.   A timed valve actuator switches the
gas chromatograph carrier gas supply through the tubes for a five second
period, completely eluting the vapour phase material onto the chromato-
graphy column.   This ensures that the organic materials are in a suffi-
ciently small plug to be resolved by the column.   Broad classification of
the eluted species was  carried out using a Hewlett-Packard 5930A mass spec-
trometer  fitted with a  gas chromatographic inlet and linked to the H.P.
5932A data system.
                            Carrier Qaa f ram
                           ' Flow Controller
                                     doted Loop for
                                     Normal Q£. Operation
                                     end Stand By
                                         Off
                                            Tub* A
                                       Tube to to Sampled
                              Tub* to be Sampled
                                Subsequently
    Figure 1.   Timed Elution Gas Sampling Valve Assembly,  Schematic.

-------
          o-s-i
          0-7
  Quantity
par. 5 aacond
 akitton ug
          0-6
       KEY
O    Aircraft Exhaust
X     DtoMl Exhautt
•     Patrol Exhauat
                                                                                               Ethylbafizatw +
                                                                                               m ,p-Xrl«n»i
                                                                                                             90        100
                                                                                                               Tuba contanU |ig
                                                                                                                                            120
                             Figure  2.    Typical Elution Gradients  of Various Hydrocarbons.

-------
                                   1392
                     Hydrocarbon Levels in Urban Air
Kerbside street air analyses at typical sites
TABLE: 3
Typical Street Air Hydrocarbon Levels
Pinstone St.
Sheffield
HYDROCARBONS
Benzene
Toluene
Ethylbenzene)
)
•(pXylenes )
o-Xylene
p-Ethyltoluene
Trine thylbenzenes
n-Decane
nsm'3
270
550

400

184
270
112
40
ppb
85
146

92

43
55
23
7
Exhibition Rd.
London
W50~3
576
1210

850

290
280
150
60
ppb
179
322

192

67
57
31
10
are given
in Tab
Mobile Sampling
M4 Motorway
-3
1100
2500
400

2700
1060
920
840
1060
ppb
344
665
92

624
245
188
171
182
                          (Notes to Table 3)
      Pinstone Street  is a major shopping street in Sheffield, heavily
 contained by large buildings.   At the time of sampling traffic flows
 were 131O vehicles per hour, 19£ of which were diesel fueled.   There
 was a fresh blustery  wind blowing  along the street direction.
      Exhibition Road  is a wide dual carriageway with tall buildings
 either side.   At the time  of sampling (in the central reservation)
 there was no wind, hot dry  weather prevailing.   Traffic flows were
 1922 vehicles per hour, with a 22$ diesel composition.
      Mobile eanpling  on the M4 Motorway was performed in a platoon
 travelling at 45-5O aph,  the sampling vehicle maintaining the recom-
 mended braking distance.    There was a slight cross wind.   Sample
 inlet was positioned  adjacent to the interior ventilation intake on
 the vehicle bonnet.
Comparison of the individual aromatic hydrocarbon levels  obtained at
the Sheffield site, with  their odour threshold values,  (Table 2)
indicates  that the xylenes  were present at  about a quarter  of their

-------
CO
     50
     30
     20
     15
     10
                                                       8     9    10    11    12    13
14    15
Tim* mini
               Figure 3*   Fluctuations  in  Kerbside Carbon Monoxide Levels During
                           Hydrocarbon Sampling at Pinetone Street, Sheffield.

-------
                                1394
                              TABLE 4
                  BASIHGSTOKE INCINERATOR EMISSIONS
POLLUTANT



fl^Wm3
TOLUENE .
START-UP PROCEDURE
Ignition
'1/11/73
190
820
warm-up
21/11/73
350
550
230
490
NORM4L OPERATION
Primary air onlj
20/11/73
300
450
280
280
Primary +
22/1^73
330
700
SHUTDOWN

29/11/73
340
70
260
100
odour threshold levels.  Olfactory response is extremely rapid and
ideally fluctuations should be monitored on a much shorter scale.
This at present is not possible when monitoring individual hydrocarbons
at these levels.  However some indication of probable fluctuation can
be obtained from the carbon monoxide levels which were monitored
simultaneously (Fig.3).  Here peak values of up to four times the mean
were found at the Sheffield site.  Assuming good correlation between
carbon monoxide and the hydrocarbons it is apparent that the levels
of aromatic species peak over their threshold levels.  Fluctuations
in kerbside levels are undoubtedly caused by turbulent mixing processes
diluting the exhaust, rather than by diffusion hence the lower diffusion
rates of aromatic species relative to carbon monoxide are of minor
importance.  Similatly the mean levels found in Exhibition Rd., London,
were higher and thus it is to be expected that odour thresholds were
exceeded for a greater proportion of the sampling time.
                     Municipal refuse incinerator emissions
     The levels of aromatic hydrocarbons emitted by a modern municipal

-------
                                  1395
waste incinerator were examined.  The only species found to any
appreciable extent were benzene and toluene, other aromatics being at
very low levels.  Sampling was performed in the stack base after
electrostatic dust precipation and water scrubbing.  The levels found
(Table **) were no greater than those found in a heavily trafficked
street.  There thus appears to be no reason for undue concern about
benzene emissions from incinerators of this design.
     It should be noted that although the benzene level was almost
independent of incinerator mode, toluene varied greatly.  We believe
that increased adsorption of toluene by particulates must occur as the
system cools, the toluene being re-emitted during warm-up.  Secondary
air may cause premature flame quenching.
                    Hydrocarbon levels inside vehicles
     The levels of aromatic hydrocarbons in the passenger air-intake
of a car were examined.  The sample was taken at the bonnet air-intake
of a car forming part of a traffic platoon travelling along the M4*
motorway.  It is apparent from the results (Table 3) that the levels
of some aromatic species are permanently in excess of their odour
thresholds, although olfactory fatigue will obviously occur.

                    Possibility of Health Hazards
     Of the aromatic species investigated in street air only^ benzene
has been linked with definite health hazards.  It is a suspected
carcinogen and is known to accumulate in bone marrow.  However levels
of benzene occurring in street air are about two order of magnitude
lower  than those recommended as maximum allowable concentrations over
an 8 hour period, in industrial establishments.  Even the highest
level  measured inside the vehicle moving in a traffic stream was only
3.5S6 of the accepted industrial level.  Exposure to benzene as an air
pollutant from traffic does not at present appear to be a significant
health hazard, although nothing is known of the effects of long term
exposure to these levels.   In addition the  presence of other exhaust
derived materials may exacerbate the toxic  effects.

-------
                                1396
References:


1.   ARTHUR D. LITTLE INC.   Chemical  Identification of the Odor
     Components in Diesel Engine Exhausts.   Final Report (Year 3) to
     Coordinating Research Council and Environmental Protection Agency.
     June 1971.

2.   VOGH, J.W.   Nature of Odor Components in Diesel Exhaust.
     J.  Air Pollution Control Association 19, 773-

3.   GERARDE, H.W.   The Toxicology and Biochemistry of Aromatic
     Hydrocarbons. Elsevier Monograph,  Elsevier Publishing Co. I960.

1+.   PERRY, R. & TWIBELL, J.D.   The Collection and Analysis of Volatile
     Hydrocarbon Air Pollutants Using  a Timed ELution Technique Linked to
     a. Computer Controlled Mass Spectrometer.   Biomedical Mass
     Spec trome try 1, 73
5.   PILAR, S. and GRAYDON, W.F.   Benzene and Toluene Distribution in
     Toronto Atmosphere,  Environmental  Science and Technology 7,  628
     (1973).

6.   GROB, K. and GROB,  G.   Gas-Liquid Chromatographic-Mass Spectrometric
     Investigation of C.-  - C_n Organic  Compounds in an Urban Atmosphere.
     J.  Chromatography  62, I (1971).

7.   PERRY, R. and TWIBELL, J.D.   A Time Based ELution Technique for the
     Estimation of Specific Hydrocarbons in Air.   Atmosphere Environment
     7,  927 (1973).

8.   LEONARDOS, G. , KENDALL, D. and BARNARD, N.   Odor Threshold
     Determination of 53 Odorant Chemicals, J. Air Pollution Coritrol
     Association 19, 91  (1969).
                             DISCUSSION
                                                    o
MERIAN (Switzerland)
      I'm interested  in the environmental pollution by hydrocarbons
and  its relation to  ecotoxicology (see  Chimia  28  (May 1974)  and
paper, of 2 July 1974  at the Environmental Medicine and Biology
Congress, Paris) .  In your ..Table 3 you  mention concentrations
of O,3 to 1,1  mg Benzene/m , of O,6 to  2,5 mg  Toluene/m   and of
O,6  to 3,8 mg  Xylenes/Ethylbenzene/m  in Sheffield and London.

      A.P. Altshuller,  W.A. Lonneman and K. Grab found only about
O,O4 mg Benzene/m ,  O,O6 to O,15 mg Toluene/m   and 0,02  to O,15
mg Xylenes/m   in Los  Angeles Azusa, New York Manhattan,  New York
Lincoln Tunnel and in Zurich.   G. Guiochon found  even much lower
concentrations in Paris.

      Do you have an  explanation why you found  about 10 times higher
concentrations in the UK.

-------
                             1397
TWIBELL  (U.K.)

     I would like to emphasise that we were looking for extreme
cases and the street air samples were taken at the kerbside at
a height of only 5 feet.  We were primarily interested in the
effects of aromatic hydrocarbons on the pedestrians in the street,
and therefore we sampled at nose level.  Other authors tend to
be rather vague about their sampling site details, but Grob and
Grob (J. Chromatogr. 62 1-13  (1971)) state that they found 54
ppb. benzene and 39 ppb toluene at a site in their university
campus BOO metres from the nearest road.  I therefore believe
that our highest kerbside levels of 179 ppb benzene and 322 ppb.
toluene are not at all out of agreement.

     The motorway samples are higher as the air entering a
vehicle obviously contains a poorly diluted portion of the ex~
haust of the vehicles in front.  Sampling was undertaken in the
air intake  (interior ventilation system) of a vehicle moving
rapidly in heavy traffic on an urban motorway.  The intention
was to investigate the levels of aromatic species which motorists
may be subjected to under these conditions.  The relatively high
level of decane reflects the proportion of diesel fueled vehicles
overtaken during the sampling.


BRAMAN  (U.S.A.)

     Have you noticed any decomposition of the chromasorb Io2
during heating?


TWIBELL  (U.K.)

     The temperature of desorption was  13O C.  At this temper-
ature no decomposition of the chromosorb occured.  Preliminary
experiments showed very heavy contamination by styrene and other
species.  A ten day soxhlet extraction  using five changes of
cyclohexane proved to be an adequate clean-up procedure.  No
further  problems were encountered.  The recommended temperature
limit for this material is    25O C.
BARRATT  (U.K.)

     We  have  considered  using  a  trapping  system  for  the  analysis
of  surgical environments where there  is assumed  to be  harmful
pollution by  volatile  anaesthetic  agents.   Can you describe  how
you have evaluated  the performance of your  analytical  system
and how  the calibration  has  been carried  out?

-------
                            1398
TWIBELL (U.K.)

     The collection efficiency of the system was evaluated
during tube calibration, by means of a gas sampling value loop
situated in the line after the sampler.  Periodic analyses of
the exit cases were made in order to check that the retention
volumes of the species under observation had not been exceeded.
At the sampling rates involved (^ 0.51/min.) aromatic species
were completely retained up to approaching their retention vol-
umes (9 litres could be sampled before benzene began to break
through).

     Tube calibrations were carried out using a range of dilutions
of auto exhausts, down to the limit of quantitative G.C. Detection
Although held in plastic bags, bag losses were overcome by
analysing the gases at the sampling tube entry, after the par-
ticulate filter.  Several analyses were performed to check hom-
ogeneity of the samples mixture.  Rotameters and wet gasmeters
were used to calculate the total volume hence total quantities
absorbed.  Calculation of total amounts collected at lower con-
centrations was by several extractions of the packing with chlor-
inated solvents and G.C. analysis of solutions, after timed
elution.

     The results of the third elution were used to calibrate the
technique, the first elution never being representative.

-------
                             1399
           A SYSTEM FOR THE RAPID IDENTIFICATION OF
               TOXIC ORGANIC POLLUTANTS "iN V'ATER

   WILLIAM T, DONALDSON/ MIKE H, CARTER AND JOHN M, MCGUIRE
Analytical Chemistry Branch, Southeast Environmental Research
Laboratory, Environmental Protection Agency, Athens, Georgia,
USA
ABSTRACT

     A system has been developed that will identify, within a
feu hours, any of eight thousand organic compounds in water at
concentrations as low as 0.01 ,ug per liter.   A one-liter sample
of water is extracted with an appropriate, water-immiscible,
organic solvent.   After the extract is concentrated to 100 to
1,000 microliters, a portion is injected into a gas chromato-
graph coupled to a computer-controlled mass spectrometer.   As
separated compounds pass through the mass spectrometer, spectra
are recorded on magnetic disks frequently enough to obtain sev-
eral spectra for each compound.

     Compounds are identified from  their mass spectra by compu-
terized spectra matching.   A central computer  library contains
spectra for eight thousand compounds.   The computer selects
and lists the compounds whose spectra are similar  to that of
the unknown, in order of their degree of similarity.   The com-
puter also assigns a numerical value for the similarity between
the unknown and each listed spectrum from the library, giving
the analyst a high degree of confidence in the  reliability of
the identification.   Further confirmation of the  compound's
identity can be made by comparing gas chromatograph retention
times and by obtaining infrared absorption spectra or chemical
ionization mass spectra.

-------
                              1400
     With approximately twenty-five mass spectrometers located
throughout the United States, the laboratories of the Environ-
mental Protection Agency can contribute to the continuous re-
cording of the geographic frequency distribution of toxic pol-
lutants.   When a. spectrum is referred to the central computer
together with the location of the sampling site from which the
corresponding compound was obtained, the computer automatically
records the location for each identified compound.

     The library of spectra is being upgraded continually by
carefully elucidating the structures of previously unidentified
organic water pollutants.   High-resolution mass spectra, infra-
red spectra and nuclear magnetic resonance spectra are used in
the initial identification of pollutants whose spectra are not
in the computer library.   Once their identities are certain,
the spectra are added to the library.

     The computer usually identifies over half of the compounds
in samples from a wide variety of industrial waste effluents.
In many cases the total cost for an identification is less than
ten dollars.

-------
                                   1401
1.  Introduction
    In any study of the relationship between observed human health effects
and exposure to environmental pollutants, one must be certain that the
effects he observes result from the cause that he attributes them to.
Therefore he must have knowledge of all components in the systems he is
studying, not just those whose effects he is concerned with in the study.
    During the past few decades tens of thousands of organic chemicals
have been introduced into the aquatic environment as a result of the
chemical technology explosion.  With so many organic compounds to consider
in samples of water from streams that receive industrial and municipal
waste effluents, agricultural run-off and atmospheric fall-out, it is
mandatory that the water be analyzed qualitatively for the presence of
specific toxic pollutants.  The system described in the Summary to this
paper (Webb et. al. (1)) should prove to be a powerful tool in the evalua-
tion of health effects, as it has already proven to be in studying ecologi-
cal effects in the aquatic environment (Alford (2)).
2.  Identification of Carcinogens
    To demonstrate the applicability of the system to the identification of
compounds of significance to health-effect studies, a synthetic sample was
analyzed.  Four carcinogens (selected from the OSHA list of carcinogens',
July 27, 1973) were added to a sample of Oconee River water to provide a
concentration of approximately 50 gg/1 for each carcinogen.  The river was
sampled a few hundred feet downstream from a municipal sewage effluent.
The stream, therefore, was typical of those that must be considered in
assessing potential health effects of municipal water supplies — particu-
larly those in which sewage may be recycled to the supply.
    The sample was extracted with methylene chloride, concentrated by
evaporation, and injected into a Varian Model 1400 gas chromatograph
equipped with a 6 ft., 0.25 in. (O.D.) column, packed with 3% SE 30 on Gas
Chrom Q.  The effluent from the gas chromatograph flowed directly into a
Finnigan Model 1015 quadrupole mass spectrometer with a Systems Industries
data system.

-------
                                   1402
    Figure 1,  the resulting computer-reconstructed gas chromatograph (RGC),
shows the four separated compounds that .were  subsequently identified as  the
      ROC
                 4-NITtOIPHENYL
                                                         3,3-DICHLOROSENZIDmE
                                         N-(2-FLUORENYL)ACETAMIDE
                                                 -MB tie «•  » M «• «o «s «B *» a
     figure 1 : Computer reconstructed gas chromatograph.
carcinogens added to the sample.  Mass spectrum number  320 was selected as
the optimum spectrum for identifying 4-nitrobiphenyl.   The computer was
instructed to plot spectrum number 320 with spectrum number 317 subtracted
(to eliminate background ion  fragments).  The computer-purified spectrum
is shown in figure 2.  Mass 199 represents the singly-charged, unfragmented
molecular ion for nitrobiphenyl.
        SPECTRUM  32O
  B     4-NITROBIPHENYL
  R.
  8
  ft.
  8.
  a.
  ?.
  3.
                                                     M
4JU
  SB  60  71
                                                                  I
    ZO  9Q  10
        MX e
                                    ..

                                                                   »
       Figure 2 : Computer purified spectrum

-------
                                  1403
    A digital description of this spectrum (containing the  relative
amplitudes  for significant raass-to-charge ratios) was transmitted to a
second computer for comparing with the spectra of 8,000 compounds in the
computer's  library.  The computer located 8 spectra that were  similar to
the spectrum representing the unknown compound.  •The computer  print-out in
figure 3 lists four of the compounds  in order of their spectra's similarity
to that of  the unknown.  Note that the similarity index (a  quantitative
assessment  of spectra similarity) for 4-nitrobiphenyl is significantly
higher than that of the compound whose spectrum was second-most  like that
of the unknown.  In a similar manner  the computer also correctly identified
each of the three other carcinogens.
  SIECTRUM NO.  320
  RIVER WATER
    8  HITS
  4-NITROBIPHENYL  (SERL)-WNR DR~  199 C12.H9.N.02 BSR  0518
  FILE KEY=  9020
  SIMILARITY INDEX=  0.854

  5-BROMOACENAPHTHENE 232  C12.H9.BR API 1526
  FILE KEY=  3872
  SIMILARITY INDEX=  0.210

  ANTHRAQUINONE  (SEWL)~T C666 BU  IUJ~  208 C14.H8.02 BSR 0381
  FILE KEY=.  8865
  SIMILARITY INDEX=  0.176

  ACENAPHTHALENE 152 C12.H8 DOW  0925
  FILE KEY=  2050
  SIMILARITY INDEX=  0.157
   Figure 3 : Computer Print-Cut
     When  only compounds of known spectral characteristics  are of interest,
 the dedicated computer can produce  a  limited mass reconstructed gas
 chromatogram (IMIGC), which contains  responses from only those compounds
 containing pre-selected characteristic ion-fragments.   Figure 4 shows an
 industrial effluent's LMRGC in which  only compounds containing an ion
 fragment  of raass-to-charge ratio 149  (typical of phthalates) were observed,

-------
                                     1404
. in
  80
i«i
140
  20
           LMRGC  m/el49
                                                      •292
                                                  ~-J(-~.
     0  20  40  60801001201401601802002202*0260280300320340360580
                                SPECTRUM NUMBER
      0  20 40  60  U 100  120 1*3 160  l«  203 220 2«  260 280 300  320 340 360  380
                                SPECTRUM NUMBER
  Figure 4  » Limited mass reconstructed gas chromatograph (LNRGC)
             only for  ion fragments of mass-to-charge ratio 149
             (typieal  of phthalates) and computer reconstructed mass
             gas  ohromatograph (RGC)'of an industrial effluent.
In the conventional  RGC  (lover part  of the photograph)  OC peak  292 was
obscured by other peaks.   Techniques such as  the  IHROC  allow detection of
compounds at concentrations as low as 0.01 yg/1.
    International cooperative  programs are now being  developed  for
expanding the spectral library, which should  result In  a system that is
indispensable in the assessment of environmental  health effects.
                               REFERENCES
1.  WEBB, R. G.; GARRISON, A.  W.; KEITH,  L. H.; McGUIRE, J. M.,  "Current
    Practice in GC-MS Analysis of Organic8 in Water", Environmental
    Protection Agency Research Report,  EPA R-2-73-277,  (August  1973).
2.  AIFORD, A. L., "Environmental Applications of Advanced  Instrumental
    Analysis: Assistance  Projects, FT 72", Environmental Protection  Agency
    Research Report EPA 660/2-73-013,  (September  1973).

-------
                               1405
                          DISCUSSION
PERRY (U.K.)
1.   You indicate that analysis is carried out on 1 liter of
water containing O.O1 ,ug/liter of pollutants of which you only
used a fraction,  does this mean that mass spectrometric analysis
can be carried out on 1 ng of material?

2.   Can you define the spectral search routine and the meaning
of "similarily index"?


DONALDSON  (U.S.A.)

1.   Yes.  We have obtained satisfactory mass spectra when  less
than 1 nanogram of material was introduced to the gas chromato-
graph.

2.   Briefly, the search routine takes advantage of the infor-
mation redundancy of mass spectra and is based only on the  two
most intense peaks in every 14 atomic mass unit slot.  There
are four main steps:

     1.  Screening based on molecular weight range.
     2.  Screening based on most intense peak of the unknown
         spectrum

     3.  Pre-searching based on spectrum family.
     4.  Calculation  of the similarity index and ordering of
         best matches based on peak-by-peak comparison of the
         unknown  to those reference  spectra that passed the
         pre-search.

     The "similarity  index" is a numerical index  (ranging be-
tween  zero and  one) of the relationship between  (1)  the average
ratio  of corresponding peak intensities in the unknown and  in
the reference and (2) the fraction of unmatched peaks in both
spectra.   It simply describes how similar the spectrum in the
computer file is  to the  spectrum of  the unknown compound.
 HARRISON (U.K.)

 1.    What procedure did you use for extraction of organic mat-
 erials from water?

 2.    What was the efficiency of the extraction procedure and
 how did it vary with the concentration of solute in water and
 the level of suspended solids?  Our own experience has shown
 that efficiencies under certain circumstances may be as low as
 20%.

-------
                               1406
3.   Which compounds are the carcinogens referred to?  Were not
the levels used in the synthetic samples unrealistically high
and hence liable to give unreliable results for efficiency of
extraction?


DONALDSON (U.S.A.)

1.   The sample was extracted from the water, without pH adjust-
ment, with methylene chloride in a separatory funnel.

2.   It was not the intention of this paper to optimize extrac-
tion conditions for the four compounds selected for demonstration.
We did not study extraction efficiencies since we were inter-
ested only in identification and not quantification.  Our ex-
perience in extraction is similar to yours.  A great deal needs
to be done to improve extraction techniques.

3.   The carcinogens were 4-aminobiphenyl, 4-nitrobiphenyl,
N-(2-fluorenyl) acetamide and 3, 3-dichlorobenzidine, all sel-
ected from the OSHA list of carcinogens.  Yes/ the levels were
unrealistically high.  These compounds were selected to show
that carcinogens could be identified by a computer from a group
of 8,OOO compounds.  The technique, therefore, should be con-
sidered for application in studying health effects of drinking
water.  As stated  in answer to your previous questions, no attempt
was made to optimize extraction and concentration conditions,
so the larger quantities were selected to allow identification
even with poor extraction efficiencies.  You are stressing a
very important point:  one can not identify compounds if he can't
extract them.  By  no means would I recommend the extraction
technique used here as optimum.

-------
                              1407
       VERGLEICHENDE SCHWERMETALLANALYTIK MIT HILFE DER
       POLAROGRAPHIE UNO ATOMABSORPtlON - MOGLICHKEITEN
              UND GRENZEN DER REPRODUZIERBARKEIT

      W, EGELS/ G, CR6SSMANN, H, RETHFELD UND D, SEIFERT
Landwirtschaftliche Untersuchungs- und Forschungsanstalt,
Joseph-Konig-Institut, Mxinster/Westfalen, BRD
KURZFASSUNG

     FUr die Spurenanalyse von Sohwermetallen wurden verschiedene
Untersuchungaverfahren entuiakelt, wobei siah u.a. inverapolaro-
graphische und atomabsorptionsflammenapektrometrisQhe Methoden
bewahrt haben.   Jedoah ist die Vergleiahbarkeit dei* vd>n ver-
schiedenen Inetituten eravbeiteten UntevBUchungswerte h&ufig
infrage gestellt, weil es noch an allgemein verbindliahen Stan-
dardvorschriften fehlt.   Am beispiel der Bestimmung von Blei und
Cadmium mit Hilfe der o. genannten Verfahren wivd auf verschiedene
instvumentelle Einflussfaktoren Bezug genomment die nicht vern-
aohlassigt werden diirfen, wenn die Reproduzierbarkeit gewti.hr-
leistet werden soil.  Eierbei sind folgende Grunds&tze besonders
zu beaehten;

1.   Einhaltung optimaler Meaabedingungen,
2.   Erkennung und Beseitigung alter Storeffektef
3.   Kurvenkorrektion auf den Linearbeveiah,
4.   Auswertung mffglichst mit Hilfe von Eichkurvensystemen oder
     Einheitseiahkurven.

     Bei Vorliegen dieser Voraussetzungen, die dem Analytiker zwar
prinzipiell bekannt sind, aber einer einheitlioheren fiandhabung
bediirfen, kann die Reproduzierbarkeit merkliah verbessert werden.

-------
                              U08
Es kommt daher in den Standardvorschriften niaht nur> auf die
alleinige Angabe des Messprinzipes an, sondern es miissen auch
diejenigen Angaben in den Arbeitsvorschriften enthalten sein,
die fur die Einhaltung der durch das Messprinzip bedingten
gunstigen Messkonzeptionen wesentlich sind.
ABSTRACT

     Various test methods for the analysis of heavy metals  have
been developed, of which inverse palaeography and atomia absorp-
tion/flame epectrometry have been among  the most effective.
However,  the comparability  of experimental values obtained  by
various institutes is  often questioned because  there  are still
no generally binding standard regulations.   Taking as  an example
the determination of amounts-of lead  and cadmium by means of the
above-mentioned methods, reference  is made to various instrumen-
tation factors which must not be neglected if reproducibility  is
to be ensured.   Particular attention must be paid  to the follow-
ing principles:

I.   Measurements must be made  under  optimum  conditions,
2.   All  interference  effects must  be detected  and  eliminated,
3.   Curve  corrections must be  made as  regards  the  linear  region,
4.   If possible,  calibration  curve systems  or  unified calibra-
     tion curves  should be  used to  evaluate  the results.

      If these  conditions,  with  which the analyst is familiar in
principle but  which require a more  uniform application, are
 satisfied,  the reproducibility can be perceptibly improved.
 Thus what is important with regard to standard regulations is
 not only  that the measuring method should be laid down, but that
 they must also contain the information required in order to main-
 tain the  conditions favourable to  the application of the method.

-------
                              U09
1. Elnleitung
Wenn man die historische Entwicklung der Untersuchungspraxis
in der Spurenanalyse von Schwenaetallen zuruckverfolgt, so
laflt sich feststellen, daB die herkQmmlichen chemischen Ver-
fahren (z.B. Gravimetric, Titration U.S.) vor etwa 15 Jahren
durch physikalisch-chemische Methoden abgelSst und damit in
mannigfacher und bedeutender Weise verbessert warden. In In-
stituten mit routinemafligem Untersuchungsbetrieb wie z.B. in
der Landwirtschaftlichen Untersuchungs- und Forschungsanstalt
(LUFA) MUnster stiegen die Analysenzahlen hierdurch in die-
sem Zeitraum von wenigen Hundert auf jahrlich ca. 100.000 an
(Abb.1). Vielleicht aus verstfindlichen Griinden verfahrt man
gerade in Routinebetrieben allzu oft eingleisig, d.h. daB
ein bestimmtes ausgewahltes Analysenverfahren und dieses m6g-
lichst im Serien- oder Massenbetrieb zur alleinigen Anwendung
gelangt. Schon seit langeren Jahren streben wir aus verschie-
denen GrUnden zweigleisige Analysenmethoden an, u.a. auch,
urn durch Vergleiche die Moglichkeiten und Grenzen der Repro-
duzierbarkeit besser zu erkennen und urn dem jeweils lelstungs-
fahigeren Verfahren den Vorzug geben zu kfinnen. Auf diese
Weise werden hier schon lange Inverspolarographie und Atom-
absorptionsflammenspektrometrie nebeneinander betrieben. Als
Ontersuchungsmaterialien dienen Kulturpflanzen, Futtermittel,
B5den und Staube, samtlich mit Oder ohne Immissionsbelastung
u.a. durch Blei und Cadmium.

2* Ausftihrungen zur Methodik
2.1 Inverspolarographie
Die Empfindlichkeit der inverspolarographischen Methode
reicht auch unter Einhaltung optimaler HeBbedingungen im
allgemeinen aus, urn bereits die natUrlich vorkommenden Ge-
halte an Blei und Cadmium in den genannten Untersuchungsma-
terialien ermitteln zu kQnnen, im Gegensatz zur AAS in der
Flamme, die diese Lelstungsfahigkeit nicht oder nur bedingt
besitzt. Bel der Inverspolarographie laBt sich mit wenigen
Mefipunkten ein umfassendes Eichkurvensystem aufstellen -

-------
                                 U10
           Potorographie, AAS
           Antony B^chictajng
 Urtersuchuigvn
.
                                                        Abb.  2
     Abb.  3
                                                      Abb.  4

-------
                            1411
Anlage:  Tabelle  1
Methode bzw.
Material
Blei (Pb)
(
Inverspolaro-
graphie
AAS i.d.
Flamme
Weidelgras:
ohne Immission
n = ca. 100
mit Immission
n = >1300
Futtermittel:
n = 358
Getreide
n = 175
Sojaschrot
n = 149
Fischmehl
n m 34
Indus tri e s taut
n = > 2000
Cadmium (Cd)
C
Inverspolaro-
graphie
AAS i.d.
Flamme
Weidelgras:
ohne Immissior
n = ca. 100
mit Immission
n = >1300
Futtermittel:
n = 358
Getreide
n « 175
Sojaschrot
n = 149
Fischmehl
n = 34
Indus triestaut
n m >2000
Nachweisbereiche / Ermittelte Gehalte
n




H
n

,

i
000000X000000
0
Gehaltsspanne
) 0
n v ,
.11. ;


01 0
inmmTnTfiinTnr


.1
*
aOOOOOOO


Verlagerung des MeB-
bereichs durch Ver-
diinnen leicht moglich


DOOOOOOC
\> 0.


3X0
3OOOOOOC
DOOOOOOC
3OOOOOOC
oooooooc
oooooooc
oooooooc

OOOOOOOC
OOOOOOOC
C
OOOOOOOC
01 0
nunmmOOOC

JOOOOOOX
100X0000
IOOOOOOO
ooooo
lOOOOOOO
1 1|D 160 ppm
r
OOOOOOOOmmmmmmmTi
mmmmmmmmnnnnonnr

OOOOOOOC
3OOOOOOC
3OOOOOOC
30O
Xoooo
DOXOOOOC
3OOOOOOC
1 1 1C
)OOOOOOO
IpmnnnnniTnin

lOOOOXOOC
ioooXooo<
iooXoooo<
ioooXooo<
tOOOOOOOl
>ooooooo<
1

3OOOOOOC
3OOOOOOC
3OO
3
3OXOOOO
DOOOOOOC
oooooooc

1X000
ooooXoo<

oooooooc

OOOOmmn
'OOOOOOO

>ooooXoo
i 100 ppn
IOOOOOOOC
IniTnin
DOOOOOOOOOOOOOOOm

t

Verschiebung des
Mefibereichs durch
Konzentrieren der
Analysenlbsung nur
bedingt moglich

oXoooooi
j
*
3OOOOOOOI

)OOO
mmmmmOOOOOOOOOOOmmmmmmm n.n. = nicht
min. Optimaler max. nachgewiesen
MeSbereich

-------
                             1412
bestehend aus den Eichgeraden der verschledenen Empfindlich-
keitsstufen (Ix10~6 bis 5x10"10 A/mm) -, das fiir Jedes Ele-
ment individuell verschieden 1st und im Ubrigen den Vorzug
hat, dafl man die Konzentration einer Ldsung an verschledenen
Punkten dieses Eichkurvensystems messen kann (Abb. 2 u. 3).
Dies 1st eine wesentllche Auswertungshilfe, da die Jewells
korrespondlerenden Punkte des Eichkurvensystems In elnem
bestimmten Korrelationsverhaitnis zueinander stehen. Dadurch
kBnnen auch die nach elnem Testwert zu erwartenden Peakhblien
namentllch 1m Hinblick auf die optlmalen HeBbedlngtmgen voraus-
berechnet und 1m Ubrigen ferner etwaige auftretende Storeffek-
te frtlhzeitiger erkannt werden.
Xs 1st aber zu berUcksichtigen, daB die Elemente mlt niedri-
gerem Halbstufenpotential die Grundllnlen der nachfolgenden
Elemente und dadurch auch ihre konzentrationsbedlngten Peaks
verschieben. Zwar kann man diese slch ergebenden PeakerhBhun-
gen genau berechnen, da sie von der jeweiligen Konzentration
der Blemente und Ihrer Halbstufenpotentialdlfferenz abhangig
slnd. Bel der DurchfUhrung von Routineanalysen 1st das Je-
doch ein erschwerendes Moment, well die in der Auswertung
Ublichen Hilfslinien (z.B. 1) die Abklingkurve oder 2) die
Horlzontale oder 3) die Minimumtangente) leicht zu Fehlbe-
rechnungen ftlhren ktSnnen (Abb. 4). Urn die Inverspolarographie
reproduzierbar anzuwenden, mufl der Schreiber unmittelbar
nach Erreichen jeder Peakspitze durch kurzzeitiges Abschal-
ten des Spannungsvorschubs auf die Grundlinie zurUckgefUhrt
werden (Abb. 5). Auf diese Weise kann man bei jedem Element
mlt verschledenen» d.h. den Jewells gUnstlgsten Empfindlich-
keitsstufen messen, und die Auswertung der Peaks ISBt sich
auBerdem nach vorberechneten allgemeingliltigen Tabellenwer-
ten vornehmen* Zwischen den Zahlenrelhen kann Interpoliert
werden*
Systemabweichende Zahlenwerte bel Peakvermessungen weisen auf
Stttreffekte hin. Bvtl. durch Elsen hervorgerufene Peakver-
zerrungen oder -depressionen soilten besser mlt Ascorbin-

-------
                              1413
saurezusatz als durch Methylisobutylketon beseitigt werden,
well dadurch ein Arbeitsgang  zu  ersparen ist und zusatzliche
FehlermSglichkeiten ausscheiden.

2.2 Atomabsorption i.d. Flamme
Diese Methode 1st aufgrund der leichteren Bedienungsweise
und der MSglichkeit der automatischen Beschickung des Cerates
mit ProbenlSsungen beliebter  und erlaubt eine grSflere MeB-
frequenz pro Zeiteinheit. Auch 1st der Anwendungsbereich
evtl. bei Skalenspreizung, Brennerverstellung und Benutzung
verschiedener Wellenlangen sehr breit, jedoch fur die Ge-
haltslagen von Pflanzenmaterial ohne gr58ere Immissionsbe-
lastung meist nicht empfindlich genug, so dafl ihr Einsatz,
wenn man das Prinzip der optimalen Meflbedingungen befolgen
mochte, haufiger fraglich bleibt. Unter diesen Gegebenheiten
kann nur dann eine gUnstige Reproduzierbarkeit erwartet wer-
den, wenn auch hier die vom MeBinstrument ausgehenden Ein-
fluflfaktoren nicht willkUrlich zur Geltung gelangen. Die
PehlermSglichkeiten sind besonders grofl im nichtlinearen
Teil der Eichkurven, der damit wenigstens ftir die Routine-
analyse kein Bereich der optimalen Meflbedingungen ist. Ftir
den nichtlinearen Bereich lief ern die Hersteller von Plam-
menphotometern als ZubehSr u.a. Kurvenkorrektoren,  wobei
Eichpunkte mit einer Abweichung bis zu 50 96 auf den Linear-
bereich projiziert werden, Messungen mit Cadmium (Cd) haben
aber z.B. ergeben, dafl bei Auftragung von Absorption gegen
beliebige KonzentrationsverhSltnisse (C_/Cmov) stets gleiche
                                       H
Kurven erhalten werden, die ungef&hr erst beim Konzentrations-
verhaltnis 0,5 von der Geraden abweichen (Abb. 6 u. 7). Da-
durch ist es mSglich, auch ohne Kurvenkorrektor im linearen
Bereich zu arbeiten, wenn das Gerat mit der doppelten maxi-
mal zu messenden Konzentration auf die Absorption 1,0 ein-
gestellt wird. In Anbetracht des groflen MeObereiches der AAS
- namentlich bei Brennerverstellung - ist die Einengung des
jeweiligen Mefibereiches urn den Faktor 2 im Sinne einer ver-

-------
                                     1414
                                  Abb. 5
                      A* AAS t d_
       0*    09 •
       «•    4.S-
       y.    ^.




»•
E
OjB1

(M-

52-
(WB











/



-
-.












'
H -




/
^





i




/











/
/










i
'












/
/
/^
•





•









f
/
^







mo« ^
'
-


^







—
            QOtt 02   Qi IS 06   Ql   (0

           . 10	
           • so —-
           • too---
           -500- -
-20 p
-«0
-200
-WOO




' M,


Q56



.
~~




lOgtimoltr
HeOSifSch






''
apis
mMr -
illung i
0- 02
!• Q7
5' V
o*






^




'
/




,
/






/
i



Q2 Qt Q




,
'









,
-------
                            1415
besserten Reproduzierbarkeit durchaus hinzunehmen. FUr den
Routinebetrieb genligt die Auf stellung einer Einheitseichkurve
mlt entsprechender Aufteilung des gesamten Nachweisbereiches.
Im Falle der Brennerdrehung (Lichtwegverkiirzung) 1st hier -
bei sonst gleichen Versuchsbedingungen - elne Angabe Uber
die (leicht meBbare) Winkelverstellung erforderlich. Dieses
Prinzip ist auch bei der Bestimmung anderer Elemente zu empfeh-
len, so dafl hier ebenfalls ein Eichkurvensystem aufgestellt
werden kann. Die Vernaltnisse liegen gUnstiger als beim Cad-
mium, wenn mehr als die Hfilfte der Eichkurve linear verlauft,
und ungUnstiger, wenn die Absorptionskurve schon weiter un-
terhalb 0,5 vom linearen Verlauf abweicht wie z.B. beim Zink
ca. 0,4 oder bei Blei und Kupfer je ca. 0,3.

3. Schluflfolgerungen
Vergleicht man Brauchoarkeit und Einsatzfahigkeit der beiden
Untersuchungsverfahren (Inverspolarographie und AAS i.d.
Flamme) fUr die Spurenanalyse von Schwennetallen am Beispiel
von Blei und Cadmium, so kommt man zu dem Ergebnis, daB zwar
beide Methoden ftlr die Anwendung in der Umweltanalytik grund-
satzlich geeignet sind, jedoch mit der Einschrankung, daB
ihre wechselseitige Benutzung zu empfehlen ist. Infolge der
unterschiedlichen HeBprinzipien und der dadurch bedingten
eigenen instrumentellen Abhangigkeitsfaktoren decken sich die
optimalen MeBbereiche nur teilwelse, so daB nicht selten so-
gar ausschlieSlich eine der beiden Methoden zur Wahl steht
(vgl. Tabelle 1). Der optimale Nachweisbereich gewfihrleistet
aber allein nicht das Mafl der Reproduzierbarkeit, da hier-
fUr ferner von entscheidender Bedeutung ist, dafl (mehr) Ver-
einheitlichung im technischen Vorgehen bei der Erzeugung und
Auswertung der Mefldaten erreicht wird. Zur Verbesserung der
Reproduzierbarkeit gehoren entsprechende Angaben - auf die
im Text hingewiesen wurde - in die Untersuchungsvorschriften
namentlich, wenn diese rechtsverbindlichen Charakter haben
sollen.

-------
                              1416
                        DISKUSSION
VALENTA  (Bundesrepublik Deutschland)

I) Wir haben festgestellt, dass bei der Messung des Bleige-
   halts 1m Blut mit der inverspolarographie die Methode der
   Standardzugabe vorzuziehen ist, well dabei der Einfluss der
   organischen Matrix eliminert werden kann. Haben Sie bei
   Ihren Messungen einen solchen Einfluss der Matrix nicht
   festgestellt?

2) Die Messung der Peakhohe von der Nullinie des Stromes ist
   slcher am genauesten. Die Voraussetzung dafiir ist aberv
   dass der Reststrom bei verschiedenen Proben immer derselbe
   ist. Haben Sie das uberpriift?

3) Die Beeinflussung der benachbarten Peaks/ wie von Blei und
   Kadmium  kann man weitgehend durch Anwendung der Differen-
   tialpulspolarographie  beseitigen.


RETHFELD  (Bundesrepublik Deutschland)

I) Wir untersuchen hauptsSchlich Pflanzenmaterial. Die Ab-
   schaltemethode  (d.h. Abschalten des Spannungsverschubs nach
   Erreichen der Peakhohe,  der Stromkreis wird nicht unter-
   brochen) haben wir zunachst an Standardlbsungen ausprobiert.
   Die Schwermetallgehalte in Pflanzen sind sehr hoch im Ver-
   gleich zu tierischen Organen. Dadurch ist die Storung nicht
   so ausschlaggebend.

2) Bei uns handelt es sich meistens  urn grosse Serien ein-und
   derselben Substanz. Der Reststrom bleibt dann annahernd
   konstant.

3) Die Abschaltung des Spannungsvorschubes wird bei uns noch
   von Hand vorgenommen. In der Landesanstalt fur Bodennut-
   zungs- und Immissionsschutz, Essen 1st die Abschaltung auto-
   mat isiert. Die Differentialpulspolarographie steht uns
   nicht zur Verfugung.

-------
                              1417
         SPURENNACHWEIS SCHADLICHER ELEFENTE IN LUFT,
            WASSER UNO BIOLOGISCHEN MEDIEN TIT DER
                   RONTGENFLUORESZENZANALYSE

       LIESELOTTE BEITZ, JOACHIM HAASE, NORBERT WEICHERT
Siemens Aktiengesellschaft, Karlsruhe, BRD
KURZFASSUNG

     Zur Untersuchung der Hirkung der Umweltverschmutzung auf
den Menschen ist die ahemisahe Ubervaahung der Bereiohe - Luft-
Wasser-Boden - sowie die Kontrolle pflanzliaher und tierischer
Produkte von grundlegender Bedeutung.

     Neuere Untersuchungen haben ergeben, dass die RGntgenfluo-
reszenzanalyse - als zerstttrungsfreies Analysenverfahren - ev-
folgreieh zur Ldsung der oben angesprochenen Pvobleme eingesetzt
werden kann.

     Zur chemischen Uberwachung der Luft werden Spurenelemente,
die an Aercsole gebunden sind, auf geeignete Filterpapiere ange-
reiehert.  Ein Vergleieh mit Standardproben gestattet eine quan-
titative Analyse mit 1 bis 10% relativer Genauigkeit.   Die Nach-
                                                         2
veisgrenzen liegen - je nach Element - bei 2 bis SO ng/cm  Filter-
flache.

     Bei der Wasseranalyse werden uber Anreicherungsverfahren
- Je naoh Element, und Anreicherung - Naahweisgrenzen von 0,l.ug/l
bis 50.ug/l erreicht.

     Die Analyse von Schadmetallen im PfZanzenmaterial  erfolgt
an getrocknetem, zerkleinertem und tablettiertem Material. Die

-------
                              1418
erreichbaren Nachweisgrensen liegen bei 0,1 bis 0,6 ppm. Bei
Veraschung dee Materials wird diese Grenze urn den Faktor 6
reduziert.

     Untersuchungen an getrockneten, pulverisierten und tablet-
tierten Fleischproben ergaben /fir Pb, As, Eg und Cd Ifachweis-
grenzen von 20 bis 30 ppb  (0,02 bis 0,03 ppm).
ABSTRACT

     For the investigation of the effect of environmental pollu-
tion on man, chemical monitoring of air, water and earth and
also checks on vegetable and animal products are of fundamental
importance.

     Recent research has shown that X-ray fluorescent analysis,
being a non-destructive analytical method, can be used success-
fully to solve the above-mentioned problems.

     For chemical monitoring of the air, trace elements combined
with aerosols are concentrated on suitable filter papers.   A
comparison with standard samples permits quantitative analysis
within a relative accuracy of 1 to 555.   The detection limits
                                      2
are within the range of 1 to 30 ng/cm  filter surface, depending
on the element.

     In water analysis, concentration processes enable detection
limits of  0.1,ug/l to SO ,ug/l to be achieved, depending on  the
element and the concentration.

     The analysis of noxious metals in  vegetable matter is  car-

-------
                               1419
ried out on dried, crushed and pelletized  substances.    The det-
ection limits attainable are between 0.1 and  0.5 ppm.    If  the
substance is incinerated, this limit is reduced by  a  factor of
6.

     In tests on dried, crushed and pelletized meat samples the
detection limits were 20 to 30 ppb  (0,02 to 0.03 ppm)  for Pbf
As, Hg and Cd.

-------
                           1420
1. Elnleitung
   Zur Untersuchung der Wlrkung der Umweltbelastung auf den
   Nenschen 1st die chemtsche Oberwachung der Bereiche-Luft-
   Wasser-Boden- sowie etne laufende Eontrolle pflanzlicher
   and tierischer Produkte von grundlegender Bedeutung.
   Keuere Untersuchungen haben ergeben, dass die Rontgen-
   fluoreszenzanalyse (RFA) ale zerstSrungsfrei arbeiten-
   des Analysenverfahren erfolgreich zur LBsung der oben
   angesprochenen Probleme eingesetzt werden kann.
   Wesentliche Vorteile der RPA Bind eine kurze Analysenzeit,
   eine vollst&ndige Automatisierbarkeit sowie die MSglich-
   keit der Archivierung des untersuchten Probengutes.

2. Die Analyse achadlicher El entente in der Atmoaphare
   Zur Analyse der Luft werden Spurenelemente, die an Aero-
   sole gebunden sind, auf geeignete Filter £lj angereichert,
   indem definierte Luftvolumlna durch diese Filter gesaugt
   werden. Derart mit Staub belegte Filter ko'nnen ohne
   Nachbehandlung im Spektrometer gemessen werden [2j .
   Die Eichung des Terfahrens erfolgt mit Standardproben,
   die wle folgt prapariert werden: Auf die in der Praxis
   zur Staubabscheidung verwendeten Filterpapiere (Whatman
   oder Millipore) werden 5o bis loo /ul Eichlb'sung gege-
   ben. Die Hessung der Standardproben unter optimalen Be-
   dingungen ergibt lineare Eichkurven.
   Kach der Bichung ist es notwendig, Korrekturfaktoren fUr
   die zu analyaierenden Elemente zu finden, welche die
   unterschiedliche Absorption der gemessenen RQntgenstrah-
   lung an den mit EichlSsungen und den mit Stauben beleg-
   ten Filtern berUcksichtigt. Die Bestimmung dieser Korrek-
   turfaktoren K erfolgt Uber eine gesicherte chemische
   Analyse des mit Staub belegten Filters. Die mit der RFA
   erhaltenen Analysenwerte C ergeben sich dann zu:

-------
                            1421
   CStaubfilter = CEichung X K ™/*&™  Filterflache

3. Die Wasseranalyse
   Die oben beschriebene Piltermethode kann auch zur Ana-
   lyse des Wassers eingesetzt werden. In dlesem Pall wer-
   den loo/ul Wasser auf die Filterpapiere gegeben. Die
   Tabelle 1 zeigt die mit dieser Methode erreichbaren
   Nachweisgrenzen fur Aerosole und Wasser.
   Die Nachweisgrenzen lassen sich durch Anreicherungaver-
   fahren reduzieren: lurch Eindampfen einiger liter Was-
   ser und Tablettieren des RUctcstandes erhait man die in
   Tabelle 2 angegebenen nachweisbaren Grenzkonzentratio-
   nen* Eine weitere Anreicherung der Schadstoffe erhalt
   man durch chemische Abtrennung der Hauptkomponenten
   (Ca- und Mg-Yerbindungen) und Vermessen des Niederschla-
   ges (Tabelle 2).

4. Die Schadtne tall best imam ng in pflanzlichen Produkten
   Die Analyse von Schadatoffen in Pflanzenmaterial er-
   folgt an getrocknetem, zerkleinertem und tablettiertem
   Material.
   Zur Herstellung von Eichproben werden einem chemisch
   analysierten Ausgangsmaterial Eichlosungen zugegeben,
   welche die zu analysierenden Elemente in abgestuften
   Konzentrationen enthalten. Das so entstandene Suepen-
   sat wird homogenisiert und getrocknet. Dae getrocknete
   Material wird bei Zugabe eines organischen Bindemittels
   erneut homogenisiert und zu Tabletten verpreeet..
   Bild 1 zeigt die aus Eichkurven Standardabweichungen Sx
   fiir die Elemente Pb, As, Hg, Se, Zn und Mn als Mass fur
   die Analysengenauigkeit mit
       Sx -TU^L.   x ioo£ relativ
            m Cx

-------
Tabelle It   Kit der RPA errelohbare Nachwelegrenzen Cg acMdllcher Elemente an Aerosolen
             und 1m Waasec mlt Flltermetbode
Og           Na     Mg     Al     St     S     01     K     Ca     V     Fe     Zn    AB   Cd   Pb


Cg (LBBung)
(ng/cn )
_^_^_-^__^_^^_^_^^^_^______^___-^_^i^__-__^_______^_-____^__^^___.^^_^__^.,^___^^__^__^___1^_^_^_^^^_^_^_^^_^_
.1^^__>^___-.^_^..^.^_^__—_^___

0L(ng/m5)   23oo    32o    44     25     29     9o     4,9   5,7   3o     52     69   38o  19   12o



Ow (ppm)      52     7,2   o,94   0,55   0,64   2,o    o,ll  o,13   o,67   1,2   1,6   8,5 o,43  2,7



C^    bezogen auf 1 m5 Luft  (ohne Eorrektur K)

C^    bezogen auf loo/ul  Wasser

-------
Tabelle 2
Nachweiegrenzen In Waaser nach Anrelcherung
Element
Na
Mg
S, K, Ca, )
Ci(, Pa )
Fe, Ju,
Zn, Co,
Pb, As,
Mo, W
Measzelt
sec
loo
loo
4o
4o
Na chweisgr enze yUg/1
Eindampfen Abtrennen (HgS-Niederachlag)
lo
5
ca 1
o,l ... o,5 o,ol . . . 0,05
                                                                                                     NJ
                                                                                                     W

-------
                          1424
 10
10

I
                             As
       mCx
0.1                1
Andysengenouigkeit in Pflonzenmoterio1

                     Bild 1
                                     10
                                         C(ppm)
100
 Nz s  die in der Zeit T an der Konzentration C  gemeasene
       Impulszahl
 m  =  Steigung der Eichkurve
 Nach Bild 1 betragt die Standsrdabweichung im Bereich
 1 bis lo ppm  lo bis 1% relativ.
 Die Eichkurven sind auch hier - vie allgemein in der
 Spurenanalyse - linear.
 In der Tabelle 3 sind die mit dieser  einfachen PrSpara-
 tion in einer Messzeit T = 2oo sec erreichbaren Grenzkon-
 eentrationen eingetragen.
 Zur Herabsetzung der Nachweisgrenze bis  in den ppb-Be-
 reich mUssen Anreicherungsmethoden wie Veraschung oder
 Extraktion nerangezogen werden.

-------
Tabelle 5   Nachweiegrenzen und AnalyBengenaulglcetten an getrocknetem Pf lanzeraaaterial
Element
Ma
Ou
Zn
Se
Pb
As
Mefizeit
(sec)
200
200
200
200
200
200
Analyser^
<
10
10
8
1
2
6
pnautglceit *)
,ppm)
±
±
±
±
±
±
0,3
0,3
0,24
0,15
0,08
0,18
Nachweisgrenze *)
•pp'nv
0,4
0,4
0,36
0,13
0,2
0,5
      *)   far S *= 99,7

-------
                             1426
   Bine Pb-Bestimmung an getrocknetem und veraschtem Mate-
   rial ergab an getrocknetem Material eine Nachweiagreuze
   Cg von o,2 ppm, an veraachtem Material eln Cg von
   Of033 ppm in T = 2oo sec. Durch Verlftngerung der Meae-
   zeit kann Cg wie folgt verbeaaert werden:
                    ir^r
       Og,  =  Cgn     -i-
            =  Nachweisgrenze bet elner Messzeit T,

       Cg2  =  Nachweisgrenze in der Messzeit T2

5. Die Analyse tleriacher Produkte
   Ton hochaktuellem Interesse 1st die Schadmetallbest1m-
   mung an tierlschen Frodukten:
   Hler hat alch folgende Preparation bewOhrt:
   Ca 2o g des Produktes, e.B. eine Fleischprobe,  werden
   grob zerkleinert, mit 2o ml Wasser versetzt and homogeni-
   siert. Das Homogenlsat wird durch Yakuum- oder  Gefrier-
   trocknung  entwUssert. Dae  getrocknete Material wird
   pulverisiert und kann mit  oder  ohne  Bindemittel zu Tab-
   letten verpresst werden. Diese  Tabletten kOnnen direkt
   im Spektrometer  vermessen  werden.
   Zur Herstellung  von  Standardproben werden  dem chemisch
   analyslerten Material nach der  Grobzerkleinerung 2o ml
   SchwermetallsalzlSsungen in abgestuften Konzentrationen
   zogegeben.
   Die weitere Preparation erfolgt wie  einganga beachrieben.
   Zur Eliminierung von ffatrizeinflUssen £4j  bei wechaelnder
   fleischmatrix kann die  Bichung  mit einem NQ-Wert
   ("background") als innerer  Standard durchgeftihrt werden.

-------
Tabelle 4
Element
Pb
Pb
As
Hg
Sb
Cd
flff = 2 	
Nachweisgrenzen fUr Scnadmetalle In Pleisch
linie T B 2 oo e T = 4oo a
Let 0,0? o,o49
Lfl^ 2 o,o75 o,o53
K < o,o35 o,o25
L °L o,o65 o,o46
LflT o,55 o,39
Ko^ o,o76 o,o54

Tel OOO 8
o,o3
o,o3
o,ol5
o,o3
o,25
o,o3
o . N Untergrundimpulazahl
m                m    Stelgung der Elchkurve

-------
                          1428
              10
                             o Kolbslcber
                             • Schweineteber
                             • Kalbfleisch
                             • Schwoneikssch
                 0        1       2
                            	-ppm Hg
             Hg-Efchung m» !<, als inncrcrn Standard
                         Bild 2
Bild 2 zeigt eine derart  ermittelte Eichkurve  zur Hg-
Bestimmung  an Fleisch and Leber.  Aus Bild 2 geht  hervor,
dass eowohl Kalb- als auch  Schweinefleisch - die  sich in
ihrem Fettgehalt wesentlich unterscheiden - mit einer
Eichkurve erfasst warden  konnen.  Das gleiche gilt fiir
Kalbs- und  Schweineleber.
Tabelle 4 gibt schliesslich die Nachweisgrenzen ftir Pb,
As, Hg, Sb  und Cd wieder, die mit der einfachen PrSpara-
tion der Trocknung in 2oo,  4oo und 1 ooo sec.  Messzeit
erhalten wurden.

-------
                          1429
Literatur

 jlj     L.Beitz, G.Rb'nicke:
         "Spurennachweis schadlicher Elemente in Luft und
         Regenvjasser"              ,
         Proc. 3 rd. Int. Clean Air Congr. 73, C 32,
         VDI-Verlag Diisseldorf.

 £2]     J.V. Gilfrich, P.G.Burkhalter und L.S.Birks:
         11 X-Ray Spectrometrie for Particulate Air
         Pollution - A quantitative Comparison of Techniques"
         Analytical Chemistry, Vol. 45, Nr.12 (1973)
         S. 2oo2

         E.Forschner, W.Wildanger, L.Beitz, J.Haase,
         L.MUller: "Schadmetallbestimmung in Fleisch mit
         Hilfe der Rb'ntgenfluoreazenzanalyse".
         FleiBchwirtschaft, H.3   (1974), S. 529

 f 4 1    R.Plesch:
         Private Mitteilung; Veroffentlichung in Vorbereitung,

-------
                               1430
                           DISKUSSION
DAMS  (Belgien)

      Ich habe eine Frage bezuglich des Berichtigungsfaktors,
der bei der  Analyse von auf Filtern gesammelten Aerosolen er-
forderlich 1st. Dieser  Berichtigungsfaktor kann bei leichten
Elementen wie Na, Mg,  Al usw. sehr hoch, bis zu IO sein. Der
Faktor 1st abhangig vom Filtertyp, von der Eindringungstiefe
in den Filter und von der Absorption des Staubes selbst. Die
Eindringungstiefe der Aerosole in den Filter hangt wiederum ab
von der Luftgeschwindigkeit  und dem Filtertyp, aber auch von
der Partikelgrosse und der Verteilung von Masse/Grosse des Ele-
ments als Funktion der chemischen Zusammensetzung. Meines
Erachtens wurde ein "Treib-"(puff) Verfahren zur Ausarbeitung
der Nonnen einen Tell der Probleme  losen, wenn eine Aerosol-
Standardprobe homogen auf einem Filter verteilt wird.  Waren
Sie in der Lage, konstante Berichtigungsfaktoren zu ermitteln?


BEITZ  (Bundesrepublik  Deutschland)

      Die Grosse der Korrekturfaktoren bei der Analyse von auf
Filtern gesammelten Aerosolen  ist abhangig vom analysierten
Element, vom Filtertyp und vom Staubsammelgerat.

      Bei Konstanthaltung des  Filtertypes, des Staubsammelge-
rates und der  Staubsammelbedingungen  (Druck, Geschwindigkeit
usw.) ist der  Korrekturfaktor  abhangig von der Absorption des
Staubes.

      Fur Schwefel  fanden Bonnevie-Svendson  und Folio vom
Institut fur Atomenergie in  Kjellar  Norwegen, einen Korrektur-
faktor K = O,66.

      Zu gleichen Ergebnissen  kamen G. Ronnicke, Aerosolmess-
stelle Schauinsland* 78OI   Schallstadt-Wolfenweiler, BRD und
D.  Klockow,  Institut fur anorganische Chemie, Freiburg/Brg.  BRD,

      Ronnicke und  Klockow  haben in jiingster Zeit  auch  Korrek-
turfaktoren  fur weitere  Elemente bestimmt.  Zahlenwerte  sind
z.Zt. noch nicht veroffentlicht.

-------
          UNTERSUCHUNG DER WIRKUNGEN AUF DEN MENSCHEN
                      EXPOSURE MONITORING
                    MESURE DE L'EXPOSITION
                    MISURA DELL'ESPOSIZIONE
                      METING VAN EXPOSIE
VQvsi.-tZfnd.er  - Chairman - President - Presidente - Voorsitter

                   G.C. BUTLER (Canada)

-------
                              1433
         ETUDE PREPARATOIRE EN VUE DE L'ETABLISSE!«ENT
           D'UN PROGRAMME COORDONNE DE SURVEILLANCE
            DE LA CONTAMINATION DE L'HOWE PAR LES
                      PRODUITS CHIMCUES

                          E, FOURNIER
Centre de Toxicologie, HOpital Fernand-Widal, Paris, France
RESUME

     Le projet H.M.P. a'efforae d'&tudier les intoxications sub-
chroniques et chroniques liSes & I'ent>ironnement chimique moderne,
les intoxications aiguSs &tant e'tudie'es par ailleurs.

     Le projet doit porter BUT des observations de tres longue
dur£e a forte dispersion Saologiquet dona difficiles et cofiteuses.
C'est oependant 1e seul moyen de d&finir des param&tres de sur-
veillance et de valider des observations cliniques rgparties au
hasard.

     Le projet oomprend la surveillance r&guli&re d'Schantillons
de population reprSsentatifs d'un ensemble suffisant3  la sur-
veillanae sp^aifique de certaines populations particulierement^
exposees ou particulie~rement sensibles:  Stude professionnelle,
pharmacovigilance, groupes humains sentinelles en partioulier
allevgiques.

     Une banque de donnSes toxiaologiques par oolleote aleatoire
eat une organisation indispensable a la reaherehe des ootnai-
denoes suspeotes.   Elle partioipe a la coordination d'un projet
qui tend 3 r&pondre aux inquietudes manifestoes devant le "danger
chimique" par I'ensemble des populations a civilisation industria-

-------
                              1A34

ABSTRACT

     The BMP is aimed at studying chronic and sub-chronic pois-
oning caused by the chemicals in the modern environment^ acute
poisoning being dealt with in another study.

     The project must be based on observations over a very long
period and with a vide ecological distribution, and so they will
be difficult and expensive.   This is, however3 the only way of
defining surveillance parameters and of validating random clini-
aal observations.

     The project involves the regular surveillance of population
samples representative of an adequate total populationt as well
as specific surveillance of certain groups which are particularly
exposed or particularly sensitive:  professional studies, phar-
macological control^ pilot groups of people - particularly those
with allergies.

     A bank of toxicological data obtained by random sampling is
indispensable for research into suspicious coincidences.   This
databank will help to coordinate a project which reflects the
disquiet about the "chemical danger" felt by all people in in-
dustrialised countries.

-------
                                    1435
I/ L'etablissement d'un programme coordonne de surveillance de la contami-
nation de 1'horame par les polluants chimiques constitue la base d'une in-
formation scientifique en vue de 1'evaluation objective des risques que
fait peser la pollution sur la sante publique,
   1 - I/ Le projet H.M.P. doit eviter la confusion des secteurs ou phases
de la toxicologie clinique :
Phase 1 : Intoxications aiguSs : Accidents, suicides, crimes.
Phase 2 : Intoxications subaigugs et chroniques : Maladies professionnelles
          Maladies medicamenteuses. Teratogenese. Allergie.
Phase 3 : Intoxications chroniques : Cancerogenese. Mutagenese. Vieillisse-
          ment. Flu ctuations allergologiques et imraunologiques. Impregna-
          tions toxiques.
   A ces phases correspondent des organisations particulieres  :
Phase 1 : Les Centres Anti-Poison ont deja repondu a la premiere demande.
Phase 2 s Les Centres de Medecine du Travail, les Centres Anti-Poison, les
Centres de Pharraacovigilance peuvent s'organiser pour repondre ensemble a
la deuxierne demande.
Phase 3 : Chacun des centres de la Phase 2 peut participer a un travail
d'ensemble.
   1 - 2/ Difficultes particulieres de realisation :
   La troisieme phase, specifique du projet H.M.P., ne devrait etre explo-
ree que progressivement et avec beaucoup de prudence pour trois raisons
essentielles  :
_ la tres longue duree necessaire a 1'observation reellement chronique,
oblige a tenir compte d'une foule d1interferences dont 1'effet propre peut
etre superieur a ceux que  1'on pourrait attribuer a 1'environnement. Com-
mencer la phase 3  sans assurer la securite scientifique de la  phase 2,
exposerait a  des erreurs  tres grossieres.
- la forte dispersion ecologique des conditions de vie et I1interference
avec des habitudes deviantes jouent un role essentiel en pathologic chro-
nique  : nutrition  anormale, alcoolisme, tabagisrae, absorption  reguliere de
medicaments.
_ la rarete des operations regutterement conduites dans cet esprit  de recher-
che est  due a la difficulte et au cout d'operations portant  sur  des centai-
nes de references  acquises sur chacun des  individus  etudies  grace  a la de-
finition des  parametres  biologiques.

-------
                                     1436
   1 - 3/ Aspects psycho-sociologiques
   11 serait inutile  de prevoir d'aussi importants projets et de les doter
d'une validite sientifique sans prendre en consideration en merae temps les
consequences psycho-sociologiques de telles entreprises. Les consequences
d'une etude toxicologique font toujours apparaitre une certaine forme de
responsabilite judiciaire, une certaine possibilite de controle technologi-
que et de modification de 1'utilisation, des exigences nouvelles de contro-
le et de prevention ; elles suscitent beaucoup de discussions.
2/ Tout projet H.M.P. doit suivre un plan rigoureux et comprendre :
- une recherche progranmee specifique du projet H.M.P. et d'une substance
  chimique precise ;
- une action de coordination des centres de Phase 2 sans modifier leur acti-
  vite specifique ;
- 1'adjonction a chaque groupe coordonne d'une activite H.M.P.
   Apres une periode de fonctionnement, de nouvelles ameliorations peuvent
etre proposees.
2/1La recherche exhaustive des maladies nouvelles effectivement liees a une
contamination chimique est 1'un des objectifs du projet H.M.P. Il ne peut
Stre realise. D*ou 1'idee d'une Banque des appels et des demandes d'informa-
tions & propos de coincidences suspectes ou de maladies inexpliquees ...
(Banque de collecte quasi-aleatoire - B.C.A.). L'accumulation des donnees
aleatoires peut faire apparattre des coincidences assez regulieres pour de-
venir d^monstratrices d'un phenomena multifactoriel de toxicite distribuee
au hasard (sperergique) et nous pensons que c'est le seul moyen a mettre en
oeuvre actuellement par H.M.P** en debutant I1operation initiale par la
phase 2 (enregistrement aussi organise que possible des affections des ou-
vriers "anormalement" exposes, des etres humains particulierement sensibles
(sujets sentinelles), des reactions adverses aux medicaments, aux pestici-
des,  aux produits de maison). L'identite des troubles cliniques a pexmis
le diagnostic de toutes les intoxications chroniques collectives recentes,
ce qui justifie 1'espoir d'un resultat dans cette direction de recherches.
2/2Surveillance reguliere d*4chantilions de population
  .. La surveillance reguliere d*un echantillon de population est une autre
vole d'abord de H.M.P. (Phase 3). Ses buts doivent Stre precises, ce qui
suppose :
            a) la definition de la population
            b) la definition de 1'echantillon

-------
                                    1437
            c) la description des limites du projet
            d) la recherche des resultats previsibles
   La definition des groupes temoins dans cheque pays permettrait :
- de fournir des renseignements acceptables pour decider de la surveillance
de la population et de la redalion de mesures* preventives.
- de fournir un document de base sur la contamination reelle de chaque po-
pulation permettant dans une certaine mesure de comparer les types alimen-
taires, les types de contaminations chimiques de base liees a 1'ecologie,
au type de civilisation, leurs effets sur le frequence de certaines mala-
dies tres courantes.
- d'aboutir par le moyen indirect de la surveillance systematise^, reperto-
ries tous les trois ans, a une etude extensive portant sur environ 10 OOO
sujets, groupe global important mais a peine suffisant pour apprecier 1'ap-
parition de maladies rares ou comparer la frequence de maladies communes.
- ce projet se situerait par rapport aux methodes courantes de surveillance
comme un axe de reference sans pour autant eliminer les autres voies d'ob-
servation et les autres etudes qui devraient s'y referer dans la mesure ou
leur impact statistique se rapprocherait de la surveillance de un pour
25 OOO habitants, en poursuivant des buts paralleles.
- Il faudrait en outre isoler des groupes humains "anormalement" exposes
(phase 2), ou anormalement receptifs. Leur etude doit etre organisee des
que possible au titre de la phase 2. Les sujets "anormaux" : les ouvriers
special  ement exposes, les organismes sentinelles par anomalie congenitale
ou acquise, doivent etre rapportes au groupe ou ils se trouvent, celui-ci
faisant souvent 1'objet d'une surveillance medicale particuliere.
      3 / Realisation du  programme
   Toute operation H.M.P. devrait ensuite se faire en trois temps, apres le
choix des sujets, des parametres biologiques utiles et comparables, et de
I'echantillon de la population a etudier :
3/1   preparation et collecte des documents :
   Aucun document acceptable ne peut etre recueilli par des personnes in-
competentes. D'ou 1'idee formelle d'eliminer du temps de preparation et de
collecte des documents toxicologiques toute organisation non habilitee par
une competence technique, c'est-a-dire par une experience  suffisante con-
fronted a celles d1autres laboratoires analogues.
3/2    etude critique des resultats bruts : observation et  dosages.
3/3   validation des resultats retenus et correlations avec d'autres opera-
tions  portant sur le meme sujet. Formation des Laboratoires.

-------
                                   1438
- d'abord definir les laboratoires et groupes cliniques competents ou faci-
liter I'acces a la competence de ceux qui souhaitent partiqper au program-
me H.M.P.
- puis proposer les sujets a traiter en commun et definir les collabora-
teurs possibles, les parametres retenus, les methodes acceptees, le type de
collects organ!see.
   Les laboratoires de Toxicologie subaigug (phase 2), assez nombreux, sont
proches des services de Medecine du Travail, zone elective de collecte de
ce type de desordres, des Services Hospitaliers pour ce qui concerns les
accidents therapeutiques ou I1allergic aux substances chimiques. Des labo-
ratoires specialises existent dans les grands centres de medecine d'entre-
prisey dans certains departements hospitalo-universitaires : Centtes anti-
poison, Service de Pathologic Industrielle, Services de Pharmacovigilance.
   L'activite de 1'ensemble se complete normalement par des etudes socio-
psychiatriques touchant les toxicomanes, I'ergonomie, les deviants alimen-
taires, et par des etudes epidemioLogiques visant a demontrer le danger de
nouveaux produits manipules par grandes quantites.
   Les laboratoires de Toxicoloeie chronique (phase 3), essentiellement
celle de I'environnement, seraient typiques d'une activite epidemiologique
greffee sur des etudes cliniques et experimentales, mais ik sont exception-
nellement orientes vers des etudes de toxicologie humaine.
   Assez paradoxalement, les laboratoires de toxicologie des grands insti-
tuts medico-legame peuvent jouer un role essentiel dans ce domaine en per-
mettantt grace au grand nombre d'autopsies de sujets accidentes probable-
ment sains, de fonrnir des echantilions representatifs de la population re-
gionale normale ou tout au moins controlable, et d'effectuer les mesures
necessaires.
   La collecte peut aussi se faire dans les centres de pathologic (ou
d'anatomo-pathologie) des hopitaux universitaires importants. Selon les
cas, le laboratoire de toxicologie medico-legale ou le laboratoire de chi-
aie  medicale hospitaller ou universitaire, assurera les dosages.

   Le benefice reel a tirer d'un projet H.M.P. sera en fait une modifica-
tion de I1esprit des etudes toxicologiques en favorisant les discussions
entre personnes competentes dans les domaines differents et en permettant
d'exposer 1*ensemble des problemes aux responsables gouvernementaux.

-------
                               1439
                           DISCUSSION
WASSERMANN (Israel)
     Les modeles que 1'auteur suggdre pour 1*Evaluation des
risques dus 3. la pollution de 1' enviropnement sont fort interes-
sants.  Quelie sera la liste des prioritSs qui sera consid&rSe
en ce qui concerne des problemes a aborder et des populations
a etudier?


FOURNIER  (France)

     Les propositions qui ont £t€ faites tendent surtout 3 fiviter
une certaine confusion entre les differents secteurs de toxicologie.

     Les probldmes a aborder en prior ite1 dependent:

     - pour  la toxicologie sub-aigue de Involution de la techno-
logic Industrielie et des produits mgdicamenteux,

     - pour la toxicologie chronique, des produits non dSgrada-
bles essentiellement d'origine mine"rale, des produits susceptibles
d'un effet cumulatif en particulier cancferogene ou mutag^ne.

     Un ^change de vues sera done n^cessaire entre les differents
specialistes de toxicologie afin que les choix portent avant
tout sur des produits re" el lenient manipules ou apparaissant en
quantitS notable de maniSre persistante ou encore susceptibles
de provoquer des cancers spontanes ou exp^riroentaux.

     Les populations a Etudier sont:

     - pour la toxicologie subaigue ceux qui sont particuliSre-
ment exposes : ouvriers pour les produits industrials, malades
pour les medicaments.

     - dans le domaine de la toxicologie chronique des e"chantil-
Ions de population a suivre pendant des p€riodes suffisamment
longues.

-------
                              1441
        METHODOLOGIE POUR L'EVALUATION DE (.'EXPOSITION
           DE L'HOWC RESULTANT DE LA CONTAniNATION
              DES ALIMENTS PAR LES METAUX LOURDS,
                ETUDE DE OUELQUES CAS CONCRETS

              R, BITTEL, A, GARNIER, G, LACOURLY
Association Euratom-CEA, Niveaux de pollution du milieu ambiant
Departement de Protection, Service de Protection Sanitaire,
Fontenay-aux-Roses, France
RESUME

     Dans le cadre d'une Association entice les Communautes Euro-
peennes et le CEAt une methodologiet inspires par  les recherches
anterieures dans  le domains radioactif^ a ete mise au point en
Vue d'evaluerf dans toute situation^ les quantitSs de metaux
lourds -Lng&v&es par I ' homme en fonation des niveaux dans le mi-
lieUj des regimes alimentaires et des caraoteristiques des trans-
fevts, en tenant  compte des concentrations et des  dilutions qui
peuvent intervenir lors des differents stades du transfert:
production* transformation et distribution des aliments.   La
meme methodologie permet, inversement, de determiner, dans chaque
cas, la capacitS  limite du milieu.   Compte tenu de la complexity
des differents param^tres intervenant dans les transferts, des
etudes de sensibilitS  apparaissent indispensables  pour dSfinir
les recherches a  entreprendre en priorite.

     Pour concretises  les conceptions precedentest on a applique
cette methodologie aux transferts des metaux  lourdes  (Hgf Cd>
Pbf Cut Znf en particulier) dans des chatnes  alimentaires aqua-
tiques et continentales.

-------
                               1442
ABSTRACT

     On the basis of previous research in  the radioactive  sphere,
a methodology has been developed by Ian Association of  the  Euro-
pean Communities and the  CEA, with a view  of evaluating  quanti-
tatively in any given situation, the heavy metals ingested by
man, as a function of levels in the environment, diet  and  trans-
fer characteristics, and  taking into account concentrations and
dilutions which may occur at the different transfer  stages viz.
production, processing and distribution  of foodstuffs.   Conver-
sely, the same methodology can be used to  determine, in  each
case, the limiting capacity of the  environment.   Given  the
complexity of the different parameters occurring during  transfer
processes, sensitivity studies appear to be essential, in  order
to define which research  should be  given priority.

     In order to make the previous  ideas more concrete,  this
methodology was applied to the transfer  of the  heavy metals (es-
pecially Eg, Cd, Pb, Cu and Zn) in  the food chains,  both in water
and in  the soil.

-------
                                   1443

1. INTROJUCTIOH

      II est bien connu que les radionucleides rejetes dans le milieu sont
susceptibles de parvenir a 1'ttomme, pour une grande part, par la voie de
la chalne alimentaire. Le cheminement des polluants radioactifs des
points de rejet jusqu'a I1 organe critique de ^Homme k fait 1'objet d'e-
tudes tres pousse'es, qui ont servi de bases a I1 evaluation des niveaux
adnissibles dans le milieu et a celle des taux maximaux d'introduction
des polluants dans I1organisms a partir de la dose maximale admissible
pour le groupe de population considere. Plus generalement, il est possi-
ble de relier le niveau x d'un polluant radioactif dans le milieu au
debit de dose D delivree a 1'organe critique de 1'Homme. Les relations
font intervenir les parametres physiques, chimiques, biologiques, 3co-
nomiques dans las transferts* Les recherches effectules dans le cadre
de 1*Association EURATCK-CEA "Niveaux de pollution du milieu ambiant"
ont permis de faire 1'inventaire de ces parametres, et, dans divers cas
concrets, d'evaluer I1 incidence de leur variabilite sur le d£bit de dose.
Ce travail a permis d1eclairer I1importante notion de "capaoite radiolo-
gique limit e "d'un milieu rScepteur, en tenant conrpte des diff^rents
stades du transfert: production, transformation et distribution des ali-
ments, di&tStique (l).
      H est apparu comme souhaitable de s'inspirer de la methodologie
mise au point dans le domains radioactif en vue de chercher a evaluer,
en toutes situations, les quantites de metaux lourds inger€es par
l*Homme, en fonction des niveaux dans le milieu. Les examples qui vont
suivre sont eopruntes a des etudes realises en collaboration avec 1*As-
sociation et divers laboratoires.

2. TRANSFBRT PU MILIEU AOX PRODUITS BRUTS DE L'ALMEOTATIOy IE L'HCHME
2.1. Transfert des metaux lourds de 1'eau d1irrigation aux vegetaux
      On a en particulier etudie le transfert du cadmium de 1'eau au riz
irrigue par submersion, en utilisant des rizieres artificielles miniatu—
ris£es, au Laboratoire d'Ispra de la Division de Biologie d'Euratom (2).
On a constate que le facteur de transfert P, eau -^-endosperme, du
cadmium

-------
                                  1444
                           Cd dans 1* endosperms
                           Cd dans 1'eau d1 irrigation
eat de 1'ordre de 60, Men supe'rieur & oelui du zinc ( F £^ 8). Cette
experimentation souligne I'inte'rSt de recherches sur la contamination
de prodoits cere'aliers par le cadmium*

2«2» Transfert des mfrtaiiT loprds de l*eau aafr orftanismes aguaticiu.es
      Une recherche de longue haleine est actuellement poursuivie a Nice
au Centre d'Etudes et de Recherches de Biologie et d'0c6ano graphic Me-
dical e (GBRBGM) sur le transfert de mdtaux lottrds dans des chatnes
trophodynamiques marines. Lee Elements sont, comme dans la re"alite"?
introdnits sous forme lonique simple et sous forme associSe a divers
conposants organiquee naturels et artificiels. Le premier echelon marin
est containing par 1'eau, le second, a la fois par 1'eau et le premier
6chelon qui eat consomme1 comme aliment* Le dernier Schelon comport e un
mammifere, la souris, considdr^e comme consommateur final des chalnes
trophiques, Le tableau 1 donne un ezemple des re" suit at s obtenus sur une
chalne neritique a mollusques (3). Des recherches sont actuellement
entreprises a Ispra en vue d'6tudier lea transferts des mStaux lottrds
(Eg, Cdf Zn, Fb) de 1'eau, auz organismes duly aqui coles (facteurs de
transfert}
                           concentration dans jL'orgjanisme (poids frais)
                           concentration dans 1'eau
 3.  TRAHSFERTS DES PRODOITS SHUTS ATO PRODUITS AU STAKE IE LEUR CONSCM-
          PAR L'HCMME
      Divers aspects ont  6tS  envisages* la transformation et la prepara-
tion des denre'es alimentaires, les processus de commercialisation et  de
distribution,  enfin des enqu^tes sur les regimes alimentaires dans la
Communaote' 5urop 6enne*

3.1.  Transferts des produits  bruts aux produits au stade de leur  consom-
      mation _p_ar_l .'Homjite
      Ces etudes ont 6t6  entreprises en. collaboration -avee 1'Institut
Hational de la Recherche  Agronomique,  le Ministers de 1 'Education

-------
                                  1445

Phytoplancton
Moule
Cu
1000
120
Zn
7000
1200
Pb
•
5000
150
Cr (VI)
100
40
Hg
3000
100
 TABLEAU I  -
ORDRE HE GRANDEUR KES PACTEURS BE TRANSFERTS 13ES METAUX
LOURDS MNS UNE CHAINE NERITIQUE A MOLLUSQUES (elements
introduits h 1'Stat ionique simple)
TABLEAU n  -  TAUX DE SECRETION MNS LE LAIT DBS ELEMENTS INQERES PAR
               LA VACHE (rapport dee quantites secret^ee, par jour et
               par animal, aux quantites ingerees, par jour et par ani-
               mal)
       Pe -. 0,009
       Zn : 0,08

       Sr ! 0,03
                                 Cd  s 0,06
                                 Pt>  : 0,1
                                 Mn  t 10,0010

-------
                                  1446
Rationale et le Laboratoire de Biologie Vegetale du CEA a Grenoble.  On a
etudi£ en particulier les transferts des metaux lourds dans la chalne:
aliment du betail, lalt, produits laitiers: I1experimentation a montre
que , pour tous les elements envisages, 1*animal consitue un filtre
effioace entre la pollution chimique du milieu et le lait (tableau II)
(4).

3«2. Comnercialisation et distribution
     Le mouvement dee prodoits alimentaires, bruts ou transformed, est un
parametre important a. prendre en consideration, soit qu'il aboutisse a une
dilution des produits issus d'une zone susceptible d'etre contaminee, soit,
au contraire, qu'il aboutisse a une ooncentrationt sur certains marches,
de produits provenant de zones interess^es par des re jets industriels et
urbains.
     Un modele perraettant d'estimer I1 influence des circuits de collecte,
de transformation et de distribution des produits alimentaires BUT le
facteur de dilution d'une contamination intervenant au stade de la produc-
tion agrioole est actuelleoent a I1 etude et sera test£ sur des examples,
car il doit etre adapts aux problfemes particuliers poses par chacune des
grandee categories de produits alimentaires a 1'un ou 1'autre des stades
de leur cheninement.
     En ce qui ooncerne le stade de la collecte, I1accent a d€ja et£ mis
sur les produits cerealiers, notamment le b!6 et la farine* H a etS
montre, par exeople, qu'en France, les zones d* exportation du b!6 6taient:
le Centre (Orleans), la Picardie et la Champagne, les regions importatri—
ces etant les zones urbaines industrialisees; Paris constitue la plaque
tournante de la commercialisation de la farine (3)* Pour les produits
laitiers, on s'appuie sur des etudes realisees sous 1'Sgide du Ministfere
de 1*Agriculture (6), fournissant un panorama de 1'Industrie laitiere,
qui montre une certaine specialisation des grandes installations dans tel
ou tel type de fabrication, selon la region ou elles sont implantees*
Pour les mouvements des produits fruitiers et ISgumiers, des donnees
sont obtenues aupres du Centre Francais du Commerce Exterieur.

-------
                                   1447
3.3. Aspects alimentaires et diete'tiques
      Une enquSte realisee par CHESTA dans diverses regions permet  de se
faire une idee de la variability des regimes alimentaires dans les  pays
de la Communaute Europe'enne (7). On a constate que, contrairement
peut-Stre aui macroelements indispensables,* les apports alimentaires re-
latifs aux oligoelements indispensables sont relativement peu variables
d'une region a 1'autre. Kais le probleme des elements polluants reste
pose.
      Les aspects dietetiques n'ont pas pu etre abordes par I1Association.
Kais le Service de Recherches Toricologiques et Ecologiques du C.E.A. a
entrepris des etudes experimentales sur le devenir des radionucleides au
cours de la preparation des aliments, recherches dont on peut s'inspirer
dans le cas des me'taux lourds (8).

4. CONCLUSIONS
      II apparalt tres interessant de poursuivre, dans le meme esprit,
les recherches encore sporadiques effectu6es sur les transferts de
me'taux lourds. Des modeles inspires de ceux mis au point dans le cas de
polluants radioactifs permettent d'evaluer les quantit^s D ingerees d'un
element, en I'unitfi de temps, si on connalt les quantit^s Qi de chaque
vecteur alimentaire i ingere en 1'unite de temps, les facteurs de trans-
f ert P. milieu  >" aliment et le niveau de contamination X du milieu:
      1                  T|
                   D "X \  Q. Fj  (n vecteurs alimentaires).
                         i
On pourra determiner le niveau limite Xp dans le milieu, si on peut
apprecier la quant it e maximal e ingerable D^>. De premieres etudes de
sensibilite sur D montrent que les parametres les plus sensibles sont
ceux qui modifient les valeurs des facteurs de transfert, en particulier
les parametres impliques par la transformation, la commercialisation, la
distribution et la preparation culinaire des aliments.

-------
                                  1448
                          R EFEREHCES
(l)  LACOURLY G. - Problemes pos4s par I1 evaluation de la capacite ra-
     diologique limite d'un milieu ricepteur - Principes  genSraux
     d'approche. La Radio6cologie appliquee a la protection  de  1'Horome
     et de son environnement. C.E.E..  Luxembourg 1972  (EUR 4800 d.f.i.e)
     459-475.

(2)  Association EURATOK-CEA "Pollution du milieu anbiant" - Rapport
     Annuel 1973f en preparation.

(3)  AUBERT M.f BITTEL R. et alii - Utilisation d'une  chains tropho-
     dynamique de type nSritique & mollusques pour I1etude des  transferts
     des polluants metalliques - 6eme  Colloque International d'Oceano-
     graphie Medicale (Porto-Roz, 1973), a parattre dans  la  Revue intern.
     Oceanogr. medicale.

(4)  BRUAHT C. et alii - Etude de quelques mStaux lourds  dans la  chaSne
     herbe-produits laitiere au moyen  de I1activation  neutronique et  de
     I1absorption atomique - AIEA, Utilisation des techniques nuoleaires
     dans I'fitude comparative de la contamination des  aliments  et de
     1'environnenent (HelBinki. aottt 1973) SH 175/4*

(5)  Association 10RA7GM-CEA "Niveaux  de contamination radioactive du
     •ilieu ambiant et de la chain e aliment aire - Rapport Annuel  1968
     (HJR 4411 f), 1970.

(6)  Kinistere de 1*Agriculture et du  Developpeoent rural (France),
     Direction des Industries Agricoles et Alimentaires - L*Industrie
     laitiere en France, collecte et transformation (1970-1971).

(7)  CRESTA K.t LAGOURLT G. - Determination des regimes alimentaires
     des populations de onze regions de la CommunautS  Europeenne  en
     vue de 1* etude des niveauz de contamination radioactive -  Ketho-
     dologie - RSsultats des enquStes  familiales - Rapport CEA  R  2979
     et HJR 2?68 f (1966).

(8)  BOVARD P., DEXXAS J.t DISDIER R., GRADBT A. - Estimation de  la
     contamination de la ration alimentaire; Stude du  transfert du
     cesium et du strontium a travers  la plante jusqu'au  produit
   *  consommable - AIEAt Environmental Contamination by radioactive
     materials. Vienne (1969) 113r-124.

-------
                               1449
                         DISCUSSION
ANDERSON  (U.S.A.)

     Qu'est-ce qui a produit les facteurs de concentration pour
les moules?  Etaient-elles exposees a des algues contaminees
ou a de 1'eau ou aux deux a la fois?
BITTEL  (France)

     Le phytoplancton a 6te cultive dans de 1'eau de mer con-
taminee.  Les moules ont ete placSes, pendant une semaine, dans
cette meme eau et ont recu, comme aliment, les micro-algues
pre"ce"demment contaminees.  La contamination des moules est done
& la fois directe, par 1'eau, et indirecte, par les aliments.

-------
                               1451
        MERCURY PATHWAYS TO MAN AND "IN VIVO" CONTENT OF
              THE POPULATION OF THE NT- AMIATA AREA

         P, CAGNETTI+, L. CIGNA ROSSI+, G. F, CLEMENTE*
                      AND G, P, SANTARONl"1"1"

+  Laboratorio Radioattivita Ambientale CSN Casaccia, CNEN, Roma
   Italia
++ Istituto Nazionale della Nutrizione, Roma, Italia
ABSTRACT

      In this paper have been discussed and compared the results
referring to the two main routes of Eg 'intake, -inhalation and in-
gestion, and the average amount due to each route has been esti-
mated.

      Furthermore, both the Eg concentrations in some human tissues
(blood, hair) and the Hg excretions (urine, feces) from selected
subjects have been measured.   Such subjects were divided in three
groups according to the degree of their exposure to Hg contamination.

      In order to assess the ratio between Hg contamination levels
in the environment and in man, some correlations have been carried
out among the Hg concentrations, found "in vivo" in the various
biological samples, and Hg amounts introduced in man through in-
gestion and inhalation.

-------
                              1452
1.  Introduction
     In a previous paper, Bombace M.A. et al  [ij , the Hg dis-
tribution and environmental concentration have been studied in
an area selected owing to both its mineral resources and indus-
trial activity.
     In the present paper the study has been mainly focused on
some group of human population, exposed to different levels and
routes (ingestion and inhalation) of Hg contamination.
     Among such groups, those more heavily exposed introduce Hg
mainly through inhalation of either Hg vapor  (people working
near the furnaces where the metallic mercury is extracted) or
mineral dust  (miners).
     In spite of the many papers  (Wilcox et al.^2 , Dinman 3',
Nordberg and skerfving [4J) dealing with the relationship existing
in man between the ingestion of organic mercury and the corres-
ponding level in biological samples (blood, hair, urine) , few
data (Smith et al. [5J )  are available on such relationship due
to inorganic Hg contamination from vapor or dust.

     The Hg intake due to the environmental exposure has been
examinated separately from that due to the working exposure.
     A direct assessment of the Hg vapor inhalation due to the
professional exposure was not possible* in any case, the Hg
concentration in the air, inhaled during the working hours by
the subjects employed near the furnaces, has been estimated by
applying to the urine and blood Hg concentrations in such subjects,
the correlation found by Smith et al.[sj.
     The subjects of the control group have been selected in
order to be exposed to Hg intake from the environment only with-
out any professional exposure.

2.  Materials and Methods
     The subjects selected for this study were divided into
the following groups:

-------
                              1453
- Group A:  seven male individuals  (29-54 years aged) working
            near to the furnaces, where the metallic mercury
            is extracted; such subjects are those involved in
            the heaviest professional exposure, owing to the Hg
            air concentration presumably higher in this point
            than in any other of the mine.
- Group B:  thirteen male miners (4O-45 years aged) working at
            the extraction of the Hg ore.  The inhalation ex-
            posure to Hg vapor is less for this group than for
            group A and the main exposure for this group is due
            inhalation or ingestion of the ore's dust.
- Group C:  eight control individuals of both sexes, without any
            professional exposure,  18-30 years aged, selected
            among the relatives of  the subjects considered in
            groups A and B.
     Blood  (lOcc), hair  (few mg), 24 hours urine and feces
samples have been taken from every  subject.  The total meals
eaten by  the subjects of groups A and B have been  collected
during  the  three  days preceding  the excretion collection date.
Every precaution was taken in order to avoid contamination dur-
 ing  samples collection.
     Each sample  was  measured by the non destructive neutron
 activation  analysis method described  in a previous paper
 (Bombace  M.A.  et  al[lj).
     The  treatment  of  the  samples varied  according to  their
 nature.   Few  grams  aliquots  of  feces  were  irradiated after  a
 preliminar  sterilization*  in a  gamma  irradiation facility
 (137Cs  source);  the urine  volumes  (20 cc)  were irradiated with-
 out any previous  treatment.   The blood samples have been separ-
 ated by means of  a centrifugation,  in two parts,  the serum  and
 the red and white cells, which  were directly irradiated "in toto"
     The  total diets have  been  homogeneized and sterilized  in
 a gamma irradiation facility as the feces samples, and then
 liophilized;  a small aliquot (2-4 g)  was then irradiated.   The
 human  hair  were irradiated after a careful and repeated washing
 in aceton and ether.
 *Dose  to  the  samples __ 5 MRad

-------
                              1454
     The average Hg inhalation for the control group C, was
calculated by estimating the air Hg concentration in the region
around the mine, by means of the Pasquill formulas and of the
following parameters:
a)   the wind rose, referred to the different stability cate-
     gories, was obtained by analyzing both the anenometers data
     for the period march-april 1974 and the clouds mass on the
     site.
b)   the main Hg discharge point was the furnaces stack  (4O m
     high).
c)   the average Hg discharge from the stack was estimated as
     1 mg/sec by means of many different air filtration exper-
     iences around the mine plant.
     The average air Hg concentration, obtained with such
     calculation, is clearly referred only to the analysed
     period, and can be considered as only indicative of the
     order of magnitude of the yearly average value.

3.  Results and Discussion
     In table 1 and 2 Hg concentrations in the analyzed samples
are reported.  The data show a clear correlation of the Hg
content with the degree of exposure.  As the blood is concerned,
the individual Hg concentrations for subjects of group A are
much higher than those reported for the normal population by
Skerfving [4^ and GoIdwater et al. [V]; group B on the other hand
resulted much less contaminated, with 1O subjects over 13 having
Hg blood concentrations less than 5 ppb: such a value can be con-
sidered as a limit value for the 75%-77% of the non exposed
population  (Goldwater et al. [el ; Skerfving [4J).
     The blood Hg concentrations found in the groups A subjects
can be related to the chemical  form of the contaminant: the
highest values are,  indeed, measured in the serum, with a serum/
red and whith blood  cells ration almost equal to a value of 2
 (tab.3).  Such a result could be attributed either to the high
reaction speed of inorganic mercury with the seric proteins

-------
                                          TABLE 1
                           Hg concentration in blood  (ng/ml)
Number of
Subjects
7
13
8
Group
A
B
C
Blood
mean + S.E. range
59 +11 37 4 123
* <5 v 36
<5 <5

mean +
93 +
*
<5
Serum
S.E. rancre
27 46 T 256
< 5 T 53
<5
Red and white
mean + S.E.
44 + 4
*
< 5
cells
range
32 + 63
£5*28
<5
* 3 samples only with Hg concentrations ->5 ng/cc
                                           TABLE 2
                         Hg concentration in urine, feces and hair
Number of
Subjects
7
13
8
Group
A
B
C
urine (ng/ml)
mean + S.E.
251 + 41
4.7 + 1.8
3.4 + 1

range mean
155 T 426 3
<1 * 24.8 1
< 1 T 9.0 0

+
.5
.2
feces
S.E.
+ 1.1
+ 0.5
.25+ 0.07
( /ug/g)
' range
0.6 -5- 8
0.07^- 6
<0.02* 0

mean
.9 25.
.O 4.
.55 1.
hair (
+ S.E.
0 + 6.1
0 + 0.8
8 + 0.4
/ug/g)
' range


7.6 + 50
1.4 f
0.9 +
8
4


.0
.8
.5
                                                                                                   Ui

-------
                             1456
(Guirgis et al. [?] ) , due to its strong degree of ionization, or
to the solubility of the Hg vapor in the lipid fraction  (Clarkson
et al. [a]).  On the other hand, a Hg concentration higher in red
and white blood cells than in serum was observed, when organic
mercury is introduced through a fish diet, according to  Dirman
[3] and Skerfving [4] .  The subjects of group A have also  a much
higher urine and hair Hg concentration than the other groups
(tab.2); this  result is less evident for the Hg concentrations
in feces which show an high degree of variability (tab. 2) .
      As the average ratios, reported in table  3 for  group A,  are
concerned, the following considerations can be done:
a)    the urine/blood ratio is in good agreement with that reported
      by SmithL5]for similarly exposed subjects
b)    the  hair/blood ration is slightly higher  than  those (181 +
      385)  reported by Dinman [3] for  subjects  exposed  to  organic
      mercury,  assumed mainly through the diet.
      Therefore it  could be argued  that, exposures to a  vapor
      of  metallic mercury produce  concentrations  in  some body
      tissues,  as hair,  higher  than exposures to  organic mercury
      as  considered by Dinman fcj
 c)    as  the  ratio  between  the  fecal and the urinary excretion
      (both as  24 hours  total excretion)  is concerned,  no data
      are available in literature.   The  results reported for
      group A seem to show a fairly similar Hg excretion, through
      both feces and urines,  in case of exposures to a metallic
      mercury vapor.
      Other results reported in table 2 and 3 for the groups B
 and C show no indication of a pronounced exposure as for group
 A.  On the other hand, the data of group B indicate a certain
 degree of Hg  contamination, probably due to inhalation  and/or
 ingestion of mineral dust: such a contamination, which  is poorly
 metabolized in lungs and in the gastrointestinal tract  is dis-
 played by the high Hg level in feces only.

-------
                                       TABLE 3

         Group A - Mean values for individual ratios among Hg concentrations
                           in some tissues and excreta
Ratio
serum/red and white cells
urine/blood
hair/blood
feces/urine*
Mean + S.E.
1.97 + 0.36
4.4O + 0.51
441 + 110
1.05 ± Q.16
Range
1.29 * 4.02
3.12 * 7.22
105 * 940
0.70 * 1.53
* 24 hours fecal and urinary excretion
                                                                                                       Ln
                                                                                                       -si
                                       TABLE  4

Urinary, fecal and total Hg excretion compared with the Hg intake through the diet and the
                                inhalation  { ug/day)
Nurrber
of
Subjects
7
8
6

Group

A
B
C
^ ^—
Urinary excretion

mean ± SE
265 ± 49
6.0 ± 2.8
3.8 + 0.8

range
149 ^ 468
1.3 v 24.6
1.3 T 6.1
Fecal excretion

mean ± SE
332 ± 111
66.9 ± 15.8
21.5 ± 4.3
	 — • •. a i •

rarv
104 v
12 v
12 T
'etion

range
104 T 714
12 v 130
12 4- 34
Total excretion

irean ± SE
562 ± 132
73 ± 15
25.2 ± 3.4

range
235 v 1183
14 * 129
18.2 T 35.3
Intake
Diet
mean ± SE
25 ± 6
25 ± 6
25 ± 6
range
6.3 v 98.2
6.3 T 98.2
6.3 T 98.2
Average
inhalation
1000
0,2 *
0,2 *
      *  inhalation due to non-professional exposure,

-------
                              1458
     The hair Hg level for group B are slightly higher than those
reported for the non exposed population  (Dinman J3_ / Skerfving
CO)» while those for group C are in very good agreement with
the normal natural values.
     With reference to the Hg introduced through  ingestion and
inhalation, the following results were obtained:
1)   the average amount introduced by any subject through the
     diet resulted to be 25 /ug of Hg  per day
           *                '
2)   the inhaled amount due to the environmental  exposure, for
     the population living inside the area of 1 km radius from
     the stack  (almost all the considered subjects are living
     in that area), resulted to be nearly 0.2,ug  of Hg per day
                                              —i      3
     (average Hg concentration in air =  1O.1O  J  /ug/m ; air
                                   3       r -\
     inhaled during 24 hours = 2O m   (ICRPL9J).   Such an amount
     is clearly negligible when compared to  that  introduced
     through the diet.
3)   the quantity, due to the professional exposure, inhaled by
     the subjects of group A, was estimated  to be 1 mg of Hg
     per day, on the basis of an air Hg  concentration of O.I
     mg/ra   (air inhaled during 8 hours work  day = 10 m   (ICRP 9  ).
     Some provisional conclusions can be drawn from the values
reported in table 4, as the 24 hours total excretion and the
total Hg intake are considered:
1)   the main Hg intake for group A subjects is due to the Hg
     inhaled during working hours, being negligible the contri-
     bution due to the diet.  Furthermore the Hg  inhaled as a
     vapor is easily metabolized and accumulated  in some tissues,
     the total excretion being less than the total intake  (see
     tab. 4).  As a consequence fairly high  levels were found
     for those  subjects in  blood, urine  and  hairs.
2)   the group B subjects seem to be exposed to a random intake
     of mineral dust, which, being poorly metabolized in the
     lungs and  in gastrointestinal tract, are more easily  ex-
     creted through the feces.  For such a reason the feces/urine

-------
                             1459
     ratio for this group is higher than that found in other
     groups.  The dietary Hg intake for group B is not as
     negligible as for group A.
3)   a substantial equilibrium between total intake (mainly
                                    •
     dietary) and total excretion has been found for group C
     subjects.  As a consequence the Hg levels in the body
     tissues of this group are fairly identical to those found
     in the normal population
4)   the feces/urine ration is function of the kind of exposure;
     such a ratio reaches a value of 1O in group B; it is lower
     in group C than in group B by a factor of 2, and finally
     lower by a factor of 5 in group A than in group C.
     In any case it has to be considered that all the average
data reported in tab.1-4 are mainly indicative of the order
of magnitude, due to both the poor number of subjects examined
and the individual variability of the analyzed quantities; for
such a reason in any table the range of such quantities is also
reported.

Acknowledgments
     The authors are indebted to Dr. R. Rossaro, physician of
the mine of Abbadia S. Salvatore  (Siena) and to Dr. Pincelli
Health Officer of Abbadia S. Salvatore  (Siena) for their cooper-
ation which made possible the present research.

Ref er ence s

I-   BOMBACE M.A., CIGNA ROSSI L., CLEMENTS G.F.,  ZUCCARO
     LABELLARTE G., ALLEGRINI M., LANZOLA E., GATTI L.;
     Ricerca Ecologica sulle Zone Mercurifere del Monte Amiata.
     Igiene e Sanita Pubblica, XXIX,  191  (1973)
2.   tflLCOXW Jr. K.R., ; Mercury  Levels in a Sample of  Michigan
     Residents, in Environmental Mercury Contamination.  Hartung
     R. e Dinman B.D. Editors pag. £J2, Ann. Arbor Science Publ. ,

3.   DINMAN B.D., HECKER L.H., ; The dose-response relationship
     resulting from exposure to Alky Mercury Compounds, in :
     Hartung R. e Dinman B.D. Editors, pag. 29O, Ann. Arbor
     Science Publ.,  (1972)                  	

-------
                              1460
4.    SKERFVING S., :  Normal Concentrations of Mercury in Human
     Tissue and Urine in: Mercury in the Environment; Friberg L.
     e Vostal J. Editors, Chap. 6, pag. 109, CRC Press (1972)

5.    SMITH R.G., VORWALD A.J., PATIL L.S., MOONEY Jr. T.F.,:
     Effects of Exposure to Mercury in the Manifacture of
     Chlorine, Ann. Ind. Hyg. Assoc. J. 31, 687 (197O)

6.    GOLDWATER L.J.,  LADD A.C., JACOBS M.B.,; Absorption and
     Excretion of Mercury in Man: VII, significance of Mercury
     in Blood, Arch.  Environ. Health j>, 735 (1964)

7.    GUIRGIS H.A., STEWART W.K., TAYLOR W.,; The Binding of
     Inorganic and Organic Mercury Compounds (HG 203) to
     Constituents of Normal Human Blood, in: Environmental Mer-
     cury Contamination, Harting R. e Dinman B.D. Editors,
     pag. 239, Ann,Arbor Sciences Publ. (1972)

8.    CLARKSON T.W., et al.: Biological Effects of Mercury
     Compounds: Discussion Paper in: Environmental Mercury
     Contamination, Harting R. e Dinman B.D. Editors, pag. 239,
     Ann Arbor Science Publ.,  (1972)

9.    I.C.R.P. n.2 Pergamon Press, Oxford  (1959)

-------
                               1461
             ACCUMULATION AND RELEASE OF  PETROLEUM
             HYDROCARBONS BY EDIBLE MARINE ANIMALS

                 J, W, ANDERSON AND J..M, NEFF

Department of Biology, Texas A & M University, College  Station,
Texas, USA
ABSTRACT

     The objectives of this investigation were to determine the
petroleum hydrocarbon composition of the tissues of oil-exposed
marine animals and to determine the rates of petroleum hydro-
carbon release when animals were returned to oil-free sea water.

     When exposed to dispersed #2 fuel oil in sea water, oysters
rapidly accumulated a wide spectrum of different hydrocarbpns.
The mono- di- and tri-methylnaphthalenes were accumulated to
higher concentrations in the oyster tissues than were the other
di- and tri-aromatic hydrocarbons or individual paraffins.  When
oysters were returned to oil-free sea water, they released 90%
of the accumulated paraffins in 24 hours.   Naphthalenes were
released to background levels in 28 days.

     When exposed to water-soluble fractions of #2 fuel oil,
clams, shrimp and fish rapidly accumulated naphthalenes in their
tissues to levels one or more orders of magnitude higher than
the exposure concentrations.   When returned to oil-free sea
water, they released the naphthalenes to low or undetectable
levels in from 6 to 14 days.

     These results are discussed in relation to the potential
health hazards arising from the consumption of oil-contaminated
marine animals.

-------
                                    1462
1.  Introduction
    Pollution of the marine  environment by petroleum and petroleum hydro-
carbons  (PHCs) may pose a  health harard to consumers of fish and shellfish
products.  Petroleum is a  complex mixture of  thousands of different hydro-
carbons  (HCs) and related  substances, some of which are known to be highly
toxic and/or carcinogenic.   Several  investigations have shown that when
marine animals are exposed to oil-contaminated sea water they rapidly ac-
cumulate PHCs in their tissues, often to levels many times higher than
those in the exposure water. Blumer e_t aJL.  fl] reported that oil-contami-
nated oysters, Crassostrea virginica, retained the accumulated PHCs in
their tissues permanently.   However, more recently it has been shown that
when oil contaminated oysters £. virginica (Stegeman and Teal, [2]) or
mussels  Mvtilus edulis (Clark and Finley, [3]) were placed in oil-free
sea water, they rapidly released the accumulated  HCs.  Little information
was provided by these investigators  concerning the PUC composition of the
tissues  of oil-exposed animals.  Such .information is essential in predic-
'ting potential health hazards arising from the consumption of oil-contami-
nated marine animals.  An  Investigation was  therefore initiated dealing
with the PRO composition of oil-exposed marine animals and the relative
rates of release of different PHCs when the  animals were returned to oil-
free sea water.
 2-  Methods and Results
     A group of oysters was exposed to dispersed 12 fuel oil in a 46 1 flow
 through exposure tank for 8 hours.  The dispersed oil reached a concentra-
 tion of approximately 400 ppm in the aqueous phase of the exposure tank.
 After the exposure period 3 oysters were sacrificed and their soft tissues
 pooled for HC analysis.   The remaining oysters were placed in large aquar-
 ia containing filtered oil-free sea water.  At different time intervals
 up to 672 hours (28 days) 3 of these oysters were sacrificed for tissue
 hydrocarbon analysis.  Unexposed oysters from the same population were al-
 so analyzed to ascertain the background levels of PHCs present.  Quanti-
 tative determinations of  tissue PHCs were conducted by gas chromatograph-
 ic mass spectrometric methods (Warner, [4]).  The results of this experi-
 ment are summarized in Table 1.  The unexposed oysters contained a total
 of 2.4  ppm of naphthalene and alkylnaphthalenes, the di- and tri-methyl-
 naphthalenes being the dominant forms present.  After 8 hours exposure  to
 dispersed 12 fuel oil, the  animals  had accumulated a total of 312 ppm

-------
              Table I.  Concentrations of different petroleum hydrocarbons in the tissues of
                        oysters Crasaostr«a virginica after 8 hours exposure to a 12 fuel oil-in-
Time
(hours)
Exposure
0
8
Depuration
3
6
24
120
672
water dispersion and at different times following return to oil-free
Petroleum Hydrocarbon Concentration (ppm. JIK/B wet wt.)
n-P

—
235

156
68
18
10
___
N

0.2
14.7

12.0
7.3
6.5
8.2
_—_
1-MN

0.1
8.7

8.4
5.1
5.7
4.7
___
2-MN

0.3
15.0

12.0
7.3
7,6
6.8
0.1
22.

1.0
21.8

22.7
13.2
14.8
13-4
0-5
TMN

0.8
9.1

10.8
5.7
9.5
4.9
0.9
B.

	
0.3

0.3
0.1
0.2
0.1
	
MB,

	
0.5

0.4
0.2
0.2
0.1
	
I

	
1.0

0.7
0.4
0.5
0.2
	
M7

	
1.2

0.7
0.2
0.7
0.1
	
DBT P.

	 	
0.3 1.9

0.3 1.3
0.1 0.6
0.2 1.2
0.1 0.4
	 	
sea water.
MP

—
1.9

1.3
0.6
1.3
0.4
	
DMP

	
0.3

0.2
0.1
0.3
0.2
*
Tote

2.4
312

228
109
67
54
1.5
*n-P, n-paraffins; N,  naphthalene;  1-MN,  2-MN,  1 methyl- and  2-methylnaphthalene; DMN, dimethylnaphthalene;
 TMN, trimethylnaphthalene;  B, biphenyl;  MB.  methylblphenyl;  F,  fluorine, MF, methylfluorlne; DBT, dibenzo-
 thiophene; P, phenanthrene; HP, methylphenanthrene;  DMP, dimethylphenanthrene.
                                                                                                                     ON

-------
                                     1464
  100
a
a
   10
o
c
o  1
u  '
3
                                 9-
      1    24     10   20 30   O r1    2    4     K>   20  40    100 200 400
             Exposure           7/me  ( hour$ )        Depuration

   Figure 1.   The concentrations of  total  naphthalenes  (TNs) In  the tissues
   of clans,  shrimp and  fish at different  times during  exposure  to water-
   soluble fractions of  12 fuel oil  and  following  return  to oil-free  sea
   water.
        —A  Clams, Rangia cuneata.  exposure  concentration  3.4  ppm TNs.
             Shrimp, Penaeus aztecus.  exposure concentration 0.7 ppm  TNs.
             Fish,  Pundulus similus. exposure concentration 1.9  ppm TNs.
 of PHCs in their tissues.  The dominant HC type present at this time was
 the normal paraffins.  These were represented on the gas chromatogram by  a
 large unresolved envelope and a series of resolved peaks representing the
 C12~C23 8eries °£ n-paraffins.  The dominant resolved n-alkane present was
 n-pentadecane (5.4 ppm).  The different naphthalenes were present at con-
 centrations ranging between 8.7 and 21.8 ppm while the other di- and trl-
 aromatic UCs were present at much lower concentrations.
     When the oysters were returned to oil-free sea water they rapidly re-
 leased their accumulated oil UCs.  More than ninety percent of the  n-para-
 ffins were released in 24 hours.  The aromatic HCs were released much more

-------
                                      1465
slowly.   From 24 to 672 hours Che naphthalenes were the dominant HCs pre-
sent In the oyster tissues.  After 672 hours of depuration only small a-
mounts of mono-, dl- and tri-methylnaphthalenes remained.  In similar ex-
periments with other marine species and with three other oils, South
Louisiana crude, Kuwait crude and Venezuelan buaker C, the pattern of hy-
drocarbon uptake and release was similar to that described above.  In all
cases, naphthalenes were accumulated to high concentrations by the organ-
Isms and were retained longer than other aromatics or paraffins.
    These results indicate that the concentration of total naphthalenes
(TNs) can be used as an indicator of petroleum contamination of marine
animals.  In subsequent experiments, accumulation and release of TNs by
clams, shrimp and fish were investigated.  An ultraviolet spectrophotome-
tric method for the determination of tissue naphthalenes was used (Neff
and Anderson, in preparation).
    Groups of clams, Rangia cune_ata_, brown shrimp Penaeus aztecus and fish
Fundulus similus were placed in synthetic sea water containing different
concentrations of the water-soluble fraction of //2 fuel oil.  The animals
were exposed to the water-soluble fraction for 2 to 24 hours and then
placed In oil-free sea water.  At different time intervals during and af-
ter exposure some animals were sacrificed and the concentration of TNs  '
in  their tissues was determined (Figure  1) .
    The patterns of uptake, retention and release of TNs were somewhat
different for the three species of animals examined.  The shrimp accumu-
lated TNs extremely rapidly.  The maximum concentration of TNs  (43 ppm)
was reached after only one hour of exposure.  The shrimp then began  to
release the accumulated compounds back to the exposure water.  The clams
accumulated the materials more slowly but continued to do so at a rela-
tively constant rate for the 24 hour period.  In 24 hours the clams accumu-
lated an average of 34 ppm TNs in their  tissues.  During the 2 hour expo-
sure period to a #2 fuel oil water-soluble fraction, the fish accumulated
TNs very rapidly to average levels of 43.2 ppm.  In all three cases the
maximum tissue concentration was one or  more orders of magnitude higher
than that of the exposure water.
    When returned to oil-free sea water, all three species rapidly released
the accumulated TNs from their tissues.  The fish and shrimp released  the
PHCs much more rapidly than did the clams.  All three species were able to
reduce  tissue naphthalenes to low or undetectable levels within  14 days.

-------
                                   1466
In subsequent experiments clams and fish were shown to completely depurate
accumulated hydrocarbons within 3 to 4 weeks.
3.  Conclusions
    When marine animals are exposed to oil-contaminated sea water, they
rapidly accumulate a wide variety of different petroleum hydrocarbons in
their tissues.  The mono- dl and tri-methylnaphtnalenes are the individual
petroleum hydrocarbons accumulated to the highest concentration in all spe-
cies and with all the oils so far tested.  These are also the PHCs retained
for the longest time when the animals are returned to oil-free sea water.
However, these aromatic hydrocarbons are released, reaching undetcctable
levels within 2 to 4 weeks.  The pattern of uptake, retention and release
of the carcinogenic HC, benzo[a]pyrene, has been shown to be similar to
that of the TNs in clams (Anderson et al., unpublished observation), mus-
sels (Lee et al., (5]) and fish (Lee et al., [6]).
    Naphthalenes have also been shown to be the most toxic component of oil
to marine animals (Anderson e_t. al., [7]) and are known to have substantial
human toxic!ty (Gerarde [8]).  Since marine animals would be expected to
accumulate these HCs to very high levels following an oil spill, consump-
tion of recently contaminated marine animals could pose a health hazard to
man.  However, rapid release of these compounds from marine animals probab-
ly also takes place in the natural environment, rendering the animals fit
for human consumption after a relatively short period of time.  The rapid
release of accumulated hydrocarbons by marine animals also indicates that
under natural conditions, food chain accumulation of PHCs would not occur.
This conclusion is supported by the observation of Burns and Teal [9], that
there was no relationship between tissue PHC concentration and trophic lev-
el among marine animals in the Sargassum community.

-------
                                     1467
                                References
 1. Bluner, M.,  Souza,  G.,  Sass, J., "Hydrocarbon pollution of edible
    shellfish by an oil spill",  Mar. Blol.. 15, 195 (1970).
 2. Stegenan, J.J., Teal, J.H.,  "Accumulation, release and retention of
    petroleum hydrocarbons  by the oyster Grassesfrea vlrelnlca",  Mar.
    Blol., 22,  37 (1973).
 3. Clark, R.C.  Jr., Finley,  J.S., "Uptake and loss of petroleum  hydrocar-
    bons by mussels (Mytllus  edulls) In laboratory experiments",  unpub-
    lished manuscript,  (1974).
 4. Warner, J.S., "Semiquantltatlve determination of C,_ to C3Q aliphatic
    and aromatic hydrocarbons in marine organisms," Anal. Chan.,  (in pressX
 5. Lee, R.F.,  Sauerheber,  R., Benson, A.A., "Petroleum hydrocarbons:
    Uptake and discharge by the  marine mussel Mytllus edulis",  Sci.,  177,
    344 (1972).
 6. Lee, R.F.,  Sauerheber,  R., Dobbs, G.H., "Uptake, metabolism and  dis-
    charge of polycyclic aromatic hydrocarbons by marine fish", Mar.  Biol..
    17, 201 (1972).
 7. Anderson, J.W., Neff, J.M.,  Cox, B.A., Tatem, H.E., "The effects of oil
    on estuarine animals: Toxicity, uptake and depuration, respiration"
    in symposium. Pollution and  the Physiological Ecology of Estuarine
    and Coastal Water Organisms, sponsored by E.P.A. and the Belle W.
    Baruch Coastal Research Institute of the U. of S. Carolina, November
    1973 (in press).
 8. Gerarde, H.W., Toxicology and Biochemistry of Aromatic Hydrocarbons.
    Elsevler Publishing Co.,  New York, 329 pp., (1960).
 9. Burns, K.A.  and Teal, J.M. "Hydrocarbons in the pelagic Sargassum
    community",  Deep Sea Res.. 20, 207 (1973).
     The authors Dish to  acknowledge  the  support  of this research
by the  American Petroleum Institute and the National  Science
Foundation  (Grant  So. GX  37344).

-------
                              1468
                          DISCUSSION
PICHE  (Canada)
     Has Dr. Anderson investigated the nature of the depuration
process which he described?  Is it the result of a simple wash-
ing out or rather through assimilation and conversion to other
forms in the metabolism of the organisms studied?
ANDERSON  (U.S.A.)

     In fish  (Lee et al. , 1972) and more recently in crabs
 (Corner et al. , 1974) the metabolic break-down of naphthalenes
has been demonstrated.   In molluscs this conversion has not
been found and the longer time to complete depuration by these
animals may indicate that "washing out" is the only mechanism
available.
PACKHAM (U.K.)

     Has any  work been done  on the  tainting  of the organisms  as
a result of fuel oil contamination?  Is  there any evidence  that
a dangerous degree of contamination would make the organism
inedible?


ANDERSON  (U.S.A.)

     Several  of us in,the US working with hydrocarbons  and
marine animals have considered conducting a  study relating  chem-
ical analyses of tissues to  animals selected as being tainted by
a taste panel.  Persons experienced in this  type of  study tell
roe that there is considerable variability between the hydrocarbon
taste capabilities of individuals.  I would  estimate that most
individuals could detect levels between  O.Sppm and l.Oppm of
these aromatic compounds  (naphthalenes).


LAO  (Canada)

1.   What  is  the concentration level of  BaP  that you found  in
the sample?

2.   Could you describe the  analytical  technique  (e.g.  GC con-
ditions) you  use for determining  hydrocarbons.

-------
                              1469
ANDERSON (U.S.A.)

 14  I should have noted that our studies with BaP utilized
C   - labelled material and therefore all data were derived from
analyses of radioactivity.  The clams were exposed to a concen-
tration of O.O3O ppm and attained levels of approximately 3 ppm.
We observed a constant decrease in tissue levels of animals
maintained in clean sea water.  The radioactivity reached back-
ground levels at about 48 days.  It would be well to verify
radioactivity data with chemical analyses, but as you know, these
determinations are extremely expensive and time consuming.


BITTEL  (France)

     Can a correlation be established between the accumulation
of petroleum products in molluscs and the amount of fatty matter
they contain?


ANDERSON (U.S.A.)

     We have not correlated lipid content of our animals with
the levels of accumulated hydrocarbons.  From the study by
Stegeman and Teal  (1973) there does appear to be a relationship
between the maximum extent of accumulation in oysters and their
lipid content.   In their study the oil was supplied over a
period of 50 days, while our exposures were quite short-tern
 (hours).  To control for differences in  lipid content and repro-
ductive state all animals used in a given experiment  are collec-
ted from the field at the same time.  It is possible  that in-
dividual differences in the hydrocarbon  content of exposed bi-
valves may be due to variations in lipid content, but on short
exposures the amount of time spent open  and filtering is per-
haps more important.

-------
                           MODELLE
                           MODELS
                           MODELES
                           MODELLI
                           MODELLEN
Vorsitzender - Chairman - President - Presidents - Voorzitter
                          B. PRINZ
                (Bundesrepublik Deutschland)

-------
                              1473
           NONLINEAR MODELS IN MULTIVARIATE ANALYSIS
                APPLIED IN EXPOSURE ASSESSMENT

                         V, STEPANEK

Czechoslovak Research Centre for Environmental Pollution Control,
Prague
ABSTRACT

     In the analysis of data on human exposure to environmental
pollutants there is a need for techniques seeking a solution
between a set of input variables and a categorized dependent
Variable (e.g. health effects).   If the dependent variable is
stochastict we are concerned with its expectation as a function
of the variables defining the conditions.   We restrict ourselves
to the polynomial response surface y =   (X-3...3X ) and find the
estimates of the parameters of the model.   Furthermore an esti-
mate of the variance is obtained and the adequacy of the model
is tested.

     Experimental designs from which the coefficients of the
model can be estimated are discussed and their standard forms are
shown.   It is emphasized that we need not  limit ourselves to
designs that are simple factorial and that  we can use replicate
results only for some sets of the input  variables to obtain an
unbiased estimate of the residual variance.
     Two or more designs  can be compared  by  comparing  the  accuracy
of estimation of    (Xlf...,X ) expressed  by  the  variance of the
estimator.   The comparison is sometimes  simplified  by the use
of rotatable designs^ in  which the  variance  of  the estimator
depends only on the  "distance" of the  point  (X^3...fX  ) from the
centre of the design.
     The response model will be illustrated  on  examples of med-
ical examinations in selected groups  of population.

-------
                             U74
  1. Introduction
       The exposure-effect relationships offer the applica-
tions of a variety of statistical methods. Some of them re-
quire the use of the technique of random processes while
others can be plausibly expressed by the adoption of suit-
able methods of statistical multivariate analysis.
     This paper is aimed at models with a set of explanatory
environmental variables such as climate conditions, age, sex,
educationt housing, social conditions etc. available for the
exploitation of aultivariate data. Appropriate analysis is
applied once the mathematical model for the data has been
specified*
   2. The Theory
      2*1  The Response Surface
      The model used is
where /xii»*2i»"* *xai*yi^ represent the values of q "inde-
pendent" defining variables xltx2,.*.,x  and one "dependent"
stochastic variable y- for the i-th individual in a random
sample of size p* The z£s are assumed to be mutually inde-
pendent and normally distributed random variables with ex-
pected value zero and variance G?* The A *s are unknown
constants. The parameters D. account for any departures from
linearity* In order to be able to test whether the D^s are
zero we have to obtain an unbiased estimate of Q>. This can
be done only if we have available more than one observation
y. for at least one set of values of the controlled variables
zli'"*">zoi" Tnua ve hOT* tne modification of our model
                       q
              = flo +
                ' °
with at least one n^ greater than 1. We can introduce N=^"
     Then the mean is

-------
                           1475
and so
           vik " y"i. = zik ~ *i.
Since the difference yik - y^ is expressed without rest in
terms of the errors z's, the "Within Arrays" mean square
                                        2
can serve as an unbiased estimator of G*  having /N - k/ de-
grees of freedom.
     The remaining "Between Arrays" sum of squares can be
split into two parts, one that is due to multiple linear
regression and one that reflects the departure from linea-
rity with q and /p-q-1/ degrees of freedom respectively.
The ratio of the latter and the "Within Arrays" mean square
is used for the test of the existence of the D?s and is com-
pared with the critical values of the F-distribution with
(/p-q-1/t/N~-p/) degrees of freedom.
     In most cases of estimating health effects of environ-
mental pollution we can restrict ourselves to the model
when y^ is a polynomial in the x»s not proceeding beyond the
terms of second order. Such a model can be written as
                                 .2 A
with i=l,2,...,p;k=l,2,... .r^ /ni> 1 for at least one i/.
     Since the x's are controlled  variables we will treat
the model with the technique of multiple linear regression.
In the appropriate analysis of variance the ratio of
to 'the "Within Arrays" mean square is used as an F-test for
the adequacy of the model./ b^ and b-. are the fitted values
of fli and /?£•» the coefficient at  the sum is calculated
asp-£l + q + q+ 4q(q - 1)J« Obviously p has to be

-------
                             1476


  greater than ^(q+l)(q+2.). /

      2.2  The Experimental  Design
      Our model can be considered as a second-order design
for which it is possible to  obtain the estimates of the /? ' s
A 3 complete factorial is such a design /Cochran and Cox £1J/
and the terms  /&X.J and fii**^* correspond to linear end
               f t) V      *  VV V
quadratic main effects of the j-th factor while the term
           corpe8P°nd8 to *ne interaction between the j-th
and m-th factors.
     The estimators of the coefficients and their variances
simplify by appropriate linear scaling of the controlled
variables if the levels of the factors are equally spaced.
     It should be noted that we are not bounded to designs
that are simple complete factorials. If we admit the equally
spaced levels of the controlled variables, we need not have
the identical number of replications at each level. For
examole, we can extend the design by adding two or more ob-
servations at the point around which the design is centered.
We can then use these replications to provide an unbiased
estimate of  or and to test the adequacy of the model.
     In assessing experimental designs for fitting response
surfaces we are interested in the accuracy of estimation of
the expected values of the dependent variables y^. It is na-
tural to measure it by the variance of the estimators
This is a function, V/x^, ...,x /, say, of xlt...,x . Its
form depends on the experimental design. Two or more designs
can be compared by means of this function. This comparison
is sometimes simplified by the use of rota table designs
     •' ,            '  .                  ^^^^^^^^^^^^^^^
for which V/x,,...,x/ is a function of  4* x ^ only, i.e.
                    q                   0*1 J
it depends only on the "distance" of the point /x^i •••»*-$/
from the centre of the experiment. It is useful to have such
designs available /Box [2 J/.
     For second-order designs with two factors the rotetabi-

-------
                            1477
lity is achieved by arranging the experimental points in the
vertices of a regular polygon centered around the origin(0,0)
and at this origin /at least two observations here/. Apart
from these designs there are rotatable designs with equally
spaced points on each of two circles centered at the origin*
The radii of the circles have to be in definite ratios /de-
pending on the numbers of the points on each circle/ for the
designs to be also orthogonal /in the sense of providing
uncorrelated estimators of the regression coefficients/*
These values can be found in Box and Hunter f3j*
     It can be shown that in the general case a completely
balanced rotatable design of second order can be constructed
using the following combinations of values of q controlled
variables /again the centre of the design is at the origin/*
First we take all 2q points with coordinates /+c,0,..,0/,
/0,ic,0,..,0/,...,/0,0,.,,-cA Then we add the 2^ points
/*d,*d,..,^d/ to these points. The constants c and d are
chosen so that      .
                    f •«
     These designs do not provide an estimate of residual
variance unless at least two observations are added, cor-
responding to values of the controlled variables at the ori-
gin /0.0...,0/.
   3* The Data
      The response model will be illustrated on examples
of various medical examinations. The complete data cannot
be shown here for the limited space of the manuscript.

   REFERENCES
   1   Cochran and Cox, Experimental Designs, 2nd ed.,
                        Wiley,1957
   2   Box,G.E*P., "The Exploration and Exploitation of
                   Response Surfaces", Biometricst 10 /1954/
   3   Box and Hunter, "Multi-Factor Experimental Designs
                       for Exploring Response Surfaces",
                       Annals of ^ath.Statistics. 28 /1957/

-------
                              1478
                          DISCUSSION
JACOBSEN (U.K.)
     Have I understood correctly that your model pre-supposes
that the explanatory, "independent" variables are subject to
control?

     If so, does this not imply that the model is irrelevant
to most situations in environmental health research, where the
hypothesised explanatory variables are, characteristically not
subject to control by the experimenter?


STEPANEK (C.S.S.R.)

   *  In environmental health research we encounter both situations,
those with controlled explanatory variables and those lacking
the control.  My model is of course applicable only in the first
case, which is clearly stated in the assumptions made in the
first section of the paper.  This is applicable to all methods
of those fields of multivariate analysis I tackled in my paper.

     The situations, in which the model can be adopted, are
frequent in the studies on exposure-effect relationships and
are mentioned in the paper.

     In the other case of uncontrolled stochastic variables an
approach using methods of structural analysis is required but
we are mostly limited to the linearity of the models.

-------
                               1479
   A LABORATORY MODEL SYSTEM FOR THE STUDY OF THE ACCUMULATION
        OF CHEMICALS IN THE AQUATIC FOOD-CHAIN, EXEMPLIFIED
                    ON ff-HEXACHLOROCYCLOHEXANE

                J,H, CANTON, P,A, GREVE, W, STOOFF
National Institute of Public Health Bilthoven, The Netherlands
ABSTRACT

     The accumulation and elimination behaviour ofr'-hexachloro-
cyclohexane (HCH) was studied on a laboratory scale food chain
model, consisting of an alga-species  (Chlorella pyrenoidosa),a
crustacea-species (Daphnia magna) and a fish-species  (Lebistes
reticulatus, or "guppy").   The direct accumulation by these
species from the water was studied as well as the accumulation
through the food chain.

     Accumulation and elimination took place within 96 hrs for
all three organisms tested',  from water with low concentrations
of \-HCH accumulation took place even more quickly^ viz. in 3 hrs.

     Accumulation factors (defined as ^ncentration in the organism}
                                      concentration in the water
ranged from 60 to 350.   The accumulation factors were concen-
trationdependent for the daphniae and the guppy's where higher-
accumulation factors were found for higher concentrations of
 1 -HCH.   These concentrations were close to the EC ,.^-values
                                                    oc/
(based on paralysis and mortality) found for these  organisms.
The accumulation factor was independent on the '.-HCH concentration
for chlorellat for which organism no EC,Q-value could be estab-
lished (i.e.  lOppm, the solubility of '-HCH in water).

     The reproduction study with daphniae gave the  most sensitive
response:  this was used to come to a proposed ecological limit for
  -HCHf which limit frequently is exceeded in Dutch surface waters.

-------
                                    1480
Introduction
Accumulation of persistent chemicals in the food chain Is an Important
factor In the evaluation of potential hazards to men caused by the pre-
sence of such chemicals in the environment. In order to efficiently study
this accumulation, model systems on laboratory scale must be developed
which give indications on the accumulating properties of a given chemical,
so that, with due precautions, an estimate can be made of the consequences
if that chemical (deliberately or not) is allowed to come into the
environment*
     The ultimate aim of the study is the development of standard test
methods for the evaluation of the environmental hazard of substances, as
is required 'more and more in national and international legislation.
     In this p
scribed, viz.:
In this paper a relatively simple fresh water  model system Is de-
     Chlorella pyrenoidosa    <——   water
             4     (*)         (1)
     Paphnln magna           <—^   water
             I    (5)         <*>
     Leblstes reticulatus     <—r-   water
      (the numbers  In this  scheme  refer to the paragraphs (1) to (5) in the
       Experimental Part, Section  I).
      As a model substance  cr-hexachlorocyclohexane was chosen, as little
 was known about this substance  in literature and as its behaviour was al-
 ready being investigated in our Laboratory on pigs and rats. Furthermore,
 the substance  is frequently found in concentrations of 0.1 - 0.5 micro-
 grams per liter (ppb)  in the Rhine river and its tributaries (Greve
        "  '                 EXPERIMENTAL

 SECTION I;  ACCUMULATION AND ELIMINATION STUDIES
 (1)  Water <—r Chlorella
 Accumulation
 Samples of ca ICr  cells in 100 ml were exposed to three  different concen-
                                                A*
 trations of cr-HCH, viz. 0.01, 0.0? and 0.80 ppm  . After lj>, 30 and 180
 *  A similar model system can be given for salt water. This system will
    not be discussed here in order to keep the paper  In the limits re-
    quired*
    The solubility of a-HCH is 10 ppm at 22 C, which  temperature was also
    maintained during; the accumulation and elimination studies.

-------
                                   1481
minutes the cells were collected by centrifuging during 5 minutes at 3OOO
rev./min. and analyzed for Of-HCH by GLC after extraction with acetone/
petroleum-ether and subsequent clean-up over alumina.
The results are summarized in fig. 1.
     From this figure it can be derived that trie average accumulation fac-
plasma is separated from the cell walls with the aid of an X-presser the
concentrations of a-HCH In these fractions have the ratio 6 : 1. As the
ratio of the fat content in these fractions can be calculated to be about
2.5  : 1 (Northcote e.a. ^~2 7 and Lewin /~3 7 there seems to be a prefe-
rence for accumulation in the cytoplasma, rather than in the cell wall.
This would indicate not only adsorption, but also absorption.
                    wo-  r
                    1
                      J!i
                      111
                      31 /
                      •"i
                      J',
                      >i>
                              30
 180
(aln.)
 pig, i Accumulation of o-HCH by Chlorella pyrenoidosa from Mater with
        different concentrations of a-HCH
    •   concentrations in the algae after exposure to water with 0.80 ppm
        a-HCH
    A   Idem with 0.05 PP« a-HCH
    •   idem with 0.01 ppm a-HCH

-------
                                    1482
Elimination
After having reached the  equilibrium levels the algae were transferred in-
to fresh water  in order to study the elimination pattern. After 1$, 30 and
l8o minutes the algae were collected again and analyzed. The water was
also analyzed:  90$ of the original amount of O-HCH was recovered indica-
ting a negligible decomposition during the experiment.
     The results  are .summarized in fig. 2. From fig. 1 and 2 it can be
derived that both accumulation and elimination take place within 15
minutes.                            ,
                      pp>'» K-HCH 111 the
                      100-
                         I
                         .1
                          \
                          \
                       10
                           I
                           t
                           I
                         -I
                         .\
                          \
                         - I
                           1
                         - I
                           \
                          t
                          I
                          I
                          I
                           I
                            15
30
    180
tUt (Bill.)
  pjg, 2 Elimination of cr-HCH by ChloreUa pyrenoidosa contaminated with
         different concentrations  of a-HCH into fresh water
      U  concentratiore in the algae after transfer into fresh water;
         starting concentration 110 ppm (cf Fig.  !:•)
      A  idem;, starting concentration 11 ppm  (cf  Fig. 1: A)
                starting concentration 1.5 ppm (cf Fig. !:• )

-------
                                     1483
 (2)  Water <—- Daphnla
           ^^^^••^
 Accumulation
 Samples  of ca 500 animals  (1 week old) in  10 liter were exposed to the
 same concentrations of a-HCH as  mentioned  above. Each experiment was car-
 ried out in duplicate. After 3,  24, 48 and 96 hr*s (or more often if de-
 sired) the animals were collected and analyzed for a-HCH. The concentra-
 tion of  a-HCH had to be kept constant by transferring the animals into
 fresh solutions of a-HCH every 48 hrs, as  by that time the concentration
 of a-HCH had dropped to about 8o£ of the original value because of the
 volatility of a-HCH.
 The results are summarized in fig. 3.
     From  this figure an accumulation factor (. concentration in daphniae.
         ,                                       concentration in water  ;
of about 60 can be derived for the O.O1 and 0.05 ppm dosages; for the
0.80 ppm dosage however this factor is about 350. This could possibly be
interpreted as a beginning intoxication unabling the animal to "keep out"
the a-HCH. This is supported by the acute-toxiclty study (section II)
where 0.80 ppm was found to be the EC Q-value on mortality and paralysis.
     Prom fig.  3 it can also be seen that the equilibrium level was
 reached  in 3 hrs for the lower dosages, but after 48 hrs only for the
highest  dosage.
Elimination
After having reached the equilibrium levels the animals were transferred
 into fresh water. After 3, 24, 48 and 96 hrs the animals were collected
 and analyzed, as was the water. 95# of the original amount of a-HCH was
 recovered, so that also in this experiment practically no a-HCH was lost.
     The results are also summarized in fig. 3.

 (3)  Water <—— Guppy
 Accumulation
Samples  of 125 fishes (3 weeks old) in 10 liter were exposed to the same
 concentrations of a-HCH as mentioned in (1). Each experiment was carried
 out in duplicate. After 0.5, 3, 24, 48 and 96 hrs 20 animals were collect-
 ed and analyzed for a-HCH. The concentrations of a-HCH had to be kept
 constant by transferring the animals into fresh solutions of a-HCH every
 48 hrs.
 The results are summarized in fig. 4.

-------
                                1484
            ppn's  K.-HCH In the cj»phni»*
            300
            200
Fig. 3 Accumulation and elimination  of a-HCH by Daphnia magna
       accumulation:
   m  concentration in the daphniae after exposure  to  water with 0.80
       ppm a-HCH
   A  idem with 0.05 PP
-------
                                   1485
         ppm's x-HCH in the guppy's
         200*
                                                     100
                                                time (hrs.)
  Fig.  4 Accumulation and elimination of a-HCH by guppy's
         accumulation:
         concentration; in the guppy's after exposure to water with 0.80
         ppm a-HCH
     4  idem with 0.0? ppm a-HCH
     •  idem with 0.01 ppm a-HCH
         elimination:
         concentrations in the guppy's after transfer into fresh water;
         starting concentration 160 ppm
         idem; starting concentration ?.4 ppm
         idem; starting concentration 1.4 ppm

                                              .  concentration in the g
     From this figure an accumulation factor  (• concentration in water
of about 140 can be derived for the 0.01 and O.Op ppm dosages; for the
0.80 ppm dosage however this factor is about 2?0. Also here  therefore a
tendency to higher accumulation factors can be  observed when the  concen-
trations applied come close to acute-toxic levels  (cf  (2^  and section II)

-------
                                    1486
Elimination
After having reached the equilibrium levels the guppy's were transferred
into fresh water. After 3, 24, 48 and 96 hrs 20 animals were collected and
analyzed. 98-100£ of the original amount of a-HCH was recovered from the
water.
     The results are also summarized In fig. 4. Prom this figure it can be
derived that during the accumulation experiment the equilibrium level was
reached in 3 hrs for the lower dosages, but after 96 hrs only for the
highest dosage level. Also the elimination took place more slowly at the
highest dosage level.

 (4)  Chlorella —> Daphnla
Chlorella's were exposed to water with 0.01 and 0.05 PP° a-HCH; these al-
 gae were  fed to daphnlae which were kept in solutions of the same
 concentrations. The control group was fed with algae which were not equi-
 librated with a-HCH In beforehand. The equilibrium levels were reached in
 5 hrs.
      The results are summarized  in table I. From .this table one can see
 that the levels In the daphnlae  fed with contaminated algae were only
 somewhat higher than in the control group. These results are not unexpec-
 ted , as the non-contaminated algae will accumulate within 15 minutes
 their equilibrium-amount of a-HCH, so that effectively  the "control"
 daphnlae were also fed with contaminated algae.

 (5)   Daphnla	> Guppy
 Young daphnlae of contaminated parents, which had reached their equi-
 librium levels after exposure to 0.01  and 0.05 ppm, were fed to guppy's
 (3 weeks old). The concentrations of a-HCH  in the water In which the
 animals were kept were also 0.01 and 0.05 Ppn. The equilibrium levels in
 the guppy's were reached In 24 hrs.
      The results are summarized In table  II.  From this table a clear dif-
 ference in equilibrium levels can be seen between the guppy's fed with
 contaminated and non-contaminated food (the levels In the guppy's fed with
 contaminated daphnlae is 3-4 times higher!).  In  this respect therefore the
 guppy shows another accumulation behaviour  than  the daphnla (4).
      The control fishes were fed with  Tetrarain,  which food was experi-
 mentally proven not to accumulate any  a-HCH.

-------
                                    1487
TABLE I
Accumulation of a-HCH by daphniae fed with contaminated (respectively non-
contaminated) algae and kept In solutions with two different concentrations
of a-HCH  (average concentrations in the daphniae given + the standard
deviation; n = 8)
    concentration of
    a-HCH in water:
    0.01 ppm
  O.Oy ppm
  Daphniae fed with:
  contaminated algae
  non-contaminated algae
0.68 + 0.1? ppm
0.50 + 0.13 ppm
3.2 ^ 0.6 ppm
2.2 ^ 0.5
TABLE II
Accumulation of a-HCH by guppy's fed with contaminated (respectively non-
contaminated)   food   and kept in solutions with two different concentra-
tions of a-HCH  (average concentrations In the guppy's given + the standard
deviation; n = 6)
concentration of
a-HCH in water:
Guppy's fed with:
contaminated algae
non-contaminated food °
0.01 ppm
5.4 + 0.6 ppm
1.4 + 0.3 ppm
O.Op ppm
27.0 + 3.2 ppm
6.2 + 3.5 ppm
TABLE III
Summary of the toxlcological data on a-HCH determined in the study
organism
Chlorella
pyrenoidosa
Daphnia
magna
Lebistes
reticulatus
parameter
***>
"So
^50
""jo
criterion
growth
mortality
+ paralysis
reproduction
mortality
+ paralysis
average value found with
9556-confidence limits
> 10 ppm*
1.0 (0.8 - 1.2) ppra
0.10 (0.0? - 0.19>ppm
0.8 (0.6 - 1.2) ppm
time of
exposure
48 hrs
48 hrs
12-21
days
48 hrs
 * The solubility of a-HCH in water under the  circumstances of the experi-
   ment is 1O ppm.
 • As food Tetramin was  used  which proved to accumulate no orHCH

-------
                                    1488
SECTION II:  ACUTE AND SEMICHRONIC^ TOXICITY TESTS
Apart from accumulation and elimination studies,  toxicological tests were
carried out in order to evaluate the ecological effect of a-HCH.
     Acute (48 hrs) tests were carried out with Chlorella, Daphnia (1 day
old) and Guppy (3 weeks old) in order to estimate EC,--values. The results
                                                     y*
are sunraarlzed in table III.
     The BCcQ of a-HCH on Chlorella, based on growth effect, was higher
than the solubility in water (10 ppm).
     Daphniae and guppy's showed paralysis after exposure to a-HCH. This
criterion, combined with  mortality,  was chosen as response for the deter-
mination of an BC5Q-value.
     For daphnlae the reproduction proved to be a still more sensitive
criterion: from  fig. 3 it can be derived that the EC-Qon reproduction is
                    •noer  of young daphnlt* p*r pirent
                    (ctiav1it1v*1y)
                     50

                     «0

                     X

                     20

                     1C
                           K) H 12 0 14 ft 16 17 18 19 20 2»
                                  *g* of the parents (d»ys)
  Fig.  5 Reproductive capacity of Daphnia magna as a function of the a-HCH
          concentration in the medium
      O   blank
      A   0.01  ppm a-HCH
      a   0.05   "
0.20 PP«
i.oo  «
2.00  "
                                                      -HCH

-------
                                  1489
about 0.10 ppm (therefore the 0.80 pptn dosage level was not applied  in the
studies mentioned in section I (4) and (5)* as under these circumstances
not enough young daphniae could be obtained for the analysis of a-HCH).
     From the BC^.-value on the reproduction of daphniae (or better  the
EC,_-value, which can be obtained by graphic extrapolation) a tentative
ecological limit for a-HCH can be derived by applying two safety-factors
of 10 (one for extrapolation to the total ecosystem, to other criteria and
to other organisms; one for possible synergistic effects between a-HCH and
other chemicals), in analogy to the approach common in food-toxicology.
     Applied to an EC---value of 0.005 ppm for the reproduction of daphniae
this approach yields a tentative ecological limit of 0.05 ppb for a-HCH.
This value clearly is lower than the "normal" content of a-HCH in e.g. the
Rhine river (0.1 - 0.5 ppb).
 Literature
      GREVE, P.A., "Pesticides  in the river Rhine",  Sci.Total Environ.,
      1, 173-180 (1972).
      NORTHCOTE, D.H., GOULDIMS, K.J., HORNE, R.W.,  "The  chemical compo-
      sition and structure of the cell wall of Chlorella  pyrenoidosa",
      Biochem.J., 70, 591-597 (1958).
      LEWIN, R.A., "Physiology and biochemistry of algae",  pp. 557-368,
      Acad.Press, New York, USA (1962).
                               DISCUSSION

 PRINZ  (Federal  Republic of Germany)
       Did  you make any attempts to develop a mathematical model
 to your experimental  investigations  about accumulation and
 elimination of orHCH?

 CANTON  (Netherlands)
       We didn't  make any attempts in  that direction as  yet.

-------
                              1491
        MEASURES OF ASSOCIATION OF SOME AIR POLLUTANTS,
     NATURAL IONIZING RADIATION AND CIGABETTE SMOKING WITH
                        MORTALITY RATES

              R, C, SCHWING* AND G, C, MCDONALD**

+  Societal Analysis Department, General Motors Research Labora-
   tories, Warren, Michigan, USA
++ Mathematics Department, General Motors Research Laboratories,
   Warren, Michigan, USA
ABSTRACT

     Two methods aye employed to estimate the association of
hydrocarbons, sulfur compounds, nitrogen compounds, natural ion-
izing radiation, and cigarette snaking with some age stratified
and disease specific United States mortality rates for white
males.   The first method is based on a ridge regression tech-
nique and the second on a sign constrained least squares analysis.
The measure of association between these environmental factors
and mortality are quantified as elasticities,  i.e., the indi-
cated percentage change in the average mortality rate correspon-
ding to a 1% change in the average level of the environmental
factor.   Elasticities are estimated for age specific and disease
specific mortality rates, and these values are then aggregated
and compared to estimates corresponding to total mortality rates.
Overall, consistent results are obtained using the above methods
for sulfur compounds and cigarette smoking.   Many of these re-
sults differ considerably from corresponding results obtained
from the ordinary least squares regression analysis, highlighting
the need for applying the appropriate estimation methods.   In
addition to the variables already specified, these analyses take
into consideration the following groups of explanatory variables:

-------
                         1492
Climate - Precipitation, January temperature, July  tempera-
          ture , humidity, and solar radiation.
Socioeconomic - Age, education, sound housing, population
          per household, population density,  % non-white,
          % white-collar, income, and city size.

-------
                              1493
1.  INTRODUCTION
     Multiple regression analyses on large data banks of
mortality rates and pollution measures have indicated an
association between environmental pollution levels and increased
mortality.  Lave and Seskin, Hickey, et al. and Carnow and
Meier are among several authors who have recently studied the
chronic health effects of pollution by means of multivariate
regression analysis including as explanatory variables a
number of interdependent urban factors which affect health.
Neyman has illustrated pitfalls in multipollutant, multi-
locality studies.  Specifically, he pinpoints the problem of
incomplete comprehensiveness of the set of pollutants studied
and the subsequent results in the following statement:  "if
the study involves a certain numbers  of pollutants, say
P..P_,...,P , but neglects another pollutant P_.that happens
to be important, the conclusions regarding P,,P_,...,P ,
                                            -L  ^      S
suggested by even very highly significant findings, may be
completely misleading."
     To address this criticism of previous studies, we have
included in this study a rather broad  (but still incomplete)
list of explanatory variables.  Both gaseous and particulate
forms of pollutants from both mobile and stationary sources
are included in this study along with natural ionizing
radiation, smoking, four climate and ten socioeconomic variables.
     Rienke has pointed out that in many studies where data is
not obtained from a well designed or controlled experiment,
as is the case in air pollution studies involving socioeconomic,
climate and other uncontrolled variables, nonorthogonality requires
that estimation of individual effects be handled by techniques
other than ordinary least squares.  He suggested that ridge
regression, as first described by Hoerl, provides a promising
method for avoiding distortion due to nonorthogonality.  Recently,
McDonald and Schwing have provided an example illustrating
the differences between ridge regression estimates and least
squares regression estimates of  the coefficients in a model
relating air pollution to a total mortality rate.

-------
                              1494
     This paper is primarily concerned with point estimates
of the association of several indices of mortality with
pollution based on a multiple linear regression model
containing 23 explanatory variables.  A description of the
variables is contained (in the next section.  Least squares,
ridge regression and sign constrained methodologies are
described in Section 3.  One of four results sections, (4.4,
Aggregation of Associations) is presented here.  Section 4.1,
Disease Specific Results; Section 4.2, Age Specific Results;
and Section 4.3, Age Specific Results for the Lung Cancer
and Arteriosclerotic Heart Disease Category are not presented
here due to lack of space.
     It is appropriate to emphasize that any regression model
has numerous limitations.  A linear model, as used in this
study, can be considered as only an approximation to a more
complicated underlying model.  This approximation has
greatest validity in the neighborhood of the point where each
variable assumes its mean value.  Further, in any epidemiologies1
study such as this, the explanatory variables cannot be
controlled and thus may be associated in unknown ways with
other hidden but influential variables.  Averaged explanatory
variables and aggregated response variables do not completely
define the exposure patterns and mortality patterns of the
large populations in our study communities.  Migration habits,
diagnostic biases, exposure histories, synergisms and personal
and dietary habits have not been quantified and therefore
are not usually included in global studies of this type.  Most
important, correlations or associations between environmental
factors and health effects do not prove causation.  One can
only infer the degree of association.  Nevertheless, the model
is frequently used as a tool in quantifying the association
of certain environmental factors with health indices such as
mortality rates.

-------
                             1495
2-   DESCRIPTION OF VARIABLES CONSIDERED
     To refine our estimates of the associations of certain
environmental factors with health, three groups of mortality
data for the years 1959-1961 from Duffy and Carroll are studied
as dependent variables.  Briefly, they'are:
     1.  The highest fifteen specific white male disease
         categories which make up the bulk  (65%) of the total
         white male mortality rate.
     2.  Total white male and age stratified white male
         mortality rates.
     3.  Age stratified white male mortality rates due to
         arteriosclerotic heart and coronary and age
         stratified mortality due to lunq cancer, the largest
         and third largest white male disease categories,
         respectively.
Preference is given to using the white male category of the
mortality rates since these rates are usually based on
large populations with a large number of deaths, and are thought
to represent a wide exposure to the various environmental
conditions.
     A total of 23 explanatory or "independent" variables
are used in this study.These  socioeconomic, climate, pollution,
cigarette smoking, and natural ionizing radiation variables
are described in the complete text of this paper.  Summary
statistics of the explanatory variables are also presented
for the 46 Standard Metropolitan Statistical Areas (SMSA's)
in this sample.  Many of the explanatory variables tend
to be highly correlated.  The degree of correlation is particul-
arly high among pollution variables, in part because prevailing
weather factors often determine whether or not pollutants
emitted into a community accumulate or disperse.  For example,
the correlation coefficient between some pollutants is as
high as 0.877.

-------
                              1496
     It is believed that the associations are affected not
only by the long-time average concentration of a pollutant
but also by its physical and chemical form as well as by its
pattern of short-time variation.  Consequently, to investigate
the possible effects of such pattern variation, the mean
and minimum of some of the pollutants are included as separate
variables in the regression analyses.  Difficulties in fitting
the linear model which arise as a result of high correlations
and possible redundancies among explanatory variables are
diminished by the use of ridge regression and sign constrained
least squares regression.
     Though the methodologies employed are intended to
diminish the difficulties encountered when collinear variables
are included, some difficulties or uncertainties in interpretation
remain due to the fact that certain variables serve as
surrogates for variables not included separately.  For example,
the hydrocarbon emissions from automobiles are so closely
linked to carbon monoxide and tetraethyl lead that the
hydrocarbon pollution potential can be considered a surrogate
for these compounds as well.  The natural ionizing radiation
index is related to altitude, a possible factor in mortality.
Finally, the smoking index is possibly correlated with other
stress related habits including alcohol and coffee consumption.

3.    METHODOLOGY
     Multiple linear regression techniques have played
a prominent role in studies of the association between air
pollution and mortality  (and/or morbidity) rates.  The
mathematical model used in these analyses is the multiple
linear regression model, y = xfi -f e.  The quantity y is the
vector of n observations on the dependent variable, x is the
(n by p) matrix of p explanatory variables for the n observat-
ions, 3 is the vector of unknown regression coefficients, and
e is the vector of errors.  The error terms are assumed to
be uncorrelated with a mean equal to zero and common  (unknown)
variance.  The variables are assumed to be transformed so

-------
                              1497
the x'x is in the form of a correlation matrix, where x'
denotes the transpose of the matrix x.  Having specified
this linear form as a basic model, three methods of
estimating the standardized regression coefficients will be
considered:  the least squares method, a ridge regression
method, and a sign constrained least squares method.
     Ridge regression, the second method applied in this
paper, is an estimation procedure based upon § (k) =
(x'x 4- kl )  x'y, k > O, and as such defines a class of
estimators where k is a scalar index and I  is the  (p by p)
identity matrix.  Note that §(O)  is the usual least squares
estimator denoted simply by g.  The ridge estimator minimizes
the residual sum of squares subject to a constraint on the
length of the estimator, V3' (k) B (k) .  In fact, the length of
B(k) is a decreasing function of k.  For k > O, the estimator g (k)
is biased; however, the total variance, i.e., the sum of the
variances of the individual coefficients, decreases as k increases.

     As a third method of arriving at point estimates of regression
coefficients,  the residual sum of squares was minimized subject
to inequality constraints on the seven explanatory pollutant
variables:  hydrocarbon potential, sulfur dioxide potential,
NO2, minimum sulfate, mean sulfate, minimum nitrate and mean
nitrate.   These coefficient estimates were constrained to be
non-negative.   In all cases,  the unconstrained least squares
solution violated one or more of the constraints.  Since
the restricted solution must then lie on at least one of the
violated boundaries, the data was fit subject to all (2)  possible
combinations of variable delations.

4.   RESULTS - POLLUTANTS, PADTATTOM AND SMOKING
     The results from this study for nine of the explanatory
variables are given in three groups of white male mortality
rates:   the fifteen high disease categories; the age stratified
total mortality rates; and the age stratified lung cancer
rates and age stratified arteriosclerotic heart and coronary
rates.   The nine explanatory  variables of concern here are

-------
                             1498
the seven pollution variables, the natural ionizing radiation
variable, and the smoking index variable.  The "measure" of
association presented in this study is an elasticity, i.e.,
the estimated percent change  (either increase or decrease)
in the average mortality rate of the given disease  (and/or
age) category corresponding to a one percent increase in
the average of the given explanatory variable, holding all
other variables fixed at their average values.  The elasticity
of a given mortality rate and explanatory variable is obtained
using the regression coefficient estimates.

     4.1  Fifteen high Disease Categories for White Maples -
          presented in complete text.
     4.2 Age Stratified White Male Mortality Rates -
          presented in complete text.
     4.3  Age Stratified White Male Lung Cancer Rates and Age
          Stratified White Male Arteriosclerotic Heart and
          Coronary Rates - presented in complete text.
     4.4  Aggregation of Associations - To assess the relative
          magnitude of the associations of the explanatory
variables.  The associations of three sulfur pollutants, SO_,
minimum sulfate, and mean sulfate, are grouped together as
are the associations of three nitrogen pollutants, measured
NO_, minimum nitrate, and mean nitrate.
     Similarly, the age specific elasticities are multiplied
by the age specific rate and by the fraction of white males
in the specific white male age group.  These appropriately
weighted elasticities are aggregated over the ten age groups
and over like pollutants to assess the net change in total
white male mortality associated with a 1% change in specific
variables or groups of variables.
     Three different data sets have been used to estimate
the associations of explanatory variables, individually or
as groups, with white male mortality:   (1) the sum of the

-------
                              1499
effects on the 15 most prominent disease categories  (assuming
pollutant association with mortality is confined to the
fifteen disease categories),  (2) the sum of the effects
on each of the 10 age categories and (3) the total mortality
data.  The three data sets indicate that a 1% decrease in
the sum of sulfur compounds is associated with decreases in
average white male mortality rates of  .028, .022 and  .025%
respectively using the ridge method, and .045, .034 and  .031%
respectively using the sign constrained method.  The  sign
constrained estimate was consistently higher in each  set b,
approximately 60%.  These data also indicate that a 1%
decrease in the nitrogen pollutants studied is associated
with decreases in average white male mortality rates  of
.020, .009 and .007% respectively using the sign constrained
method while the ridge method provides associations of
opposite sign with values -.005, -.007 and -.006%.  In
these analyses two positive associations are indicated on
mortality with the hydrocarbon variable.  All other estimates
of the association are either negative or zero.  The  fact
that sulfur compounds consistently show associations  greater
than nitrogen compounds is consistent with another study
which compares these compounds.

     The estimates resulting from a 1% change in ionizing
radiation are highly irregular.  An increase in mortality
as large as .014% is associated with radiation using  the
age summation and sign constrained estimate while a decrease
in mortality of .044% is associated with radiation using the
disease summation and ridge estimate.  Recall that in the
sign constrained method, only the pollutant coefficients
are constrained; radiation and cigarette smoking are  not
constrained.
     Using the same three data sets, a 1% decrease in
cigarette smoking corresponds to decreases in white male

-------
                            1500
mortality rates of  .081,  .092 and .096% with the ridge
method and .069,  .106 and  .11% using the sign constrained
method.  A decrease in the mortality rates corresponding
to an increase in cigarette smoking is indicated for some
disease categories.  These negative quantities are as yet
inadequately understood and deserve further investigation.
Although they might be the result of a "beneficial" effect
of smoking, it seems more  likely they would be, in part,
the result of a reallocation of deaths among the various
disease categories  (e.g.,  increased mortality from arterio-
sclerotic heart at an early age reducing the possibility
of death from cerebral hemorrhage later).  There are, of
course, other possible explanations.  For example, the
linear, additive model may not be entirely appropriate
and/or the data base may be deficient in some not yet
specifically identified ways.
     Aggregates from the age stratified white male lung
cancer and age stratified white male arteriosclerotic
heart mortality rates were calculated.  As in the total
mortality comparisons, the various estimates of the
association of mortality with the sulfur pollutant group
are all positive.  In the  lung cancer category they are
.016,  .041, .017 and .019%.  In the arteriosclerotic heart
disease category they are  .025, .026, .027 and .005%.  Sign
discrepancies again exist  for the associations of mortality
with nitrogen compounds.   The values are -.007, .017, .001
and .000% in the  lung cancer category and -.035, .000,
-i038 and .000% the arteriosclerotic heart disease category.
Generally the associations obtained for the hydrocarbon
pollutant are close to zero, while the corresponding values
for the radiation variables are uniformly negative.  The
values for the association of these diseases with cigarette
smoking are within very narrow bands.  For lung cancer the
associations are  .278, .289, .279, and.302% for a 1% change
in the smoking variable and for arteriosclerotic heart
disease they are  .180, .193, .208 and .228% for a 1%

-------
                             1501
change in the variable.

5.   CONCLUSIONS
     In order to somewhat substantiate estimates of
association, many checks should be made Within a given
analysis.  For example, the internal consistency of the data
itself needs to be examined.  The sensitivity of the parameter
estimates needs to be investigated as either the objective
function (in this study, the residual sum of squares)
and/or method of estimation is modified.  In this study,
consistency of the data and corresponding estimates is
considered by examining disease and age stratified mortality
rates, and then comparing corresponding aggregated results
with the appropriate total mortality results.  The
sensitivity of parameter estimates has been examined
by considering two methods of biased estimates - namely,
a ridge regression technique and a sign constrained least
squares method.
     As noted, there is excellent agreement among the
estimates of the cigarette smoking index elasticity.
The six estimates range from .069 to .110.  Likewise the six
estimates of the elasticity for sulfur compounds also
fall within a relatively narrow range .022 to  .045.  Nitrogen
compound elasticities  (-.007 to .020) vary substantially
in magnitude and sign,  leaving great uncertainty for this
group.  In these 40 regressions, one disease specific rate,
endocarditis and myocardial degeneration, provides a substantial
association of the hydrocarbon index with mortality.  A
comparable association  is not found in the age stratified
or total mortality estimates.

      Based on the data  used in this study, we  conclude  that:
 (1) increased concentration of sulfur compounds are associated

-------
                              1502
with a general increase in the total white male mortality

rate, (ii) increased consumption of cigarettes are also

associated with increases in the total white male mortality

rate, and (iii) associations for nitrogen compounds, the

hydrocarbon index, and ionizing radiation are dependent

on methodology and data stratification.


     A complete report of this study (including a thorough

description of variables, summary statistics, illustrations,

description of results, and a bibliography)  is available
from the authors.
                         DISCUSSION
JACOBSEN  (U.K.)
 1.   What was the correlation coefficient between the variables
 NO   and HC illustrated in your slide?
  Jv

 2.   Would you agree  that there is  a general philosophical
 difficulty in the use of the "ridge" repression method  (apart
 from the problem of biased estimation)?

     It seems to me that "ridge" regression is an attempt to
 "orthogonalise" data  that are inherently non- orthogonal.  How
 can  one expect to extract information  from a data-matrix which,
 by definition, does not contain the required information?


 SCHWING  (U.S.A.)

 1.   The illustration was taken from our earlier publication on
 ridge  analysis.  I'm  sure the correlation coefficient in that
 study  was greater that O.9 and therefore, we see a dramatic
 effect.  In  the data  reported on today, the correlation coef-
 ficients were substantially  lower,  of  the order of O.8.   The
 effects, though less  dramatic, still exist.

 2.   The objective in applying ridge regression is not  to extrao
 information  which does not exist.   The objective is  to  illustrate
 and  avoid  pitfalls  in the  interpretation  of  least  squares  coef-
 ficient estimates obtained from nonorthogonal data.

-------
                              1503
STEPANEK (CSSR)

     Was the model chosen with regard to the large number of
controlled variables or was the linearity of the model tested
at least with respect to some of the controlled variables?


SCHWING (U.S.A.)

     Our 23 explanatory variables were not controlled in this
study.  Although our previous study considered linear and log
functions of some uncontrolled variables, the results reported
today were obtained with a linear model only.

-------
                               1505
           BODY BURDEN OF POLLUTANTS:  COMPARTMENTAL
                       STOCHASTIC MODELS ,

                         A, H, MARCUS

University of Maryland, Baltimore County, Baltimore, Md, USA
ABSTRACT

     The transport of trace pollutants and radioactive tracers
through the body is usually modeled as a multiaompartment linear
differential system with constant exchange rates.   Specific
activity curves in constant exchange rates*   Specific activity
curves in constant environments must then necessarily be mixtures
of exponential functions of time.   Environmental health stand-
ards can thus be set in terms of averaging times appropriate to
the time-scale for exponential decrease ("biological half-life").
Two serious defeats in this concept are exhibited.   First, when
uptake rate depends on the remaining uncontaminated volume in a
compartment, the half-life for uptake decreases with increasing
concentration so that brief exposures to high concentrations may
do less harm than predicted by the usual compartmental model.
This is illustrated by the elimination and uptake of carbon mon-
oxide in rats and humans.    A second difficulty is that specific
activity curves (here interpreted as the probability density for
time of transport of pollutant) are often power functions of
time rather than exponential functions, especially for bone-
seeking elements.   The linear compartmental model is inadequate
and an alternative model,  with transitions governed by a Markov
process and times between transitions determined by a random
walk process, appears to be needed.

-------
                                   1506
1.   Introduction
     A very difficult problem in establishing safe dose levels for environ-
mental pollutants is that the concentration of a pollutant to which an
individual is exposed may fluctuate significantly over times comparable to,
or much shorter than, the characteristic lifetime of the pollutant in the
organ of interest.  The ability of an organ (or perhaps the whole body) to
excrete a contaminant in due course must be explicitly taken into account
in order to estimate the maximal safe dose.  This is usually done by es-
tablishing safe levels based on appropriate averaging times.  For example,
the maximal allowable concentration of carbon monoxide (CO) in the U.S. is
35 ppm for a one-hour average, and 9 ppm for an 8 hour averaging time.  The
averaging tine concept leads to mathematical and practical difficulties in
assessing fluctuating concentrations [1] [23 [3] [43-  If fluctuations in
pollutant concentration occur much more rapidly than the biological half-
life of a critical organ, it is claimed that the organ will not respond
much to these fluctuations - that is, a brief exposure to a very high con-
centration will not cause much harm.  However, this conclusion depends on a
linear model of physiological kinetics which has several serious defects,
as we shall presently demonstrate.  That is, even though brief high expo-
sures do not violate long averaging time standards, they may do much less
damage than expected.
2.   Compartmenta1 Models
     We assume that the body can be divided into n subsystems or compart-
ments, depending on the pollutant.  For carboxyhemoglobin (OOHb) in the
blood due to 00 exposure, we could use n=l (arterial blood) or n=2 (say
lungs, blood; alternatively [5] blood in which CO is bound to Hgb, and mus-
cle or liver in which 00 is bound to other hemeproteins).  For lead and
other boneseekers, we may need n=3 [63:  (mainly) blood, (mainly) soft
tissue, and (mainly) skeleton.  The basic compartmenta1 model is developed
by many authors (see [73 for references).  Let X,(t) be the amount of pol-
lutant in compartment j at time t.  Let V., a constant, be the maximum
volume in compartment j.  Let A. be the endogeneous production rate of
pollutant in compartment j.  Let C,(t) be the external concentration to
which j Is exposed at time t.  We may thus incorporate the effects of dif-
ferent pathways to different organs e.g. lead exposure by both breathing
and eating lead.  Let X, A, C be the corresponding n-vectors, say X1,
X' =(X,	X ).  Let K, L be n x n matrices, and let LC be an n x n

-------
                                    1507
diagonal matrix in which the diagonal term LjjC. is the uptake rate per
unit volume in organ i when exposed to C..  It is usually postulated that
            dx/dt = A + LCV - KX                                      (1)
                  = B - KX                                            (2)
where           B = A + LCV                       r                   (3)
the general solution is
     X(t) = exp(-Kt)X(0) + f exp(-K(t-u)B du                         (4)
                            o
allowing B = B(u) to also depend on time if (as is the case in a real en-
vironment) C(t) depends on time t.  In the usual clinical situation of ex-
posure to a constant environment, we have in general  for constants F.., G.
     X,(O = £  F, , exp(-G.t) + X.(CD)                                 (5)
      i     j=i  ij       J      1
where G  is an eigenvalue of K  (assumed all distinct  herein).  When n=l,
this is  the usual  "half-life" phenomenon.
     In  reality, the uptake  rate depends on the remaining volume in com-
partment j, which  is V. - X., so that instead  of equation (1) we ought to
write
            dX/dt  = A + LC(V-X) - KX                                  (6)
                   = B  - MX
where B  is  the same as  (3)  and
                M  = LC + K                                            (7)
hence                         t
     X(t) = exp(-T(t))X(0) + J  exp(T(u)  - T(t))B(u)  du              (8)
                              o
The random  time T  is  defined by
              T(t)  =  J  M(u)  du                                         (9)
                     o
The basic  representation is different.   In (4), X(t)  is a linear filter of
the concentration C;  in (8), X(t)  is  not  a linear filter of C since the ex-
ponential  elimination rate depends  also on the previous exposure history.
     This  difference is not important when C. is very small; it is impor-
 tant in some  practical situation such as  the elimination and uptake of car-
 bon monoxide. Applying (8) for a one-compartment (n=l) model for COHb  in
 rats and in humans, I found [8] that  (8)  gives an excellent description of
 the data,  including the highly consistent observation that the titnescale
 for elimination (C=0) namely 1/K^, is always longer than the timescale for
 uptake as a result of exposure to C ppm CO,  namely l/Mj^ = ^/^n + LilC^'
 This exhibits a larger assimilative capacity than predicted by (4) since

-------
                                    1508
pollutants are eliminated more rapidly at higher concentrations than at
lower concentrations.  (See Sec.  4 for some examples.)
3.   Power Laws
     A second and most serious difficulty is that much data on radioactive
tracers and trace contaminants is not at all like (5) but is much better
described in practice by a power law
            Xn(t) = F^'*"1                                           (10)
Bone-seeking elements - calcium, lead, strontium are of special environmen-
tal significance - exhibit this power-law behavior [9] [10] [11 ].  A
number of models have been proposed to explain the observed power-law beha-
vior  [123 [13]  [14],  I have developed [15] a flexible 4-parameter random
walk model which, upon adjustment of various combinations of parameters,
can fit all of  the proposed distributions for the time of passage of a
molecule through a compartment - gamma, power-law (skew stable) and expo-
nentially modified skew-stable.  To relate these to various compartments
requires a fundamentally different orientation, in which the allocation of
molecules to various compartments is modeled a Markovian transition pro-
cess, and the residence time between transitions can have any distribution.
This  formalism  is known as a Markov renewal process; its consequences will
be developed elsewhere.  It is sufficient to note here that no variant of
the usual compartmental model (1) yields power-law behavior like (10),  A
safe dose-exposure relation cannot therefore be built on the averaging-
time concept.
      The new formulation of the  problem  is  that  the  specific  activity in
 the ultimately observed compartment  n, say  X (t), will be  taken  as  the
 first passage time probability density function  if  a molecule of pollutant
 is immediately excreted from compartment.   If molecules  recirculate indef-
 initely without being excreted,  then X (t)  may be taken  as the renewal
 density (i.e.  recurrence density) function for compartment n.  We will  now
 consider only excretable molecules.   The Markov  renewal  postulates  are:
      (a) Transitions of a molecule from  compartment i  to compartment j  oc-
          cur independently of each other,  with no memory of earlier tran-
          sitions — i.e., as a Markov process with  transition probability
          matrix P = ((P  )).
      (b)  The passage time from i to j, say T, ,,  is  a random variable whose
          distribution depends on i and j,  but is independent  of  anthing
          e Ise.

-------
                                   1509
     It is conceptually easy -- but difficult in practice -- to derive the
distribution of the first passage time to compartment n given P and the
distribution of the T...  In general, there is no way to obtain power law
behavior without having a power law transition distribution for some essen-
tial precursor transition time T...  This can be^ obtained readily for the
bone-seeking elements, using the concept of a continuously expanding cal-
cium pool developed by BURKINSHAW et. al. [12]; see  [15) for details.  The
main idea in  15  is that the exchangeable calcium pool -- the total num-
ber of calcium molecules available for subsequent transition or elimina-
tion — is increasing essentially linearly with time.  This will automatic-
ally decrease the proportion of contaminant available for transition, but
at a much slower rate than the essentially exponential decrease required
by the constant exchange rate model expressed by (1).  This can be readily
extended to derive power law transition distributions for random numbers
of transitions as well as single-step transitions.
4.   Example: One-compartment Model for Carbon Monoxide
     In  [&]  I used data by PETERSON et. al.  [lo] and MONTGOMERY et. al.  [l?]
to determine parameters for the one-compartment (n = 1) model (8) with
constant exposures at C ppm CO.  X.(t) was the observed concentration of
carboxyhemogloblin in the blood of the subject.
                                   TABLE I
             Parameters for the uptake and elimination of CO by rats
     C (ppm)      KU (min"1)    MU (min"1)   X^OD)    B^/C (min"1)
      150           0.02241       0.02105      0.1622      2.27 10"5
      250           0.02046       0.02294      0.2368      2.17 10"5
      500           0.01968       0.02674      0.4351      2.37 10"5
     1000           0.01964       0.04167      0.6305      2.63 10"5
                                  TABLE II
         Parameters for the uptake and elimination of CO by humans
     Subject     K-. (min"1)     M,  (min"1)   X  (oo)    B    (min"  )
                  L L              Li.            L         L L      f
        I          0.002624       0.002820    0.07756     2.19  10"
        2          0.002737       0.002833    0.08114     2.30  10"4
        3          0.002386       0.002699    0.08556     2.31  10"
     (The subjects were inactive young men exposed to 50 ppm  CO.)
     The model not only proved self-consistent, it also provided an ex-
cellent fit to the data.  Note that in general, M  is  larger than  K..

-------
                                   1510
                              .  References
1     MARCUS, A.H., "Air pollutant averaging times: Notes on a statistical
      modeIV, Atmospheric Environment. 7, 265-270 (1972).

2     GOLDSMITH, J.R.,. TERZAGHI, J.,  HACKNEY,  J.D., "Evaluation of  fluctu-
      ating carbon monoxide exposures", Arch.  Environ.  Health, 7, 647(1963)

3     ROACH, S.A., "A more rational basis for air sampling  programs",  Amer.
      Indust. Hygiene Assoc. J.., 27,  1-12 (1966).

4     SALTZMAN, B.E., "Significance of sampling time  in air monitoring",
      Air Pollution Control Assoc. J., 20,  660-665 (1970).

5     LANDAW, S.A., "The effects of cigarette smoking on total body burden
      and excretion rates of carbon monoxide", J. Med., 15, 231-235 (1973).

6     RABINOWITZ, M.B., WETHERILL, G.W., KOPPLE, J.D.,  "Lead metabolism in
      the normal human: Stable isotope studies", Science»  182, 725-7 (1973).

7     JACQUEZ, J.A., Compartmental Analysis In Biology  and  Medicine. Amer-
      ican Elsevier, New York, 1972.

8     MARCUS, A.H., "Uptake and elimination of carbon monoxide: A one-
      compartment model", (submitted  for publication),  (1974).

9     NORRIS, W.P., TYLER, S.A., BRUES, A.M., "Retention of radioactive
      bone-seekers", Science. 128, 456-462  (1958).

10    ANDERSON, J., OSBORN, S.B., TOMLINSON, R.W.S.,  WEINBREN, I.,  "Some
      applications of power law analysis to radioisotope studies  in man",
      Phys. Med. Biol., 8, 287-295 (1963).

11    WISE, M.E., OSBORN, S.B. , ANDERSON,  J. , TOMLINSON, R.W.S. ,  "A sto-
      chastic model for turnover of radiocalcium based  on the  observed
      power  laws", Math. Biosci.. 2,  199-224 (1968).

12    BURKINSHAW, L., MARSHALL, D.H., OXBY, C.B., SPIERS,  F.U., NORDIN, B.
      E.G., YOUNG, M.M., "Bone turnover model based on  a continuously ex-
      panding exchangeable calcium pool", Nature. 222,  146-148  (1969).

13    WISE, M.E., "Skew probability curves  with negative powers of  time and
      related random walks in series", Statistica Neerj...  25,  159-180(1971)

14    WISE, M.E., "Interpreting both long term and short term power laws in
      physiological clearance curves", Math. Biosci.  (to appear), (1974).

15    MARCUS, A.H. , "Power laws in compartmental analysis,  Part  I:  A uni-
      fied stochastic theory", Math.  Biosci. (to appear), (1974).

16    PETERSON, J.E., STEWART, R.D., " Absorption and elimination of carbon
      monoxide by inactive young men", Arch. Environ. Health, 21, 154(1970).

17    MONTGOMERY, M.R., RUBIN, R.J., "The effect of carbon monoxide inhala-
      tion on in. vivo drug metabolism  in the rat", J. Phannacol.  Exper.
      Therap.. 179, 465-473  (1971).

-------
                              1511
    ANALISI DI REGRESSIONE DEI VALORI DI PIOMBO NEL SANGUE
                 SULL'ESPOSIZIONE ATMOSFERICA

              L. FAVRETTO, L, FAVRETTO GABRIELLI

Istituto di Merceologia, Universita di Trieste, Italia
RIASSUNTO

     In questo lavoro 6 di-scuesa una funzione modello  iperbolica
in alternative! a quelle gia proposte per  I'analisi di  regressions
dei valori di piombo net sangue sull'esposizione atmosfericat net
tentative di migliorare I'interpolazione  con un modello  piu ri-
apondente alia realta fisica della relazione dose-risposta.  . Come
esempio $ considerata I'applicazione a  dati della  letteratura.
ABSTRACT

     This paper discusses a  hyperbolic  model  function as  an alter-
native to those already proposed for  regression  analysis  of lead
levels in blood as a function of atmospheric  exposure.    The aim
is to achieve improved interpolation  with a model  corresponding
more closely to the physical reality  of the dose-response rela-
tionship.   Data taken from  the literature are used in sample
applications.

-------
                              1512
1.    Introduzione
      Da qualche tempo sono disponibili dati medi di concentra-
zione di piombo nel sangue, rilevati in campioni provenienti da
aree urbane, assieme alia stima dell'esposizione atmosferica al
piombo particellare proveniente dai gas di scarico degli auto-
veicoli (The Working Group  on Lead Contamination [1]). Con I' ajs_
sunzione che i campioni appartengano ad una popolazione avente
un apporto costante nella dieta, i dati sono stati interpolati
con una funzione esponenziale (Goldsmith e Hexter [ 2] ). Altri
dati si riferiscono a valori di esposizione notevolmente piu
precisi dei precedent!, perch§ ricavati da esperienze di labora-
torio, nelle quali viene simulate e controllato nel tempo 1' everi
to inquinante (Knelson, Johnson, Coulston, Golberg   e Griffin
[ 3] * Cole e Lynam [4]). II modello logaritmico si adatta a de-
scrivere la concentrazione di piombo nel sangue in funzione del
tempo di esposizione, quando quest'ultima viene mantenuta ad un
 valore costante (Knelson et alt 13]). Recentemente e stata pro-
 posta la funzione iperbolica quale curva interpolante (Favretto
 Gabrielli {5]) . Tale funzione contiene un parametro di satura-
 zione, in accordo con il raggiungimento di una concentrazione
 costante di piombo nel sangue dopo un tempo prolungato di espo-
 sizione (Cole e Lynam [ 4] ) .
       In questa ricerca vengono confrontati i modelli gii pro-
 posti con il modello iperbolico, riportando come esempio la di —
 scussione dei dati di Goldsmith e Hexter [2]  e di Knelson et
 al. 13] .
 2.     Analisi di regressione
       L'analisi di regressione e stata effettuata utilizzando
 un programma di minimi quadrat! per la stima dei parametri non
 lineari (Marquardt, Baumeister, Sheldon e Stanley t6]) su un
 calcolatore CDC 6200 del Centre di Calcolo dell'Universita di
 Trieste. Come indice di adattamento dei dati alia funzione mo-
 dello 6 stato considerate lo scarto quadratico medio dei valo-
 ri osservati da quelli calcolati.

-------
                                1513
3.     Risultati  e discussions
      La tabella I riporta  i valori dei parametri calcolati m£
diante regressione non lineare  dai dati di  Goldsmith e Hexter
[2]  con vari  modelli. La  figura 1 riporta come  esempio 1'anda-
mento della  curva calcolata assumendo il modello iperbolico.
Tale modello, scritto in  forma  di equazione "tipo Langmuir", e
stato gia  utilizzato per  descrivere 1 *assorbimento di piombo da
parte del  tessuto osseo  (Mac Donald, Nusbaum,  Ezmirlian,  Barbe-
ra, Spain  e  Rounds [7],  Gitelman e Neumann  [8]).
                              TABELLA I
      Eaempio di regressions non lineare dei valori di piombo  nel sangue
      (y, ug Pb/lOC ml di sangue.) eui dati di esposizione atmosferica (x,
      Ug/m'  di aria;. 39 ooppie di o&eervazioni tratte da Goldsmith e Hex-
      tar [ 2] . «, acarto quadratics/ media.
                                Parametri              „
       Funzione modello                                 •
                           a        b        c

       y - axb *           18.1       0.239       -     ± 3.0

       y - a +  b log x      17.8      13.9        -     * 3.5
       * Proposta da Goldsmith e Baxter nella forma linearissata:
                       log^y - logi|a + b
       La  funzione iperbolica a tre parametri  descrive l'andamen_
 to dei  punti nella stessa  misura di quella  esponenziale,  come
 appare  dal  confronto dei valori di s, mentre  con la funzione  l£
 garitmica 1'adattamento e  minore. Inoltre nella funzione  iperbo
 lica  1'intercetta c =  15.7 ug Pb/lOO ml  di  sangue, stabilisce
 1'apporto della dieta  e di altre fonti,  che determinano il  li-
 vello  base  di piombo nel sangue. Kehoe [9]  e  Azar, Snee e Habi-
 bi [10J  sostengono che 1'influenza del piombo atmosferico sul
 tenore  di piombo nel sangue risulta inferiore a quella  supposta
 in base alia relazione di  Goldsmith e Hexter: il modello  da noi
 proposto consent^  di  interpretare  i  dati differenziando il cori

-------
                                1514
                                      Figura 1

                      Regrei»ion« della concentrazione del  pionbo nel (an
                      gue (y. ug Pb/100 •!)  «ull'eipo»ition« atmoiftrica
                      (z.|if Pb/»' di aria) con la funiione  iperboliea.
                      Lo ccarto quadratico aedio del daci dalla curva in-
                      terpolante * indicate  del tratteggio. Dati tratti
                      da Cold»«ith e Hester  ( 2) .
                 10
20
30
4O
tribute  della dieta  (supposco  costante) dal1 *esposiziona atmo-

sferica.  A basse esposizioni  (z -*•  0)  y tende  all' espressi one

y "  c  +  abx,  cioe  la coacentrazione  di piombo  nel  sangue varia

quasi  linearmente  con 1'esposizione,  in accordo  con le osserva

zioni  di  Goldsmith [llj.  Unfultima caratteristica  del modello

e il parametro di  saturazione  (b).  In teoria  un  limits di  sa-

turazione potrebbe rappresentare il  piombo  nel  sangue in equi-

librio sia con quello presente in  altri tessuti  di deposito,

che  con  una certa  esposizione media.  11 modello  fisico giusti-

fica il  livello di saturazione, dato  che tale  livello si ritr£

va nell'esposizione  prolungata ad  una concentrazione costante

di polluente.

       Le  osservazioni di  Rnelson et  al. [3] possono essere  uti_

lizzate  per tale controllo: i dati hanno il pregio della pre-

cisione  nella variabile indipendente  (tempo),  precisione che

-------
                             1515
rappresenta uno del requisiti per una corretta applicazione
dell'analisi di regressione; viceversa nei dati raccolti da
Goldsmith e Hexter, eterogeneita campionaria ed imprecisione
nella stima di x a basse esposizioni 
-------
                                         1516
                                      TABELLA II
        Analiei di regreaoione non lineare (eu N dati)  dell'aumento del con-
        tenuto di. piombo nel sangue (y. mg in 4.9 litri di sangue)  eull'au-
        mento del aariao totale di piombo nell'organismo (x,  ogj  a  due espo-
        sizioni.  Dati tratt-i da Knel&on et al. [ 3) .  s,  saarto quadrat-ico me-
        dia.
 Funzione modello
                   Esposizione: 3.2 Mg/m5, N " 10   Esposizione:  10.9 yg/m3,  N - 16
 y « a + b log  x *
       abx
      1 * ax
0.205    0.182   ± 0.038

1.09     0.387   + 0.034
0.326     0.556   * 0.070

0.336     1.17    ± 0.067
 • Propoata da Kneleon et al.  [ 3] .
1-Or
                                                              10-9   g/rn
                                                  Figura  2
                              Regressione  dell'aumento  del  contenuto  di  piombo
                              nel  sangue  (y.  yg  Pb/100  ml di  sangue)  sull'aumen-
                              to  del  carico  cotale  di piombo  nel1'organismo  (x,
                              mg  Pb)  a  due esposizioni  atmosferiche.  Lo  scarto
                              quadratico medio  dei  daci dalla  curva interpolan-
                              te  e  indicate  dal  Cratteggio. Dati  tratti  da Knel-
                              *on  et  al. [ 3] .
                                                                 B
                                                         10

-------
                              1517
= 39.6) risulta circa 20% maggiore  del  valore  sull'iperbole

{per x = 10.9, y = 32.8).

4.    Conclusion!
      La presente ricerca vuol  essere  soltanto un suggerimen-
to tendente a sottolineare  1'utilita del  modello iperbolico

nell'analisi di regressions  della  concentrazione di  piombo
nel sangue sul1'esposizione  atmosferica.  Ulterior!  esperienze
di esposizione continua,  anche  a  livelli  piu elevati,  potrebbe_
ro consentire di stabilire  con  tnaggior  precisione 1'andamento

della relazione empirica  piombo nel sangue-esposizione atmo -
sferica continua, che non la raccolta  di  dati  in popolazioni

urbane, dove sorgenti di  variazione incontroi1abili  oscurano

1'andamento stesso.
Ricerca. eseguita con -it  contribute  del Consiglio Nazionale
delle Ricerche
                         Bib 1 lograf i_a

      THE WORKING GROUP ON LEAD CONTAMINATION,  "Survey  of  lead
      in the atmosphere of three urban communities",  U.S.  Dept.
      of Health, Education, and Welfare, Public Health  Service
      Publication N. 999-AP-12  (1965).
      GOLDSMITH, J. R., HEXTER, A. C., "Respiratory  exposure
      to lead: epidemiological  and experimental dose-response
      relationships", Science,  158, 132  (1967).
      KNELSON, J. H., JOHNSON,  R. J., COULSTON, P.,  GOLBERG,  L.,
      GRIFFIN, T., "Kinetics of respiratory  lead  uptake in
      humans", Proc. Intern. Symp. Environmental  Health Aspects
      of Lead, Amsterdam, October 2-6, 1972,  p. 391,  C.E.C.,
      Luxembourg  (1973).
      COLE, J. F., LYNAM, D. R., "I.L.Z.R.0.'s  research to de-
      fine lead's impact on man", Proc.  Intern. Symp.  Environ-
      mental Health Aspects of  Lead, Amsterdam, October 2-6,
      1972, p. 169, C.E.C., Luxembourg (1973).

-------
                             1518
 5     FAVRETTO GABRIELLI L.,  "Problem! igienico sanitari",  in
      Favretto L. , Pertoldi  Marietta G.,  Favretto Gabrielli L. ,
      "Piombo nell*ambiente:  element! per un giudizio",  in
      stampa (1974).
 6     MARQUARDT,  D. W.,  BADMEISTER,  T. Ill,  SHELDON,  J.  A.,
      STANLEY, R.  M. , "Least-squares estimation of nonlinear
      parameters", E. I. Du  Pont de  Nemours, Engineering Dept.,
      Wilmington,  Delaware (1964).
 7     MAC DONALD,  N.  S-, NUSBAUM, R. E.,  EZMIRLIAN, F.,  BARBE-
      RA, R. C.,  SPAIN,  P.,  ROUNDS,  D. E., "Mechanisms in ske-
      letal accumulation of  ions", Arch.  Biochem. Biophys., 43,
      118,(1953).
 8     GITELMAN, H. J-, NEDMAN, W. F., "Lead-hydroxyapatite  in-
      teraction",  The University of  Rochester, Atomic Energy
      Project, UR-551 (1959).
 9     KEHOE, R. A., "Lead intake from food and from the atmo-
      sphere", Science,  159,  100O (1968).
10     AZAR, A., SNEE, R. D.,  HABIBI, K.,  "Relationship of com-
      munity levels of air lead and  indices of lead absorption"
      Proc. Intern. Symp. Environmental Health Aspects of Lead,
      Amsterdam,  October 2-6, 1972,  p. 581, C.E.C., Luxembourg
      (1973).
11     GOLDSMITH,  J. R.,  "Epidemiological bases for possible
      air quality criteria for lead", J. Air Foil. Control As-
      soc., 19, 714  (1969).

-------
                              1519
        SUR UN MODELE EPIDEMIOLOGIQUE, STDCHASTIQUE ET
             COMPARTIMENTAL SIMULANT L'APPARITION
                ET L'AGGRAVATION DE LA SURDITE

                      M, T, HO, E, QUINOT
Institut National de Recherche et de Securite  (INRS), Paris,
France
RESUME

     Les resultats d'enquSte audom6triques-sonometriquesf dont
on dispose & I'heure aotuelle, sont he'te'rogSnes.   En effett le
eeuil de deficit auditif & partir duquel la surdit£ est prSsum4et
varie:  2SdB dans la Reoommandation ISO P/1999 et 35dB dans le
tableau franyais des Maladies Professionnellest par exemple.
Le deficit moyen pratique lui-m§me donne suivant le cas plus ou
mains d'importance a la frequence conversationnelle 1000 Hz.
II en r&sulte que la frequence F(tfb) des surditSs profession-
nelles, apparue dans une population expose'e a un niveau moyen de
bruit b decibels (A) pendant la duree (t) ann&est diffSre fonda-
mentalement d'une enquSte d I'autre.   Le niveau admissible de
bruit devient alors inpr&cis.

     Le modele stoohastique oompartimental propose permet d'ex-
pliquer ces disparit6s en s'inspirant d'une m&thode 6pid£miolo-
gique qui a fait sa preuve dans le domaine d'une autre maladie
professionnellet la siliaose.

     Le paramStre fondamental de oe module est la probability
d1 aggravation du deficit moyen pratique d'un sujet exposd d un
niveau de bruit donne pendant un temps donne.   Les deficits
auditifs sont regroupSs en classes et le module s'apparente d un
processus markovien de diffusion d'un aompartiment dans un autre

-------
                              1520
     La probability ainsi d&finie est une femotion du niveau
moyen de bruit.   Elle doit permettre de determiner la frequence
dea surdites d craindre pour une dur&e d1'exposition quelconque
et de fixer le niveau moyen admissible de bruit.

     La validation du modele ne pourra Be faire aompl&tement
qu'au jour oft seront connus les rSsultats de l'enqu§te entre-
prise par I'INRS dans I'industrie frangaiee:  cependant, pour
fixer l&s id&es, le modele a &te confront^ avec succSs avec
quelques donnees aotuellement disponiblee.   Dans ce cadre3 une
tentative a ete faite pour £tablir la relation entre I'intensite,
la sensation sonore  et la probability d'aggravation du deficit
auditif.
ABSTRACT

     The results of audiometry and sound-level studies currently
available are heterogeneous.   In fact, the hearing deficiency
threshold, beyond which deafness is considered to exist, varies:
SSdB according to ISO standard R/1999 and 3SdB in the French
table of occupational diseases.   The practical mean deficiency
itself accords varying degrees of importance to the conversational
frequency of lOOOBz.   The result is that the frequency F(t,b) of
occupational deafness in a section of the population exposed to
a mean noise level b in decibelst over a given period of time (t),
differs fundamentally from one survey to another.   The admissible
noise level, therefore, is very flexible.

     The proposed compartmental stochastic model provides an ex-
planation of these disparities, based on an epidemiological
method uhich proved useful in the case of another occupational
disease, silioosis.

-------
                              1521
     The fundamental parameter of this model is the probability
of aggravating the practical mean deficiency in a subject exposed
to a given noise level over a given period of time.   Auditory
deficiencies are grouped according to categories and the model
has some points in common with a Markov process of diffusion from
one compartment into another.

     The probability thus defined is a function of the mean
noise level.   It should make it possible to determine the fre-
quency of deafness which can be expected for any given period
of exposure and to fix a mean admissible noise level.

     The model cannot be considered thoroughly tested until the
results of  the investigation undertaken by  the INKS in industry
in France are known.   However, to obtain an indication of its
efficiency  the model was used successfully  with some currently
available data.   Within this framework an  attempt was made to
establish the relationship  between intensity3  sense of hearing
and probable  aggravation of auditory  deficiency.

-------
                                   1522
1.  Introduction
    Les examens audiometriques et sonometriques permettent de determiner
le deficit auditif moyen pratique des sujets d'une population exposee a un
mime niveau de bruit. Celui-ci est malheureusement defini de fac.ons diver-
ses. En France, on considere la moyenne des deficits aux trois frequences
conversationnelles 500, 1000, 2000 hertz en donnant a. La seconde un poids
double des deux autres. La recommandation ISO R/1999 preconise de son cote
la moyenne equiponderee des trois deficits. Quand le deficit moyen prati-
que depasse une certaine valeur seuil, fixee a 35 dB en France mais a 25dB
dans la reconmandation ISO, il est convenu d'y voir le signe de la surdite.
    II en resulte que la frequence F(t,b) des surdites professionnelles
apparues dans une population exposee 3 un bruit de niveau b pendant la du~
rSe t differe fondamentalement d'une enquete a 1'autre. Le niveau de bruit
admissible devient alors imprecis. Aussi est-il interessant de rechercher
des parametres lies plus directement 1 1'etiologie et aux conditions d1ap-
parition de la surditl professionnelle et d'introduire en particulier une
notion probabiliste plus fine que la simple prevalence F(t,b).
    Dans ce qui suit la population exposee a un certain niveau de bruit b
sera decrite au moyen de classes de dificits auditifs limites superieure-
ment, par exemple, a 15, 25, 35,...dB. L'effectif de chaque classe sera
alors 1'Element d'une matrice ligne :
    N(t,b) - (N,, N2	Nk	Vt,b   '                        (1)
indexee par les parametres t et b. En considerant comne unite 1'effectif
de  la population totale, la prevalence F.(t,b) des sujets atteints de sur-
dite professionnelle au seuil i de deficit auditif sera egale a  :
    Fj  (t.b) - I   Nj                                               (2)
L*incidence pourra etre  represented  par  la probabilite R.(t,b) dt de pre-
senter  une surdite entre les  epoques t et  (t+dt) pour un  sujet que  1'on
sait non  atteint 1 1'epoque t :
                   dF.(t,b)       dZ.(t.b)
    H.(t.b) dt  -  —	:  —	                           (3)
                    1 - Fi(t,b)     dt             ,
ou  Z.(t,b) est la fonction auxiliaire   Log 	            (4)
    1                                        (I  - F.(t.b))
    Les figures  1 et 2 donnent  les valeurs de F. et  Z. pour  deux enquetes,
d'une part celle de MORIN portant sur 149  forgerons  exposes  a des bruits
de  1OO  dB environ avec des pointes frequentes & 115  ou  120 dB  (i -  1,2,3
correspondent &  des  seuils de 15, 25, 35 dB), d'autre part celle de la re-

-------
                                  1523
        PREVALENCE DE LA SURDITE AUX SEUILS 15.25.35 dB
              EN FONCTION de la DUREE d1 EX POSITION
                 POUR UNE MEME AMBIANCE SONORE
               (D'apres les resultats de  J MORIN }
                                               25        30              0
                                                Dur«» d'«ipo«it)on [«nn««f]  t
FIGURE.1

-------
                                  1524
  PREVALENCE DEL A SURDITE AUSEUIL25dB EN FONCTION DELA DUREE

            D' EXPOSITION ET OU NIVEAU MOYEN DU BRUIT

            (D'apri* !•• re'sultats d'enquitet  contcnua dan*
                    la Recommandation  I.S.O R/1999J
                                                        (niv*«u d* bruit 90 dBA)

                                                                      0/25-
FIGURE 2

-------
                                    1525
commandation ISO R/1999 (i =1 correspondant au seuil de 25 dB ; b = 90,100,
105,110 dBA).
    La probabilite par unite de temps Jl.(b,t) varie avec le temps d'exposi-
tion et le niveau b, sauf d'ailleurs pour de faibles valeurs de seuil.
    Le but du modele propose est de determiner une probabilite elementaire
qui rende compte des valeurs de F.(t,b) et Z.(t,b), tout en ne dependant
que de b.
2.  Modele stochastique compartimental
    Le modele qui suit est donne a priori. II sera justifie ulterieurement
par ses consequences et par la comparaison avec les donnees des enquetes
audiometrie-sonorcetrie.
    L'etat d'une population a I1instantt  etantcaracterise par la matrice
ligne (N(t,b)) :
    (N(t,b)) - (Nj, N2, N3,..., Nk,...,Nn) t,b                     (5)
on suppose que la probabilite de passage de la classe de deficit auditif k
2 la classe suivante (k+1) (probabilite d'aggravation) entre les epoques t
et (t+dt) est egale a p (b) dt. Cette quantite est une fonction du rang de
la classe de depart et du niveau moyen de bruit auquel a etc expose le su-
jet. Tout se passe dans ce modele comme si tous les sujets places a 1'epo-
que t dans le compartiment correspondant 3 la classe k de deficit auditif
"diffusaient" entre t et  (t+dt) dans le compartiment k+1 (et non dans un au-
tre) avec la probabilite p (b)dt. Le processus de changement d'etat de la
population exposee au bruit est continu, stochastique et markovien. Dans le
cas ou P, (b) est independant de k, il est de plus poissonnien. II sera de-
        K.
nontre d'ailleurs que 1'examen de ce cas particulier est pratiquement suf-
fisant. Les sujets sont repartis entre les differents compartiments suivant
une distribution d'ERLANG generalisee. La probabilite a (t,b)dt pour un su-
jet d'atteindre le compartiment k entre les epoques t et (t+dt) est definie
par le produit de composition  :  _     ,   v
    ak(t,b) = /^ V^T.b) dT  . e  k~1       ,                      (6)
somme des produits de la probabilite d'atteindre le compartiment k-1 entre
les epoques T et T+dr par celle de le quitter au terme d'une periode compri-
se entre  (t-t) et (t+dt-T>.
Ces produits de composition sont facilement calculables dans le cas poisson-
nien ou p = p  - p2 -...- p. «...  (distribution d'ERLANG classique) :
       (t,b)  .   _!	 pfe-«  tk~2  e-Pt   f (k >  ,)(
     k            (k - 2)  !

-------
                                     1526
Le nombre de sujets qui, a 1'epoque t, n'ont pas atteint le compartiment k
est egal d'une part a la somme N.(t.b) + N2(t,b) + ...+ N   (t,b) et d'autre
part a :
    /^ afc  (t,b) dt                                               (8)
On tire de cette egalite la valeur de N. (t,b) :
    Nk(t,b) - (*t (ak+1(t,b) - afc(t,b)) dt ,                       (9)
avec N (t,b) « e Pl
Dans le cas poissonnien, il vient en particulier :
    N2 - e~pt pt  ;  N3 - jy e~ptp2t2 ...                         (10)
    Dans le cas general, on montre que les valeurs donnees par  (9) diffe-
rent peu de (10) tant que les ecarts de p.  a leur moyenne p sont faibles.
Aussi les  (10) pourront-elles etre utilisees en premiere approximation pour
determiner la frequence des surdites dans chacun des compartiments k a 1'e-
poque t.
    En fixant le seuil de surdite entre les classes de deficit  auditif k
et k+1 et en posant X - pt, la fonction Z s'ecrit :
                                                                  (II)
                          £. t            \*-  ' J *
II vient en particulier :
    Z, - X    Z2 - X - Log(H-X)    Z3 - X - Log(l+X+ j X2)        (12)
et en inversant les fonctions Z., Z_, Z- :
    X - ^(Zj) - *2(Z2) - «3(Z3)                                  (13)
3.  Validation du modele
    Le modele ci-dessus pourra etre valide quand 1'Institut National de
Recherche et de Se'curite sera en possession des resultats de  1'enquete au-
diome'trie-sonometrie qu'il a entreprise. II en sera de meme du  choix des
compartiments eh nombre et en extension. Cependant, il est interessant de
preciser la methode de validation en utilisant les resultats  des enquetes
cities au  $ 1.
    Dans le travail de Horin, en adoptant quatre compartiments  (0-15,
15-25, 25-35, 35 dB et plus), on constate que  les relations (13) sont veri-
fiies quels que soient i et t pour une meme valeur de p egale § 0,11. Ain-
si, les  courbes experimentales Z et F de la figure  1 sont correctement
intetpr€t£es avec le seul paratnStre p qui caracterise le niveau de bruit
dficrit. Les risultats obtenus a partir de la recommandation ISO R/1999
sont d'un  autre ordre. La forme quasi lineaire des courbes de la figure 2
montre en  effet qu'au seuil 25 dB et avec la definition du deficit auditif

-------
                                    1527
pratique adopte par 1'ISO n.(t,b) = dZ/dt ne depend que de b, ce qui per-
met de prendre 25 dB conme limite du premier compartiment (i-1). La defini-
tion du deficit pratique moyen de cette Recommandation donne a la frequence
2000 hertz un role preponderant. On sait par ailleurs que c'est au voisina-
ge de cette frequence qufapparaissent les troubles auditifs precoces (dans
la gamme 500 a 2000 hertz). Aussi, ne faut-il pas s'etonner si les compar-
timents qui rendent compte de la prevalence en fonction de la duree d1expo-
sition sont differents de ceux qui ont ete utilises dans 1'enquete de MORIN.
    La figure 3 montre d'autre part que II..  ,R(t,b) = p(b) depend de b, en
suivant une loi logarithmique :
    Log p{b) - Ab + B ,                                           (14)
ou A et B sont des constantes respectivement egales a 0,0389 et - 5,414.
    En revenant a la definition du decibel en fonction de 1'intensite sono-
                                          — 16         2
re I et de 1'intensite de reference I  (10    watts/cm ), la relation (14)
s'ecrit pour un bruit de IOOO Hz  :
    p(b) - 3,9 x  10"6 x  (I/Io)0'39                                (15)
    Les probabilites p(b) d'apparition ou d'aggravation sont multipliees
par 2,5 environ quand 1'intensite est multiplied par 10. La relation (12)
exprimant Z.(t,b) en fonction du produit p.t, diviser p(b) par un coeffi-
cient k et multiplier simultanement t par le meme coefficient ne change
pas la prevalence, quel que soit d'ailleurs i. En tenant compte de (14),
on pourra done ecrire :
    Z.(t.b) = Z.  (t x 2,5n  , b -  lOn)                             (16)
    Depuis les travaux de STEVENS il est admis que la sensation sonore
(sonie) varie conme une puissance de 1'intensite sonore, soit pour un
bruit de  1000 Hz  [3] et [4]  :
    S =  to"1'2 (I/I )°'3                                          (17)
                   o
    La comparaison de (15) et  (17) montre que la sensation sonore et la
probabilite p(b)  sont proportionnelles en premiere approximation. Cette
relation est satisfaisante pour l'esprit et donne un sens intrinseque a
la probabilite p(b).
    La limite superieure du niveau moyen de bruit admissible sera detenni-
nee a partir des resultats d'une enquete extensive en choississant 1'ende-
nie residuelle acceptable, AMOUDRU [5], HO et QUINOT [fi|.

-------
1528
••
9
•




I VALEURS DE TTib)= p
«W*M»'
« ,0'apres les reiultats d'«n<
contanus dans la R«comm
I.S.O. R/1999)
4
3
2
:,e-
a
8
7
0
5 _J
./
3
2
FIG
0,001





/
/
/



URE 3
80


/
/
/






•B


(b)
V '
^uetes
andation












90







/
Jr








95





/
/










too



/
/
.











1O5
/
/
/














110 & 115


-------
                                  1529
  REFERENCES

1  Estimation de 1'exposition au bruit durant  le travail en vue de la pro-
   tection de 1'audition - Recommandation  ISO  R/1999  (mai 1971) -.

2  MORIN, J., Recherche sur les  alterations de 1'appareil auditif provo-
   quees par le travail industrial  en milieu bruyant.
   Travnil Humain janvier-juin 1965 - N° I et  2-.

3  Expression de 1'intensite physique et subjective d'un son ou d'un
   bruit - Recommandation ISO R/131  (1959) -.

4  STEVENS, S.S., The measurement of loudness.
   J. Acoust; Soc.  Am., 27 pp 815-829 (1955) .

5  AMOUDRU, C.,  La  notion de concentration maximale admissible. Journee
   medicales des Charbonnages  de France. Paris - 6 et 7 novembre 1969.

6  HO, M.T., QUINOT,  E..Apparition  et aggravation de la surdite dans  une
   population exposee au bruit - Simulation par un modele epidemiologique,
   stochastique  et  compartimental -.
   Note Documentaire 895-75-?^-  (Cahiers de Notes Documentaires  de
   1'INRS  (Institut National de Recherche et de Securite)  -  Paris  -
   France)

-------
                              1531
         EVALUATION DE LA CHARGE CORPORELLE EN MERCURE
                   PAR SIMULATION NUMERIQUE
             R, MAGNAVAL"*", R, BATTI+/A, BOUVILLE
+  Association Euratom-CEA, Fontenay-aux-Roses, France
+•*• Dfipartement de Protection, Fontenay-aux-Roses, France
RESUME

     On lvalue la quaniite de mSthyle de mercure  et de mercure
inorganique absorbe"e par la population apr&s  consommation  de
poisaons.   A partir dea analyses effectuSes  sur  lea poisaona
marine p§che"s en Europe, on prend comme teneur mSdiane 0.3 et
0.7 ppm de mercure total, les tro-C8 quarts   de  la concentration  en
mercure Stant sous forme me'thyle'e.   L '^valuation tient  oompte
aussi de la aonsommation de poissons,   Elle  est  simul&e de deux
facons a partir des enquStes alimentaires  effeotueea antevieure-
    f d'une part, en fonation de I'age des oonsommateursy  d'autre
    f en prenant une distribution de la aonsommation par repas
r^partie normalement autour d'une valeur moyenne.   Dans oe eas,
on etudie I'influence de la frequence de consommation du poisson*
elle est introduite dans le calcul  sous forme de  variable  al£a-
toiresimulant les cas de consommations rapproche"es ou espac$es.
Pour aalculer la charge corporelle  dans ces divers cas en  fonation
du tempst on utilise un module  d un  seul compartiment pour I'ab-
sorption du mSthyle de mercure  (absorption 100%,  pSriode 70 jours)
et pour la forme inorganique  (15%,  pSriode IS jours).    Les ni-
veaux decontamination et leurs amplitudes de variation sont discu-
tSs, dans un certain nombre de  oas  traitSs par simulation  nume-
rique sur calculateur,  les charges  corporelles en mercure  different
par un facteur de s&curite" infSrieur d. dix en regard des niveaux
toxiques repertories au Japon.

-------
                               1532
ABSTRACT

     An assessment is made of the amount of methylmeroury  and
inorganic mercury the population absorbs by eating fish.   Using
analyses oarried out on European salt-water fish as a basis, the
median content is taken to be O.Z and 0.7 ppm total mercury,
three-quarters of the mercury concentration being in the methyl
form.   The assessment also takes account of fish consumption
which is simulated in two ways, using previous surveys of food
consumption:  firstly in terms of the age of the consumer and
secondly using consumption distribution by meal based on an
average value.   In this case the effect of the frequency of fish
consumption is studied, and is introduced in the calculation in
the form of a random variable simulating consumption at frequent
or infrequent intervals.   In order  to calculate the time-depen-
dent body burden in the various cases a model with a single com-
partment is used (absorption 100%, period 70 days) for methyl
mercury absorption, and,for the inorganic form,  (15%, period 15
days).   Contamination levels and their range of variation are
discussed and in certain cases processed by digital simulation
by computer;  the mercury body burdens differ from the  toxic
levels recorded in Japan by a safety factor of  less than ten.

-------
                                    1533
INTRODUCTION
Le phenomene de concentration d'un certain nombre de polluants metalliques
ct organometalliques dans les divers maillons de la chalhe alimentaire jusqu'a
1'homme, a ete particulierement bien montre pour le mercure. MSme si le
rejet de ce polluant n'entrafhe de nuisances qu'i une echelle locale, la toxicite
du methyle de mercure est suffisamment severe a faible teneur pour justifier
une etude g6nerale de la charge  corporelle apres ingestion de ce compost.
De toute fa?on, il y a disaccord entre les auteurs, sur la quantity du mercure
qui est sous  forme methyiee dans 1'organisme (1)  (2).  Nous sommes essentiel-
lement guides dans ce travail par 1'approche qu'en ont fait, pour la  premiere
fois SPEAR et WEI (3) en proposant une etude simuiee pour calculer le residu
de methyle de mercure res tan t dans 1'organisme.  On se propose de g6n6raliser
cette methode a la fraction organique et inorganique du mercure tout en n'en-
visageant, com me ces auteurs,  que le poisson comme seul vecteur  de conta-
mination. Nous utilisons,  pour cela, une methode de simulation numerique
Bar calculateur digital (4).

Hypotheses utilis6es
La courbe de repartition de la teneur du mercure  dans le poisson est proche
d'une distribution log normale. Nous pr£ferons  introduire une  distribution
observee par CUMONT (5) sur 360  poissons. La teneur mecliane en  mercure
est 0.3 ppm (Fig.  1). Cette m6me distribution a 6te utilisee avec une teneur
mediane de 0.7 ppm. Dans tous  les cas, la fraction methylSe repr^sente 75%
de la teneur en mercure.
La consommation de poisson par repas est represented par une loi normale
dont la valeur moyenne est 200 g et 1'ecart type 40 g.  La frequence  de consom-
mation du poisson  est introduite dans le calcul,  sous forme de loi binomiale
tvec paqs  1/2; la. valeur m^diane represente  les cas de consommation rap-
prochee (2 repas par semaine),  normale (1 repas par  semaine), espacee
(1 fois par mois).

-------
                                 1534
              n=360
                                                 TENCUR en Ha ppm
                                                (J'opre's  CUHONT)
    Fig. 1 REPARTITION DES TENEURS EN MERCURE DANS LE
           POISSON.
Pour une region fortement consommatrice, nous avons d'une part, utilis6
la quantity consomm6e en fonction de I'Sge, de 1'ordre de 75 g/jour (6,  7)
d'autre part, 250 g de poisson par repas, le tn6tne 6cart type et 2 repas
par semaine. Le modele m£tabolique ne met en  jeu qu'un seul comparti-
ment.  Le m6thyle de mercure est absorb6 £ 100%,  et sa peViode biologi-
que est de 70 jours, le mercure inorganique est absorb^ a 15% et sa p6rio-
de biologique est de 15 jours.
Evaluation de la charge corporelle
La variation de la charge corporelle de 6 individus est suivie sur une pg-
riode de deux annees (Fig. 2).  L'equilibre est pratiquement atteint au bout
de 150 jours dans tous les cas, mais les ecarts  entre les teneurs minima
et maxima atteignent 35%, 10% provenant de la dispersion des concentrations
en mercure dans le poisson.
Le cas d'une region fortement consommatrice a ete etudi6 a partir des don-
nees de CRESTA sur la Vendee (France) (7) le niveau critique atteint est
discut£ par la suite.

-------
CHARGE   CORPORELLE

    mq d. mercure
      •j
                                      Teneur du poUion en mtrourvi
  - 200 3  . Z

         - 1 f»l»
3 .  -     _ 1 u,
                                                                                   .•*••%.•-.     ..   7
                                                                               -.-..•    	•• '-.ft
                               ZOO             300             400             500

                     Fig. 2 VARIATION DE LA CHARGE CORPORELLE EN FONC-


                            TION DU TEMPS.


-------
                                 1536
Charge corporelle et niveaux toxiques
WEI et SPEAR (8) ont determine la charge corporelle totale correspondant
aux teneurs en mercure relevees sur differents echantillons biologiques de
onze cas en d6but d1 intoxication ulterieurement letale: 54 mg, 87 mg,  90 mg,
91 mg, 94 mg,  114 mg, 118 mg,  118 mg,  165 mg,  292 mg, 292 mg et d'autres
cas de troubles nerveux correspondant a une charge corporelle  de 60 mg. En
appliquant un facteur de securite de 10, la charge corporelle critique  sera
de 5.4 a 6 mg.
La recherche des maxima de charge corporelle en mercure sur un echantillon
de population de 1000 individus dont la quantite hebdomadaire ingeree  corres-
pond a la dose tolerable de 0.3 mg propos6e par TO.M.S. (9),  nous montre
qu'un individu sur mille atteint une valeur proche de la charge critique, soit
5.2 mg (Fig. 3). Dans tous les autres cas, elle est tres inferieure. On a vu
que le niveau critique egal ou superieur £ 6 mg est atteint pour  une region
fortement consommatrice (7) et pour une teneur en methyle de mercure de
0.75 ppm dans le poisson. L'extension du modele a la ration complete (10)
entra the rait un doublement de 1'ingestion du mercure total (11) mais la for-
me inorganique y est preponderate (12) et on a vu qu'elle est rapidement
eliminee.

CONCLUSION
Nous avons calcuie la charge corporelle en mercure, resultant  de 1'inges-
tion de poisson en tenant compte de la distribution statistique d'un certain
nombre de parametres:
teneur du poisson en mercure, quantite ing6ree par repas, espacement des
repas.
Par cette methode, nous avons mis en evidence une amplitude de variations
de la quantite de mercure retenue de 35% chez le mdme individu.  La charge
corporelle  en methyle de mercure est preponderante et 1'extension du modele
a la ration complete ne modifierait pas sensiblement les valeurs obtenues
pour cette forme chimique.

-------
                                  1537
        Fig. 3 REPARTITION DES MAXIMA DE LA CHARGE CORPO-
               RELLE EN MERCURE SUR 1000 INDIVIDUS.
150
too
 50
                                                             1&
   1,99
                              3«7
               5,19
                                          CHAR6E CORPORKU-E  «n nu
                                                                 4
     Teneur du poiison en m«rcure ; 0,7 ppm   (0,5 ppm de meHxylmerture)
     ConsomKialion :  200 q . 2
par «em»ine
3,7 m   aUt
Je
       m«iture
     Charge eorporelle maximum :  m
 La recherche des maxima sur un echantillon de 1000 individus consommant
 du poisson deux fois par semaine en moyenne,  contenant 0.7 ppm de mercure,
 montre qu'une charge corporelle de  5.2 mg est atteinte avec une frequence
 de 1/1000 alors que les cas d1 intoxication ont et6 observes pour une charge
 estimee de 54 a 60 mg.

-------
                             1538
                    BIBLIOGRAPHIE
(1)   BIRKE G., JOHNEL A.. PLANTIN L., SJOSTRAND B.,
     SKERFINV1NG S., WESTERMARK T. Studies on humans exposed
     to methylmercury through fish consumption.  Arch.  Environ.
     Health 25, 77-91  (1972).
(2)   KEVORKIAN J.,  CENTO D., UTHE J., HAGSTRIM R.  Methyl-
     mercury content of selected human tissues over the past 60 years.
     Am. J. Pub. Health  63, 11, 931-934(1973).
(3)   SPEAR R.C., WEI E.Methylmercury toxicity: a probalistic assess-
     ment Workshop on mercury in the Western Environment. Oregon
     State University,  lip. Feb. 6 (1971).
(4)   MEYER H.A . Symposium on Monte Carlo methods. John WILEY
     and sons (1956).

(5)   CUMONT G., VIALLEX G.  Contamination des poissons de mer
     par le mercure. Rev. Intern. Oceanogr. Med. 28.  95-126(1972).

(6)   O.E.C.D. Food consumption statistics 1954-1968 Paris (1970).
(7)   CRESTA M., LEDERMAN S., GARNIER A.,  LOMBARDO E.,
     LACOURLY G. Rapport Euratom EUR 4218, 285 p. (1969).
(8)   WEI E., SPEAR R.C.  The fatal dose of methylmercury in man.
     J. Am. Med. Ass. 216, 8. 1347 (1971).
(9)   O.M.S. Evaluation de certains additifs alimentaires et des con-
     taminants: mercure, plomb et cadmium.KiSer. Rapp. Tech. n.505,
     12-17 (1972).
(10)  SCHELENZ R., DD3HL J.F. Quecksilver in Lebensmitteln unter-
     suchungen on taglicher gesamtnahring. Z.  Lebensm. Unters.
     Forsch.  153, 151-154(1973).

(11)  UNITED KINGDOM,  Working party on the monitoring of foodstuffs
     for mercury in food. H.M.S.O.  (1971).
(12)  BACHE C.. GUTENMANN W.,  ST. HOHN L., SWEET R.,
     HATFIELD H., LISK D.  Mercury and methylmercury content of
     agricultural crops. J. Agr. Food Chem. 21, 4, 607-613(1973).

-------
   ZUSA'TZLICHE BERICHTE





   SUPPLEMENTARY PAPERS





 RAPPORTS SUPPLEMENTAIRES





 RAPPORTI SUPPLEMENTARI





SUPPLEMENTAIRE DOCUMENTEN

-------
UNTERSUCHUN6 DER WIRKUNGEN AUF DEN MENSCHEN





           HUMAN EFFECTS STUDIES





       ETUDES DES EFFETS SUR L*HOMME





       STUDI DEGLI EFFETTI SULI_'UOMO





 ONDERZOEKINGEN NAAR EFFECTEN BIJ DE MENS

-------
                              1543
         THE USE OF CANADIAN MEDICARE DATA IN DEFINING
           HEALTH EFFECTS OF THE ENVIRONMENT ON MAN

                        C,  A, R, DENNIS

Prairie Institute of Environmental Health, Regina, Saskatchewan,
Canada
ABSTRACT

     With the development of computer technology health data can
be used to identify effects of the environment on health.    The
Canadian medicare data base, comprised of provincial components,
is composed of the number,  cost and diagnostic code of the phy-
sician services that each resident receives.    Analysis of such
computerized information allows comparison to be made between
populations with regard to the number and cost of physician ser~
vices for defined diagnostic codes or code groupings.

     Saskatchewan medicare data has been used to define the ef-
fects of the agricultural environment on health with particular
reference to accidents and the effects of agricultural machinery,
grain dust and agricultural chemicals.   Medicare data was used
to identify 112 persons treated for methomyl intoxication in 1971;
subsequent interview of the physician and patient allowed for the
^Valuation of diagnosis and insecticide-handling practices.

     The main limitation to the use of medicare data stems from
the fact that health insurance agencies may have no defined res-
ponsibility for the preventive aspects of health, care and there-
fore no commitment to make full use of their data to define
health effects of the environment on man.

-------
                              1544
     The use that can be made of medicare data to define the ef-
fects of environment on health appears initially to be dependent
not only on the ability to precisely define a particular health
hazard and the resulting health effectt but also on the unique-
ness of the health effect,   Medicare data has special value
where precise definitions cannot be formulated but where the data
can still be used to support* or not* a hypothesis that a parti-
cular environmental hazard contributes to the incidence of a
health effect in a population.

-------
                                  1545
1.   INTRODUCTION
     While most health agencies necessarily have been governed by
economic interests in the design and utilization of their health data
bases, nevertheless it is becoming apparent that with the development of
computer technology, this same information can also be used to identify
health problems which may, upon evaluation, contribute to a greater
understanding of environmentally induced health effects.
     The characteristics of optimum quality health (and accident)
statistics may be enumerated as follows:
     (a) The information on which the statistics are based should be true.
     (b) The statistics should relate to total populations.
     (c) The statistics should be detailed enough to identify problem
     areas but simple enough to ensure assembly.
     (d) The statistics should allow access back to individual records
     for detailed evaluation of problem areas.
     (e) The statistics should allow valid comparisons to be made with
     regard to time, place and population,
     Canadian medicare data, particularly the data from Saskatchewan,
possesses in large measure the characteristics outlined.
.2.   HEALTH CARE DATA IN CANADA
     Canada has a national health care data base composed of various
provincial components.  All provinces produce certain information which
is somewhat comparable such as in the area of vital statistics.
Hospitalization data are also compiled nationally.
     The advent of universal medical care insurance coverage in Canada
created a vital component in the health care data base in regard to the
definition of environmentally induced health effects.  Basically,
whenever an insured resident receives a physician's service the physician
is required to identify the diagnosis on a billing card which he renders
to the provincial Insurance agency in order to get paid.  Information
relating to each service becomes a part of that person's "patient
history file".  In a given population the patient history files include
a record of the number, cost and diagnostic codes of physician service to
that population.  In addition they will indicate where the services were
provided as well as Individual characteristics of each patient and allow
comparison with other populations.  Each province produces statistics
from the medicare system for its own purposes as well as to satisfy
federal requirements.  Some produce more than others and some have It

-------
                                   1546
stored over a longer period of time.  The province of Saskatchewan has
been a pioneer in the field of health care delivery systems in Canada
and medicare data have been computerized since 1962 in that province.
Use has been made of it in defining the effects of the agricultural
environment on farmers with particular reference to accidents; tractor
vibration and tractor noise; the effects of continued exposure of grain
buyers to grain dust; and the evaluation of the acute toxic effects of
certain agricultural chemicals.
3.   EXAMPLES OF THE USE OF SASKATCHEWAN MEDICARE DATA
     3.1 Toxic effects of methomyl.  In 1971, when a heavy infestation of
the Saskatchewan rape seed crop by the Bertha army-worm necessitated the
widespread use of the insecticide "methomyl", a project was designed to
Identify and evaluate cases of methomyl Intoxication.  A computer program
modification was introduced for three weeks which identified all
physician service claims for Code 989 (ICDA - eighth rev.).  Cards were
then scanned manually to identify claims for methomyl intoxication and
112 persons were identified.  Each patient and the attending physician
was subsequently contacted personally for evaluation of exposure,
diagnosis, effect and treatment.  A similar program was carried out in
1972 but for a longer period of time.  19 patients were identified in the
first three weeks, 38 in total.  The program not only allowed speedy
identification of patients, (3-4 weeks from date of physician service to
computer identification) but also allowed for the evaluation of causes
of illness and insecticide-handling practices.
     3.2 Health of Farmers.  Health surveys of farmers in Saskatchewan
have indicated a positive correlation between hearing loss, back
disorders, accidents; and increasing farm size (which implies increased
use of farm machinery).  An evaluation of medicare data using a random
sample of Saskatchewan farmers supports the hypothesis that increased
exposure to farm machinery is associated with increasing number and
costs of total physician services  (Table I) and services relating to
specific health effects.
4.   DISCUSSION
     While the value of health statistics derived from medicare data
obviously depends upon the ability and efficiency of the physician who is
providing the diagnosis, it should also be noted that they are based
upon information recorded for payment purposes.  Thus, where a choice

-------
TABLE I
                                   1547
                  NUMBER AND COST OF PHYSICIAN SERVICES
                     TO FARMERS BY AGE AND SPECIFIED
                     FARM SIZE, 1963-1969 INCLUSIVE.

Age Group
(in 1969)

25-34
35-44
45-54
55-64
65 and over

Number of Farmers
<900
acres
458
691
921
734
437
900 +
acres
49
163
173
73
17
Number of Services
per Farmer
<900
acres
17.9
21.7
24.0
28.2
36.0
900 +
acres
17.9
23.9
26.2
39.9
44.7
Cost of Services
per Farmer
< 900
acres
$118.00
145.20
165.40
189.20
263.00
900 +
acres
$115.40
168.60
163.90
276.20
342.90
NOTE:  1 acre - .405 hectare
 must be made between two legitimate diagnoses and their treatments, a
 decision may be made for economic reasons which may prove misleading for
 research purposes; by using large populations these research limitations
 are minimized.  A further limitation to the use of medicare data is that
 only one diagnosis is recorded for payment purposes and a decision often
 oust be made among several reasonable possibilities.  The most important
 limitation  stems from the fact that often the health insurance agencies
 have no direct responsibility for the preventive aspects of health so
 that the commitment to make available medicare data to define the
 interactions between man and his environment may be absent.
      Notwithstanding the limitations, this data source has great
 potential in defining man/environment effects, particularly when the
 health characteristics of different populations need to be compared to
 identify the health aspects.  For example, health effects associated
 with increasing population size and density: infectious and parasitic
 diseases, accidents and poisonings, congenital anomalies and various
 health effects peculiar to occupational groups.  (Table II.)
      The use  that  can be made of medicare data to define the effects of
 environment on health appears at first glance to be dependent not only
 on the ability to  precisely  define a particular health hazard and  the

-------
                                    1548
TABLE II
         COST OF INSURED SERVICES PER 1000 PERSONS (CANADIAN $)
      FOR SELECTED DIAGNOSTIC CATEGORIES AND SELECTED POPULATIONS,
          SASKATCHEWAN MEDICAL CARE INSURANCE COMMISSION, 1971.
DIAGNOSTIC
CATEGORY

I Infectious &
parasitic diseases
V Mental Disorders
Xiy Congenital Anonalies
XVII Accidents, Poisonings,
etc.
TOTAL SERVICES
RESIDENCE GROUPS

RURAL
(POP. 253,822)
$
719
825
286
2,598
29,863

2 MAJOR CITIES
(POP. 267,328)
$
1,113
3,183
521
2,952
48,466
REGISTERED
INDIANS
(POP. 33,478;,
$
3,616
809
529
5,242
43,552
  resulting health effect,  but  also on the  uniqueness  of  the health effect.
  While precision is to be  desired, the special value  of  medicare data Is
  particularly apparent where precise definitions  cannot  be formulated
  and where the data can still  be used to support  (or  not, as  the case may
  be) a hypothesis that a particular environmental hazard contributes to
  the incidence of a health effect (or type of health  effect)  in a
  population.
       I wish to acknowledge the co-operation of the Saskatchewan Medical
  Care Insurance Commission and its staff for facilitating my research
  into the use of medicare data in defining the effects of the environment
  on health.

-------
                              1549
        THE RELATIONSHIP BETWEEN LEAD AND HYPERACTIVITY

                           0, DAVID

State University of New York, Downstate Medical Center, Brooklyn,
New York,  USA
ABSTRACT

     Hyperactive children were compared with, a non-hyperactive
control group on two measures that reflect the presence of body
lead - blood-lead levels and urine levels after challenge by a
single oral dose of the chelating agent penicillamine.   Hyper-
active children had significantly higher values on both measures
than did the controls.   More than half the hyperactive children
had blood-lead levels in the range considered to be raised but
not "toxic", and 60% of post-penicillamine urine levels were in
the "toxic" range.   It is concluded that there is an association
between hyperactivity and raised lead levels;  that a large body-
lead burden may exert consequences that have been hitherto unre-
aliseds  that the definition of what is a toxic level for blood-
lead needs re-evaluation^  and that physicians should look for
raised lead levels in children with hyperactivity.

-------
                                   1550
     Introduction

     As most readers know, hyperactivity is one of many terms used to
describe a complex, multifaceted syndrome.  It is characterized by the
fundamental behavioral manifestation of a high level of motor activity
and is usually coupled with a short attention span, low frustration
tolerance and hyperexcitability.

     The question of causality regarding hyperactivity is quite fluid.
Retrospective studies commonly show that hyperactive children have in
their background a much higher incidence of such events as abnormalities
in the mother during the gestational period; difficult delivery histories;
and postnatal difficulties such as head trauma, encephalitis, meningitis
and other encephalopathies.  Prevalence estimates of this and closely
related disorders vary, but if we hew closely to the prevalence of
hyperactivity per se, approximately 5Z of children in the U.S. alone may
be suffering with this disorder.

     The literature on lead, before our work, contained much that indicat-
ed a lively suspicion held by many authorities regarding associations be-
tween minor brain dysfunction and low lead levels.  It was clear, however,
that no study provided a definitive answer to the question:  Is there a
relationship between asymptomatic lead poisoning or lead poisoning less
severe than that causing encephalitis, and mental Impairment?

     Our investigation was designed to answer as well as we could the
following question:  Do hyperactive children have larger body lead stores
than non-hyperactive controls?  The operational hypotheses tested were
(1) Hyperactive children will have significantly higher blood lead levels
than non-hyperactive controls; and (2) Hyperactive children after chal-
lenge by a heavy metal chela ting agent will have dramatically higher urine
lead levels than controls subjected to the same challenge.

     Patients and Methods

     Children were classed as hyperactive or non-hyperactive on the basis
of a doctor's diagnosis, a teacher's rating scale (the Connera^scale) and
a parent's rating scale (the Wherry^Weiss-Peters scale).  If the three
Indices were not in accord, the two scores that did agree were used to
designate the child hyperactive or non-hyperactive.  If only two scores
were available, the teacher's scale and then the doctor's diagnosis took
precedence, in that order.

     Those children who were diagnosed as psychotic or who had evidence of
significant neurological disease were excluded from this study.  Children
with hyperactivity and an event in their background thought to be a cause
of this condition were considered separately from those with no such his-
tory.  These events were broken into "possible" causes and "highly prob-
able" causes, and these groups were analyzed separately.  Utilizing the
above exclusion process, the following groups emerged: (1) A hyperactive
group without psychosis or neurological disease and with no evidence of an
event known to be associated with the development of hyperactivity ("pure"
hyperactive); (2) The hyperactive group with a "highly probable" cause for
their hyperactivity; (3) The hyperactive group with "possible" cause for
their hyperactivity; and  (4) A group of children whose highly probable
cause for hyperactivity was a history of lead poisoning.  All children in

-------
                                   1551
this group had been treated at least five years previously (with one excep-
tion).  Five of the eight children in this group were just hyperactive*
three had neurological conditions of miId-to-moderate diffuse brain damage,
one with mental retardation.

     The fifth group was a non-hyperactive control group who were judged
on the same criteria as that for the hyperactive group.  (See Table I)

     Blood lead determinations (two) were done by the New York City,Board
of Health lead laboratory by the atomic absorption method of HesselT
Urine lead levels were done by the same laboratory on an overnight sample
collected after a 250 mgm Penicillamine provocation dose.

                     TABLE  1  -  DETAILS  OF  GROUPS  STUDIED

Group


"Pure" hyperactive
Highly probable cause
^
for hyperacti vi ty
Possible cause

for hvperacti vi ty
History of lead poisoning
Controls


No.

&

9


11

8
37
Age (yr.)


Mean
7.8

7.3


7.7

7.3
7.6

S.D.
2. Ok

2.25


1.28

1.73
2.*
Range
3^-11

U-10


6-10

3^-9
3*4-12


M/F

43/12

8/1


9/2

8/0
27/10
Hear, hyperacti vi ty
score

Teacher
3.2

2.8


3.1

3.5
1.6
Parent
3.2

2.7


2.1

3.4
0-8
     Results

     Results comparing the normal control group to each of the four hyper-
active groups is presented in Table II.  Mean differences in blood and
urine lead levels between the controls and 3 of the 4 hyperactive groups
show that the non-hyperactive children have significantly lower blood and
urine lead levels than do the hyperactive subjects.

     The exception to this was the lead levels in the highly probable
cause group.  On the two measures utilized, they showed no significant
difference from the controls.

     The range and frequency distribution of these findings are present in
Table III, in the light of what is now considered to be normal, marginal
and toxic levels.  A normal blood lead value is here defined as 24.5 ug/
100 ml or less.  A normal urine lead level is defined as 80 ug/liter or
less.

-------
                                      1552
TABLE I I—BLOOD-LEAD AND POST-PENI Cl LLAMINE URINE LEAD LEVELS IN FOUR GROUPS

        OF HYPERACTIVE CHILDREN COMPARED WITH NON-HYPERACTIVE CONTROLS

Group
"Pure" hyperactive
History of lead
poisoning
Highly probable cause
of hyper act I vi ty
Possible cause
for hyperact 1 vl ty
Controls

Blood-lead
No. tested
&
8
9
8
37
Mean S.D.
(ug./lOO ml.)
26.23^8.41
M.06±12.18
22»89± 6.60
29.9^17.35
22.l6i9.59
*
P
< 0.01
< 0.001
N.S.

-------
TCBLE MI-FREQUENCY OF RAISED BLOOD-LEAD AND POST-PENI Cl LLAMINE URINE-LEAD LEVELS
Group
"Pure" hyperactive
History of lead
poison! ng
Highly probable cause

for hyperactl vl ty
Possible cause

fnr hyperactl vl ty
Controls
No. of children vvl th bl
(ug./100 ml.)
5-14-5
4 (7%)
0


0


0

4 (11%)
15-24-5
22 (41%)
0


5


3

23 (62%)
25-34-5
19 (35%)
3


3


3

6 (16%)
ood-lead levels
n range:
35-44-5
7 (13%)
3


1


1

4 (11%)
45-54-5
2 (4%)
0


0


1

0
55-64-5
0
2


0


0

0
No.
0-80
!9 (38%)
0


4


2

19 (82%)
of chl Idren v*i th urine
(ug./l) in range:
81-160
19 (38%)
0





4

3 (13%)
161-240
4 (8%)
1





1

'
> 241
8 (16%)
3





2

1

-------
                                  1554
     Discussion

     Authorities have commented on some erroneous concepts regarding
"normal" blood lead levels.  A point commonly made is that most studies
equate the lowest level diagnostic of clinically manifest lead poisoning
with the upper limits of normal.  Unfortunately a level not associated
with overt clinical evidence of toxicity does not have to be normal.
Symptoms nay be overlooked for instance in low level toxicity because no
one yet knows what to look for.  At this point most authorities agree that
a blood lead value below 24.5 ug/100 ml is normal and a value above 55 ug/
100 ml is abnormal (although not necessarily treated) .  Our findings sug-
gest that the arguments for considering any lead elevation above 24.5 ug/
100 ml as dangerous, should receive serious attention.

     A massive single dose of lead can result in death or severe brain
damage, and in many cases a large dose of lead ingested over time can also
lead to very severe brain damage.  It seems reasonable to infer, therefore
that raised levels of lead (not necessarily in the toxic range) present
over a long period could be responsible for the minimal brain damage that
may be present in the hyperactive syndrome.  Our finding of raised post-
penicillanine urine-lead levels in combination with the higher blood-lead
values Indicate that many hyperactive children have had increased body-
lead stores for a long time.  It is conceivable that one consequence of
this constant minimal poisonous assault is hyperactivity.  If we hold
lead exposure responsible for some cases of hyperactivity (an assumption
only), we might predict the following t

     (1)  There would be little difference in lead value between the nor-
mal control group and the "highly probable" cause group (i.e. the group
with a likely cause for hyperactivity other than lead.)
     (2)  The "pure" hyperactive group and the "possible" group would both
show raised lead levels.  The "possible" group differs from the "highly
probable" group in the crucial distinction of a convincing other cause
for hyperactivity being absent.

     These predictions are supported by the results.
          eight hyperactive children who had also a history of lead poison-
 ing had raised blood and urine lead levels at  the time  of the study.  All
 but one had been treated with chelating agents at least five years pre-
 viously.  This finding Indicates that the treatment  of  these lead-poisoned
 children may  not have  been extensive  enough and/or that follow-up proced-
 ures had been unsuccessful.  It may also  mean  that hyperactivity hereto-
 fore thought  to  be a relatively common consequence of lead poisoning is
 not necessarily  a consequence at all, but a condition that is dependent
 on continuing elevations of body lead.

      Might the lead levels recorded be a  consequence of the child's hyper-
 activity rather  than a hyperactivity  cause?  This Interpretation is not
 supported by  the finding that the "highly probable"  cause hyperactive
 group do not  show Increased blood or  urine lead levels. There  are,
 however, too  few children in that group to arrive at a  definite
 conclusion.

-------
                                    1555
REFERENCES.

(1)  Conners, C.K., A teacher  rating scale  for use  in drug studies  with
    children.  American Journal  of  Psychiatry. 1969, Vol.  126  (6),
    pp. 884-888.
(2)  Wherry, J.S., Developmental Hyperactivity.   Pediatrics Clinics oij
    North America.  Aug.  1968,  Vol.  15 (3),  pp.  581-601.
(3)  Hessel, D.W.,  A simple and rapid determination of lead in blood.
    Atomic Absorption Newsletter,  1968,  Vol.  3,  No.  3, pp. 55-56.
(4)   Lin-Fu,  J.S.,  New England Journal of Medicine.  1972, 236, p.  702.

-------
                              1557
    UNTERSUCHUNGEN AN 12,000  PATIENTEN  UBER DEN EINFLUSS
    DER LUFTVERSCHMUTZUNG  UND DES  ARBEITSPLATZES AUF DIE
            HAUFIGKEIT DER CHRONISCHEN  BRONCHITIS

                     U, SMIDT  UND G,  WORTH
Krankenhaus Bethanien, Moers,  BED
KURZFASSUNG

    Die bish.eri.gen  epidemiologischen Untersuchungen Uber den
finfluss der  allgemeinen  und der berufsbedingten luftverachmut-
zung auf die Hitufigkeit  chronischer Atemwegserkrankungen wurden
meist  an Personen  durchgeftlhrt,  die noch im aktiven Berufsleben
Stehen.    Daher ist  dort  nur selten mit fortgesahrittenen Kvank-
heitsstadien  zu fechnen.

    Wir haben aus den  Kvankengeschichten aller Patienten der
Inneren Abteilungen  von S Krankenh&usern dee linksniederrhei-
nischen Steinkohlenbergbaureviers im Kreis Moers aus den Jahren
1S7I und 1972 folgende   Angaben herausgezogen:
    Gesahlecht
    Krankenkasse  (Knappschaft/Nichtknappschaft)
    Lebensalter
    Rauptdiagnose
    Begleitdiagnose:  ohronisahe Bronchitis
    Silikoee
    Behandlungsdauer
     Wohnort  und Strasse mit Hausnummer
    entlassen/gestorben
Von den  so ausgewerteten 12.084 Patienten wurde die Baufigkeit
der Saupt- und Nebendiagnose "chronische Bronchitis" vergliehen
zvischen

-------
                               1558
     - Mannern und Frauen
     - Knappachafta- und Nichtknappechaftaversicherten
     - Patienten mit verschiedenem Vohnort.

     Ddbei vurden zundchst alle Patient en mit Silikoae eliminiert
und auaaerdem auf vergleichbarea Lebenaalter geachtet.   In alien
Krankenhaueern war die Eaupt- und Nebendiagnoae chroniache Bron-
chitis bei den knappschaftsversicherten Mdnnern eignifikant
haufiger als bei den nichtknappechaftlich versicherten Mannern
(im Mittel 1814% : 10,4).    Die Bdufigkeit bei den Frauen lag
niedriger und zeigte in vergleiohbaren Alteregruppen keine aig-
nifikanten Haufigkeitaunterechiede zviachen den beiden Veraioherten-
gruppen.   Diea aprioht gegen einen  Einflue8 dea  Wohnortea, der
bei den Knappachafteveraieherten im  Mittel dichter an den Quellen
der allgemeinen Luftveraohmutsung liegt.   Auoh fur Patienten aus
verachiedenen Wohnorten mit  underaohiedlioher Luftveraokmutzung
lieaa si oh ein Bitufigkeit aunt er 8 chied der chroniachen Bronchitis
nicht atatiatiach aichern.
 ABSTRACT

      Up to the present  time,  epidemiological studies  on  the  ef-
 fect of general air pollution and of occupational expoaure to
 polluted air on the incidence of chronic diaeasee of  the respira-
 tory tract have mostly  been carried out among persons still  in
.active employment.    For this reason advanced stages  of  the  dis-
 eases are seldom to be  expected there.

      We extracted the following data* for the years 1971 and 1972,
 from the medical records of all in-patients in five hospitals  in
 the Moers district of the coal-mining area located left  of the
 lover Rhine:
      Sex
      Sickness benefit fund (miners' provident fund or other)
      Age

-------
                              1559
     Main diagnosis
     Secondary diagnosis:  chronic bronchitis
     Silicosis
     Duration of treatment
     Postal address
     Discharged/died
Among the 12,084 patients investigated, the frequency of chronic
bronchitis as the main and secondary condition was compared be-
tveen:
     - men and women,
     - miners' provident fund members and other patients,
     - patients coming from different localities.

     First, all silicosis patients were excluded, and compara-
bility with regard to age was taken into consideration.   In all
hospitals, chronic bronchitis as the main and as the secondary
condition was significantly more frequent among male members of
miners' provident funds than among other male patients (average,
18.4% as against 10.4%).   The frequency in women was lower, and
in comparable age groups showed no significant difference between
the two insured groups.   This suggests no influence by the place
of residence, which in the case of those insured with the miners'
provident fund, was usually located closer to the source of the
general air pollution.   Also, there was no statistical difference
nith respect to the incidence of chronic bronchitis between pa-
tients from different localities exposed to different levels of
atmospheric pollution.

-------
                            1560

 Methodik
 EB  warden alle Krankengeschichten von denjenigen Patienten
 ausgewertet,  die auch im Einzugsbereich der Krankenhauser
 wohnen und zur Behandlung, nicht zur Begutachtung ins
 Krankenhaus katnen. Da jeder Stationsarzt Patienten beider
 Versichertengruppen behandelt, iat eine unterschiedliche
 Diagnostik in beiden Gruppen unwahrscheinlich.
 Die knappschaftsversicherten Manner sind zu iiber 8o#
 Bergleute (gewesen) und a omit beruflich staubexponiert,
 wahrend fur die nichtknappschaftlich Versicherten eine
 berufliche Staubexpoaition bei der Industriestruktur des
 Kreises Moera unwahrscheinlich ist. Es gibt keine Hinweiae
 auf unterschiedliche Rauchgewohnheiten zwischen beiden
 Versichertengruppen.
 Ergebniaae
 Da  Baa Lebenaalter der Patienten mit der Diagnose "chro-
 nische Bronchitis" in alien Gruppen iiber 60 Jahre llegt,
 sei hier die  Haufigkeit bei den Fallen ab dem 55» Lebens-
 Jahr fur die  H auptdiagnoae "chron. Bronchitis" gezeigt :
 MannerFrauen
 Knappschaft Nichtknappschaft   Knappschaft Nichtknappschaft
   10,254          df696           ^,29i           1,996
 Bei dieser tJberhoufigkeit der knappachaftaversicherten
 Manner kbnnte man vermutent dafi Bergleute mit einer
 Bronchitis eher von den Hauaarzten zur stationaren Behand-
 lung eingewiesen werden ala Nichtbergleute, wenngleich
 dieae Unterstellung durch nichts zu belegen ist*
 Dieser Verdacht ist aber erst recht unbegriindet bei der
 Nebendiagnose  "chronische Bronchitis",  deren Haufigkeit
 in  den  verschiedenen Alteragruppen in Abb.1 fur die Manner
 und  in  Abb«2  fiir  die Frauen aufgetragen iat.  Hierbei sind
 alle  Falie  aufier  denen mit  der Hauptdiagnose  chronische
 Bronchitis  erf a lit,  jedoch nur  die  nicht  verstorbenen
 Patienten,  urn  Falle  mit  einer  prafinalen Stauungsbronchitia
auszuschlieBen. Die  grdBere Bronchitishaufigkeit  der knapp-
 schaftsversicherten  Manner  kommt auch hier  eindeutig zum
Ausdruck*

-------
                                1561
Abb
KHaufigkeit der Nebendiagnose-Chronische Bronchitis-
       bei stationar behandelten Mannern
      40-
      35-
      25-
      20-
      15-
      10-
        5
     aNichtknappschaft (n-3237)
     •IKnappschaft (n-1702)
                                                               Alter
            15 20  25 30 35  40 45  50 55 60  65 70 75  80 85
Abb.  2   % Haufigkeit der Nebendiagnose »Chronische Bronchitis*
       40-
       35-
       30-
       25-
       20-
       15-
       10-
        5-
        bei stationar behandelten Frauen

        i=i Nichtknappschaft(n-3544)
        m Knappschaft (n=l560)
                                                                   Altei
             15 20 25  30 35  40 45 50  55 60  65 70 75  80 85 90

-------
                            1562
Fur die Letalitat, das Lebensalter und die Behand lungs dauer
haben sich in den Mittelwerten keine Unterschiede  zwischen
Mannern und Prauen Oder zwischen den beiden Versicherten-
gruppen ergeben.
Diskusaion
B ei der Auswertung wurden alle Patlenten rait  einer Silikose
aus dem Vergleich ausgeschaltet, so daB die Silikose nicht
die Ursache der grofieren Bronchitishaufigkeit  bei  den
Bergleuten gegeniiber den Nichtbergleuten sein  kann«
Da andrerseits bei den Frauen die beiden Versichertengruppen
keine signifikanten Haufigkeitsunterschiede zeigten, ergibt
sich auch kein Hinweis auf eine durch den Wohnort  bedingte
Uberhaufigkeit der Bronchitis,  jedoch ist keineE der in
Betracht kommenden Wohnorte aIs staubfrei zu bezeichnen*
Die Ursache der grofieren Bronchitishaufigkeit  bei  den
knappschaftlich versicherten Mannern wird in ihrer beruf*.
lichen Staubexposition gesehen, so daB die dadurch beding—
ten Einfliisse auf die Lunge nicht kanger als geringfugig
angesehen werden konnen*
Da sich in Lebensalter, der Behandlungsdauer und der I eta—
litat aber keine Unterschiede ergeben haben, ist zu schlie-
B«n, daB der Verlauf der Erkrankung von der speziellen Ur-
sache unabhangig ist und eigengesetzlich verHauft*
Literatur.
A, Mohamad: Die Haufigkeit der  klinisch behandlungsbedurf-
tigen Bronchitis bei knappschaftsveraicherten  und  nicht-
knappachaftsversicherten Mannern und Frauen.
Dissertation, Bonn,  1974

-------
                              1563
    ECO-GENETICS:   HOST VARIABILITY IN HEALTH EFFECTS OF
                     ENVIRONMENTAL AGENTS

                       GILBERT S,  OMENN

U.S. Atomic  Energy  Commission, and Division of Medical Genetics,
University of  Washington, Seattle, Washington, USA
ABSTRACT

    An important  variable  in assessing the effects of environ-
mental agents  upon living things is the genetically determined
variation  in enzymes,  cell  surfaces,  and immune responses of the
host.   Such genetic variation may influence spread of agents
vithin the body, specific tissue vulnerability, rate of conver-
sion of agents to  active (toxic) forms or metabolism to inactive
forms.   Therefore,  individuals may be at relatively high risk
for mutagenic, carcinogenic,  teratogenic, or inflammatory effects
of common  agents for which  no statistically significant harm can
lie detected in overall epidemiological investigations.

    Certain inherited human disorders are known to make healthy
persons more vulnerable to  infectious or physical agents.   Indi-
viduals deficient  in the plasma protein alpha--antitrypsin are
mich more  susceptible than  normals to develop emphysema from
chronic cigarette  smoking and general environment.   Many males
of Mediterranean or Negro ancestry are deficient in the enzyme
glucose-6-phosphate dehydrogenase (G6PD), determined by a gene
on the X chromosome.   A long list of oxidizing agents can induce
severe hemolytic anemia in  those with G6PD deficiency.   Well-
eharacterized  genetic variation in acetylating and esterolytic
tnzymes causes severe adverse reactions from anti-tuberculous
and muscle-relaxing drugs.    Individuals with certain abnormal

-------
                              1564
hemoglobins OP e-rythrocyte enzyme defects are more susceptible
to other drugs and carbon monoxide.   Variation in cellular mem-
brane8t detected by his to compatibility testing^ has been corre-
lated with susceptibility to diseases probably triggered by
viruses or by ultraviolet radiation.   Variation in enzymes of
the DNA repair mechanisms may greatly influence mutagenic effects,
Abnormalities in humoral and cellular immunity increase the risk
of bacterial and fungal infections^ as well as of neoplastic
transformation*

     Other types of vulnerability result from the action of the
host on the agent.   The risk of developing cancer of the lung
is much higher among people who have the genetic trait of high
inducibility of aryl hydrocarbon hydroxylase (an enzyme that
forms carcinogenic epoxides from aromatic hydrocarbon compounds
in cigarette smoke and other emissions) compared with individuals
whose enzyme level is increased very little by such exposure.

     One may anticipate practical screening methods to recognize
individuals at much higher than "normal" or "acceptable" risk to
specific agents in defined occupational or geographic environ-
ments.

-------
                                  1565
From the field of pharmacogenetics,  striking examples have been recognized
of differences in the way different  individuals respond to drugs or are
affected adversely by such agents.   An  important variable in assessing the
effects of environmental agents upon living things  is the genetically-
determined variation in enzymes, cell surfaces, and  immune responses of
the host.  Such genetic variation may determine or  influence the spread of
agents within the body, specific tissue vulnerability, rate of conversion
of agents to active (toxic) forms, or rate of metabolism to inactive forms.
Thus, individuals may be at relatively  high risk for rautagenic, carcino-
genic, teratogenic, or inflammatory  effects of common agents even when no
statistically significant harm can be detected in over-all epidemiological
investigations. Several excellent reviews of pharmacogenetics exist tl-AJ .
l-  High vulnerability to direct action of exogenous agents
   1.1 Alpha^-anti-trypsin £Q.
     The serum alpha-globulin fraction  is largely comprised of alpha,-
anti-trypsin, also termed protease inhibitor.  The concentration of this
acute phase-reactant serum protein increases under various physiologic
and pathologic conditions, such as pregnancy, inflammation, injection of
typhoid vaccine, and administration  of  estrogens.  Individuals with the
deficient ZZ phenotype are known to  be  predisposed to early-onset of
severe pulmonary emphysema; presumably  the role of anti-trypsin activity
is important in limiting inflammatory responses.  Heterozygotes o£'  MZ
phenotype and SZ phenotype probably  also have significantly impaired
response to inflammation.  Approximately 10% of the general population are
heterozygotes for variants of anti-trypsin activity.  Smoking appears to
hasten development of chronic obstructive pulmonary disease in heterozy-
gotes as well as homozygotes for deficient phenotypes.  It is likely that
these individuals also face higher risks of complications  from a whole
array of other inhaled agents in the occupational and general environment.

   1.2  Glucose-6-phosphate dehydrogenase (G6PDO.
     This enzyme is essential for the pentose-phosphate shunt pathway,
generating energy and reduced pyridine nucleotides (NADPH)  to maintain
the integrity of red blood cells.   In the Negro type of G6PD deficiency
the level of G6PD activity is about  15% of normal.  This activity is ade-
quate for the red cells, unless an oxidative stress is introduced.   Oxi-
dizing drugs or foodstuffs or other agents can provoke a severe breakdown
of red blood cells with resultant anemia and hemoglobulinuria.   It  is
important to realize that reticulocytes and young erythrocytes have higher
G6PD activity than do older cells,  so an assay of G6PD after the old cells
have been destroyed will give a spuriously "normal" value  for G6ED  activ-
ity.   About 1 in 10 Negro males is  G6PD-deficient.  In the Caucasian type
of G6PD deficiency found among males of Mediterranean ancestry,  the level
of activity is only 1 to 5% of normal.  The list of oxidizing agents which
can precipitate hemolysis in these  patients is even longer  1 .   Oxidizing
agents other than drugs can be dangerous for these individuals,  including
naphthalene in moth balls.
     About 17. of people are carriers for an entirely different enzyme
deficiency,  metheraoglobin reductase, rendering them susceptible to
cyanosis from some of these same common oxidizing drugs,  such as dapsone,
chloroquine,  and primaquine.

-------
                                  1566
   1.3  Heavy-Metal Poisoning Hi.
   Organic mercury compounds have been responsible for a syndrome of
neurologic and psychiatric signs called Minaraata disease.  In two large
outbreaks in Japan, there has been considerable variability in the level
of exposure at which Individuals developed symptoms.  An analogous situa-
tion may hold for itai-itai disease, thought to be due to cadmium poison-
ing.
   1.4  Adverse effects of foodstuffs.
   Fava- beans can precipitate severe hemolysis in certain individuals with
the Caucasian type of G6FD deficiency.  Migraine can be precipitated by
the tyramine in various cheeses or by phenylethylamine in chocolates,
presumably due to low levels of monoamine oxldase activity in susceptible
individuals (Youdim & Sandier, personal communication).  The lactose in
milk causes intestinal discomfort and diarrhea in large numbers of people
who have lactose intolerance after the age of weaning (75% of Blacks,
90-951 of American Indians and Orientals).  Foods high in saturated fats
may grossly elevate the serum cholesterol and triglyceride levels in
individuals with hereditary hyperlipidemias.  Gluten in wheat produces
intestinal complications in persons predisposed to celiac sprue.  Nodular
goiter and thyroid adenomas occur more frequently in non-tasters of FTC
(phenylthiocarbamide), presumably since these persons are unable to recog-
nize such goitrogens in the diet.  Final"y, effects of poorly characterized
food additives and of various allergenic substances in the environment
must be borne primarily by those genetically predisposed.
    1.5  Radiation damage and mutagenesis.
    Patients with the rare inherited disorder xeroderma pigmentosum lack
activity of a normal DNA repair enzyme.  Ordinary exposure to sunlight
produces so much ultraviolet-induced damage that multiple cancers develop
on  the skin.  This repair mechanism normally provides a  "threshold" below
which radiation-induced damage may be  corrected.  It is  likely  that varia-
tion of  this and of related enzyme steps in the repair processes of cellsi
will be  found and will be correlated with risks of damage from  defined
exposures.
    i.6   Biological agents in  the environment.
    It  is well known clinically  that individuals vary in  the  likelihood  of
contracting  infectious diseases  in  the face of common  exposure.  The inte-
grity  of skin, respiratory, urinary, bowel, and other  portals of entry  is
very important, as are the  immunological responses  to  such agents.  Defects
in humoral or  cellular immunity predispose to  specific kinds of infections;
in fact, the type  of organism involved is often a clue to  the  type  of under-
 lying  immunological defect.   The widespread polymorphism of  histocompati-
bility types,  representing  antigens on cell surfaces,  now appears to be
correlated with widely varying risks  of developing  particular  diseases.
Of current  interest  is  the  finding  that HL-A27, an  uncommon  type, is found
 in extraordinarily high  proportions of patient groups  with rheumatoid
 spondylitis, Reiter's  syndrome, and uveitis—three  disorders of obscure
 origin likely involving  altered immune response.

 2.  Differences  in rate of conversion of  agent to active (toxic)  form
      The enzyme  aryl hydrocarbon hydroxylase  (AHH)  activates a variety  of
 exogenous hydrocarbon compounds, including many drugs, insecticides,  ster-
 oids,  and chemical carcinogens.  AHH catalyzes the  initial step in  the
 microsomal oxidation of polycyclic hydrocarbons,  forming active epoxides.
 The great variability in AHH enzyme levels induced  by carcinogens applied
 to human lymphocytes cultured in vitro has now been demonstrated to be
 controlled by a single gene [VJTsince such hydrocarbon carcinogens  are

-------
                                    1567
present in cigarette smoke, it is likely that high Inducers are subjected
to much higher levels of epoxides than are low inducers with the same
amount of inhaled smoke.  The general population falls into three groups:
high inducers (9%), intermediate inducers (44%), and low inducers (47%).
Shaw and his colleagues have now found that high inducers are enormously
over-represented among patients with lung cancer and that low inducers
are only rarely found with lung cancer.  This finding, if confirmed, should
stimulate general testing of smokers, as well as a search for other conver-
sion pathways that may determine differences in susceptibility to other
types of carcinogens.

3.  Differences in rate of tnactivation of toxic agents,
   3.1  Pseudocholtnesterase deficiency--Suxamethonium sensitivity.
     Individuals completely deficient in plasma pseudochoiinesterase acti-
vity (about I in 2000 Caucasians) have prolonged muscular relaxation and
cessation of breathing when administered the common muscle relaxant
Suxamethonium (succinylcholine) for surgery.  An altered enzyme is respon-
sible for this Life-threatening risk.
   3.2  Acetylator phenotype.
     About half of the Caucasian and Black populations are "rapid" and half
"slow" acetylators of numerous drugs and other agents in the liver.  Only
10% of Orientals are slow acetyLators.  Slow acetylators accumulate higher
concentrations of active drug than do rapid acetylators, thus being suscep-
tible to side effects at dosage levels that do not affect rapid acetylators
adversely.
4.  Metabolic Interactions of drugs or agents taken together
     Chemicals found in man's environment that alter the action of other
drugs by stimulating microsomal enzyme activity include halogenated hydro-
carbon insecticides, urea herbicides, volatile oils, polycycllc aromatic
hydrocarbons, coloring agents, nicotine and other alkaloids, food preserva-
tives, and substances such as safrole, beta-ionone, xanthines, flavones,
and organic peroxides that occur in food.  Other chemicals inhibit'these
enzyme systems in the liver, including organophosphorus insecticides,
pesticide synergists of the methylene-dioxyphenyl type, carbon tetrachloride,
ozone, and carbon monoxideDQ.

      These examples of genetically-determined variation in susceptibility
 to drugs and other environmental agents should demonstrate that host
 variability is a potentially very significant factor both for research
 investigations of the health effects of particular agents and for concerns
 about public and corporate responsibility.
  References

  1.   World Health Organization Technical Report Series No.  524,  "Pharmaco-
         Genetics".  1973.

  2.   MOTULSKY,  A.G.,  "History and current status of pharraacogenetics".
         Proc.  4th Intl.  Congr. Human Genetics,s  Paris, 6-11 Sept., 1971.

-------
                                      1568
3.  VESELL, E.S., "Advances in pharmacogenetics".  Progr. Med. Genetics.
        9, 291 (1973).

4.  OMENN, G.S., MOTULSKY, A.G., "Pharraacogenetics".  Yearbook of Drug
        Therapy. 5-26 (1973).

5.  KUEPPERS, F., "Annotation: Alpha,-Anti-Trypsin", American Journal of
        Human Genetics, 25, 677 (1973).

6.  KELLERMAN, G., LUYTEN-KELLERMAN, M, SHAW, C.R., "Genetic variation
        of aryl hydrocarbon hydroxylase in human lymphocytes", American
        Journal of Human Genetics, 25, 327,(1973).

7.  CONNEY,.A.H., and (BURNS, J.J., "Metabolic interactions among environ-
         mental chemicals and drugs", Science, 178, 576 (1972).

-------
                              1569
           THE EFFECTS OF PARTICULATE AIR POLLUTION
                    ON EXPECTATION OF LIFE

                T,  J,  THOMAS, AND D.  W.  MALONE

Battelle,  Columbus  Laboratories, Columbus, Ohio, USA
ABSTRACT

     A technique was developed by which the effect of particu-
late air pollution on expectation of life could be quantified
and thus utilized in regional air-pollution planning.   The meth-
odology was then applied to a regional air shed in order to dem-
onstrate the loss of human functionality incurred for various
•emission-control practices.

     Expectation of life, broken down by age, race, and sex, and
calculated from observed mortality rates, was related to parti-
culate air pollution using regression analysis applied to data
collected for the year 1960.   The observations for the regres-
sion were based upon standard metropolitan statistical areas.
Many causal variables were included in the regression analyses,
vhich required compensation in the regression procedure for the
nonorthoganality of the causal variables.   This compensation
"procedure increases the confidence of the results and greatly
decreases the effect of correlated causal variables on the re-
gression coefficients of the pollution variables.

     An adequately documented regional air shed was selected as
tfce test region.   Air-pollution concentrations over the region
for three proposed sets of air-pollution-control regulations
were utilized along with the demographic characteristics of the
region to integrate per-capita losses over the age, race, sex,
aid location variables.

-------
                              1570
     The results of the methodology applied to the test region
were used to address several hypotheses.   Two conclusions
reached were:

     (1) Expectation of life is an important parameter to meas-
         uring the effect of air pollution on mortalityt and has
         the advantage of being commensurable with economic loss.

     (2) The change in expectation of life due to particulate
         air pollution is quite large.   An average loss rate for
         the test area is approximately 3 days per year per
         capita.

-------
                                    1571
1.0          Introduction

             It has been conceded by the Council on Environmental Quality  -*
that the ambient-air-quality standards are based upon an inadequate base of
information.  While many industries claim the standards are much too strict^
the other extreme might also be true.  Also, owing to the variability of
weather conditions, the existence of any (other than zero) average ambient
level of pollution will guarantee that there will occur, at some time, an
episode serious enough to cause noticeable increases in morbidity.  This
implies that there is no "safe" level of pollution, meaning that standards,
such as those defined by the EPA, should be set according to socioeconomic
trade-offs, rather than through determination of safe levels.  The work
reported in this study was addressed to these considerations.

2.0          Regression Analysis of Expectation of Life and Particulate
             Air Pollution

             2.1  Data Sources
             The expectation of life for members of a population is estima-
ble from the mortality experience of the population.  Duffy and Carrol*--']
have tabulated the mortality rate for the years 1959-1961 by age group,
race, and sex for each standard metropolitan statistical area  (SMSA).
These mortality data can be used to evaluate expectation of life with
techniques proposed by Chiang^J, with a slight modification to account for
the averaging of 3 years of mortality data.
             Lave and Seskin^] have utilized socioeconomic, pollution,
and climate variables to calculate the effect of air pollution on mortality.
These data, representing some  24 potential causal variables defined for
each SMSA, were kindly provided by Mr. Seskin.

             2.2  Regression Procedure

             The 24 causal variables defined were utilized in a multi-
variate regression analysis to develop the relationship of expectation of
life(by age, race, and sex) to particulate concentration.  The causal
variables were found to be highly correlated with each other.  Under this
condition, a regression analysis yields estimates of coefficients which
are unbiased (accurate), but which have a large variance (imprecise).
Thus, it is possible for estimated coefficients to be quite different from
the true coefficients.
             Hoerl and KennardL5]  have presented a technique called ridge
regression which alleviates this problem by the addition of a  small value
to the diagonal entries of the correlation matrix of the causal variables.
With this technique, the estimates of the coefficients have a much smaller
variance, while becoming only  slightly biased.  The application of ridge
regression thus improves the likelihood that the coefficient estimates
are in the neighborhood of their true values.
             It is possible to extend the ridge-regression procedure to
include weighting.  In this study, the case weights were based upon the
variance in the estimates of expectation of life.
;             Two separate regression models were applied.  In  the first,
expectation of life was related to particulate concentration.  In the
second, the Naperian logarithm of concentration was used.

-------
                                   1572



3.0          Application of Regression Results

             3.1  Test Area

  • .  •        A regional air shed for which demographic and industrial data
were known was needed to apply the results of the regression model.  The
area selected was a fully documented environmental planning regional
derived from an actual city.                   •
             Three sets of emission controls were postulated for the city.
,In the first, no control equipment was placed on any particulate source
(except where particulate recovery was an economic necessity).  In the
second,/extremely good equipment was placed on all industry, and residen-
tial heating was converted from coal to natural gas.  The third was a
repeat of the second, except that the city power  plant was left uncon-
trolled.    :
             3.2  Results

             Air-pollution concentrations over the city were estimated for
these three cases via a climatic  air transport model.  The concentrations
were then coupled with demographic data and regression results to produce
the losses in expectation of life due to particulate concentrations.  The
change due to enission controls was calculated from the difference between
the losses for emission controls to the loss for no emission controls.
This change, which represents an increment in expectation of life to be
achieved (conceptually) at the end of the life span, was apportioned on a
yearly basis by division by the nominal expectation of life.
    ,    /.These calculations were performed for the three emission
control sets and the two (linear and log) models  of the regression analysfa.
The results show that the benefits to be achieved by emission controls
(over the no-control case)  expressed in man-days  per capita per year are
as follows:

  Emission Controls               Linear Model         Loearithmetic Model
Extremely Good                        3-08                    2-73
Uncontrolled Power Plant              0.46                    0.66

       -      It is seen that a good set of emission controls applied over
the city would result in an increased life span of about 3 days per capita
per year.  This increase is essentially an increase in productivity, and
thus is commensurable with economic loss.

-------
                                  1573
4.0                        LIST OF REFERENCES

     Council on Environmental Quality,  "Environmental  Quality"
     GPO, Wash., D.C., 1972.
[2]  Duffy, E. A., and Carroll, R. E., United States Metropolitan Mor-
     tality. 1959-1961. PHS Publication No. 999-AP-39, U. S.  Public
     Health Service, National Center for Pollution Control,  1967.

[3]  Chiang, C. L., Introduction to Stochastic Processes in Biostatistics.
     New York, Wiley, 1968, pp 189-219.

     Lave, L. B., and Seskin, E. P., "Air Pollution and Human Health",
     Science. Vol 169, No. 3947, Aug 21, 1971, pp 723-733.

[5]  Hoerl, A. E. and Kennard, R. W., "Ridge Regression:  Biased
     Estimation for Nonorthogonal Problems", Technoroetrics Vol 12,
     1970, pp 69-82.

[6-1  Bell, J., and Brenchley, D., Environmental Engineering Teaching
     Reference Community. School of Civil Engineering, Purdue University,
     1971.

-------
                             1575
          ILLNESS DUE TO ENVIRONMENTAL POLLUTANTS IN
                    A PHYSICIAN'S PRACTICE

                        G, L, WALDBOTT

Warren, Michigan, USA
ABSTRACT

     Practicing physicians are frequently confronted with illness
due to environmental pollution, but many lack the diagnostic
tools for its recognition.   Relatively little information is
available to the physician on the clinical aspect of such dis-
eases in contrast to an abundance of literature pertaining to
the biochemical and epidemiological approach.

     This paper attempts to bridge this gap.   Data have been
gleaned from acute pollution episodes and from chronic low-grade
exposure to environmental pollutants which leads to slowly pro-
gressive illness.   The patient's history, physical and labora-
tory findings are correlated with available environmental, data
and elucidated by brief case reports.   The modus operandi of
respiratory pollutants is discussed in distinction to systemic
airborne poisons which enter the system via the gastrointestinal
tract.   The target organs of the major environmental pollutants
;are presented diagramatically and the diagnostic features of
illness due to air pollution are tabulated.

-------
                                    1576
        Practicing physicians are frequently confronted with illness due to
environmental pollution; however many lack the tools for establishing the diag-
nosis.  In contrast to an abundance of literature pertaining to the biochemical
and epidemiological aspect, relatively little information is available on the
clinical aspect of environmental diseases.  This paper constitutes an attempt
to bridge this gap.
        In order to recognize an environmental disease,  the patient*s history,
clinical and laboratory findings must be correlated with data gleaned from
acute pollution episodes and from long-term, low-grade exposure to environ-
mental pollutants.
        History:   Data on meteorological conditions of the region in which the
patient resides such as the prevailing winds, humidity, rainfall must be incor-
porated into the history.  The topography of the area and the distance of the
residence from sources of pollution should be taken into account.  The onset of
the illness must be correlated with the official report on weather conditions
and with information available from monitoring for air pollutants.  The pre-
sence or absence of symptoms during vacation periods, their intensity outdoors
compared with that in enclosed places, the simultaneous onset of identical
symptoms among other residents of the neighborhood constitute clues concern-
ing environmental diseases.  The effect of smoking on the patient's condition
should be explored.
        Examination;  The physical examination must take into account that
many environmental pollutants can enter the system through the gastrointes-
tinal tract and the skin,  not necessarily through the respiratory organs and
that most environmental diseases exhibit a multiplicity  of symptoms.  Targets
of the most common pollutants are represented in Figure I.  In the evaluation
of the symptomatology, the patient's personality should also be considered and
the possibility of malingery be ruled out.
        Laboratory Data; The accompanying table presents some of the major
laboratory features encountered in illness due to air pollution.  Analysis of
hair, uzine and blood for suspected contaminants are helpful in determining the
degree of exposure to atmospheric pollutants but  they are not by themselves

-------
                                      1577
sufficient to determine the offending agent.  Chemical assays of biopsied

material are often of value.
                               ArMnic Cadmium
                                           O*gamc *Md

                                               Cartxn monoudt
                  Ftuono* i
                 S*tmnirfl<
               Mangann*
               Mignwum
                BWfftum
                   Zinc
Figure I  -  Main Targets of Major Air PoUutants
                        Laboratory Features in Diseases Caused by Major PoUutants
POLLUTANTS
Arsenic
Asbestos

Barium
Beryllium
Cadmium

Carbon monoxide
Cobalt
DDT (chlorinated
hydrocarbons)
Fluoride*

Hydrogen sulfide
Lead

Manganeve
Mercury
Nitrite*
Organophosphates
Pulmonary irritant*
Selenium
Vanadium
BLOOD
Arsenic T *


Potassium A *
Uric acidT
Alkaline phosphataset

Carbaxyhemoglobin*
Cobalt-


Alkaline phosphataset
Fluoride r
Sulfhemoglobin
Red blood cells (stippled)
Delta aminolevulinic acidi
Lymphocytect
Mercuryt uric acidT
MethemoglobinT
Cholinecterasel


Cholesteroll serotoninT
URINE



Potassium t

Microglobulint sulfurt
Glucose in urine




Fluorides^

Coproporphyrin Hit , leadT
Delta aminolevulinic acidT



NitrophenolT

Selenium T

OTHERS
Hair, urine (As t)
Sputum, ferruginous
bodies

Sputum, Schauman bodies
Nails (Cd t )



Fat (DOT T)

Hair, bones (Ft)

X-rays of bones,
abdomen
Hair(Mnt)



Chest x-ray


= decrease.
       * T= increase;

 Table I  -   Laboratory Features in Diseases Caused by Major Pollutants

*From Waldbott, G.  L. ,  Health Effects of Environmental Pollutants,

 The C. V.  Mosby Company,  St.  Louis, 1973.

-------
                              1579
          ASSESSMENT OF THE HEALTH EFFECTS CAUSED BY
                 CHRONIC EXPOSURE TO SOLVENTS

                       PIETRO U, CAPURRO
Union Hospital of Cecil County, Elkton, Maryland, USA
ABSTRACT

     About 120 people living in a small valley in Maryland have
been chronically exposed to a mixture of solvent vapors since
1961, when a chemical plant in the valley began operations.   A
study of the effects of this exposure was begun when many resi-
dents, including children, complained of headaches, abdominal
pain, and other symptoms.   Laboratory tests on these patients
showed abnormal glucose tolerance and insulin curves, abnormal
lipase levels, and abnormal respiratory function.   Diver and
pancreatic scans were likewise abnormal.   Meanwhile, gas chro-
jaatography identified the vapors from at least 25 different sol-
vents in the valley atmosphere or from the water of a creek
running through the valley.   Nine of these compounds were found
in the blood of the patients.   Mortality records for the period
1967-74 have revealed an unusually high incidence of cancer
among the valley residents, with particular emphasis on malignant
lymphoma and pancreatic carcinoma.

-------
                                  1580
1. raTROMJCTION

      For several years,  residents of a small  valley  in Maryland have
complained of disagreeable odors in the atmosphere* These odors come
from a small chemical plant which began operations in the valley in
1961; the plant recovers solvents from various chemical waste products.
Ho more than 120 people have lived in the valley for  at least 2 years
during this period. When many of the residents,  including children,
began showing signs of actual physical distress, we undertook a study
of the chemical pollution in the valley and its possible effects on the
health of the people. Typical complaints received from the residents were:
pain in the upper abdomen, chest, and extremities; pain underneath the
ribs, often radiating to the left side and back; burning of  the eyes,
nose, and throat; headaches, dizziness, fatigue, irritability, and absent
mindedness.

2. EVIDENCE OP POLLUTION

      Analyses, mainly by gas chromatography,  were made of samples of the
following* air in the valley, water from a creek running through the
plant area, the plant grounds, and blood from residents of the valley.
The list of compounds identified in these samples follows; the letters
A,W,0 and B designate the kind of sample (air, water, ground, and blood)
in which the compound was found.
Benzene                    A,W,Q,B        Acetonitrile             G
Toluene                    A,G,B          Acrylonitrile            A,G
Xylenes            .        G              Acetone                 A,W,G,B
Phenol                     H,G            Methyl ethyl ketone      A,W,G,B
Hitrophenol                A,G            Methyl i-butyl ketone    A,H,G,B
Hitrochlorobenzenes        A,G            Formaldehyde             A,W,G
Ghlorobencenes             A,G            Kethanol                 A,tf,G
Methyl chloride            A,B            i-Propyl alcohol         A,H,G
Methylene chloride         A,W,G,B        Butanol                 A,W,G
Carbon tetrachloride       A,G,B          Amyl alcohol             A,H,G
Trichloroethylene          A,W,G           i-Amyl alcohol         A,W,G
Perchloroethylene          A,G             Furan                  G
Tetrachloroethane          A,G,B           Tetrahydrofuran        W,G

-------
                                    1581
Acetic acid                  H,G       Ethyl acetate               A,G

      These compound were present in mixtures.  Several other compounds
were detected but not identified.

3. ASSESSMENT OF HEALTH EFFECTS
      Statistical analysis of clinical data shows significantly abnormal
glucose tolerance and lipase levels in residents of the valley, compared
with patients in the hospital. The probabilities of finding these diffe-
rences by chance were 0.05 and 0.01, respectively. We found that eleva-
ted lipase levels, rather than amylase, were the most significant indica-
tion of acute pancreatic damage due to exposure to solvents. In addition,
residents of the valley showed abnormal insulin curves, usually with a
depressed level and delayed peak. None of the pancreatic scans made on
ten residents of the valley was completely normal; they generally showed
a low count with uneven uptake. Abnormalities were found in liver function
studies, ICD levels, PCOp, and bicarbonate levels. Kany of the abnormali-
ties disappeared with removal of the patient from the polluted area. A
decreased level of zinc in the blood and/or an increased urinary excretion
of zinc were noted.
      During the seven year period beginning late in 19671 the death rate
in the valley community has been 2,2 times the rate in the County in which
the valley is located. The death rate due to cancer has been at least 7
times that in the County. Most of the valley people who died of cancer
had lived there for several years, or from birth. Pour deaths were due
to malignant lymphoma; an increased death rate due to carcinoma of the
pancreas was also found. The probabilities that these deaths would have
Occurred during this period by chance are about 0.00001 for the lymphomas
and 0.006 for the pancreatic carcinomas*

4. ACKMOMLEDGMEHT
      Thanks are due Dr. Elois KAILIN and Dr. N.E. DAY for help in
statistical evaluation of the data.

-------
                                  1582
                          R E F E R E N C E S
1.  CAPURRO, P.U.| "Effects of chronic exposure to solvents caused by
    air pollution?1! Clinical Toxicology.  3(2),  233 (1970).

2.  CAPURRO, P.U., "Effects of exposure to solvents caused "by air
    pollution with special reference to CC1.  and its distribution
    in air", Clinical Toxicology. 6(l), 109^(1973).

3.  CAPURRO, P.U*, ELDRIDOE, J.E., "Cancer mortality in a community
    subject to air pollution by solvent vapors", Clinical Toxicology.
    7(4), in press (1974).

-------
                               1583
      INCIDENCE DE CERTAINES MALADIES CHEZ LA POPULATION
           DE DEUX CENTRES FORTEMENT INDUSTRIALISES,
           COMPARATIVEMENT A TROIS LOCALITES TEMOINS

      M,  DELEANU, N, W, GHELBERG/ I, LENGHEL/ N, MOLDOVAN

Institut de SantS Publique et de Recherches MSdicales, Cluj ,
Roumanie
RESUME

     La morbidity a ete etudiee comparativement chez la population
de 2 centres dont I 'atmosphere est intensivement pollute par des
industries diverses,  et chez la population de 3 villes temoins
dont I'atmosphere urbaine est exempte de pollution intense et
epecifique.   Les donnees primaires sur I'incidence dea affec-
tions ont ete extraites des rapports mensuels des circonscrip-
tions medicales.    L 'etude comprend une periode de ? ans (1963-
1969).

     La pollution atmospherique a ete aonsideree oomme etant un
facteur favorisant de I'incidenoe de certaines maladies seulement
lorsque I'incidence de celles-ci dans le centre industriel de-
passait 2,0 fois  pour lea enfantst ou 2,5 fois pour leo adultes
I'incidence des mSmes maladies dans lee looalit&s temoins.

     Les re suit ats de I'etude montrent que:

     Dana un centre a puissante industrie de metalurgie non-
ferreuse et chimique, une incidence aignificativement plus ele-
vSe a 6t6 const atee pour certaines maladies reepiratoires, car-
diovasculaires, digestives, neurologiquees  psychiques, hematolo-
giques et du metabolieme - chez les adultes;  de mSme pour cer-
       maladies respiratoires, hematologiques et le rachitisme
       lea en fonts (0 - 16 ans).

-------
                               1584
     Dans un centre a Industrie de mat&riaux de construction et
chimique, une frequence accrue a ete constatee pour certaines
maladies respiratoires* cardiovasculaires et digestives - chez
lea adultea, aussi bien que pour certaines affections respira-
toires et le rachitisme - chez les enfants.
ABSTRACT

     The incidence of disease Mas compared in the population of
two centres in which the atmosphere ie severely polluted by var-
ious industries, and in the population of three control towns
free from severe or specific atmsopheric pollution.   The pri-
mary data on the incidence of the diseases were taken from the
area monthly medical reports.   The study covers a seven-year
period (1963 - 1969).

     Atmospheric pollution was considered to be a factor favour-
ing the incidence of certain diseases only when their incidence
in the industrial centre was more than twice (in children), or
more than 2.5 times (in adults) that of the same disease in the
control areas.

     The results of the study show that:

     In an important centre of the non-ferrous-metal and chemical
industries there was a significantly higher incidence of certain
respiratory, cardiovascular, digestive, neurological, psycholo-
gical t haematological and metabolic diseases in adults, and like-
wise of certain respiratory and haematological diseases and rick-
ets in children in the age range 0 - 16.

     In a centre of the building materials and chemical industries
there was a greater incidence of certain respiratory, cardiovas-
cular f and digestive diseases in adults, and of certain respira-
tory conditions and rickets in children.

-------
                                   1585
1. IBTROKJCTION

      La pollution atmospherique revSt des aspects diffe"rents pour chaque
localite, en raison de la nature, de la quantite et du caractere des eli-
minations nocives, de mSme qu'en fonction de certaines conditions meteo-
rologiquee et topographiques locales. II n'est done pas depourvu d'inte-
r8t d1analyser si, dans des localites a diverses industries et intensites
de pollution, il existe des differences dans la morbidite de la popula-
tion. Le but de telles rechercbes est celui de savoir si la pollution
atmospherique constitue un facteur de risque concernant certaines affec-
tions, s'il y a une implication particuliere de la pollution dans 1'etat
de BantS de la population dans lee divers centres industriaux.
      Dans le cadre du present travail la morbidite (l1incidence des cas
nouveaux) a ete Studiee comparativement chez la population de 2 centres
dont I1atmosphere est intensivement polluee par des industries diverses,
et chez la population de 3 villes t&noins dont I1atmosphere urbaine est
exempte de pollution intense et specifique. Le travail synthetise des
donnees qui ont et6 en partie publiees (1-3)•

2. MBTHODE

      Etant donne 1'uniformite de 1'organisation de I1assistance medico-
sanitaire urbaine, la grande accessibilite (aussi bien au niveau du
re"seau sanitaire de base, qu'au niveau medical de specialite), et la
standardisation du systeme d'enregistrement des cas de maladies en Rou-
nanie, on a precede 4 I1etude comparative de certaines maladies pour
lesquelles il existe dans la litterature des donnees sur une correlation
possible avec les conditions de la pollution atmospherique. Notre etude
comparative concerne I1incidence dee affections respiratoires, cardio-
vasculaires, digestives, hematologiques, neurologiques, psychiques,
allergiques, de mStaboIsme et nutrition, les maladies des organes sen-
sor i els, le rachitisne et les tumeurs.
      Les rapports mensuels des circonscriptions medicales sur le nombre
des cas nouveaux (incidence) de maladies ont servi comme donnees primaires.
On a d'abord fait la somme des cas nouveaux de maladies enregistrees au
cours de 1'annSe respective, puis on a rapporte ce nombre a 10OO habitants,

-------
                                 1586
adultes cm enfants. L'Stude a compris one periode de 7 annees (1963-1969).
Afin de minimaliser les differences annuelles, on a calcule les moyennes
des frequences annuelles des maladies BUT la periode analysee.
      La moyenne de I1 incidence dee maladies a ete calculee de maniere
analogue pour les 3 localit6s temoins,  cette incidence etant consideree
Sgale a I1 unit6 (incidence conventionnelle * 1.0).
      La pollution atmospherique a 6te"  consideree comme etant un facteur
favorieant de 1'incidence de certaines  maladies seulement lorsque I1 inci-
dence de celles-ci dans le centre industriel depassait 2,0 fois pour les
enfants, ou 2,5 fois pour les adultes I1incidence des m&nes maladies
dans les locality's temoins (incidence conventionnelle plus de 2,0 ou
2,5).


3* RESULTATS
3*1*  Dans un centre a une puissante Industrie de me'tallurgie non-ferreuse
et ctimique, une incidence significativement plus elevee a 6te const at ee
pour les maladies suivantess
      Chee les adultee - du groupe des  affections respiratoires, la
bronchite chronique; le groupe des affections cardiovasculaires (parmi
celles-ci, hypertension artSrielle, artSriopathie coronaire chronique
et maladies des artferes); du groupe des affections digestives, ulcere
gastrique et duodenal; le groupe des maladies du sang et des organes
hematopol&tiques (parni celles-ci, les  anemies); le groupe des affec-
tions psychiques (parmi celles—ci psychoses, psychonSvroses et troubles
de caractere et de comportement); les maladies chroniques du systeme
nerveuz central; groupe des affections  allergiques, endocriniennes,  de
mfitabolisme et de nutrition (parmi celles-ci, le diabete Sucre").
      Chez les enfants - le groupe dee  affections respiratoifes (parmi
celles-ci, les infections algues des voies respiratoires supeVieures,
la grippe, la bronchite algae et la bronchite chronique); le rachitisme;
les maladies du sang et des organes hematopoi^tiques (parni celles-ci,
les anemies){ les affections psychiques (parmi celles-ci, les peychone-
vroses, les troubles de caractere et de comport ement); les maladies  des
yeirr; le groupe des affections allergiques, de mStabolisme et de
nutrition*

-------
                                   1587
3*2.  Dans un centre 6. Industrie de materiaux de construction et chimique,
une frequence accrue a ete constatee pour les maladies suivantes:
      Chez les adultes — du groupe des maladies des voies respiratoires,
la pneumonic et la bronchite chronique; le groupe des maladies cardio-
vasculaires (parmi celles-cit le rhumatisme cardio-articulaire aigu,
1'arteriopathie coronaire chronique, I1hypertension arterielle); du
groupe des affections digestives, 1'ulcere gastro-duodenal.
      Chez les enfant8 - le groupe des affections de 1'appareil respira-
toire (parmi celles-ci, la pneumonie, la bronchite algue, la bronchite
chronique, les hypertrophies des amygdales, les vegetations adenoldes);
du groupe des affections cardiovasculaires, le rhumatisme cardio—arti-
culaire aigu.
                            REFERENCES
1.  DELEANU M., ANTAL A., MOLDOVAN N., BRETTEH E., COSTIN L.f BORDAS E. :
    "Date privind morbiditatea prin afec^iuni respiratorii si mortalita-
    tea prin cancer pulmonar, in raport cu poluarea aerului" (en roumain).
    Igiena. 21, 199 (1972).
2.  DELEANU M., GHELBERG N.W., MOLDOVAN N. s "Date asupra morbidita-fii
    populatiei dintr-un centru cu Industrie chimica si metalurgie nefero-
    asa" (en roumain). Igiena, 21, 529 (1972).
3.  DELEANU M., BORDAS E., BRETTER E., COSTIN L.: "Date comparative
    asupra mortalitatii prin cancerbronhopulmonar in 10 centre urbane".
    (en roumain). Igiena, 22, 351 (1973).
4.  LEHGHEL I.f DELEANU M., RADULESCU N., MOLDOVAN N.: "Donnees concer-
    nant la morbidite" de la population dans une ville a atmosphere intense-
    ment pollute par I1Industrie chimique et celle de materiaux de
    construction ". La Sant6 Publionie. 16, 3o) (1973).
5.  GHELBERG N.H., DELEANU M.: "Aspecte ale poluarii atmosferei urbane
    si locul ocupat de plumb in cadrul acesteia" (en roumain). Revista
    Medico-Chirurgicallt. 77, 507 (1973).

-------
                               1589
              BONE MARROW CHANGES  IN ASBESTOSIS

      A, Z, EL-SEWEFY/ H, SHAHEEN  AND A.  SHAMS EL-DEEN

fce National Research Center and Kasr El-Aini Hospital,  Cairo,
IBSTRACT

    Seventeen workers from an Egyptian asbestos-cement  pipe  fac-
tory exposed to asbestos dust, and seldom  exposed  to  silica and
bnent dust for a period of 20 -  26 years  of age range from SO  -
HI years were examined clinically, radio logically*  and hematolo-
jteally including bone marrow puncture.

    The result showed a state of leucopenia,  increased  sedimen-
tation rate, the bone marrow showed a marked to extreme  degree  of
generalized hyperplasia* the outstanding feature is the  marked
bcrease in the number of retiaulum cells,  some of them  are im-
Kture and binucleated.

    These results are discussed  in the text.

-------
                                   1590
INTRODUCTION

      The inhalation of asbestos duBt produces  chronic interstitial  pul-
monary fibrosis, pleura! reaction and might predispose to bronchogenie
carcinoma or pleural mesothelioma.
      Recent studies on the serum of asbestos workers showed the presence
of hyperproteinemia, hypoalbuminemia and hypergammaglobulinemia  [1]  ,
in addition to marked changes in Igs.  A, M, and G  C2)  .  T. Warwick  [ 3]
(1970)| reported an increase in A.N.F. and rheumatoid factors but not a
significant increase in antibodies to cytoplasmic constituents.
      Similarly rheumatoid factor has been reported in asbestosis by
Pernis et al  [4]  and in silicosis by Vigliani and Pernis  [5] .
      It is likely that several different mechanisms for production  of
rheumatoid factor will be found.
      Disturbance of protein and amino acids metabolism  in asbestos!s \6 Jt
as well as hypoglycemia  [7] , in addition to hypocholesterolemia and
disturbed lipid electrophoresis in the same disease were recently
reported  [8] .
      In disorders associated with pronounced byperglobulinemia, there
is an increase of plasma cells  [9] *
      El- WaraaM et al  [10] , examined the bone marrow specimens of 7
workers having silicosis. The pattern of changes noted was in the form
of generalised hyperplasia, and in particular there were very high fi-
gures for the myeloid series, ret ionium cells and plasma cells. The
eosinophils showed a slight but definite increase,  but the erythrocytes,
megakaryocyteB, lymphocytes and monocytes were  within normal limits.
      These changes were considered to be related to silicosis since
other diseases causing stimulation of the bone  marrow had been adequately
excluded.
      Bone •arrow has not received universal attention in workers exposed
to asbestos dust*
      The evidence presented in favour of an auto-  immunological mechanism
operating in asbestosis stimulated this study of bone marrow in workers
exposed to asbestos and shoving various graded  of radiological shadows.
      Therefore the purpose was to investigate  these anticipated changes
and to clarify their significance in the diagnosis  and prognosis of
asbestosis.

-------
                                   1591
MATERIAL AND METHODS

      Seventeen workers from an Egyptian asbestos - cement pipe factory
exposed to asbestos dust (crocidolite, chrysotile and amosite)  Cll] ,
and seldom exposed to silica and cement dust, for a period of 10 to 26
years and of age range from 30 to 57 years were the subject of this
study.
      These workers were admitted to N.H.S. Helwan hospital. They were
examined clinically.  Clinical examinations of all cases revealed no
enlargement of the liver, spleen, or lymph nodes, no abdominal masses were
felt. Sputum analysis was repeatedly negative for acid fast bacilli by
direct smear examination. The sputum was also examined for malignant
cells by staining with Papanicolau stain  [12]  no malignancy was detected.
Large chest radiographs were done and the cases were classified according
to the extent of radiological findings*
      A single bone marrow puncture was performed in each case. Serologi—
cal tests for syphilis, salmonella, and brucellosis were free. The
patients were afebrile throughout the time they spent in hospital and
they had no skin lesions or itching. A special search for parasites in
urine, stool, blood and bone marrow specimens was done, but all were
negative.
      Blood samples were collected and examined for:
      I     - Total and differential white blood cell count
      II    - Hemoglobin %, R.B.Cs. count and reticulocytes
      III   - Platelet count, haematocrite value and sedementation rate
      Bone marrow was examined fort Total  and differential  count.
RESULT

I   BLOOD CHANCES

      Table I shows the findings in blood, in this table we notice a state
of leucopenia in almost all cases ranging from 3000 - 5*600, a slight but
persistent eosinophilia 5 - 20$, the hemoglobin % was within normal
limits 12.5 - 16 gas. %. Reticulocytes, hematocrite and platelets count
trere within the normal range.

-------
                                                    TABLE I  (Blood Picture Changes in Asbeitosis)
Dane
K*H.+
n.H.
M.M. «
A.A.+
V
A.A.
M.O.*
'+
E.F.
M.E.+
A. A.
S.Z.
J.
G.H.
r.t*
A.S.
S.M.
A.R.*
Age
40
50
50
57
49
52
51
55
49
32
40
49
40
33
49
49
40
Period of
exposure
26
22
13
26
25
20
22
20
21
13
10
11
22
16
10
11
23
Extent of Physical Hb. J
X-ray signs gnu 2
+ +14
+ 44 14
+ + 13.5
444+ ++4 13
+ 4+ 13
+++ 44 13
4++ 44+ 12
++ 444 12
4+ 44 13
44+ 44 14
44+ + 14
13
4+ 44 14.5
4+ 44 13
+ +14
+ +16
* | t 1 .!.!_> 1 IT
' ' T ' ' "'* r AV
R.B.Cs.
48
5
4.8
4.7
4.5
4.7
4.0
4.0
4.9
4.7
5.2
4.7
4.8
4.4
4.7
5.1
4.4
T.L.C.
in 1000
3.8
5.4
4.2
3.8
3.8
4.2
3.2
3.9
4.9
3.4
4.6
4.6
5.6
3.4
4.0
3.8
3.0
B.
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
0
E. St.
8 6
6 2
104
144
6 6
200
9 2
162
106
12 2
6 6
5 3
6 4
8 4
6 10
14 6
12 4
Sq. L. M.
42 368
56 342
54 302
60 18 4
48 328
20 528
51 325
38 368
60 22 2
54 266
52 360
66 20 6
58 22 10
20 58 8
68 14 0
J4 386
42 348
Platelet
in 1000
200
280
200
250
300
300
200
150
230
350
250
380
245
250
200
286
280
Retic
I
1
1
0.3
0.5
1
0.5
0.4
2
0.8
1
0.8
0.7
0.6
0.3
0.2
0.2
0.6
Sed. Rate
9/20
30/70
19/40
16/30
14/30
18/40
40/80
20/40
8/20
30/60
22/45
11/24
3/5
20/40
12/26
8/17
40/80
+  on top of initials means that they are repeated in  the 2 tables.
+  to -H-H-S the degree of severity of either the x-ray  or the physical signs and synptoms.
  Figure J,4 and 5 are examples for the severity of  the x-ray.
                                                                                                                                                             VD

-------
                                  1593
      Sedimentation rate using the Westergreen method was in almost all
cases higher than normal,
      Bone marrow showed a marked to extreme degree of generalised
hyperplasia.
      The myeloid series showed a marked hyperplasia of its elements.
There is a definite increase in the number of eosinophils.  The  erythroid
series showed a normal process of maturation of its elements which are
present in normal numbers. The megakaryocytes are present in normal
numbers and showed a normal process of platelet separation. The lympho-
cytes and monocytes are present in fairly normal numbers and showed no
morphological anomalies.
      There is an increase in the number of plasma cells some are imma-
ture and binucleated (table II),  figure  1.
      The outstanding feature is the marked increase in the number of
reticulum cells, some are binucleated similar in morphology to  Dorothy
Reed cells, Figure  2.

                          T*.
             Figure l.i   Immature and Binucleated plasma cells

                  •  AHB
               Figure 2,:  Binucleated reticulum cells

-------
                                                TABLE II  (Bone Marrow Changes in Asbestos!s)
Nane
K.M

M.H.
M.M.
A.A.
S.I.
A.A.

H.D.

M.I.
E.F
M.E.
G.H.
F.L
A.S.
A.R.
S.M.

A.M.

T.C.
in 1000
3ZO

165
320
200
320
Itt

265

320
210
Z«»
200
95
198
286
280

360

MYELOID
B E BL.
0 *! 2

07 0
05 2
090
080
06 0

09 2

OBI
080
090
1 7 0
09 1
OA 0
060
02 0

on i

SERIES %
PRO. MYE.
5

1
2
0
3
1 •

0

0
1
1
0
0
0
0
1

1

10

18
18
9
13
11

11

15
10
13
12
9
12
5
18

15

JUV.
9

11
15
13
14
6

10

10
11
15
7
9
12
4
5

7

ST
9

15
17
17
25
21

10

17
20
20
25
21
26
21
16

22

SO.
15

15
15
20
W
17

15

6
13
13
12
1*
11
15
Z*

12

LYMPHORETICULAR SERIES.
RC. PP. L. M
2

2
3
3
3
2

3

2
4
t.
3
0
2
2
2

2

5

6
1
5
3
5

5

6
6
6
it
3
5
6
3

5

I". 0

B 0
10 0
10 0
7 0
7 1

11 2

17 2
7 1
23 3
15 2
6 2
6 1 .
20 10
12 1

1 1

ERYTHROID
ZA

15
12
l
-------
                                    1595
Figure 3.



Figure 4.
 Figure 5.

 Figure  3, 4t 5  '  Examples of the Severity of Lung Damage.

-------
                                    1596
       The total  count  of  the  nucleated  cellular elements of the marrows
 examined varied  between 95.000 and 360.000/mm . (normal marrow 50.000 -
 100.000).
       These changes  were  very similar to changes observed in the bone
 marrow in certain forms of reticulosis  and particularly in association
 with malignant lymphoma.

 DISCUSSION

       It is pertinent  to  emphasize that these bone marrow and blood
 changes were related to the inhaled dust produced in this industry
 (asbestos,  silica and  cement), thorough clinical examination and labo-
 ratory investigations  were carried out  to exclude other diseases that
 may stimulate the bone marrow i.e. specific fevers, syphilis, leukemia,
Hogkins'  disease,  Bilharziasis, malaria  and other tropical or skin
 diseases.
       Prom  the above results  it is evident that the bone marrow picture
 was definitely abnormal in the cases of asbestos!B and/or mixed dust
 pneumoconiosie investigated.
       These changes observed showed the following:
       1 - Generalised hyperplasia in almost all cases  with particularly
           high figures for  the myeloid  series, reticulum cells and
           plasma cells
       2 - A slight but definite increase in eosinophils
       3 - Erythrocytes, megakaryocytes, lymphocytes and monocytes
           were within  the normal  range.
       It is appreciated that  if lesions of the bone marrow are focal in
 distribution,the puncture as  well as the trephine biopsies cannot be
 representative of the  overall state of  marrow activity. The fact, however
 that all cases showed  more or less the  same abnormal results may be
 taken as evidence of the  diffuse nature of these abnormalities.
       There was  no correlation between  the degree of the bone marrow
 reaction and the extent of the radiological shadows in the lungs. On the
 other hand,  the  total  leucocytic count  was found in these cases to be on
 the low normal level or leucopenic in spite of the myeloid hyperplasia
 of  the bone marrow,  this mi^it be explained on the assumption that

-------
                                      1597

   leucocytes are mainly phagocytosed in the reticulo-endothelial tissue of
   the lungs, and this is assumed to take place in asbestosis and mixed dust
   pneumoconiosis as El- Warraki et al  [loj   found similar findings in
   silicosis.
         In conclusion the similarity of these findings to silicosis in
   addition to the presence of binucleated reticulum cells and the hyper-
   plasia of the bone marrow as in cases of reticulosis and lymphomas stimu-
   lates further research on the aetiopathogenesis of pneuraoconiosis.
                               References
       El-Sewefy,  A.Z,  i Hassan,  F.  1971;  J.  Egypt.  Med. Assoc,  54;  243
 [2]    El-Sewefy,  A.Z.  & Hassan,  P.  1971?  Inn.  Occup.  Hyg.  14;  25
 [3]    Warwick.  M.T.  & Haslam,  P.  1970,  Clinical Allergy 1,  83
 [4]    Pernis, B.  Vigliani,  E.G.  & Selikoff,  L.G.  1965,  Ann.  N.Y.  Acad.  Sbi.  132;  112
 pjj    Vigliani, E.G.,  & Pernis,  B.  1963f  Adv.  Tub.  Res. 12;  230
 jjfj    El-Sewefy,  A.Z.  & Hegazi,  S.  1971t  Ann.  Occup.  Hyg.  14;  29
 [j]    El-Sewefy,  A.Z.  & El-Shobaky,  P.,  1972 Ann.  Occup.  Hyg.  (in press)
 j 8]    El-Sewefy,  A.Z.  & Hanna.  T.  1972,  2nd  Pneumoconiosis  Conf.  Yugoslavia,
 u J    31 May -  2  June.
 [~9~|    Clinical  Hematology.  Maxwell  M. Wintrope'6th edit.  1967,  Henry Kimpton.
   -1    London P. 41
       El-Warraki,  3.B.  El-Gammal, M.Y. & Awny,  1965, Brit.  J.  Iniust. Med.  22;  279
jllj   El-Sewefy,   A.Z.  & Awad,  S. 1969, I.  L.O. Congress, Geneva;
       Attia,  0.  &  El-Sewefy, A.Z.,  197 2j 2nd Pneumoconiosis  Conf. Yugoslavia,
       3 May - 2  June.
   The authors wish to thank Dr. A. YOUSSED the Director of El- Nasr N.H.S.
   hospital for making this work possible.
   Our gratitude to Dr. Nehad SHAHEEN for technical assistance and to
   Dr. Mohamed HASSAKEIN for his help in the clinical examination and
   cooperation.

-------
   TIERUNTERSUCHUNGEN





     ANIMAL STUDIES





 ETUDES SUR LES ANIMAUX





   STUD I SUGLI ANIMALI





ONDERZOEKIN6EN BIJ DIEREN

-------
                              1601
      BEHAVIOUR OF RATS AND THE THRESHOLD LIMIT VALUE OF
                       TRICHLOROETHYLENE

              A, P, SILVERMAN AND HELEN WILLIAMS

Imperial Chemical Industries Limited, Industrial Hygiene Research
Laboratories, Macclesfield, Cheshire, United Kingdom
ABSTRACT

     Objective evidence concerning health effects of low doses
of a solvent like trichloroethylene (TCE) will probably continue
to come largely from tests of animal behaviour, because its main
toxic action is on the central nervous system and because the
long-term, low-dose evidence from man may be considered subjec-
tive .

     Groups of 16 male rats were exposed singly to TCE vapour at
lOOppm (S35mg TCE/m ), or 200, 500 or lOOOppm 5x6 hours a week
for about 5 weeks (12 weeks at lOOppm).   Social behaviour was
observed, usually once a week, in the S min. immediately follow-
ing exposure.   Control rats show their highest spontaneous ac-
tivity on return to the home cage after daily separation, and
TCE reduced this activity.   In a social situation, activity is
of several kinds but, unlike most centrally-active drugs, the
reduction due to TCE was fairly uniform with little evidence of
stimulation or depression of any one kind of behaviour.

     The fall in activity due to TCE was significant after 1 day
at lOOOppm.   Following lOOppm, activity fell after 1 day in one
group of rats,  but after l%weeks in another and as much as 5%
weeks in a third.

-------
                              1602
     In a test of exploration and learning in an unfamiliar cage,
thirsty rats exposed to TCE were quicker than controls to find
water and begin drinking.   The rate of exploratory movement was
unaffected.

     The present threshold limit value for TCE in Britain is
therefore not a no-effect level.   Tentative interpretations of
the effects are:

     (1) that a fall in the highest rate of spontaneous activity
         is consistent with the drowsiness and fatigue complained
         of by exposed humanst and

     (2) that a low dose of an anaesthetic might well be expected
         to lead to a slight loss of inhibitory control in an un-
         familiar situation.

-------
                                     1603
1. Introduction
1.1. Some effects of environmental pollutants seem to be considered
negligible, for in practice far more action is taken on some effects than
on others, and a theoretical distinction is implied by the American
Conference definition of a threshold limit value (TLV) in terms of
"conditions	without adverse  effects" (our emphasis).   Before it is
safe to distinguish the trivial from the harmful, we need to know as much
as possible about a) the dose and duration of exposure which produces any
detectable effect, and b) what kind of effect it is.  Though obvious, this
is easier said than done.
1.2. A particular problem concerns TLVs for the vapour of solvents like
trichloroethylene (TCE).   In the U.K. this is at present 100 ppm.  At
high concentrations, TCE is a useful anaesthetic/analgesic, while CNS-
related symptoms are also reported in surveys of people occupationally
exposed to low concentrations.   However, while symptoms like drowsiness,
fatigue, headache and gastro-intestinal disturbances are often not trivial,
they are not altogether reliable because they depend on self-report and
because they occur frequently in control populations.  Formal
psychophysiological tests are useful, provided the available methods are
relevant to the postulated effects, but TCE has cumulative effects which
make it difficult to achieve sufficiently long exposure in controlled
experiments or adequate control data in  people chronically exposed at work.

1.3. Basic objective evidence on TCE at concentrations near the TLV is
therefore likely to come from suitable experiments on animal behaviour. The
proviso is that methods are selected which will be sensitive to a variety
of possible effects and yet which will discriminate between them, and which
will also be statistically reliable.   Moreover it should be possible to
relate the results to any available information from man.   Published
evidence from animals varies considerably in value on these criteria, and
we report here an attempt to supplement this evidence.

2. Behaviour of rats exposed to TCE vapour
2.1. Methods  Groups of 16 male albino rats were housed in pairs but exposed
singly for 6-7 hours daily, 5 days a week to TCE at 100, 200, 500 or
1000 ppm.  Vapour was generated by syringe and atomiser into a metered
stream of clean air.   Each TCE group was matched with controls separated
daily into smdl holding cages, and a dummy experiment compared such controls
with a group in the exposure chambers breathing clean air.
Social behaviour was observed as described by Silverman (1971).  Paired rats
which are separated daily and returned to their home cage early in the dark
phase of a 12 hour white/dim red lightine cycle, show a very high rate of
spontaneous activity for a few minutes.  On the ethological assumption that
behaviour is composed of  recognizable elements, the acts and postures
described by Grant & Mackintosh (1963) were observed for 5 minutes.  These
occur in functional groups which are differentially sensitive to drugs.
For example chlorpromazine reduces the total activity, but within this,
disproportionately reduces all forms of social approach while increasing
escape away from another rat,without altering exploration, eating or
grooming.   A low dose of nicotine selectively reduces aggression,thus
one kind of behaviour acts as an internal control for drug effects on
another (Silverman 1966, 1971).
Social behaviour was observed in a baseline for each experiment, again after
one day's exposure to TCE, then usually after three days and weekly for five
weeks, in one experiment at 100 ppm for twelve weeks.  Controls then

-------
                                    1604
received a "crossover baseline" observation in the exposure chambers and
one day's exposure to TCE.

An "exploration-thirst" test combined a measure of motor activity with a
simple learning task.  On two successive days after overnight deprivation
of water, rats were placed singly for five minutesin an unfamiliar plastic
cage similar to their home cages.  Exploratory movements (expressed as
movements/minute)  were recorded whenever the rat paused to sniff at floor,
wall or roof.    The latency before the rat found and began to drink from a
water bottle was also recorded (as the square root of the time in sec),
separately if necessary for the first brief drink and the first of longer
than 10 sec.

2.2. Results
2.2.1. Social behaviour.  The main finding was a consistent reduction in
the sum of elements of all kinds observed in 5 minutes (the "total activity")
A dummy experiment showed that any fall in activity due to the exposure
chambers themselves was transient.  Analyses of variance (P< .05 at 1 and 3
days, P<",001 from 10 to 38 days and in XI) followed by t tests showed (see
fig. 1) that although exposed rats could not be distinguished by gross
observation, a) a concentration of 1000 ppm significantly reduced activity
after one day's exposure  b) at 100 ppm there was a similar fall, but only
after repeated exposures.   One group of rats, moderately active at first,
showed a significant reduction after Ij weeks exposure; another group,
highly active, showed a fall only after two months.  These also showed a
post-exposure rebound, by comparison with which their matched controls
reduced activity after one day's exposure.
The reduction in activity was fairly uniform among different kinds of
behaviour (by multivariate analysis of variance).  Reductions specifically
in exploration of the cage and submission to or escape from the other rat
occasionally  reached the 5% significance level.

2.2.2. In the exploration-thirst test after 2{ or 5 weeks exposure, TCE
caused rats to begin both the first brief drink or the first of more than
10 sec sooner  than controls.   Exploratory motor activity was unchanged,
averaging 13.70 movements/min for TCE-exposed rats on trial 1 against
13.40 for controls, and 15.80 against 15.68 on trial 2.   Exposed rats
never took longer than controls to begin drinking, and were significantly
earlier in trial 1 at 200 ppm and trial 2 at 10O and 1000 ppm.  Within the
five minute observations exposed rats did not drink for longer than controls.
3. Discussion
3.1. There was an interesting interaction between the concentration of
TCE, the duration of exposure, and "individual" differences between batches
of rats.  The highest concentration of TCE tested reduced total activity
after a single day's exposure, the lowest did so only after repeated
exposures and one batch of rats was much more susceptible than another.
3.2. Any gross neuromuscular disturbance would have been visible under the
close observation, and a motivational disturbance would have been detected
as a specific change in the appropriate  behavioural category.  The absence
of such findings and the moderate decline in activity, makes the  results
consistent with subjective reports in man:- reduced activity when controls'
activity is maximal could be secondary to the rats' equivalent of
drowsiness, fatigue or headache in a familiar situation.
3.3. In the exploration-thirst test, TCE allowed thirsty rats to begin
drinking sooner than controls, yet they were equally likely to interrupt

-------
                                        1605
                               Total Activity  After T.C.E.
   A)
   Mean Control Activity as %XB
        120-

        110-

        100

                 6     13    10   17  24-3031-38
                         Days
o  o Air dummy
o--Ti 100 ppm.
o—o 200 ppm.
»-•-• 500 ppm.
•—•l.OOOppm.

O  P<-05
                                        ' Trichloroethylene
Activity as percent XB, difference from control group.
        +10-
      Control
        -10-
        -20-
        -30-
                                                                         Ex - TCE
10
                                  17 24-3031-38 45
                                        Days
                                  87   XB XI
    Fig 1.  Total_activity  in  the baseline  (B)  observation of social
            behaviour and after  exposure  to  TCE 6 -- 7 hours a day
            5 days  a week.

            A) Mean activity of  the control  groups expressed  as a
            percentage of their  activity  in  the crossover  ba8eline
            (XB) observation.

           B)  Activity of rats from the exposure  chambers  in  the
            baseline  observation  and  after exposure  to TCE.
            Difference from controls  in activity  as  % XB.

            Former  controls breathed  clean air  in  the exposure
            chambers  (XB)  and were  then exposed for  1  day  to TCE (XI)

            Circled points show reductions significant  at P< 0.05
            by t test  after analysis  of variance of  the  untransformed
            activity  scores.

-------
                                    1606
drinking for a littlefurther exploration before drinking continuously for
10 sec.    Although TCE had none of the paradoxical motor-stimulant effects
of low doses of barbiturates, TCE could well have increased responsiveness
to stimuli or reduced anxiety in an unfamiliar situation as these drugs
have been suggested to do.
3.4. In rats therefore, the TLV for trichloroethylene in the UK and else-
where (100 ppn) is not a no-effect level, if methods are used which minimise
the inevitable laboratory restrictions on the animals'  flexibility of
response.  Extrapolating the present results to man,  we would expect TCE
to reduce the rate of working on a familiar self-paced task.  Some
individuals would be more sensitive than others, but there might be a
trade-off between concentration and a sufficiently continuous exposure.
In unfamiliar (emergency ?) conditions, exposed people may respond sooner;
this could be beneficial but could also represent a lack of proper caution.
The experimental work in this paper has been submitted for publication.

GRANT, E.G., MACKINTOSH, J.H. "Comparison of the social postures of some
common laboratory rodents",  Behaviour 21, 246-259 (1963).

SILVERMAN, A.P.  "Social behaviour of rats and the action of
chlorpromazine and other drugs"    Behaviour 27, 1 - 38 (1966).

SILVERMAN, A.P.  "Behaviour of rats given a 'smoking-dose' of nicotine"
Anim. Behav.. 19, 67 - 74 (1971).

-------
                              1607
       MUTAGENITATS - VORTESTE AUF CARCINOGENE WIRKUNGEN
                     VON UMWELTCHEMIKALIEN

                         H, MARQUARDT

Forstbotanisches Institut, Universitat Freiburg Brg.r BRD
ABSTRACT

     (1) From the mutational processes induced by environmental
agents usually the numerical and structural chromosome aberra-
tions as well as different types of molecular mutations on the
DNA-level are observed.   Until now only in microorganisms clear-
ly defined molecular mutations can be demonstrated.

     (2) According to recent opinions, at the beginning of oar-
oinogenesis mutational changes of the DNA take place.   In spite
of former difficulties a correlation exists between mutagenic
and carcinogenic actions of chemicals.   For thefoe reasons it
is possible to develop a pattern of mutation testst which can
serve as a prescreening method for carcinogenic effects:
a)  Direct action of the substance to be tested on strains of
Saccharomyoes oerevisiae (eukaryotio cells, genetically well
analyzed, in stable haploid or diploid strains), which show
induced suppressor mutations or induced gene conversions.
b)  Treatment of mammals and examination of the bone marrow (or
blood cultures) for chromosome aberrations.   o)  Injection of
yeasts in the ventral cavity or in the urine of treated animals
and test for induced mutations or recombinations.   d)  In vitro
test with liver miorosomes and yeasts to cover the effect of
short-lived metabolites.   B)  Treatment of yeasts with synthe-
sized metabolites, which are known to arise in the organism
treated.

-------
                              1608
     (3) The usefulness of the genetic prescreening method has
been shown with two carcinogenic compounds:  Cyalophosphamide
and Benzo(a)pyrene.   The advantages of the method consist in
a short duration of the tests, small amounts of substances used
and the possibility of testing a great number of compounds and
their combinations.

-------
                              1609
     Im Gegensatz zu den akuten Wirkungen von Umweltagentien
auf die menschliche Gesundheit werden die sogenannten SpSt-
schSdigungen weniger beachtet. Ihre Wirkungen treten erst nach
einem stummen Intervall auf: Die Xnderungen des Erbgutes (Mu-
tationen) als erster Typ wird in einer der Folgegenerationen
nach der Einwirkung sichtbar; die induzierte krebsige Ent-
artung als zweiter SpatschSdigungstyp nach Jahrzehnten.
     Bei der groflen Zahl von Umweltchemikalien, die noch nicht
auf carcinogene Wirkung untersucht sind, sowie bei der Dauer
und Kostspieligkeit derartiger Experimente 1st es zweckmSMg,
einen Vortest auf mogliche carcinogene Wirkungen einer Sub-
stanz zu entwickeln.
     Nach neueren experimentellen Ergebnissen sind bei der
krebsigen Entartung der Zellen mutative Xnderungen an der DNS
entscheidend beteiligt und es besteht eiee gute Korrelation
zwischen mutagener und carcinogener Wirkung einer Substanz. Es
ist daher moglich, aus der groSen Zahl von Mutagenitats-Priif-
verfahren ein Vortest-Muster zusammenzustellen, das einen opti-
malen SchluB von dor induzierten Mutabilit&t auf eine carcino-
gene Wirkung zulSflt. Im Gegensatz zu den typischen Mutageni-
tcltsprufungen dttrfen hier nur Teste an somatischen und nicht
an Keimzellen vervendet werden.
     In der Tabelle I sind die aufeinanderfolgenden Prttfver-
fahren des Carcinogenitats-Vortestes zusammengestellt. Nachge-
wiesen werden in Knochenmark Oder Blutzellen von saugetieren
(Chinesischer Hamster) induzierte numerische Oder strukturelle
Chromosomen-Aberrationen. Mit St&iuncn von Saccheromyces cere-
visiae (Hefe) werden als definierter Mutationstyp die Suppres-
sormutationen im Isoleucin-Valin^lGenlocus sowie als Rekombi-
nationstyp die mitotischen Genkonversionen in zwei Genloci err
fasst. Beide DNS-Snderungen sind streng miteinander korreliert
(1), sodafi in der Regel die Bestimmung der Genkonversion als
das empfindlichere System genligt.
     Die VerSnderungen der zu prtifenden Substanz im Metabolis-
mus des SMugers wird durch Injektion der Hefezellen in die
Bauchhfihle, durch Exposition mit Harn behandelter Tiere sowie

-------
                              1610
Tabelle I:
               Mutagenita'ts-Vorteste auf carcinogene Wirkung

               rait Cyclophosphamid und Benzo(a)pyren
Objekt
                Technik
                             Mutationstyp
                                            Cyclo-
                                            phosphamid
Benzo(a)-
pyren
S acchar cony ces
cerevisiae
                Direkt-
                Behandlung
                             Mitotische
                             Genkonversion
                Knochen-
                mark (Blut-
                kultur)
Behandeltes
S3ugetier
(Chines.

Hamster)
                             numerische
                             und struktu-
                             relle Chrorao
                             somen-Abeera
                             tionen
                Injektion
                BauchhSaie:
                Saccharo-
                myces

                Urin: nach
                Behandlungs-
                beginn
                Saccharo-
                myces
                             Mitotische
                             Genkonver-
                             sion
                             Mitotische
                             Genkonver-
                             sion
Saccharomyces
cerevisiae
                In vitro:
                Lebermikro
                somen +
                Saccharo-
                royces +
                Substanz
                Direkt be-
                handelt +
                syntheti-
                siertem
                ultimalen
                Carcinogen
                             Mitotische
                             Genkonver-
                             sion
                             Mitotische
                             Genkonver-
                             sion
                                              n.t.
                                                       Epoxyd)
1) Ergebnisse aus der Literatur
n.t. - nicht geprtlft

-------
                             1611
durch den in vitro Versuch mit Lebermikrosomen und zu priifen-
der Substanz einbezogen. Dieser Test wird durchgefilhrt/ um
auch kurzlebige aktive Metabolite nachweisen zu kSnnen. Sofern
die aus den verabreichten Substartzen im Metabolisraus entsteh-
enden ultimalen Carcinogene bekannt und synthetisiert sind,
werden sie durch direkte Einwirkung auf Hefezellen Oder Blut-
kulturen getestet.
     Die Brauchbarkeit dieses Vortest-Musters ist mit zwei
carcinogenen Substanzen gepriift worden: Cyclophosphamid als
alkylierende, bei direkter Wirkung auf Mikroorganismen schein-
bar nicht mutagene Substanz sowie Benzo(a)pyren als carcino-
gener, aromatischer Kohlenwasserstoff, der als fleispiel fUr
die Nicht-Korrelation von Mutagenitat und Carcinogenitat
hSufig herangezogen wurde.
     Cyclophosphamid Ifist zahlreiche Chromosomen Aberrationen
aus; seine genetisch aktiven Metabolite sind im Urin in Uber
1OO facher Konzentration als in der Lymphe vorhanden und nach-
zuweisen (2), desgleichen im Lebermikrosomen-Test (Siebert
unverttff.). Benzo(a)pyren IGst bei hOherer akuter Dosierung
Chromosomenaberrationen aus, seine aktiven Metabolite sind
nur im Lebermikrosomentest nachzuweisen, desgleichen bei direk-
ter Wirkung der synthetisierten ultimalen Carcinogene (Epoxyde
der K-Region. 3).
     Diese vorlSufigen Ergebnisse mit zwei genetisch besonders
schwierigen Substanzen lassen das ausgewHhlte Muster von Vor-
testen geeignet erscheinen, 1m Falle positiver Ergebnisse auch
eine carcinogene Wirkung der Substanz zu erwarten.

(1) Zimmermann F.K., "Induction of mitotic gene conversion by
    mutagens", Mutation Res. 11, 327 (1971).
(2) Siebert D., "A new method for testing genetically active
    metabolites",  Mutation Res. 17, 307 (1973).
(3) Ames B.N., P.  Sims a. P.L. Grove, "Epoxides of carcino-
    genic polycyclic hydrocarbons are frameshift mutagens",
    Science 176r 47 (1972).

-------
                               1613
     LES CHAINES TROPHODYNAMIQUES MARINES EXPERIMENTALES:
           METHODE D'ETUDE DES EFFETS PATHOLOGIQUES
             CONSECUTIFS AUX POLLUTIONS CHIMIQUES

                     M, AUBERT, B, DONNIER

C.E.R.B.O.M. (I.N.S.E.R.M.), Nice, France
RESUME

     Le milieu marin eat susceptible de concentrer pay I'inter-
mediaire dee differents Echelons de la biomaaae, lea produita
chimiquea qui y sont rejetSa, pouvant aboutir  ainsi  a des  phe-
nomenea  toxiquea chez  le conaommateur de  oea eapecea.

     Pour etudier ces processus, noua avona mis an point une
methode originate quit en laboratoiret utilise diversee chatnea
trophodynamiquee.

     Cette methodologiet employee ayatematiquement, permet de
meeurer lee destructions de la biomasse, maie egalement d'appre-
cier lea dangers vie-a-vie de la Sante Publique que peuvent
oreer dee phenomenes de concentration de substances toxiques.
ABSTRACT

     Marine organisms of various levels of the biomase are able
to concentrate chemicals which are disposed of into the sea, and
this can give rise to toxic effects in the consumer of these
species t

     To study these processes we have developed an original

-------
                               1614
laboratory method which makes use of various food chains.

     Used systematicallyt this method permits measurement of the
destruction of biomass* and makes it possible to assess hazards
to public health which may arise from the concentration of toxic
substances,

-------
                                    1615
          Le danger des pollutions chlmtques de la mer tlent  aux capacltes
de concentration de certains de ces polluants au niveau d'especes marines.
Ce fait peut entratner des taux eleves de ces substances souvent toxiques,
principalement dans les derniers echelons des chatnes biologiques, qui sont
habituellement livres a la consommation.
          II y a plusieurs annees, nous avons mis au point une methode ori-
ginale qui consiste a recreer en laboratoire des chatnes trophodynamiques
marines, typiques des principales biocenoses, a leur soumettre pendant des
temps plus ou moins prolonges et a des concentrations progressives les pro-
duits polluants que 1'on desire etudier et a en lire les resultats a travers
le coraportement biologique et 1'impregnation chimique de leurs constltuants.
          Cette methodologie, que nous utilisons d'une maniere systematique,
permet :
          - de raesurer la destruction produtte au niveau d'organismes-types
representant divers echelons les plus caracteristiques de chatnes tropho-
dynamiques marines. Par ce moyen, 11 est possible d'evaluer la destruction
en capital des ressources biologiques de la mer ;
          - de tenir compte de I1ensemble des phenomenes de concentration
et de biodegradation des substances chimiques soumises a ces tests et de
juger des effets biologiques induits par la toxicite initiale ou secondaire-
ment apparus, ces produits chimiques passant a travers le milieu et etant
absorbes successivement par les divers elements vivants ;
          - enfin, par 1'introduction d'un consommateur final mammifSre, de
juger des phenomenes toxiques en fin de chatne et de raesurer le danger en-
couru par I'Homme nourri de produtts marins pollues. Get aspect hygieniste
debouche ainsi sur les impacts de la pollution marine vis-a-vis de la Sante
Publique.
          C'est ainsi que nous avons mis au point quatre types de chatnes
trophodynamiques :
          - la chatne de type pelagique, c'est-a-dire  :
            phytoplancton - zooplancton - poissons - mammiferes ;
          - la chalne de type neritique 1 mollusques, c'est-3-dlre :
            phytoplancton - mollusques - mammiferes ;
          - la chatne de type neritique S crustaces, c'est-S-dire :
            mtcro-organismes marins - invertebres - crustaces - mammifSres  j
          - la chatne de type benthique, comprenant :
            micro-organismes marins - invertebres - poissons benthiques -
            mammiferes.

-------
                                    1616
          Ainsi, une chains biologique peut  etre  utilisee  a  differentes
fins :
          7) Re.cheAc.he. de.i> -ieiUx&s de.
          En mettant en contact chacun des echelons  de  la chaine  envisagee
avec le polluant  (produit chimique pur ou effluent  Industrie!)  a  differentes
concentrations, on peut determiner avec une  assez grande  precision la sen-
slbilite de chaque espece a ce polluant. On  peut ainsi  definir  pour chaque
echelon un seuil  toxique.
          Les resultats permettent done une  prevision a court tcrrae des
consequences du rejet de ces polluants en mer  sur la survie  des especes et
de definir les risques de perte en capital nutritionnel.

          2) Richnfiche. de4 e^e-ti du  to. bwde.gfiadati.on  d'un  po££uatit
          Lors de son rejet en mer, un polluant  peut  servir de substrat aux
bacteries marines et done  subir une biodegradation. Un tel  effet  peut  en-
traTner une variation de  la toxtcite  du  polluant  vis-a-vis  des differents
echelons marins de la chaine biologique.
          Les resultats rendent compte de  la  variation de la toxicite  d'un
polluant apres contact avec 1'eau  de  mer.

          3) Re.cheA.che. du.  ttia.n&^eAt zt de.  la.  c.onc.ent.tiatis'n  de 6u.bAtanc.eA
          Ayant determine  les  seuils  de  toxicite  directe d'un effluent,
nous choisissons la concentration  maxima permettant  la survie totale de
tous les echelons de  la  chaine marine :  de  cette  fac.on,  la substance to-
xique pourra etre transmise  d'un echelon a  1'autre,  et eventuellement con-
centree au cours des  differentes etapes  sans que  la  chaine biologique soit
rompue par la mort d'un  des  interraediaires.
          Les resultats  permettent d'evaluer les  risques de la consommation
par 1'Homme des produits marins et ainsi les dangers eventuels vis-a-vis de
la Sante Publique.

-------
                             1617
           VERSUCHE ZUR MEDIZINISCHEN BEDEUTUNG DES
              GLEICHFELD- ODER FARADAY-EFFEKTES;
                   BAUTECHNISCHE FOLGERUNGEN

                          J, R, MOSE

Hygiene-Institut der UniversitSt Graz, Osterreich
KURZFASSUNG

     M&use wurden unter der Einuirkung versohiedener bioklima-
tologisaher Umweltbedingungen (elektrostatische Gleiohfelder mit
Feldstarken von:  40, 200, 1000, 5000, 24 000 Vt normales Raum-
klima e-inea Labors, Faraday-K&fig) mit Hammelerythrozyten voi>-
behandelt, und ee wurde der Immnnieierungagrad nit HiZfe der
direkten Plaque-Technik naah JERNE ermittelt.   Die h&oheten
Plaquebildungszahlen zeigten aioh bei Milzen der im Gleiohfeld
lebenden Tiere, die niedrigsten Wevte wurden dagegen unter feld-
loeen Umweltbedignungen gefunden.   Dieselben Reeultate in Form
von Htfchstwerten im Gleichfeld und Tiefetverten im Faraday-Raum
lieferten auoh die Bestimmungen der jeweiligen Milzgewiahte,
Milzzellzahlen und Hdmagglutinationstiter,   MtSgliehe praktieohe
Folgerungen werden diskutiert.
ABSTRACT

     Mice previously treated uith sheep erythrooytea were sub-
jected to various bioolimatic conditions  (constant electrostatic
fields of 40t 200t 1000, 5000 and 24 000  V;  normal laboratory
environmental conditions;  Faraday cage), and the levels of im-
munisation was determined by use of Jerne's direct plaque tech-
nique.   The highest plaque formation counts were seen in the

-------
                              1618
spleens of animals living exposed to constant fields, and the
lowest in animals in a field-free environment.   Spleen weights^
spleen cell counts, and haemagglutination titres showed a simi-
lar picture^ with highest values under constant field and lowest
values in the Faraday chamber.   Possible practical conclusions
are discussed.

-------
                            1619

Da die vielfaltigen Moglichkeiten der  technischen Zivili=
sation heute so tief in unseren Alltag eingreifen, konnen
uns klimatologische Faktoren gerade dadurch stark beein=
flusaen. Bei unseren Versuchen kam es tins primar darauf an,
mit exakt reproduzierbaren Versuchsbedingungen ebenso exakt
faftbare Meflpararaeter zu erarbeiten, die, zumindest fiir ei =
nen klimatologischen Einzelfaktor, sichere Ausaagen ge=
statten. In diesem Sinne befaflten wir una mit'der Luft =
elektrizitat und hier apeziell mit dem elektroatatischen
Gleichfeld der Luft. (Gemeinsam m.Doz.Dr.Fischer).
Fiir diese Versuche an Kleintieren wurden in Zusammenar=
belt mit Experten der Technischen Hochschule in Graz
spezielle Versuchsanlagen angefertigt, (Gleichfeldkammern,
Faraday-Kammern, Kontrollkammern).
Die Unterauchungen bewiesen vorerst, daft aich unter der
Einwirkung verschiedener hochgespannter elektroatatischer
Gleichfelder bzw. im Faraday-Raum die MeRparameter Motili =
tat, Nahrungaaufnahme und Kbrpertemperatur, Saueratoff=
verbrauch von Lebergewebe, sowie Glutamat- und Succinat=
dehydrogenase in Lebermitochondrien in signifikanter Weise
anderten. Im Gleichfeld erfolgte dies in Form einer physios
logisch als durchaus giinstig zu interpretierenden Weise.
Die Abschirmung von elektrischen Feldern im Faraday-Raum
bewirkte dagegen stets eine unvorteilhafte Reduzierung
der genannten Stoffwechselmeftgrofien.
Recht vage Literaturberichte iiber giinatige Allgemeineffekte
elektroatatiacher Gleichfelder auf Infektionen beim Men=
schen veranlafiten uns zu einer experimentellen Uberpriifung
dieser subjektiv qualitativen Beobachtungen« Gruppen von
Mausen wurden dazu mit Hammelerythrozyten immunisiert.
Den erzielten Immunisierungsgrad beatimraten wir mit Hilfe
der direkten Plaque-Technik nach Jerne (Abb.1).
Die hbchsten Plaquewerte und somit der atarkste Immunia
sierungsgrad ergaben aich zwischen 1000 und 50OO V/m,
d.h. nur unter Einwirkung einea ganz beatimmten Feldsta'rke=
bereichea. Selbat die, natiirlichen Durchschnittswerten
bei ungestb'rten atmoaphariachen Bedingungen entaprechende

-------
                           1620
      Plaques
                             >3000
800
700
600
500
400
300
200
100

V/m
 N
             1
       1
             1
      • .
       J
8
(N
       to
  ii j
                               JL
                             U")
15
                                     • t
                                     tN
                         10
                     Abb.  1
      Jerne-Teat (direkt)
1.  Kolumne

2.   "      ('V': )

3.   n
                 ) = Gleichfeldkammer
                     darunter Spannung  in V/m
                     Kontrollkammer
                     (normale  Laborgegebenheiten)
             (Mil) = Faraday-Rammer
FeldBtarke von 200 V/m  fiihrte  noch,  im Vergleich mit den
Kontrollen, zu einem  aignifikanten Anatieg d«r Inniuniai*.
rung. Onter den Werten  natUrlicher Gleichfelder liegende,
niedrige Feldstarken  fuhrten zu keiner ErhShung dea Immuni.
aierungsgradea. Die Abschirmung dea Gleichfeldea in der
Faraday-Kammer bewirkte ateta weit unter der Norm liegende
Plaquewerte.

-------
                       1621
  1:256
          G   K
    V/m  40
F    G  K   F    G  K   F

   5000        24000
                  ABB.  2

Hamagglutinationstiter - je 10 Tiere pro Einzeli
                         versuchsgruppe
     G - Gleichfeldkammer
         darunter Spannung in V/m
     K s Kontrollkaramer (normale Laborgegeb.)
     F = Faraday-Komraer

-------
                           1622
Gleichzeitig mit jedem Plaquetest durchgefiihrte Bestimtnun=
gen der entsprechenden Milzgewichte, MiIzzellzahlen und
Hamagglutinationstiter zeigten einen deutlichen Trend  zur
Ubereinstimmung in Form einer generellen Zunahme  im
Gleichfeld und Abnahme unter feldlosen Umweltbedingungen.
Diese Untersuchungen iiberpriifen und bestatigen die im
Plaquetest erzielten Resultato (Abb^jj•) •
Die Ergebnisae weiaen darauf bin, dafi das naturliche
elektrostatische Gleichfeld der Erdatmosphare einen
bemerkenawerten Faktor zur Forderung der Bildung  von
Immunstoffen daratellt.
Wir wollen heute noch keine Schluftfolgerungen ziehen;
aber alles weist darauf bin, daft man in Zukunft auch bei
bautechnischen Uberlegungen diese Verbaltnisse weit mebr
als bisher mitberiickaichtigen aollte. Wir verbringen
heute einen Groflteil unaeres Lebens in geschlosaenen Rau=
men und es iat, wie unsere Verauche zeigen, durchaus
nicht gleichgultig wie dabei klimatologiache Faktoren
geandert oder abgeachirmt werden.
Selbatveratandlich haben diese Uberlegungen vordringlich
dort Gevicht, wo es urn eine optimale Beeinfluasung von
Kranken geht, die erfahrungsgemaft auf negative bioklima=
tologische Einfliisse beaonders ungiinatig reagieren -
z.B. in Intensivpflegestationen, Herzstationen, bei der
pra- und poatoperativen Behandlung etc.
leh bin mir vollig bevruftt, daft wir mit unseren Versuchen
nur einen Teil bioklimatologiacher Einfliisse erfafit
haben; die Ergebnisse sind aber derartig eindeutig»dafi
man in Zukunft die Gesamtforschung auf diesem wichtigen
Gebiet unbedingt wird intensivieren miiasen.

-------
                      1623
              BIBLIOGRAFJE

MOSE, J.R., FISCHER, G., "Zur Wirkung elektrosta=
tischer Gleichf elder , tierexperimentelle Ergebnisse."
Arch.f .Hyg.u.Bakt. , Ig4. 3?8 (1970).
MOSE, J.R., FISCHER, G., PORTA, J., "Die Wirkung
des elektrostatischen Gleichfeldes auf den Sauer=
stoffverbrauch der Mauseleber."
Arch.Hyg.(Berl. ), lg*t, 5^9 (1971).
FISCHER, G. , "Die bioklima tologische Dedeutung des
elektrostatischen Gleichfeldes. "
Zbl.Bakt.Hyg. , I.Abt.Orig. B  157.  113  (1973).

MOSE, J.R., FISCHER, G., STRAMPFER.H. ,  "Immunbio=
logische Reaktionen im elektrostatischen Gleich=
feld und Faraday-Kafig. "
Z.Immun. -Forsch. lk$, kOk (1973).

ZOLLNER, H. , FISCHER, G.,. Noch unverof f entlichte
Resultate .

-------
                              1625
   METHODOLOGY OF EXPERIMENTAL AND EPIDEOLOGICAL RESEARCH
     OF DIFFERENT ENVIRONMENTAL PHYSIOPATHOLOGICAL FACTORS

                        MAURICE STUPFEL

Research Group U.123, I.N.S.E.R.M., Physiopathological Mechan-
isms of Environmental Nuisances, Paris, France
ABSTRACT

     The multiplicity of the environmental factors and the diver-
sity of their points of impact on the existential and evolutive
states of living beings must not lead to neglect of physical,
chemiaalf  biological and statistical principles, the synthetical
application of which constitutes the basis of a methodology.
A pluridisciplinary approach is necessary but its difficulty and
its complexity should not be minimized.

-------
                               1626
Introduction
     The multiplicity of the environmental factors and the diver-
sity of their sites of impact on the existential and evolutive
states of living beings must not result in the neglect of physical,
chemical, biological and statistical principles, the application
of which constitutes the basis of a methodology.  A pluridisci-
plenary approach is necessary but its difficulty and its com-
plexity should not be minimized.

Historical
     The possible interactions between man and his environment
have always been a subject of interest.  Reclus (1) and Dubos  (2)
quote Hippocrate's "Airs, waters and places" and Montesquieu's
"de 1'Esprit des lois"; Brunhes, Vidal de la Blache and Leroi-
Gourhan in France have described the relationships between human
ethnology and geography.  According to Bernard and Ruffie (3)
the stratification of human raciation results from geographic
separation and from socioeconomic isolation ^Israelites of the
Kurdistan, Laurie (4jj.  In a more restricted but more immediate
domain, aviation, submarine and space experiments in which man
is enclosed in a limited space have lead to the study of the
effects of imposed thermal, physical and chemical constraints.
In addition to classical meteorological factors (temperature,
humidity,atmospheric pressure, static electricity and electro-
magnetic waves Tromp  (5) and Rivolier  (6)) technical progress,
since the beginning of the Victorian era increased industrial
pollution.  But awareness of these dangers has only become real
since the threat of the atomic bomb.  Radioactivity measurements
in different parts of the world have discovered unsuspected
methods of propagation through aerial, water and biological media.
Research Objectives
     Every substance or factor which is a possible risk to man's
health or comfort or which can indirectly touch his natural,
cultural or economic patrimony is considered an environmental
nuisance.  This definition, obviously of a general nature must

-------
                              1627
be practically limited to the search for the pathological effects
of contamination of water, air and food by  chmeical, physical
and biological agents.
     Most of environmental nuisances are interdependent, for
example alcohol, tobacco and traffic casualties.
     The multiplicity of environmental factors has as a counter-
part the variety of their physiopathological effects.  Choices
must be made.  Short-term objectives (acute toxicity, morbidity
and mortality, urban  factor) have mostly an applied aspect.
Long-term objectives  (penetration, metabolism, target organs,
enzymatic inhibition or induction, genetic and teratological
consequences, mutations, interactions between physical and
chemical agents) are more theoretical and require a fundamental
farsighted approach.

Technical Procedures
     Two distinct complementary methodologies are employed, one
centred on man, the other on animal experimentation.
A-Human studies
     They include geopathology, epidemiological surveys,analyses
of pollution, accidents, and occupational risks.
     Geopathology investigates the action of soils  (rural, urban
or industrial) on pathology  (malaria, Burkitt's tumor, air pol-
lution of the Maurienne valley)  and on human ecology.
     Prospective inquiries require a very large number of sub-
jects if the pathological consequence is rare, sometimes more
than 200,000 subjects.  In retrospective inquiries  the definition
of a control group  is difficult.  For each sick subject of the
survey a control of the same age, sex, socioeconomical, geogra-
phical and racial origin must be matched  (7).
     Epidemiological  surveys can be divided into surveys of
mortality and morbidity.  In France a certain uncertainty per-
sists in the diagnosis of lethality.  Morbidity criteria are
difficult: absenteeism, exposure to medical and pharmaceutic

-------
                              1628
agents.  The Swedish studies on homozygote and heterozygote
twins have enabled examinations of genetic factors in environ-
mental toxicology  (smokers and non-smokers).
     Occupational medicine necessitates a good knowledge of well
specified environmental factors and investigates immediate pre-
ventive measures.  They must be developed rapidly especially in
view of the increase in the number of new synthetic chemicals
and their potential pathology  (skin or respiratory allergies:
farmers, lungs, byssinosis, bird breeders disease).
     Certain acute episodes or accidents  (air pollution in
London in 1952, Mimamata's disease, diethylstilboestrol genital
cancers in young girls) are privileged observations.
B-Animal experimentation
     Considered as the quickest and safest way to obtain data on
acute or chronic toxicity, animal experimentation gives rise to
specific difficulties.  A 1/10,000 risk will necessitate treating
10,000 or 30,000 rats or mice, supposing that the effect is the
same in these animals and in men.  In fact teratogenic effects
must be assayed on 4 or more different species of mammals
(thalidomide).  The choice of an "animal model" imposes certain
constraints (guinea pigs for allergical studies).   Rats and mice
are genetically well determined, and can be obtained pathogen
free, but repeated blood of tissue sampling requires larger
animals.  Allometrical laws (8,9) should be taken into account
but only between certain limits  (diameters of dust particles
penetrating into the airways of different species.  On small
laboratory rodents, which have a short life-span,  it is possible
to study long-term effects of pesticides, food additives and
ionizing radiations on longevity, fecondity,  offsprings and
population dynamics (10).   They enable investigations of genetic
factors to be made which more and more are shown to have a role
in the multiple-factor etiology of many diseases (cancer, anti-
alpha- 1-trypsine and emphysema,  (11).   Within the same species
it has often been demonstrated that in addition to strain, sex
and age interfere in toxicity studies  (12,13,14).

-------
                              1629
C~Statistical procedures

     Statisticians must consider  the  possibility  of  interference

of temporal  trends  ^ultradian,  circadian  and  infradian rhythms)(15}

References

 (1)  RECLUS, E. ,  L'homme  et  la  terre.  Albin Michel publ.,  Paris
     1931, 2nd  edit.,  I,  p.43.

 (2)  DUBOS,  R., Man, medicine and environment.  Pall  Mall Press
     publ.,  G.B., 1968.

 (3)  BERNARD,J.,  RUFFIE,  J., HSmatoligie  gfeographique. Masson
     publ.,  Paris,  1966.

 (4)  LOURIE, J.A.,  Human  adaptability studies in  the international
     biological programme. PrS-adaptation et  adaptation ge'ne'tique,
     Colloque  Inserm,  Inserm publ.,  1971.

 (5)  TROMP,  S.W., Medical Biometeorology. Elsevier publ.,
     Amsterdam, 1963

 (6)  RIVOLIER,C. , RIVOLIER,  J. , Me'te'oropathologie humaine
     Cahiers Sandoz publ.,  1972,  no.  22.

 (7)  SCHWARTZ;  D.,  DENOIX,  P.P., L'enquete francais sur
     l'6tiologie  du cancer broncho-pulmonaire.  Role  du tabac.
     Sem.  Hop.  Paris,  1957,  no.62-67, p.  424-437.

 (8)  BRODY,  S., Bionergetics and  growth.  Reinhold publ., New
     York, 1945.

 (9)  STAHL,  W.R.  Physiological  similarity and modeling.  The
     application  of dimensional analysis  and  physical similarity
     theory  to mammalian  physiology.   Appleton-Century-Crofts
     publ.,  New York,  1970.

(1O)  PALMES, E.D.,  DELPUP,  J.,  Mouse  populations  : Biological
     systems for  evaluating  effects  of environmental agents.
     Trans.  N.Y.  Ac. Sci.,  1970,  _3_2,  p.381-384.

(11)  ERIKSSON,  S.,  Studies in a-antitrypsin defiency Acta Med.
     Scand., 1965,  suppl. 432

(12)  STUPFEL,  M., ROMARY, F., MAGNIER, M., POLIANSKI, J.,
     Toxicite1  aigue compared, chez la souria  male et  femelle
     de quelques  polluants atmosphe'riques: gaz d'6chappement de
     moteur  d1automobile, ozone,  ammoniac et  gaz  carbonique
     C.R.Soc.  Hiol.,  1971,  165, p. 1869-1872.

(13)   STUPFEL,  M. , MOUTET, J,P., MAGNIER,  M.,  An apparently
     paradoxical  action of aging  decrease of  acute hypoxic
     mortality in male aged  rats. J.  of Gerontology, 1974, on press

-------
                             1630
(14)  STUPFEL,  M.,  ROUSSEL,  A.,  :  Influence du sexe sur la mor
     tal! te  par 1'hypoxie,  confinement et oxyde de carbone.
     C.R.  Soc.  Biol.,  1969,  163,  p.  310-313.

(15)  HALBERG,  P.,  :  Chronobiology.  Ann.  Rev.  Physiol., 1969,  !
     p.  675-725.

-------
                               1631
       UTILISATION DE CULTURES DE TISSUS HUMAINS POUR LA
         RECHERCHE D'UNE EVENTUELLE SYNERGIE ENTRE LES
    EFFETS TOXIQUES DE DEUX SOLVANTS USUELS:  LE TOLUENE ET
                          LE BENZENE

                  D, GRADISKI, J, L. MAGADUR

Institut National de Recherche et de Sficuritfi pour la prevention
des accidents du travail et des maladies professionnelles,
Vandoeuvre, France
RESUME

     L 'adaptation de plans d'experience factoriels aux cultures
de cellules, d'origine humaine, permet I*etude des interactions
complexes entre toxiquee.   La realisation du plan se fait en
soumettant des cultures, de facon repetee, a des combinaieona de
doses des toxiques Studies.

     L'exploitation des donnees utilise la methode des moindres
cavres, en se fixant un certain nombre de conditions evidentes.
L'action des toxiques et leur interaction sont soumises globale-
ment au test F de Snedecor, avant de les comparer a la valeur
zero, pour chaque point de mesure, par le test t de Student.

     Ce plan, applique aux melanges benzene/toluene, a faible
teneur de benzene, revele, dans les intervalles de doses soumises
d. I'experience, et pour les premieres 24 heures:

     - une action benzine independante de la dose,
     - une action toludne dependante de la dose,
     - une -interaction complete benzene /toluene qui, pour cer-
taines doses caracteristiques, n'est pas significative.   Lorsqu'
il en est autrement, I'interaction se manifesto dans le sens d'une
diminution de I'effet toxique.

-------
                               1632
ABSTRACT

     The adaptation of the factorial planning of experiments to
human cell cultures makes it possible to study complex inter-
actions between toxic substances.   The cultures are exposed
repeatedly to combinations of doses of the toxic substances under
investigation.

     The data are evaluated by the method of least squares, with
a certain number of obvious conditions.   The effects of the
substances, and their interaction, are all submitted to the
Snedecor F test before they are compared at the zero value, for
each reading by Student's t test.

     This plan, applied to benzene/toluene mixtures having a low
benzene content, showed, for the doses tested and for the first
24 hours:

     - an effect of benzene independent of the dose,
     - a dose-dependent effect of toluene,
     - a complex interaction between benzene and toluene, which
at certain characteristic doses was not significant.   Otherwise,
the interaction shows itself by a reduction of the toxic effect.

-------
                                    1633
I .  Introduction
   La recherche d'interactions complexes entre les effets toxiques de deux
substances chimiques necessite le traitement statistique d'un grand nombre
de valeurs obtenues en combinant, deux a deux, un certain nombre de doses
de substances etudiees, chaque cas devant etre repete plusieurs fois.
L'emploi des cultures de cellules rend cette recherche tnoins lourde qua
lorsqu'on utilise des animaux de laboratoire, technologiqueraent et finan-
cierement. Elle peut se suffire a elle-meme, si on ne s'interesse qu'a la
cytotoxicite, ou si le comportement des cultures, pour les toxiques consi-
deres, est un bon modele de ce qui a ete observe en toxicite aigue sur ani-
mal (D.Gradiski, J.L Magadur |l|). Dans les auLres cas, elle peut servir de
modele a une experimentation animale ulterieure ou etre completee par une
simulation. L'etat actuel de la technique, telle qu'elle est utilisee dans
notre laboratoire,  est expose a propos de la recherche d'une eventuelle
interaction benzene/toluene, dans les melanges dans ces deux corps, a fai-
ble proportion de benzene.
2.  Plan experimental
   Les deux toxiques sont consideres comma deux traitements (T = toluene ;
B = benzene) dont les differentes doses representent les degres (K. doses
pour le toluene, K, doses pour le benzene). Chaque combinaison de doses est
repetee n fois.
Le parametre observe est le nombre de cellules par ml, compte a la cellule
"Nageotte". Chacune des mesures est reperee : x{ . r , 0 < i < K
Le modele adopte est le suivant :                  7   ^ J
                                                   \ 1 < r < n
  xjjr = u + a(  + bj  + Cjj + ejjr                                       (1)
ou :   y est la moyenne des temoins non traites (xu ),
      a, est 1'ecart par rapport a y, du a 1'action du toluene,
      bj est 1'ecart par rapport a yt du a 1'action du benzene,
     c^ est 1'ecart par rapport a p, du a 1'interaction benzene/toluene,
   Gjjt  est une realisation de la variable gaussienneede moyenne nulle et
        de variance inconnue.
Le probleme a resoudre s'enonce ainsi :
L'echantillon de mesure x-t:r dont on dispose, permet-il de refuter 1'exis-
tence des termes 3j,bj et GJJ , avec une probabilite de se tromper inferieure
au seuil a.
Le mode de resolution utilise la metnode des moindres carres,  en posant les
relations evidentes %  = 0 ;  b0 = 0 ; c|0  - 0  ; ^   =0.

-------
                                    1634
3. Mailipulation
   Les raises en cultures et traitements par les toxiques sont effectues
comme decrit dans un travail precedent (D.Gradiski, J.L Magadur |lj). Les
cellules appartiennent a la lignee Hela. Toutes les precautions sont prises
pour ne pas introduire un facteur de variation non controle. Les doses
employees sont respectivement les suivantes (en moles/ml)  :
Toluene : 0; 2,3. ICT9 ; 2,3. lO'8; 2,3.IO'7
Benzene : 0; 2,7.10-"; 2.7.10-10; 2,7.10'9; 2,7.IO'8; 2,7.]0'7; 2,7.IQ-6
Chaque combinaison de doses est repetee trois fois. Les comptages sont
effectues apres 24, 48 et 72 heures. Seuls les resultats obtenus a 24 h
sont discutes.
4. Resultats experimentaux et traitement statisEique
   II est impossible de fournir ici tous les resultats experimentaux, de
roeme que toutes les valeurs statistiques intermediaires. Un double tableau
presente (Fig. I) : - dans la par tie superieure, de fa^on  intuitive
. les resultats experimentaux, exprimes en % d'inhibition de developpement,
  par rapport aux cultures temoins (traits pleins),
. la surface theorique correspondant a un modele purement additif (pas
  d'interaction) (tirets) ;
                    - dans la partie inferieure :
. les valeurs trouvees pour les differents 3jet bj, termes d'action du ben-
  zene et du toluene (premiere colonne et premiere ligne),
. les valeurs trouvees pour les c(j , termes d'interaction
5. Discussion
   Le test F de Snedecor, effectue pour I1ensemble des resultats au premier
jour, associe au test t de Student, pour rejeter 1'hypothese de nullite des
aj,bj,Cjj, permet d'affirmer :
. que le benzene exerce une action sur le developpement des cellules Hela
  (p>0,95). Dans 1'intervalle de doses choisi, cette action atteint son ma-
  ximum des la premiere dose ;
. que le toluene exerce aussi une action (p>0,95) tendant vers une limite
  superieure,
. qu'il existe une interaction benzene/toluene (p>0,999). Cette interaction
  n'est pas significative pour certaines combinaisons de doses (en particu-
  lier celles contenant IO-8M/ml de benzene). Lorsqu'elle est significative
  elle s'exerce toujours dans le sens d'une diminution de  1'effet toxique
  total.

-------
                                      1635
                                                     Figure  n°  1
50% d'inhibit ion
     -15600 \-IIW X«OM  \7flH  \1000  X 21 IK X 1
                   HM  \ISZ67  V»S67  '.X -??67 X (ill
                              i, ,„ \ „ w'-N^i IM\, U
                                  "
          i  /       \     \
        \/ ai \I546/  VlS 1)6 / Vl
                                     1)6 /  Vl5 06)
                                                    OOQ
                                                                     BtMl/AMH'
                                                                     2,7  M/ml
533 X-15 933
Ltyende  :  La projection d'un point de mesure sur la case correspondante du
tableau   inf6rieui* permet de lire  : dans  aette case,  le terme 
-------
                              1637
    EXPOSURE OF LABORATORY ANIMALS TO URBAN AIR POLLUTION:
  PHYSICAL PLANT, EXPERIMENTAL DIFFICULTIES, OBSERVED EFFECTS

ANDREW L, REEVES"*"/ THOMAS F, MOONEY, JR/ AND RALPH G, SMITH"1"*"

+  Department of Occupational and Environmental Health, Wayne
   State University, Detroit, Michigan, USA
++ School of Public Health, University of Michigan, Ann Arbor,
   Michigan, USA
ABSTRACT

     Approximately 3500 animals were involved in a 5-year experi-
ment to determine the health effects of air pollution as it oc-
curs in the center of a large city (Detroit).   One-half of the
animals breathed air from out-of-doors constantly, with the air
delivered to them into chambers with as little change as possible
except temperature control.   The other half of the animals
breathed air after passage through activated charcoal and fine
pore filter banks.   Elimination of various pollutants ranged
from excellent (particulate matter, oxidants) through fair (nit-
rogen oxides, S02) to poor (hydrocarbons3 CO).   Several strains
of mice, rats, rabbits and guinea pige were exposed and studied
during their lifetimes, in some cases including breeding and
observation of progeny.

     Mortality, morbidity, body weight, body temperature, electro-
cardiography, pulmonary function, the results of periodic blood
tests including hemoglobin, hematocrit, and red blood cell count,
and starch gel zymograms of esterase, lactate dehydrogenase,  and
acid phosphatase of pulmonary and lymphnode tissues showed no
significant differences between the colonies,   Histopathologic
studies on various tissues were also unremarkable except for the

-------
                              1638
amount of particulate matter present in the lungs.   Possible
effects of borderline significance included elevated plasma
alkaline phosphatase in exposed rabbits and increased spontaneous
activity (by running wheel) in young exposed mice.   The only
pronounced and sustained difference between the colonies was in
the white blood cell count, attributable chiefly to an increase
in the number of lymphocytes i,n the exposed animals.

-------
                                   1639
      Tin1 health effects of urban air pollution were studied intensively in
rc< cnt years by means of human epidemiology[lJ and animal  experiment
[l].  During the years 19&0-65,  a study involving the latter approach was
conducted at Wayne State University in Detroit, under auspices of the
U. S.  Public Health ServicejY).

      The guiding principle of the Detroit study was to expose a colony of
experimental animals to  ambient levels of air pollution as it  occurs at a
busy downtown intersection with as little change as possible and compare
the  long-range health status of these animals to that  of an identically
maintained control colony which breathed as clean air as it was possible
to obtain.  An extensive physical plant was constructed in order to realize
these objectives (Figs. 1, 2).  Street air was sucked  into a 60 cm diam.
   Fig.  1.   Air pollution  intake duct and fan house,  top view.

-------
                                     ',
                                    *i  .


   Fig. 2.  Air pollution  intake duct, fan house,  and  laboratory
            building,  horizontal view.

duct which arched over a street at a height of 4. 4 m with 12. inlet holes.
The air passed through a fan house where it was heated or cooled accord-
ing to the season and then delivered into the laboratory building where the
animal rooms were located.  For the ambient air room the tempered air
was delivered directly through  a perforated wall.  For the clean air room,
the airstream was  passed through a set of absolute filters and activated
charcoal absorbers and subsequently delivered through another perforat-
ed wall.   The air was exhausted from each animal room through the oppo-
site perforated wall, with a second fan located in the discharge duct.  A
pressure  balance was maintained at a slightly positive level in order to
minimize outside contamination of the animal  rooms through the doors.
The system was capable of handling 100-350 m3 air  per minute,  depend-
ing on the outside temperature, resulting in 60-180 air changes per hour.
Delivery of outside air into the  animal rooms  took no more than  5 seconds
so that changing outside conditions due to varying  traffic patterns were
faithfully  reflected  in the  experiment.

      Table 1 illustrates  the degree of success of  the physical plant.
First, it can be seen that ambient air room levels resembled outside

-------
                                 1641
                           Table 1
              Instrumental Analysis of Air Quality
Contaminant
Total particulate
Total oxidant
Total hydrocarbon
CO
S02
NOz
NO
Unit
Y/mj
p. p. h. m.
p. p. m.
p. p. m.
p. p. h. m.
p. p. h. m.
p. p. h. m.
Typical Monthly Means
outside
building
95
0.5
2.6
3. 2
10
5.7
9.0
ambient
air room
91
0.6
2. 7
3. 1
8.8
6.5
11. 2
clean
air room
1.4
0. 05
2.5
3.0
5. 1
1.7
14.2
levels for all contaminants,  so that the atmosphere was indeed delivered
to the animals with no substantial change.  Second, it can be seen that
with respect to difference between the ambient air and clean air rooms,
the contaminants may be divided into 3 groups.  Substantial concentration
differential was seen with particulate matter and oxidant only.  Moderate
concentration  differential was seen with SO£ and NO2-   Essentially no
concentration  differential was seen with CO and  hydrocarbons. An anom-
alous situation existed with respect to NO, which was apparently created
during the  cleansing procedure,  resulting in higher concentration in the
clean air room than in the ambient air  room.  The two nitrogen oxides
together underwent moderate reduction of concentration during filtering.

      The experimental animals  exposed to these atmospheres included
650 mice,  600 rats, 400 guinea pigs, and 100 rabbits of various strains
in each colony.  The animals were kept in their  respective exposure
rooms throughout their lifetimes, in some cases including breeding and
observation of progeny.  Air pollution created by their own excrements
was carefully controlled by a thorough sanitation schedule.   Feeding,
watering, sample collection, and other handling was strictly identical
for the  two colonies.

      It was recognized that filtering of air involved removal of micro-
biological contamination as well, and since this  was not one  of the desired
variables the  incidence of infectious conditions had to be equalized in the
two colonies.  For this purpose,  25-50 additional animals of each species
were designated as "bacterial flora carriers" and alternated between the
exposure rooms on a daily basis.  The latter animals were otherwise not
evaluated in the experiment.  Equalization of bacterial flora was satis-
factory as  evidenced by very nearly identical mortality and gross mor-
bidity rates between the colonies for each species.

      In order to monitor subtle manifestations  of morbidity or other
physiological  responses attributable to air pollution, an animal testing
schedule was  established involving physiological measurements on a

-------
                                 1642
multichannel analyzer and blood tests including hematocrit,  hemoglobin,
cell counts, and serum electrophoresis.  At periodic sacrifices,  lung
and tracheobronchial lymph node tissues were obtained and examined
histopathologically and biochemically.   Mice had voluntary exercise
wheels in their cages with automatic revolution counting devices.

       The  summary results on all these measurements (Table 2)  showed
that the only really significant difference  between the animal colonies
was in the particulate load of the lungs  (attributable chiefly to increased
                            Table 2
      Measurable Differences Between Animal Colonies*
Parameter
Rat WBC counts. xlO3 (with lymphocyte %)
Typical Monthly Means
ambient air
8.6(78)
Guinea pig WBC counts, x!03(with lymphocyte %) 8. 2(61)
Rabbit WBC counts, xlO3 (with lymphocyte %)
Rabbit serum alk. phosphatase(KA units/1 00 ml
Rabbit lung ash, Al {p. p. m. )
Rabbit lung ash, Ba (p. p. m. )
Rabbit lung ash, Pb (p. p. m. )
Mouse voluntary activity (103 revs. /wk. )
4. 8(62)
5. 90-6. 21
249
5.6
23. 5
29
clean air
7.6(73)
7.3(56)
3.6(56)
5.73-5.77
170
2.9
10.4
27
*Lack of reproducible differences was experienced with respect to the
following parameters: Body weight; body temperature; electrocardio-
gram; pulmonary function; kind and degree of pulmonary histopathology;
plasma  electrophoresis; isozymes of esterase, acid phosphatase andlac-
tate dehydrogenase; hemoglobin; hematocrit; erythrocyte counts; and the
following elements inlung tissue:Cr, Co, Cu, Mn, Mo, Ni, Sr, Sn, Ti, V.and Zn.
 levels of aluminum,  barium, and lead in the exposed animals); and in
 the white blood cell counts (attributable chiefly to increased number of
 lymphocytes in the exposed animals).  Small increases of questionable
 significance were also seen in the serum alkaline phosphatase levels of
 rabbits and in the voluntary activity of young mice.  Unaffected were
 essentially all other  measured parameters including specifically the
 physiologic and histopathologic evaluations of animal health.

      The increased  particulate load of the lungs was a natural conse-
 quence of breathing unfiltered air and the increased lymphocyte counts
 probably reflected a  phagocytic response to this challenge.  This is a
 normal defense mechanism of the respiratory system and it appears that
 the physiologic reserves of the organism were not exhausted by the ex-
 posure.   It may be concluded that breathing ambient city air did not
 measurably impair the health of these animals.   However,  the increased
 heavy metal burden in the lungs is remarkable and the possible chronic
 effects of this accumulation were not evaluated in this study.

-------
                                  1643
                          References
[l]   Magnuson, H. J. , "Health Effects of Air Pollution"inj Air Pollution
     Manual I, 2nd ed. (P. M. Giever, Editor), pp. 33-44, Am. Indust. Hyg.
     Assocn. , Akron, 1972.
[2]   Reeves, A. L. ,  "The Effect of Air Pollution on Animal Health", in;
     Air Pollution Manual I, 2nd ed. (P. M. Giever, Editor), pp. 45-59, Am.
     Indust. Hyg. Assocn. , Akron, 1972.
J3j   Smith, R, G. (Principal Investigator), "Biologic Response of Animals
     Exposed to Urban Atmospheres Characterized by Automotive Emis-
     sions in Detroit", U. S.  Public Health Service Contract  No. Saph.
     69734, awarded 1958, concluded 1965.

-------
                              1645
        EFFETS SYNERGIQUES DES POLLUANTS PARTICULAIRES
  ET D'INFECTIONS EXPERIMENTALES TUBERCULEUSE OU ORNITHOSIQUE

            J, C, MARTIN, L, LE BOUFFANT, H, DANIEL

Centre d'Etudes et Recherches des Charbonnages de France,
Verneuil-en-Halatte, France
RESUME

     Pluaieurs types de modules experimentaux sont realises,
comportant des empousaieragea de particules mindrales diverges
associes & dea -infections de natures differentes.   Deux types
de contaminations bacteriennes sont r6aliseest soit pav voie
intro-veineuae pour I'infection tuberculeuse  (Mycobacterium
Kanaasii) soit par voie intra-peritoneale pour I'agent de I'or-
nithose.

     Les resultats obtenus montrent une aggravation indie outable
des infections pulmonaires ou generales par les poussieres,
Toutefoie oette aggravation eat oonmandee par:

     - la nature de I'agent infeotant et I'existenae ou non d'une
phase de parasitisme cellulaire obligatoire;
     - la nature et la nooivitS cellulaire propre de ahaque
pousai&rej
     - la quantity totale de particules retenues au niveau du
poumon dans le cas des infections de nature tuberouleuse.

     Les meoanismes primaires responsables des phenomdnee d'ag-
gravation n'apparaissent pas univoques.   Us eont ainei lie's
au point d'impact particulier et au mode de multiplication ap£-
oifique de chaque agent infectant d'une part et a la oytotoxioite
propre de chaque poussidre d'autre part.   lie conditionnent
ainsi le type de reaction et lee effete synergiques constates.

-------
                              1646
     Les nombreux travaux eonsacres independamment a I 'action
des polluants mineraux et aux diverses infections pulmonaires
ont permis de mieux aonnattre la nooivit£ et la pathologie par~
tieulaire li&e a ehaeun de oes agents.

     Neanmoi.ns3 il est frequent sinon habituel en pathologie
humaine, surtout dans aertaines  activites industrielles3 de voir
se combiner au niveau pulmonaire notammentt I'action d'agents
pafhogenes multiples.   Ainsi en est-il des polluants partiau-
laires mineraux et des infections.   C'est a quelques-uns de
ces effets synergiques que se rapporte oe travail.
ABSTRACT

     Various types of experimental model were constructed for
exposure to various mineral dusts associated with different  types
of infection.   Infection was induced  either by  the  intravenoue
injection of tuberculosis  (Mycobacterium kansasii) or by the
intraperitoneal injection of ornithosis.

     The results  showed  a definite aggravation of pulmonary  or
general infections by the dusts.   However,  the  aggravation
depended on:

     - the nature of the infective agent,  and the presence or
absence of a phase of obligatory  cellular  parasitism;
     - the nature of each dust  and -its ability to damage cells;
     - the total  quantity  of particles retained  in the  lung  in
the  case of tuberculous  infections.

     The primary  mechanisms responsible for the  aggravation
phenomena appear  to be  somewhat ambiguous.   For examplef  they
are  linked on  the one hand with the  point  of impact  concerned
and  the specific  mode of multiplication of each  infective  agent.

-------
                             1647
and on the other with the specific cytotoxicity of each dust.
In this way they condition the type of reaction and the syner-
gistic effects which are observed.

     Numerous individual studies relating to the effects of min-
eral pollutants or to various pulmonary infections have provided
further knowledge of the noxiousness and pathology specific to
each of these agents.

     Nevertheless,, it is the combined effects of a number of
pathogenic agents which are observed frequently if not regularlyt
notably in the human lung, in the context of certain industrial
activities.   The same is true of particulate mineral pollutants
and infections.   The present study is concerned with some of
these synergistia effects.

-------
                                    1648
1. Methodes experimentales
        1.1 - Poussieres
        Les differentes poussieres (charbon, quartz, oxyde de titane, chry-
sotile) utilise'es sont des particules inferieures a 5 microns. Elles sont
administre'es soit par empoussie°rage dans des cages CERGHAR a regulation au-
tomatique, soit en suspension dans un liquide physiologique dans le cas
d*injections pe*ritone"ales.
        1.2 - Infection
        Deux types de germes ont e"te inocule"s :
        - soit par injection intra-veineuse pour Mycobacterium Kansasii ;
        - soit par vole intra-pe'ritone'ale pour I1 agent de I'ornithose
(Chlamydia psittaci).
        Les animaux ont 6t& choisis en fonction de leurs sensibilite's aux
diffe'rents germes : cobayes pour I1infection mycobacterienne, souris IOPS-
OF pour I'ornithose.

2. Re"sultats
        2.1 - Empoussie'rase et infection mycobacte'rienne (Mycobactdrium
              Kansaaii)
        Les animaux ont e'te empoussieVe's avec des melanges charbon-quartz,
a raison de 300 mg/m3 d'air, pendant des peYiodes de temps variables, de fa-
gon a obtenir des charges pulmonaires progressivement croissantes. Deux
series ont &t& r^alis^es, comportant des melanges oharbon-quartz a faible
et forte teneur en quartz. Les re"sultats obtenus six semaines apres l'ino-
culation bact^rienne sont representds dans le tableau I. Us montrent :
        - I1 aggravation nette des lesions tuberculeuses par I'empoussie'ra-
ge pre"alable, aggravation Juge'e sur le pourcentage de la surface des le"-
sions histologiques pulmonaires et sur le nombre de myeobaote'ries viables
 re'isole'es des poumons  aprfes 6  semaines devolution ;
         - I1augmentation proportionnelle des  lesions en fonction de  la quav
 tit^ de  poussleres pr^sentes dans  les poumons ;
         - a  quantit^ ^gale, que la presence de particules a nooivit^  oel-
 lulaire  ^lev^e  (quartz)  dans un melange de poussieres aggrave de fa?on
 Importante le deVeloppement des lesions tuberculeuses.
         2.2  - Bnpoussle'rage et infection ornithosique (Chlamydia pslttacij^
         La suspension  infectante est obtenue  a partir du sao vitellin
d'oeufs  embryonn^s.  Apres centrifugation la suspension  riche en corps ele"-
mentalres est inocule"e par  voie intra-p^rlton^ale a la  souris apres titra-
tion de  l'inoculum sur cellules L en couches  monocellulalres.

-------
                           1649
                         TABLEAU I

   Influence de la change conlotique et de  la teneur en
            quartz sur 1*Infection tuberculeuse
(
, Poussieres
(
( 	
( Melanges
( falble
( teneur
( en quartz
/
( Melanges
( A
( forte
( teneur
( en quartz
%
fV
de quartz
dans les
poumons

3,2
5,1
5,7
5
6

6,2
11,5
12,8
15,7
64,5
Charge pulmo-
naire en
poussieres
en mg

13,4
26,6
34,5
62,2
83,5

38,7
23,8
33
44
34,3
Dose
de
Myc obact e"rium
Kansas!!


8.io5




5
6,5.10


Indice : Nombre )
histo. : de )
des :bacilles)
lesions: reisole"s)
, \
1,4 : 50 ) )
7 : 30 ) J
13,2 :424 )xltf)
12,8 :462 ) )
24,9 :676 ) )
\
2,2 ! 45 ) )
0,9 : 86 ) ]
5,2 : 1.295) 10?
4,2 :1.425) )
14,5 :3.765) )
                        TAWKftU II

Pourcentage de sourls survlvantes apres empoussidrage  et
       Infection par Chlamydia psittaci -  (8 U.I.)
(
(
( Orni those
( Quartz + Orn.
( Charbon + Orn.
( Charbon t^moin
( Quartz t^moin
/
\ 	
( Ornithose
( Ti02 + Orn.
( Chrysotlle + Orn.
( TiOp temoin
( Chryaotile temoin
Dose
infect ante
10-|
10
10"2
--
—


10" *
10"
10"1

PourcentaK*
au 8fe jour
80
60
30
100
100


90
85
4
100
100
i de survie )
au lOe Jour )
70 }
50 )
20 )
100 )
100 )
	 	 	 m)

90 )
74 )
4 )
100 )
100 )

-------
                                     1650
        Les poussieres minerales utilisees (quartz, charbon, oxyde de tita-
ne, chrysotile) sont broyees et triees en particules < 5 u.. Elles sont en-
suite injectees par vole peritoneale a la dose de 20 mg par animal.
        Les resultats exprimes en pourcentage de survie des animaux sont
exprimes dans le tableau II. On remarque que la seule injection de particu-
les minerales n'entralne aucune mortallte. Par centre, pour les animaux
inocules avec 1'agent de 1'omithose, I1 injection pre"alable d'une poussiere
entralne vine aggravation de la maladie ornithosique qui se traduit par une
acceleration dans l'apparition des symptomes de la maladie et par une aug-
mentation du taux de mortalite. On constate d'autre part que cette aggrava-
tion est variable suivant la nature de la poussiere utilised (tableau II),
Elle est lei moins importante pour le quartz, mineral a toxicite cellulaire
eleve*e, que pour le charbon. Ce phenomene est probablement en relation aveo
la phase de parasitisme cellulaire obligatoire de Chlaraydia psittaci,  celle
ci ne pouvant se produire en raison de la lyse importante des macrophages
perltone"aux provoque's par le quartz.

        Ces resultats, obtenus au niveau pulmonaire ou pe'ritone'al, sont
repr^sentatlfs de lraction aggravante de la poussiere sur oertaines infec-
tions e3qp^rimentales. Us demanderaient toutefois,  avant d'etre generalises
une etude plus etendue, notamment avec diffbrents types d1infections,  dont
les modalites de deVeloppement peuvent s'sv^rer extrSmement variables d'un
gerrae a I'autre.

                               Bibliographie
  POLICARD A., GERKEZ-MEUX C., TACQUET A., MARTIN J.C.,  DEVUIDER B.,
  LE BOUFFANT L.
  Influence of pulmonary dust load on the development of experimental
  infection by Itycobacterium Kansas!!
  Nature, 1967, 216, 177
  ORPILA J., LEPINAY A., VASSUER M., LE BOUFFANT L.,  MARTIN J.C.,
  DANIEL H.
  Influence des poussieres de quartz, charbon, oxyde de titane et amiante
  sur la maladie exp^rlmentale de la souris a Chlamydia psittaci
  C.R. Acad. Sc. Paris, 1972,

-------
                              1651

      LES EFFETS DES POLLUANTS SUR L'EPURATION PULMONAIRE

     L, LE BOUFFANT/ H, DANIEL/ J, C, MARTIN/ J, P, HENIN
Centre d1Etudes et Recherches des Charbonnages de France,
Verneuil-en-Halatte, France
RESUME

     It'Duration pulmonaire a et& mesurSe ahez diff^rentes  ea-
pScee animales:  chat, rat, singef eoit dans des  conditions nor-
malee en vue d'en pr&oiaer lea pyi.noi.paux me'aanismes,  aoit  sous
faction de divers polluants particulaires ou gazeux.    Deux
m^thodes comportant une inhalation unique ou re'pe'te'e de  pous-
eiere ont ete utiliee'eB:  I'une oonaiete d doeer  lee poueeidvee
dane lea poumono dee animaux sacrifice's par lots  en fonotion du
tempe;  I'autre neoeseite I'emploi d'une pouesi&re marquee  (h6-
                  59
matite marquee au   Fe)t la radioac
meeuree par des oomptagee externee.
                  59
matite marquee au   Fe)t la radi.oaoti.vitS intra pulmonaire  Stant
     Cette deuxidme methode a  servi egalement  a  faire  dee  meeuree
de retention et de capacite* depuration  chez  I'homme,  lee  doaee
de pouesi&re marque'e Stant tree faiblee  (infSrieuree & 2 micro-
curie) .

     Lee r&eultate obtenus permettent de faire lee  constatations
suivantes:
     - il exiate des differences  importantee  dans  la viteeee  d'6-
puration eel on la nature dee pouesi&res  et  aelon I'eep&oe  animale
consideree.   La difference eet peu important^ au niveau de l'&-
puration bronohique qui eet tree  rapidej par  oontre au. niveau
alv&olaire, la viteeee d'tpuration varie dans  une  large propor-
tion:  rapide ohez le chat, elle  eet tree lente  ches I'homme.

-------
                              1652
On constate en outre une variation individuelle importance au
sein d'une mime esp&ae.
     - la capacitS depuration pulmonaire mesuree ahez le rat
eat nettement affect&e par dee  inhalations anterieuree de pous-
eieres qui ont pour effet de ralentir le ph&nomene d'autant plus
fortement que la charge aoniotique est plus elevSe,
     - lea gaz irritants tela que lee vapeurs nitreuses et I 'an-
hydride sulfureux tendent a favoriaer legerement I 'elimination
dea pouaaierea intra-alv&olaires.
ABSTRACT

     'Pulmonary clearance was measured in various animal species
(cat, rat, monkey) either under normal conditions, to determine
the principal mechanisms, or under the influence of various par-
ticulate or gaseous pollutants.   Two methods were used, entail-
ing a single inhalation or repeated inhalations.   The first
consists in determining the quantity of dust in the lungs of
groups of animals sacrificed after various time intervals, and
                                                   59
the second is based on the use of a labelled dust (  Fe - label-
led haematite) and the determination of intra-pulmonary radio-
activity by external counting.

     The second method was also used to measure retention and
clearance capacity in man, using minute quantities of labelled
dust (less than 1 microcurie),

     The results obtained permitted the following conclusions:
     - The rate of clearance varies considerably according to
the type of dust and the animal species in question.   Bronchial
clearance is rapid, and the differences are slight}  the rate of
alveolar clearance on the other hand, varies considerably, being
rapid in the oat and very slow in man.   There are also consid-
erable individual variations within a given animal species.
     - Clearance capacity measured in the rat was clearly af-

-------
                              1653
fected by prior inhalation of dusts  the slotting down of the
clearance process becomes more marked as the quantity of dust
increases.
     - Irritant gases, such as nitrous oxide and sulphur dioxide,
show a alight tendency to favour the elimination of dust from
within the alveoli.

-------
                                     1654
1.  Introduotlon
          Les recherches consaore*es depuis plusieurs anne'es a I1Duration
pulmonaire ont permis d'en preolser les principaux me'canismes et d'en eta-
bllr  les loia, roettant notamment en Evidence I1existence d'une epuration
bronchique rapide et d'une Duration alve'olaire lente.
          Cette derniere presentarrt une importance partlculiere en ce qui
concerne la genese des lesions pulmonaires d'origine coniotique, 11 a paru
esaentlel de tenter d'en determiner les caracte'ristiques et les variations
indivlduelles, alnsl que d* analyser les modifications qu'elle peut subir du
fait  de diffe"rentes agressions d'origine exte"rieure ou encore de certaines
lesions de I'appareil pulmonaire lui-m§me.
2.  Me"thode exp^rimentale
          Les reoherches ont e'te' menses sur trois espeoes animales : le clHl;
le  rat et le singe, ainsi que sur 1'nomine.
          2.1 - PQuasiere3
          L'empoussie'rage pulmonaire eat r^alls^ par inhalation d'une sus-
pension de particules infeVieures a 5 JJ a la concentration de 300 mg/m
dans  des cages a regulation autoraatique. Quatre types de poussleres ont £t6
^tudl^s : le quartz, le oharbon, 1'o^rde de titane et 1'heViatite.
          2.2 - Gaz
          Deux gaz irritants ont eta" testds : les vapeurs nitreuses et I1 an-
hydride sulfureux. L1 inhalation de MDg a e"t<* r^alis^e par bouff^es de 30 mi-
nutes (concentration de 100 a 30 ppm) pendant 3 heures par Jour et l'inha-
lation de 302 a la concentration de 300 ppm a ralson de 2 heures par Jour.
          2.3 - Controle de 1* duration pulmonaire
          Les caract^ristlques d1epuration vls-a-vis d'une poussiere don-
n^e ont 6b6 determiners en suivant I'dllmlnation de la poussiere par des
dosages ohimlques dans les poumons apres sacrifice de I1animal.
          L1 Influence des gaz irritants sur la Vitesse d'epuration des pous-
sieres inhal^es a e'te* mesure'e selon le mSme prooeM^.
          Enfln, 1'dtude oin^tique de 1'epuration pulmonaire normale et des
modifloactions qu'elle sublt sous 1*action de divers faoteurs t nature de la
poussiere, charge oonlotlque, a 6t6 faite par inhalation de poussleres d'o-
                            CQ
xyde de fer Fe20, narqu^ au   Fe puis oomptage de la radioactivlte* y itttra-
pulmonaire*
3. Caraote'rigtiquee de I'epuration
          3*1 ~ Variation selon la nature des pousslerea

-------
                       1655
                      TABLEAU I

               Influence de la nature
     des poussieres sur leur vitesse d'dpuratlon
\ Nature de la
/ poussiere
{
/ quartz
( charbon
/ oxyde de titane
( hematite
! Pouroentage e'pure' \
| apres 46 Jours
* chat
! 45
: 84
! 70
: 76
rat
25
<5
23
55
i
)
)

)
\
                    TflTT,F.AU II

Taux de retention globale et vitesse d'dpuratlon
           chez I'homne. Variations indivlduelles.
(
( Sujet
(
(
i
(
(L...
[B...
(s...
[H...
(M...
I
Retention globale
en % de 1* aerosol
mis en suspension



5,9
8,5
3,8

! 7,2
!
Taux de retention :
alv^olalre (en % :
de la retention :
tot ale) :

•
90 :
68 ;
80 :
76 I
82 :
*
t •
Pe'riode de
I'epuration
alveola! re
(en Jours)


100
117
183
280
560

)
)
)
)
\
/
)
)

)
)
)

-------
                                     1656
          Les re"sultats de 1 etude effeature sur le chat et sur le rat aveo
des poussieres de quartz, de oharbon, d'oxyde de titane et d'he*matite sont
portes dans le tableau I. II apparalt une nette influence de la nature chi-
mlque des poussieres sur leur taux depuration. On observe en outre que le
classement des poussieres en ce qui oonoerne leur vltease d'^llmination n1-
est pas le mSme d'une espece a I'autre. En particulier, I'eVuration du ohar-
bon est rapide chez le chat tandla qu'elle est tres faible ohez le rat.
          3.2 - Variation selon I'egpeoe
          La difference de vitesse depuration d'une mSme poussiere selon
1'espece anlmale considered a e'te" etudie'e en utilisant de I'hernatite mar-
que"e chez le chat, le rat et le singe et une comparalson a e'te' faite avec
I'homme. L'^pxiration des voles respiratoires superieures et des bronches
apparalt rapide dans tous les cas (peViode infeVieure a 2 jours) tandis que
I1 Duration alvdolaire varie de faoon itnportante d'une espeoe a I'autre, coiv-
me le montrent les courbes de la figure 1. Les peViodes moyennes correspon-
dantes ont et^ calcule"es : 16 Jours pour le chat,  28 Jours pour le rat, 250
Jours pour I'homme, 300 Jours pour le singe.
          3.3 - Variations individiielles
          Une e"tude anterieure effeotue'e sur le chat au moyen d1 hematite
marquee avait montr^ I1existence de variations individuelles de la courbe
depuration. On a determine" par la tne\ne me'thode les variations individueltes
du taux de retention et de la vitesse depuration chez I'hotnme. Les r^sul-
tats portds dans le tableau II et ooncernant cinq sujets test^s dans des
conditions d'essai identiques montrent d'une part que la retention globale
varie dans des proportions importantes, d'autre part qu'il existe des dif-
ferences de 1'ordre de 1 a 5 dans les vitesses depuration alve'olaire.
4. Action des polluants sur la capaclte" d'^puratlon
          4.1 - Polluants partlculaires
          Letude, effectu^e sur le rat, a porte" sur trois sortes de pous-
sieres : le charbon, 1'oxyde de titane et I'he'matite. Apres Inhalation de
quantite*s variables de poussieres, la capacite* d'^puration a e'te teste*e a
I'hematite marquee. Le tableau III montre que I1 inhalation de poussieres,
de quelque nature qu'elles soient, dlminue la capacite" depuration vis-a-
vis de nouvelles partioules attelgnant le poumon.  Cette diminution est d*
autant plus marquee que la charge coniotique est plus ^lev^e. On constate
en outre que le retour a la capaoit^ normale depuration deepen! de la na-
ture des poussieres : dans le oas present, 11 n'est observe* que pour 1'oxy-

-------
                                1657
                             TABLEAU  III

              Action des polluants partlculaires sur la
                capaclte d'e*puration  pulmonaire  du  rat
(Nature du
(polluant
(
(
(
(
/
) oharbon
k
(
(
( oxyde de
( titane
(
(
/
> hematite
\
Charge conlotique
par poumon apres
empoussi^rage {mg)

2,7
8,1
30,9
4,9
11,2
20
5,4
25,6
51,6
Pe"riode de I1 Duration alve'olaire (en )
Jours) apres diffe"rents de*lais )
devolution (l) )
0
69
72
365
27
55
95
—
-
-
3 raois
M
-
-
22
47
63
21
]36
154
6 mo is
28
49


^
-

9 mois



20
28
32
19
87
124
l8 raolsx
)
~ \
)
\
277 j
)
™* ^
' )
' )
)
~ \
)
\
)
~ )
(l) periode moyenne des temoins :  25 Jours
             Influence de I1inhalation de  gaz  Irritants
   chez le rat (NOg,  S02) sur I'epuration  pulraonaire des poussieres
/Nature de la
\ poussiere
(
/ Charbon-
/ quartz (t^-
) moin : 2,9
/ mg par pou-
/ mon)
\ Oxyde de
) titane
) (temoin 4,4
/ mg par pou-
/ mon
: Nature du gaz
:et conditions
:d* inhalation
•
:
: N02
apres em-
poussi^rage

: SO
: T2
. apres em-
'poussi^rage
:
: Taux d'^pun
4 semi
Temoin

8,6*




46,6 #


at ion apres
lines
N02

22,4 %




50 %
Variation du }
taux d'dpura-)
tion )
)
)
+ 160 % I
)

)
)
)
+ 7 % )
\
i
. )
: )

-------
                        1658
% Activite dans le poumon
             Homme
             Singe  (macaque)
             Chat
             Rat
  to
20
  Bgl- COURSES  D'EPURATION  PULMONAIRE
        HOMME, SINGE (MACAQUE), CHAT,RAT.

-------
                                   1659
de de tltane.
          4.2 - Polluants gazeux
          L'dtude, effectue"e sur le rat, a porte" sur les vapeurs nitreuses
et I1 anhydride sulfureux. Les re"sultats indique's dans le tableau IV repre"-
sentent les quant it e"s de poussieres dosses dans les pournons. Us montrent
que l'inhalation de gaz irritants tend a aotiver I1 Duration pulmonaire.
On peut remarquer que I1 act ion de N0? est nettement plus marquee que celle
de S0,v malgr^ une concentration moindre.
          Une oonstatation similaire a e'te' faite aveo la fumee de tabac.
Bien que les re"sultats solent encore peu nombrenx dans ce domaine, une ac-
cele"ration de I'dpuratlon alve'olaire paratt e"galement se manifester ohez
I'homme sous I'action de la fume"e de tabac, probablement liee a la pr^sen-
ce de gaz irritants tels que I'acroltSine. Dans le tableau IIj les pe'rlodes
courtes correspondent en effet a trois sujets fumeuri- (L..., B..., S...),
tandis que les perlodes longues correspondent a deux sujets n'ayant Jamais
fXmxS (M...) ou ayant oesse de fumer depuis plus d'un an (H...).
5. Conclusion
          L'Duration pulmcnaire des polluants partlculalres est un ph^no-
mene sujet a des variations individuelles importantes, tout an molns au
niveau alve'olaire. Elle differe selon la nature du polluant Inhale" et est
affecte'e par la quantltd de poussieres d^ja deposde dans les poumons. Au-
dessus d'une certalne charge ooniotique, la restauration de la fonotion
epuratrice est lente et paratt m§me dans certains cas d^flnltlvement com-
promise. Les vapeurs iri-itantes tendent a activer l^gerement I1 Duration
alve'olaire selon un me'canlsme encore inexplique".

                               Bibliographie

1 - LE BOUFFANT L.
    Influence de la nature des poussi&res et de la charge pulmonaire sur
    I'dpuration.
    Inh. Part. Londres 1970, III, 227-257

2 - LE BCUFFANT L. - HENIN J.P. - MARTIN J.C. - DANIEL H.
    Etude eape'rimentale de 1'epuration pulmonaire. Action de I'empoussie'-
    rage sur la capacite* depuration. Application a lthonrae.
    Lille Meclioal 1972, Y£, 1091-1101

-------
                              1661
           EXEMPLE DE RECHERCHE DANS UNE ATMOSPHERE
     D'UNE SUBSTANCE CHIMIQUE CAUSE D'EFFETS PATHOLOGIQUES

     A, CICOLELLA, D, GRADISKI/ J, L, MAGADUR, B, MONCELON

Institut National de Recherche et de Se'curite' pour la prevention
des accidents du travail et des maladies professionnelles, Paris
France
RESUME

     le depistage des problemes toxicologiques qui peuvent ae
presenter en entreprise necessite un diapoaitif stir et eprouve.
Nous decrivons une methodologie d'action & propos d'une inter-
vent-ion type de 1'I.N.R.S, dans 1'-induetrie de I'emballage plaa-
tique ou eat apparue une pathologie profeseionnelle.

     Ce di.sp08-i.tif est tree progressif:  chaque niveau d'aotion
n'eet engagS que dans la meeure od le precedent n'a pu apporter
de solution an probleme souleve,   II compvend:
     - des equipes d'observateurs locaux specialises et senaibi"
       lisea aux problemes d'hygiene industrielle,
     ~ une infrastructure destinee d. recueillir et tranamettre
       I'information juaqu'aux laboratoires d'analyeea et de
       reaherches ou elle pourra $tre traitee,
     - des equipes pluridisciplinairee, ohargeee d'etudier I'im-
       portance et lea implications du probleme, ainei que de
       generalisert a'il y a Iieu3 I'obeervation fortuite et
       locale qui a ete faitef
     - des laboratoires de ahimie specialises dans le prelevement
       et I'analyee des substances industriellea prSaentes dans
       les atmospheres de travail,
     - des laboratoires de toxicologie experimentale, capables:

-------
                               1662
     — de conduire dea etudes raisonnees aur  I'animal, etudes
        destinies & fournir au preventeur dea may ens de depis-
        tage specifiques et precoces,
     — de pratiquer, aur des eahantillons humains, les analyses
        biologiquee correspondantes;
     - un groupe de travail au aein duquel medecins, toxiaologtiee
       et statisticiens peuvent interpreter lee resultats acquis
       au ooura dee etapes precedentes.

     Lorsque I'affaire eat inatruite, que dea cauaes ont pu 6tve
miaes  en evidence^ le doaaier eat remis aux autoritea oompetentea
qui pourront prendre lea meaures qui a'imposent.
ABSTRACT

     To get to the root of the toxioologioal probleme that may
arise in industry requires a reliable and tested machinery.   We
describe the procedure for an INKS type investigation in the
plastics packaging industry following the appearance of an occu-
pational disease.

     The machinery is progressive;  each stage only comes into
action if the previous stage has failed to find a solution to
the problem.   It comprises:
     - local teams of specialist observers versed in the prob-
       lems of industrial hygiene,
     - an infrastructure to collect and transmit data to analy-
       tical and research laboratories where they can be processed
     - multidisciplinary teams called upon to study the extent
       and implications of the problem,  and, if necessary; make
       generally known the random and local observations that
       have been made,
     - chemical laboratories specialising in the sampling and
       analysis of industrial substances present in working at-
       mospheres.

-------
                                1663
     - experimental toxicological laboratories capable:
     — of carrying out tests on animals -in order to provide the
        preventer with specific methods of early detection,
     — and of carrying out appropriate biological analysis on
        human samples;
     - a working party within which physicians, toxicologists,
       and statisticians can interpret the results obtained
       during the previous stages.

     When the facts have been ascertained and the causes demon-
strated, the report is then sent to the competent authorities
so that they can take the necessary steps.

-------
                                   1664
I .  Int.rQiducticm
   L'etude proposee ici presents une approche methodologique du dispositif
a mettre en place lorsque le problems de rapporter a une cause chimique
des troubles observes dans I1Industrie  se trouve pose.
Ce dispositif fait intervenir differents organismes au service de la pre-
vention :  Medecine du travail,  Caisses Nationale et Regionales d'Assurance
Maladie, Institut National de Recherche et de Securite,  entr'autres.
II ne nous appartient pas de fixer le role des organismes avec lesquels
nous collaborons. Nous nous contenterons de nous situer  par rapport a eux
dans cette approche methodologique, et de decrire, a propos d'un exemple
precis, notre action, telle que nous la concevons.
2.  Origins de 1'etude
   Le medecin du travail d'une importante entreprise constate 1'apparition,
chez les ouvriers, de troubles t>n rapport avec la fabrication des bouteil-
les en matiere plastique. Les troubles en cause consistent principalement
en affections sanguines. La medecine du travail, les Caisses Nationale  et
Regionales sont alr.rtees. La Caisse Nationale charge 1'I.N.R.S d'entrepren-
dre une enquete, dans le but de preciser la cause des incidents  observes.
3. Recherche d'un constituant de la matiere premiere de ces bouteilles pou-
   vant avoir une action nocive
   Un certain nombre de substances chimiques etant connues pour pouvoir
itre a 1'origine des manifestations observees, il importait de verifier
leur presence et leur tcneur eventuelle dans I'atraosphere des lieux de tra-
vail ainsi que celles de leurs metabolites dans les milieux biologiques.
Par ailleurs, une enquete siroilaire devait etre lancee aupres des entre-
prises voisines par leur activite.
Cette premiere intervention apporte les informations suivantes :
  Aucun des produits reconnus capables d'entrainer des affections hemato-
  logiques de la nature de celles qui ont ete decrites n'est present a une
  concentration suffisante dans les atmospheres des lieux de travail.  Cette
  constatation est corroboree par le fait que la degradation thermique des
  matieres premieres ne conduit pas, aux temperatures de mise en oeuvre,

-------
                                   1665
  a la  liberation des produits suspects,  produits  dont  les  metabolites  ne
  sont  d'ailleurs  pas retrouves dans les milieux biologiques  ;
  II y  a similitude des faits observes  dans plusieurs entreprises mettant
  en oeuvrc du chlorure de polyvinyle pour le conditionnement de denrees
  alimentaires.
Lo problt-me pose cxiste done reellement, mais sa cause ne peut etre rappor-
tee a aucune des substances connues pour produire ces effets. Une deuxieme
intervention est decidee.
4.  Mise en evidence de la substance probablement responsable
   Un inventaire soigneux des corps chimiques presents est dresse. L'un des
corps, 1'a-phenylindole,  parait suspect. II est en effet present dans 1'at-
mosphere, a une concentration non negligeable, et sa toxicite est mal con-
nue. Son etude est alors  confiee au laboratoire de toxicologie industrielle
experimentale de 1'I.N.R.S.
5. Experimentation animale
   Trois types d'experimentations sont envisages :
- 1'a-phenylindole est administre par inhalation chez le lapin, et par
  voies intraperitoneale et orale chez le rat et la souris. Des examens
  pharmacodynamiques, hematologiques, anatomopathologiques et biochimiques
  sont realises chez ces animaux intoxiques en aigue ;
- 1'effet de 1'a-phenylindole in vitro, sur une culture de cellules Hela,
  a etc recherche ;
- une etude metabolique preliminaire est effectuee sur le rat.
Ces premiers elements de recherche ont mis en evidence, chez 1'animal,
deux substances presentes i un taux anormal dans les urines.
6. Retour a 1'observation sur 1'homme, accompagnee de traitementa atatisti-
   ques.
   L'fixperimentation animale ayant etabli la correlation entre 1'adminis-

-------
                                   1666
tration d'a-phenylindole et la sortie urinaire de metabolites anormaux,
ceux-ci ont etc recherches systematiquement sur un echantillon d1environ
mi lie travailleurs, appartenant a differentes entreprises, exposes ou non,
avant et apres poste. Statistiquement, ces analyses ont etabli, chez I'hom-
me, la presence des memes metabolites, egalement a des taux anormaux. Des
differences significatives ont pu etre raises en evidence entre les diverses
categories de travailleurs soumis a investigation.
7. Diffusion de 1'information
   Le dossier coraplet de cette recherche preliminaire a etc largement dif-
fuse, d'une part vers les milieux responsables, d'autre part vers les uti-
lisateurs, afin que les mesures visant a soustraire les ouvriers a 1'inha-
lation d'a-phenylindole soient prises sans delai.
8. Conclusion
   La raise en oeuvre de ce dispositif, applique a 1*evaluation des effets
constates sur la sante a la suite d'une exposition a un polluant present
dans 1'environnement des lieux de travail, presente 1'interet d'etre pro-
gressive dans 1'utilisation des moyens et de pouvoir etre stoppee a la fin
de chaque etape, si les resultats sont juges satisfaisants.
La recherche presentee ici met en evidence quatre points importants :
- la necessaire collaboration entre medecins et laboratoires de recherche,
- le role d'equipes special!sees dans le prelevement et 1'analyse des
  atmospheres,
- 1'utilite du  recours a lfexperimentation animale et 1'exploration bio-
  chimique raisonnee,
- la necessite  d'un traitement statistique des donnees acquises.

-------
       UMWELTMESSUNGEN





 ENVIRONMENTAL MEASUREMENTS





MESURES DANS L'ENVIRONNEMENT





      MI SURE AMBIENTALI





   METINGEN IN HET MILIEU

-------
                              1669
          ASPECTS OF POLLUTION IN THREE URBAN-CENTRE
                  SITUATIONS IN GREAT BRITAIN

        R, G, DERWENT, H, N, M, STEWART, J, S, S, REAY

Air Pollution Division, Warren Spring Laboratory, Department of
Industry, Stevenage, Hertfordshire, United Kingdom
ABSTRACT

     This paper provides an introduction to the automatic pollu-
tion monitoring activities carried out by Warren Spring Labora-
tory.   A number of pollutants are studied using some of the
latest instrumental methods.   Methods of calibration and some
experimental findings are discussed relating to general urban
and traffic pollution monitoring.

-------
                                   1670
1 .   Introduction
     Monitoring of roadside pollution in European cities has extended
greatly since the 1960s.  No longer is concern concentrated only on smoke,
carcinogenics, carbon monoxide or even lead.  With the seemingly
inevitable rise in vehicle population it has become more important to
assess the urban situation with respect to a range of primary and
secondary pollutants, to examine not only the diurnal pattern of concen-
trations but also the atmospheric reactions, particularly those induced by
sunlight.  This has been made possible by the appearance of new instru-
ments which, though sophisticated, can be successfully operated in a
continuous mode to provide a detailed account of what is happening in the
air.
     In the UK we have chosen to make measurements for a period of up to
five years in the streets of five cities.  Each city has a site in a busy
and/or poorly ventilated street.  The results and trends should show the
intensity of roadside pollution and any effects of traffic flow or
legislation on vehicle emissions or fuel composition.  Already analysis of
the results shows the extent to which London experiences photo-chemical
phenomenon.  These urban measurements are linked to others in widely
scattered rural sites since the time scale of these atmospheric reactions
is such that they are not confined to the urban areas where the primary
pollutants largely originate.

2.   ExBieriment al_ _Methoda
2.1  Table 1 summarises the methods routinely used in our survey together
with their calibration techniques.  Not all these methods are operated
simultaneously at all sites in the survey but a minimum of total
particulates, particulate lead, carbon monoxide and total hydrocarbons are
measured at all sites.  Further parameters are measured, and these include
wind speed, wind direction, total solar radiation, visual range and
traffic count where appropriate.
2.2  Each sensor or instrument is allocated to a particular channel of a
10-channel scanning digital voltmeter with punched paper tape output.
This paper tape is then processed centrally by computer.  The data
acquisition system scans each sensor at five minute intervals with
computer integration of the observations to yield hourly mean values.   The
data are stored on magnetic tape files and are compared on a regular basis
with published national and international air quality standards and

-------
                                    1671
   TABLE_J_. __A oummary of the Instrumental Methods and their Calibration
Pollutant
Instrumental method
Absolute calibration
method
Carbon monoxide
Non-dispersive infra-       Gravimetric standard
red chromatography/flame    gas mixture
ionisation
Total hydrocarbons
Flame ionisation
Gravimetric standard
gas mixture
Oxides of nitrogen
Ozone
Chemiluminescence
Chemiluminescence
NO  permeation tube
Neutral potassium
iodide method
Sulphur dioxide
Coulomet ry, c on duct i-
metry, flame photometry
West-Gaeke method
Particulate lead
Filter paper collection     Atomic absorption
+ XRF analysis              spectroscopy
Total particulates
Smoke stain reflectivity
Analysed hydrocarbons     Chromatography/flame
                          ionisation
                            Gravimetric standard
                            gas mixture

-------
                                   1672
targets.
2.3  The instruments are inspected for malfunction twice weekly and are
recalibrated on a weekly basis.  The routine calibration procedure invol-
ves simpler experimental techniques which are more suited to field
operation.  The more rigorous absolute calibration procedures listed in
Table 1 are reserved for establishing methods and validation of the
routine calibration procedures.

3.   Some Aspects of the Results for Central London
3-1  A background site was established in Central London above a courtyard
so that it would sample the general level of urban pollution in the city
and not be specifically influenced by either traffic or local fixed source
emissions.
3.2  The computer results are used to calculate the daily, monthly and
yearly air quality parameters.  Seasonal effects are clearly evident in
the air quality values with total particulates, ozone, nitric oxide and
carbon monoxide showing relatively simple seasonal dependences.
3.3  Cumulative frequency distributions are particularly useful for
summarising air quality data.  The range of the data and the time above
threshold values can be readily assessed from these distributions.  Most
of the pollutants being studied showed log-normal cumulative frequency
distributions for the daily air quality parameters, daily mean and maximum
hourly mean for the day.  Notably, nitric oxide, sulphur dioxide, smoke
and methane show exceptionally good log-normal distributions.  Carbon
monoxide, total hydrocarbons, and nitrogen dioxide show deviations from
log-normality at low concentrations.   Such deviations are accounted for by
instrument sensitivity, natural background methane and oxides of nitrogen
respectively.  Ozone shows large deviations from log-normality over wide
concentration ranges.
3.1*  The pollutants which are typically associated with motor vehicles and
are emitted at street level with poor dispersion behave as area sources in
the general urban environment.  The pollutants carbon monoxide, hydro-
carbons appear with relatively low geometric standard deviations.  Sulphur
dioxide, which is associated with fixed source combustion, has a much
greater geometric standard deviation than for the area source pollutants.
Methane which is predominantly a natural background species with some
pollution contribution  from mans activities, shows the smallest geometric
standard deviation of all the air pollutants studied.

-------
                                   1673
 3.5  The logarithm of maximum concentration showed a linear relationship
 with the logarithm of the averaging period for a number of pollutants.
 This relationship was found to be useful for extrapolating to averaging
 times not easily or directly assessible from the data itself.
 3.6  Elevated ozone concentrations have been measured in Central London
 during warm, sunny spells during the months of April to September inclu-
 sive.  The highest hourly mean ozone concentration reached during 1973
 was 0.136 ppm with the levels elevated over 0,1 ppm usually for several
 hours on each occasion.  The origin of these elevated ozone concentra-
 tions is clearly photochemical and is strongly related to meteorological
 conditions.  During winter months, the ozone concentrations are much
 lower and show an opposite dependence on meteorological conditions.   Only
 on days with strong winds do ozone concentrations significantly exceed
 0.01 ppm.

 U.   Some Aspects of the Results for Kerbside Sites
 lt.1  Monitoring sites have been established in heavily-trafficked streets,
 one in each of London, Cardiff, Glasgow and Birmingham.  The air is
 sampled at a height of two metres, half a metre in from the kerb and is
pumped to the instruments for analysis in situ.   Simultaneous measurements
of wind and counts of vehicles passing the sampling point are also made.
k.2  The pollutants carbon monoxide, total hydrocarbons, particulate lead,
total particulates and nitric oxide follow the traffic flow very closely
at all times of the day.  During the early hours of the morning the
traffic flows and pollutant concentrations fall to low values.  The
correlations between hourly traffic flow and pollutant concentrations for
each day are usually highly statistically significant, (p>0.9, N » 2U).
These correlations are affected by the wind.   This effect is not generally
noticeable on an hourly basis due to'noise'  in the pollution emission
rate, but is noticeable on a daily basis.   With increasing wind speed,
the ratio of carbon monoxide to traffic flow decreases significantly on
a day-to-day basis.
U.3  The amount of pollution per vehicle flow appears to be heavily
dependent on street geometry when comparing different sites.  Such
variables as 1. the height of buildings on both sides of the street,
             2. traffic flow; unidirectional  or  bidirectional,
             3. on which side of the street  measurements are made,
         and U. the proximity of intersections,  traffic lights,

-------
                                    1674
all appear to influence the pollution vs traffic flow correlations.
k.k  The crosscorrelations between the pollutants listed in paragraph b.2
are also well-characterized on an hourly basis.  Sulphur dioxide, provides
a contrast for example, and does not correlate well with these other
pollutants.  Since motor vehicles provide only a small contribution to
total emissions of sulphur dioxide, these observations are relatively easy
to explain.
k, 5  The ratios between the other pollutants have been obtained for canyon
streets near to city centres.  Generally our ratios agree reasonably well
with other published surveys.  Further sites are planned, for example to
provide information on the influence of the street gradient on these
inter-pollutant correlations.

-------
                               1675
           ETUDE DE LA CONTAMINATION DE FOURRAGES ET
           LEGUMES PAR LE PLOMB ET LES METAUX LOUR5S
   DEL:ARTE+, R.IMPENS"*", R,KIRCH^ANN++/ p.NANGNior"1", E,FAGNIART++
+  Facultg des Sciences Agronomiques de 1'Etat, Gerobloux,
   Belgique
++ Centre d' Etudes de 1'Energie Nuclfiaire, Mol, Belgique
RESUME

     Lea auteura dSarivent la methode qu'ila appliquent dans
I'etude de la contamination de 1 'environnement et de  la chatne
alimentaire par lea metaux lourdat et tout particulierement par
le plombt le cadmium et le zinc,

     line premiere phase aonaiate a pvelevev en eitee  induetriela
et urbains, ainai qu'a proximite dee autorouteSj des  eohantillona
de Bol ainai que oertaina vegetaux tree repandus dont Plantago
ma^'or I. (plantain majeur), Liguatrum ap. (troene), Trifolium
repena L_. (trifle rampant) et Tueailago farfara £.  (tueailage).
Lee concentrations en Elements metalliquee de oee echantillone
eont determines par oeoillopolarographie A tension  imposee.

     On delimite ainai des zones a risques aoorus oil  des prele-
vementa eyatematiquea de fourrageat legumea et fruits sont faits
tout au long de la periode de vegetation.   L'evolution et I'im-
portanoe des retombeea en Znt Cd, Pb, Hi, Co et Cu  eont aui-oiea
dana oea potagera et vergera de  la vallee moaanet a proximitS
d'usinee traitant dee mineraia,   Leo reaultate eont  compare a d
oeux obtenua dana la banlieue de Bruxelleet prea d'Sonangeurs
autoroutiera et en zone t&moin,  eloignee de toute source de pol-
lution,   Cea reaultata font apparattre des riequee importants

-------
                               1676
 d1 intoxication  ahronique  des  hommes  et  animaux consommant  les
 vegetaux  contaminea.    Parall&lementt des  essais  portant sur
 l'absorptiont  la  retention  et la  toxiaite*  de  ces  elements  metal-
 Iique8f chez  lee  plantes, sont  m&n&B en laboratoire.    Les r&sul-
 tats aoquis,  d.  I'aide  de  Pb stable et de 2JZ0p,  chez  Zea mays L_.
 (mate) sont
     Enfint des essais d'affouvagement  de vaches  laitiSres  d
 I'aide de fourrages artificiellement  enriohis  en  Pb  stable  et
 9 7 0
   Pb sont en ooure.   Us ont pour but de  calculev  le  taux de
 tranafert du Pb alimentaire dans  le laitt les  urines  et  les
         f Scales,
     La discussion des r&sultats fait apparattre  les  risquee  de
contamination de la chatne alimentaire.
ABSTRACT

     Description of the methods used by  the authors for  their
study of contamination of the environment and of the food  chain
by heavy metalsf with particular reference to lead, cadmium  and
zinc.

     The first phase consists of taking  sample8t in industrial
and urban areas and in the vicinity of motorways of soil and
various common plants including Plantago major £.  (greater plan-
tain) j Liguetrum ep. (privet), Trifolium repene L. (white  clover)
and Tuseilago farfara L. (coltsfoot).    Concentrations of  metal
substances in these samples are measured by means  of oscillo-
polarography at a set voltage.

     Risk areas are thus demarcated where fodder,  vegetable  and
fruit samples are taken regularly throughout the vegetation  per-
iod,    A watch is kept on the progress  and extent  of Zn, Cd, Pb*
Ni, Co and Cu fall-out in kitchen gardens and orchards in  the

-------
                               1677
vicinity of ore dressing works in the Meuse valley.   The results
are compared with those obtained in the suburbs of Brussels near
motorway intersections and in a control area* away from any source
of pollution.   These results reveal that there is a serious risk
of chronic poisoning in human beings and animals which feed on
contaminated plants.   Parallel laboratory tests are being car-
ried out on the absorptiont retention and toxicity of these metal
substances in plants.   The results obtained using stable lead
and lead     in Zea mays L_. (maize) are discussed.

     Finally, dairy cattle are being experimentally fed on fodder
which has been artificially enriched with stable lead and lead
21°.   The aim is to calculate the rate of alimentary lead trans-
fer to the milkt urine and faeces.

     Discussion of the results shows the risks of contamination
of the food chain.

-------
                               1678
1. Introduction
    La contamination du milieu et de la chaine alimentaire par
les metaux lourds et tout particulierement par le Pb, le Zn et
le Cd, est une agression centre notre environnement relative-
ment recente, dont les repercussions biologiques ne sont enco-
re qu'imparfaitement connues. Nous d^crivons, ci-dessous, la
mSthode que nous utilisons pour suivre la dissemination des
metaux lourds  (Pb, Zn, Cd, Cu, Co, Ni, Hg, Mn, etc.) dans 1'en-
vironnement  (air, eaux de surface, sol, vSge'taux) et leurs
voies de transfert dans la chaine alimentaire  (boissons, ali-
ments) . D'autres publications decrivent plus en detail les
principaux resultats acquis  (IMPENS et al.  (1),(2),(3),
DELCARTE et al.  (4)).

2. Techniques

    Des prel^vements reguliers d'echantillons sont faits dans
tout le pays. Apres dessication a l'£tuve, les echantillons
sont mineralises par voie humide, puis extraits a 1'aide d'une
solution chloroformique de dithizone et analyses par oscillo-
polarographie & tension imposee  (NANGNIOT,  (5)). Les resultats
exprimSs en ppm/M.S.   (mg/kg de matiere sSche), ainsi que tou-
tes les caract^ristiques et coordonn^es de 1 "e'chantillon, sont
mis sous forme cod§e en me"moire d'ordinateur pour e"tre traitSs
statistiquement.  Les echantillons t^moins, non suspects de pol-
lution, proviennent d'une clairiere en foret (Domaine des
Epioux, Lacuisine). Enfin, les essais faisant appel aux radio-
nuclides ont lieu au C.E.N./S.C.K. de Mol.

3. Dispersion du plomb dans 1'environnement

    Nous suivons la dispersion des particules metalliques a
1'aide de plantes-temoins, Celles-ci sont prfisentes dans tou-
tes les stations. De culture aisge, elles se maintiennent en
Stat de v§g§tation pendant la majeure partie de I'annfie. Leurs
caractSristiques botaniques  (port des feuilles, rugositS ou
pilosit§ de l'6piderme, presence d'une cuticule) nous ont fait
choisir: Tussilag« farfara L., PIantago major L., Ligustrum sp.
et Trifolium ire|>eng L. .
    Parmi les facteurs qui modifient la dispersion du plomb le
long des autoroutes,  nous avons fait des observations sur ;
- 1'filoignement de la voie de circulation,
- les caractfiristiques du trafic automobile,
- la date du pr61evement d'6chantillons,
- les precipitations et le climat,
- le profil de la route et 1'orographie,
- la presence d'ficrans.
    A I1aide de ces plantes-temoins nous avons pu delimiter
des zones 3. risques accrus, ou des prelevements systematiques
de fourrages, de fruits et de legumes sont faits.

4. Presence du Plomb dans les fourrages, plantes industrielles
   et legumes

    Des le 15 aoflt, la plupart des echantillons d'herbes pr6-
leves dans une zone de 30m de part et d'autre d'une autoroute
moyennement frequentee (25.OOO vehicules/16h) ont des concen-

-------
                              1679
trations en plomb qui dSpassent les 1OO ppm/M.S.. La consomma-
tion prolonged de ces herbages par de jeunes animaux pourrait
etre une cause d1intoxication saturnine du bfitail. A titre
prEventif, nous avons pre'conise' 1'Elimination des produits de
fauchage des engazonnements autoroutiers les plus contaminfis.
    Des observations similaires ont &t£ faites egalement sur
d'autres plantes fourrageres: betterave, luzerne, navet, ainsi
que sur les c§r6ales, la chicorfie, la pomme de terre et les
betteraves sucriSres. Fort heureusement, pour ces plantes, les
parties entrant dans I1alimentation humaine (grains, tubercu-
les, ou racines) ont des teneurs en plomb trds faibles, tou-
jours tr£s infSrieures a celles des feuilles et tiges. Ce qui
confirmerait que la contamination plombique est essentielle-
ment due S. un d6pot superficiel et que 1'absorption de ce m£-
tal par les racines, meme dans un sol trfis containing, est tr£s
r^duite.
    Les r6sultats analytiques obtenus chez des legumes et des
fruits sont moins rSconfortants (tableau I). Les legumes folia-
c6s (6pinard, salade) ou 3 epiderme velu (haricot) sont les
plus contamine's.
Les risques d'ingestion de quantity's de plomb appr§ciables,par
ces legumes, est d'autant plus grand, qu'un lavage Snergique
(H 1'acide chlorydrique 1O%) auquel certains le'gumes ont 6t6
soumis, s'est r6v616 peu efficace pour ^liminer les de"p6ts de
plomb.
L1analyse simultan6e d'autres mfitaux sur ces echantilions n'a
pas montr^ de correlation entre de fortes concentrations en
plomb et des teneurs 61ev6es en d'autres polluants mfitalliques.
Tableau I - Contamination de 16gumes  (potager entre 3O et 35m
            de distance du Ring Nord de Bruxelles - 1973).
  EspSce     23 juil.   2O aoQt   6 sept.   4 oct.   7 nov.
Chou blanc
Chou rouge
Chou fris6
Poireau
Epinard
01
03
—
03
"
20
52
11
34
19
9
25
21
12

14
13
41
22

13
8
11
14

Rfisultats exprimes en ppra -P/b./M.S.

-------
                               1680
 5. Passage du plomb dans la chaine alimentaire

     L'^tude en serre de la contamination foliaire de Zea mays
 L.,  ag6 de un mois, a montr§ que la presque totalite du plomb
 d€pos£ sous forme de gouttelettes de Pb(NO3)2  (marqu£ par
 21OPb), reste localise dans la zone de d6pot.  Moins  de 2% sont
 transportes vers les autres feuilles et organes  de la plante.
 Par  ailleurs, les essais d'affouragement de deux vaches laitie-
 res,  3 1'aide de plantes de mais contenant  2,4 millicurie de
 2l°Pb(N03)2f  sont termines.  Les mesures en  cours permettent de
 calculer le taux du 21Opb dans le lait ainsi que le  taux de
 disponibilit£ par I1animal du plomb associS S  une ration de
 mais  fourrager consomrng en vert.  (Tableaux II et  III)

 6- Presence de m§taux lourds chez des  Iggumes  cultiv§s en zone
   industrielle (Engis)' ~~~

    Nous disposons  de parcelles,  a distances croissantes d'usi-
 nes traitant  des minerals de zinc.  Des pr§levements  reguliers
 de vegetaux (legumes  et plantes-t&noins)  et d'echantillons de
 sol y  sont  faits.
    La  grande he'tgroggn6it£  des  rgsultats d'analyse  des  sols
 est due non seulement a 1'eloignement  ou  B.  1'orientation par
 rapport aux sources de  pollution  mais  aussi  au bouleversement
 des terrains  tant par 1'exploitation industrielle  que  par les
 travaux de  gfinie  civil.
    A proximit6  des usines,  sur  les c6teaux exposes,  la  flore
 s'est cons iderab lenient  appauvrie,  seules quelques  especes re-
sistantes survivent  (dont Agrostis sp.).
    Les v6g£taux cultives dans les potagers  (ll plus de 60O m de
1'usine, dans la direction des vents dominants) contiennent
d'importantes concentrations en Cu, Cd et Zn. A t'itre d'exem-
ple nous donnons les r6sultats analytiques pour certains
d'entr'eux  (tableau W).  Nous comparons les teneurs en ces m6-
taux de diff^rents organes des plantes (feuilles, tiges. raci-
nes) en precisant si les Schantillons ont et6 laves (3 HC1 10%)
ou non. Les teneurs en elements m6talliques du sol en contact
avec les racines sont ggalement donnees.  Des prfilevements 3
date difffirente permettent de suivre 1'enrichissement progres-
sif en metaux lourds des organes foliaces soumis aux retombees
industrielles. La comparaison de laitues cultivees sous couches
vitr^es et en plein air permet d'apprecier 1'importance quan-
titative de ces retombSes actuelles.
    Dans certains cas, nous faisons subir S ces legumes conta-
minSs un traitement correspondant aux preparations culinaires
normales (epluchage, lavage, cuisson). Les diff£rent«s frac-
tions sont analys£es separ^ment. Si une partie des Elements
metalliques se retrouve dans 1'eau de cuisson, les concentra-
tions en Cu, Cd, et Zn dans les tissus consomm§s sont encore
tres elevees: ex. des racines de carotte apres gpluchage, la-
vage et cuisson contiennent 3O ppm de Cu, 3 ppm de Cd et 69ppm
de Zn  (ppm/M.S.).

-------
                                     1681
  TABLEAU II


  Excrdtion (urine,feces) et se'cre'tion  (lalt) du
  des tiges de mats fourrager  (vache  36)
                     210,
                        Pb inge're' avec
Temps e'coule' depuis
1' ingest ion
1
2
3
4
5
6
7

9

14
21
25
7o dans*
lea feces
10,95
44,16
28,21
12,17
3,89
1,29
0,25

0,02


feO,004

7. dans*
1 'urine
1.5.10"3
7.3.10"3
8,8.10"3
5.5.10"3
4.4.10"3
7.5.10"3
5.2.10"3

-------
                              1682
Tableau 3V -  Determination du Cuf Cd, et Zn sur des legumes
                           prSlevSs a Engis.

Raphanus
sativus

Lactuca
sat i va



Allium porrum


18-V-73
31-X-73
18-V-73
***
18-V-73
31-X-73
31-X-73
15-1-74
Sol
racine*
tige *
feuilles *
feuilles *
feuilles **
sol
racine *
feuilles *
sol
racine *
feuilles *
feuilles *
feuilles **
feuilles **
feuilles *
feuilles **
Cu
103
9
1O
40,1
166
257
102
70
34
80
37
46
285
295
152
54,2
7,4
Cd
24
9
5
8,9
16,8
62,3
2O
13
14
24
20
12
1
7,7
28,9
Zn
10.594
532
728
1.962
2.354
3.433
2.153
1.406
421
9.612
687
5.591
961
1.618
1.275
294
1.338
 Legende :  les rdsultats sont exprimes en ppm/M.S.
           *  e"chantilions lavfis a HC1 1O%
           ** echantillons non laves
           *tek plantes cultiv^es a 1'abri des retombees, sous
               chassis de verre.

-------
                              1683
 7. Conclusions

    Nos observations  sur  la presence du plomb  le  long des rou-
 tes et dans  les villes, ainsi que  sur  la  contamination du mi-
 lieu par  les mStaux lourds en zone  industrielle se poursuivent
 selon le  schema esquiss£  ci-dessus.
 Les sites "3. risques  accrus" sont  determines SL 1 'aide des plan-
 tes-t6moins. Celles-ci  servent  aussi d1indicateurs pour  suivre
 1'evolution  de la contamination. Les pr§lavements rSguliers de
 plantes cultivSes permettent 1'estimation de la contamination
 des fourrages et aliments d'origine vege'tale.  L'importance sa-
 nitaire de cette contamination  est appr§ciee en Studiant les
 taux de transfert sol-plante-animal  (6tape en  cours). Si la
 toxicit§  de  ces m£taux  est encore  peu  ou  mal connue, ils n'en
 presentent pas moins  un danger  potentiel. En attendant la fi-
 xation de seuils d*intoxication et de  normes admissibles, no-
 tre etude se propose  de :
 - estimer 1*importance  de la contamination du  milieu,
 - connaitre  les voies de  transfert du  milieu H 1'homme,
 - guider  les recherches 6pide'miologiques.  sur les  risques d'in-
  toxication de I'homme et des  animaux par la  chaine alimen-
  taire.
B ib1logr aph i e
1  IMPENS, R., M'VUNZU, Z. & P. NANGNIOT, "Determination du
   plomb sur les v&gStaux croissant en bordure des autoroutes",

   P roc. _Symp. Environmental Health Aspects of Le ad, Amsterdam
   2-6 oct. 1972f: 135-143.

2  IMPENS, R., M'VUNZU, Z. & P. NANGNIOT, "DStermination du
   plomb sur la vfige'tation le long des autoroutes",
   Analytical Letters, 6(3) 253-264 (1973).

3  IMPENS, R., J. DEROANNE-BAUVIN & J. BILLAU, "La contamina-
   tion plombique de 1'environnement a proximit6 des autorou-
   tes",
   Annales Gembloux, 79, 85-96 (1973).

4  DELCARTE, E., P. NANGNIOT & R. IMPENS, "La determination
   d'616ments m^talliques dans les sols et les v£g&taux en
   sites industriels et urbains",
   Annales Gembloux, 79, 141-149 (1973).

5  NANGNIOT, P., "La polarographie en agronomic et en biologic*

   Les Presses agronomiques de Gembloux  (1970).

-------
                              1685
      EFFECTS OF SECONDARY LEAD SMELTERS ON AIR, SOIL AND
     VEGETATION QUALITY AND ON LOCAL RESIDENTS IN TORONTO

      T, M, ROBERTS+/ T, C, HUTCHINSON4, R, E, JERVIS++,
        A, CHATTOPADHYAY++, J, PACIGA**, J, VAN LOON+++
                                   4/+41
                        AND R, CRUZ

4-   Department of Botany, Institute for Environmental Studies,
    University of Toronto, Canada
4-+  Department of Chemical Engineering, Institute for Environ-
    mental Studies, University of Toronto, Canada
+++ Department of Chemistry, Institute for Environmental Studies,
    University of Toronto, Canada
ABSTRACT

     As part of an interdisciplinary environmental research pro-
gramme, soils and vegetation around two lead recovery smelters
were initially screened for their lead content.   Lead acaumu-
lation in surface soil had reached ^>S% close to one smelter and
exceeded 5,000 ,ug/gram in residential gardens close to both
plants.   Although this level of contamination may have been
caused by past emissions, the lead content of foliage exceeded
1t000 .ug/gram close to one smelter, thus suggesting an ongoing
pollution problem.   The rate of lead accumulation in vegetables,
tree saplings and grass turf transplanted into the contaminated
area substantiated this conclusion.

     While concentrations of suspended lead particles in the
atmosphere near the smelters were significantly higher than at
                                                    2
an urban control site, the monthly means of 2-5,ug/m  were simi-
lar to values recorded in downtown areas of other large North
American cities.   In contrast, the lead in dust-fall was ext-

-------
                               1686
                                                            o
remely high, monthly averages being aa high as 600-1200 mg/m /
SO days in a residential' area close to one smelter.    Particle
size analysis of the lead aerosol showed the difference to be
due to large particles, probably from found-level operations,
falling out close to the smelter.   The lead-bromine ratios also
indicated that localized industrial rather than automobile emis-
sions, were responsible for 75% of the current fallout.

     Analysis of lead in blood and hair of residents living with-
in 150 metres of one smelter were compared with an urban control
group.   The major difference in the data was caused by about
25% of the children sampled having blood lead levels in excess
of 40 .ug/100 ml and hair values higher than 80 .ug/gram.   An
exponential relationship between lead in hair and blood was
established.   It appeared that ingestion of contaminated dirt
and dusts rather than paint "pica" was the major lead source.
Neurological and biochemical abnormalities were recorded in 10%
of the children sampled.

     As dust-handling rather than stack emissions appeared to
be the major source of lead contamination of the neighbourhood
it is hoped that secondary smelters in urban areas enclose such
operations and buffer zones be established around new secondary
smelters,

-------
                                      1687
      During  the  spring of 1973, surface soils and leaves from a variety of
  tree and  grass  species were collected from around metalliferous industries
  located in  Toronto in order to determine whether the biological monitoring
  techniques  used by Goodman and Roberts^provided a useful tool for identify-
  ing  areas of elevated metal fallout in a large urban area.   An  intensive
  survey of the urban background indicated that leaves of• Poa and Aaer>
  species contained less than 100 ppm. lead, and surface soils less than
  500  ppm.  A number of abnormal concentrations were found in industrial
  areas but the most significant were around two secondary lead smelters
  where vegetation and soil samples contained  500 and 8,000  ppm respectively.
                                            100   400...I,.,
   Fig. 1.  Lead  in surface soil (0-2 cm)  around two secondary lead
            smelters.
            Isopleth values are  micrograms  Pb extracted  by  hot
            HNOy'HClO^ acids per gram air-dry soil.
            A ;: Smelter A                 B  =  Smelter B
            M.I.  -.  municipal incinerator    S.I.  = sewage incinerator,
    Both smelters are located to the north of an elevated  expressway, and
   >se to municipal incinerators with residential housing to the north
             smelter A recycles car  batteries  whereas Smelter B also pro-
     5 crude lead oxide.   The pollution problems associated with secondary
         are different from  those of primary  smelters in that there are
fewer guest elements and  sulphur in  the raw material.  Although one would

-------
                                     1688
expect a mosaic of elevated lead concentrations in surface soils in an area
with such a multiplicity of sources, isopleths constructed from analysis of
samples for 150 sites in each area, indicated an exponential decrease with
distance from each smelter and with depth in soil.  Within 50 metres of the
stack of Smelter A lead accumulation in surface soils reached 4-5% and was
greater than 1,000 ppm in grass leaves.  High soil levels may be caused by
past emissions but high metal levels in foliage indicate current contamin-
ation  (Roberts and Goodman (2)).  This conclusion was tested by transplant-
ing saplings of birch (Betula alba) and maple (Aaer saacharinim) into the
contaminated area.  The rate of lead accumulation was rapid at first, foll-
owed by a slower asymptotic phase reaching 1200 ppm for birch and 300 ppm
for maple at the time of leaf-fall.  40-60% of the contamination was super-
ficial and could be removed by washing in a 1% detergent solution.  Lead
retention was greater in the birch leaves, probably because of their greater
pubescence and the large size of the stomata.  A marked decrease in product-
ivity was noted.  The physiological basis of these effects caused by lead
deposits is now being investigated.
    A number of authors (Graovac-Leposavic et^ a_l (3)>   Mclntire and Angle
(4) and Stofen  (5)) have found significant increases  in both daily lead
intake and lead body burdens of people living in high exposure conditions
close to both primary and secondary lead smelters.  As the distance  be-
tween crushing/smelting operations and the nearest residential area was

6



"fc*
T
ut
o»
or
*

o
I1
in
3
(Q
Z
~ 1
a
u

0

' \ / BmilUr *
/ v • 150m. N.W.
/ \ / Sd.±2-4
/ V J
x*\ / f
X' ' ' 1 ^
X \ ' / \
/•"' /\ / -{SJV .io~
/ Sd.*1-2 |
/ r,-' V 1
./"" '' "f

^
• Sm.lHr B
200m. N. _,
Sd,125 ^
|t 500
a

~~ 	 ^ Urban control o
-^^^^ ,^ Sd 10.48 <
0
M j j i A i
MONTHS
A
/ x
/ V
/ \
f \
I \
] \,
~ ws t v
\^H
^\ .' ° f*
^ y 8d!5
\ •*
V x
*




	 -* 	 ,._' ,-^"** """Snwl
0—«0i
•- — *
M J J A 8 0
MONTHS
  Figure 2.   Lead in Suspended Particles and Dustfall Close to Two
             Secondary Lead Smelters and in an Urban Control Area.
             Sd = standard deviation of all values recorded in each region.

-------
                                     1689
only 50-100 metres at Smelter A,  an extensive survey of lead concentrations
in tho air, the people and their  home environment was instigated.   The re-
sults have been compared with an  urban control group living in similar aged
housing.
    Monthly arithmetric means of  lead concentrations in suspended particles
collected over 24-hour periods by high-volume samplers varied between 2 and
5 ug/m3 close to both smelter A and B compared to 0.45-1.1 ug/rn^ in the con-
trol area  (Figure 2).  The daily values for the smelter site form a log-
normal distribution,  as values up to 14 ug/m  were recorded on days when
the wind was from the smelter direction.  The smelter and control daily
values were subjected to a paired-comparison test and found to be signific-
antly different at the 0.1% level.  The mean values around the smelter, al-
though high for Toronto, are not  unusually high when compared to downtown
areas of large U.S. cities.  In marked contrast, lead in dustfall was ex-
tremely high - up to 1200 mg/m /30 days - compared with less than 35 in the
control area.  Lead in dustfall also decreased exponentially with distance.
The rate of total deposition (cm/sec) obtained by dividing the lead and dust-
fall by the concentration of suspended particles, was found to be 1-2 for
the urban area and increased up to 14 in the vicinity of the smelter. This
difference was explained by particle size fractionation of the aerosol
using an Andersen Impactor,which indicated that the percentage of particles
above 3.3 microns increased from 20% in the urban control area to 70% close
to the smelter on episodal days.   The mass median diameter of airborne lead
on episodal days was 4.6±1.3 microns compared to 0.8±0.2 microns for urban
control sites.  These large particles were mainly of smelter origin since
lead in automobile exhausts is mainly submicron in size.  The automobile
contribution to total atmospheric lead burden near the refineries was estim-
ated to be about 20% on episodal  days using the Br/Pb ratio of the urban
area and the bromine concentration by the smelter to calculate the airborne
automotive lead .  Weslowski found similar changes in 13r/Pb ratios around smelters.
    The high concentrations of lead in dustfall were of particular concern
as Nordman et al^ (7)  found mean blood lead levels increased from 12 to 24
ug/100 ml as lead in  dustfall increased from 20 to 200 mg/m2/30 days close
to a secondary  lead smelter.  Analysis of capillary blood using the method
of Cruz and Van Loon (8) from 45 people living in our urban control area
gave a mean of 17 ug/100 ml for adults and children - no individuals above
40 ug/100 ml were recorded.  This baseline data has since been substantiated
in a survey of over 1,000 individuals by the Toronto Board of Health, and

-------
                                     1690
is surprising in that it is considerably lower than has been reported for
downtown areas of large U.S. cities.
    The arithmetric mean for the smelter community was significantly higher
than the control group.  The mean is skewed, however, by occurrence of a
small number of individuals with markedly elevated levels.  It is interest-
ing to note that these individuals are usually male and under the age of 14.
This observation of an upward shift in the population lead norm was borne
out by analysis of hair using the Photon Activation Analysis technique des-
cribed by Chattopadhyay and Jervis  (9).  The urban  control group had a
modal concentration of 17  ppm compared to a range of 20 to 160 ppm in the
smelter group.  This difference was not due to surface contamination as
only 10-20% could be removed by vigorous washing.  Correlation of lead in
blood and hair from 67 individuals indicated an exponential distribution -
values ranging from 10 to 60 ug/100 ml in blood and 3 to 166 ppm in hair -
fitting the equation log Y=0.7001+0.0277 X  (r=0.798».  A similar correlation
has been found by Hammer et al (10) for schoolboys living in smelter and
urban areas.  This is of considerable interest as individuals exposed to
acute doses of lead, where hair and blood lead levels fluctuate widely over
short periods, exhibit no such correlation  (Kopito et. al. (11).These observa-
tions, together with the finding that lead concentrations do not fluctuate
widely in centimetre sections along the hair of children whose blood lead
levels are between  40 and 60 ug/100 ml, suggests that their body burden is
in equilibrium with the environmental burden.  Any further source of intake
                                   TABLE I
         PERCENTAGE DISTRIBUTION OF LEAD IN CAPILLARY BLOOD AND HAIR
      OF PEOPLE LIVING NEAR A SECONDARY LEAD SMELTER AND A CONTROL AREA
                                    Blood Lead Level (ug/100 ml)
               n
 Urban Control 45
 Smelter Area  32

               n
 Urban Control 45
 Smelter Area  51

               x = arithmetric mean
X
17
33
X
17
52
0-9
9
0
0-9
27
0
10-19
64
6.5
10-19
37
22
20-29
22
25
Hair Lead
20-29
18
8
.30-39
5
41
Level
30-39
18
26
40-49,
0
18.5
(ug/gram)
40-49
0
16
50-59
0
9
50-99
0
12
60+
0
0
100+
0
16

-------
                                     1691
  would increase the body burden  from this  sub-clinical  level  to levels  at
  which clinical symptoms of lead poisoning  may appear.
      More extensive blood sampling has since  been completed and 16 children,
  i.e.  2% of those living within 500 metres  of both Smelter  A and B, have
  been given Ca EDTA treatment.   All children  showed lead excretion above
  baseline levels and 2 achieved more than 600 ug Pb within  24  hours.   7  dem-
  onstrated more than 1000 ug ALA in urine per 24 hours.   There were 5  cases
  of hyperactivity and 5 with neurological abnormalities.  3 showed definite
  lead lines in bone epiphyses.   It appears, therefore, that the increased
  lead absorption,associated with ingestion  and inhalation of dirt and  dusts
  in the Toronto situation,does  not appear sufficient to  cause  clear-cut
  symptoms of poisoning but may  cause subclinical metabolic  effects.
      It would seem that if it is not possible to have buffer zones around
  secondary lead smelters in urban areas,  both ground level  and stack
  emissions will have to be better controlled  than is the current practice.
  In addition,  air quality criteria based  on the lead content of dustfall as
  well  as suspended lead particles  are required to prevent environmental  con-
  tamination.   Levels of 100 mg/m2/30 days for lead in dustfall and a daily
  concentration of 5 ug/m or monthly geometric average of 2 ug/m^ for  lead
  in suspended particles has been suggested  for this region.
                                    REFERENCES
1.  GOODMAN, G.T.  and ROBERTS, T.M.  "Plants and soils as indicators of metals
    in the air." Nature, 231, pp. 287-292. (1971).
2.  ROBERTS, T.M.  and GOODMAN, G.T.  "The persistence of heavy metals in soils
    and  natural vegetation following closure of a smelter." Proceedings
    Seventh Annual Conference, Trace Substances in Environmental Health.
    Ed. Hemphill,  D.D. pp. 117-125.  (1973).
3.  GRAOVAC-LEPOSAVIC, L., DJURIC, D., VALJAREVIC, U., SENICAR, H., SENICAR,  L. ,
    MILIC,S. and DELIC, V. "Environmental lead contamination of Meza Valley -
    study on lead exposure of population." Proceeding of International Symposium
    on Environmental Health Aspects of Lead. Amsterdam, pp. 685-705. (1973).
4.  McINTIRE, M.S. and  ANGLE, C.R.  "Air Lead:  Relation to lead in blood of
    black school children deficient in glucose-6-phosphate dehydrogenase.
    Science, 177,  pp. 520-522. (1972).

-------
                                    1692
 5.  STOFEN, D. "Die verinreinigurg der atmospharischen  Luft  mit  blei  und ihr
     einflub auf die gesundheit der bevolkerung. "  Zentralblatt  fur  Arbeitsmedizii
     und Arbeitsschutz. 13, pp. 39-40.  (1963).
 6.  WESOLOWSKI, J.J., JOHN, W. and KAIFER, R.  "Trace  elements  in the  environment
     Advances in Chemistry Series No. 123. American Chemical  Society,  pp.  1-16.
     (1973).
 7.  NORDMAN, C.H. , HERNBERG, S., NIKKANEN, J.  and RYHANEN, A.  "Blood  lead levels
     and erythrocyte D-ALA dehydratase activity in people  living  around  a
     secondary lead smelter." Work-Environment-Health. 10. pp.  19-25.  (1973).
 8.  CRUZ, R. and VAN LOON, J. "A critical study of the  application of graphite-
     furnace non-flame atomic absorption to the analysis of trace base metals in
     complex heavy matrix sample solutions." Analytica Chimica  Acta. In  Press.
 9.  CHATTOPADHYAY, A. and JERVIS, R.E. "Determination of  Nanogram  Quantities of
     lead by Photon Activation Analysis" Radiochem. Radioanal.  Letters,  11,
     p. 331. (1972).
10.  HAMMER, D.I., FINKLEA, J.F., HENDRICKS, R.H., HINNERS, T.A., RIGGAN,  W.B.
     and SHY, C.M. "Trace metals in human hair  as  a simple epidemiologic monitor
     of environmental exposure." Proceedings Fifth Annual Conference,  Trace
     Substances in Environmental Health. Ed. Hemphill, D.D.,  pp.  25-38.  (1971).
11.  KOPITA, L., BYERS, R.K. and SHWACHMAN, H.  "Lead in hair  of children with
     chronic lead poisoning." New England J. Medicine, 276, pp. 949-953.  (1967).

-------
                              1693
       SUI VANTAGGI PRESENTATI DALL'USO BELLA FORNACE A
      GRAFITE COME NUOVA TECNICA DI ASSORBIMENTO ATOMICO
    NELL'ANALISI DELL'INQUINAMENTO DELL'AMBIENTE DI LAVORO

                     D, CASULA, G, DEVOTO

Istituto di Medicina del Lavoro dell'Universita degli Studi di
Cagliari, Italia
RIASSUNTO

     Alcuni elements metallioi (Piombo, Zinco, Cadmio, Antimonio,
Afgento, Rame, Nichel, Cobalto) presenti a livelli di tracoe in
differenti ambienti di Iavoro3 sono stati analizzati mediante
1'uao delta spettvoscopia pep assorbimento atomiao, aon 1e
nuove teeniche ehe utilizzano il fornetto a grafite.

     Vengono presi in esame i vantaggi presentati dalle nuove
metodiche di dosaggio e di prelievo.
ABSTRACT

     Some metallic elements (leadt zinct cadmium, antimony,
silver, copper, nickel, cobalt) of which traces are sometimes
found in various types of working environment, were analysed by
means of atomic absorption spectroscopy and new techniques based
on the use of the graphite furnace.

     The paper covers the working conditions* the results ob-
tained and the advantages offered by the new methods of sampling
and quantitative analysis which were used.

-------
                             1694
   La determinazione degli inquinanti atmosferici negli am-
bienti di lavoro rappresenta un parametro importante nella va
lutazione della salubrita dell'ambiente e deve essere esegui-
ta con la massima precisione possibile data la particolare im
portanza che ormai viene attribuita alia presenza anche di
tracce minime di diversi inquinanti.
   Per queste ragioni nel nostro Istituto si £ sempre voluto,
nel corso degli anni, affinare le varie determinazioni, modi-
ficando e raigliorando le tecniche operative. Riferiremo,  nel
corso di questo nostro intervento, sui vantaggi offerti  dal-
1'impiego della fornace a grafite per la determinazione abi-
tuale di tracce di metalli nell'aria degli ambienti di lavoro.
I dati che riportiamo si riferiscono, principalmente, al do-
saggio ripetuto di metalli, present! nell'aria di alcune mi-
niere metallifere della Sardegna, durante diverse fasi della
estrazione del minerale.
   L'uso di un fornetto a grafite, oltre a permettere di rile
vare tracce minime dei vari inquinanti, ci ha dato la possibi^
lita di ripetere su di uno stesso campione la determinazione
contemporanea di diversi metalli e, data 1'alta sensibilita
della suddetta metodica, 6 stato sufficiente eseguire campio-
namenti ridotti sia nel volume sia nel tempo.

METODICA
Prelievo dei campion!
   Per i prelievi dei campion! d'aria sono state impiegate le
seguenti attrezzature:
1) prelevatori automatic! Dual e Sequential Sampler (Gelman);
   prelevatori automatic! modello D P 12 (Terzano);
2) contatori registratori  (Terzano); contatori a secco (S.I.M.
   Brunt);
3) portafiltri metallic!  (Gelman) corredati di membrane Metri
   eel con pori del diametro di 45 micron  (Gelman) .         ~~
   I prelievi dei campion! d'aria sono stati effettuati in
punti opportunamente  studiati, onde poter disporre di un qua-
dro rispecchiante le  condizioni general! dell'ambiente di la-
voro: gallerie, front! di avanzamento, coltivazioni, sgombero,
fornelli, carreggi, tramogge.
   L'aria §  stata prelevata am.  1,50 dal suolo, alia veloci-
ta di 1-2 litri al minuto e per 30'-60' per ciascun filtro.

posaggio
   Per la determinazione dei metalli abbiamo utilizzato uno
spettrofotometro per  assorbimento atomico mod. 495  (Beckman) ,
corredato di un fornetto a grafite  "Massmann Cuvette"  (Beck-
man) . Per la introduzione delle piccole quantita di campione
sono state adoperate  le micropipette Eppendorf Marburg da 10

-------

a
b
c
d
e
f
Ph
Max
0,064
0,046
0,062
0,022
0,009
0,028
>
Min
0,012
0/008
0,012
0,005
0,003
0,004
Z]
Max
0,090
0,045
0,068
0,031
0,015
0,035
i
Min
0,010
0,007
0,005
0,003
0,002
0,003
C
Max
0,0010
0,0000
0,0008
0,0001
0,0005
0,0002
d
Min
0,000
0,000
0,000
0,000
0,000
0,000
st
Max
0,0007
0,0000
0,0004
0,0011
0,0000
0,0000
Min
0,000
0,000
0,000
0,000
0,000
0,000
Ag
0,000
0,000
0,000
0,000
0,000
0,000
Cu
0,000
0,000
0,000
0,000
0,000
0,000
Ni
0,000
0,000
0,000
0,000
0,000
0,000
Co
0,00
0,00
0,00
0,00
0,00
0,00
                                                                                                 V£>
                                                                                                 Ul
a) Miniera di San Giovanni  (50 determinazioni)
b) Miniera di "Su Zurfuru"  (35 determinazioni} esterno
c) Miniera di "Su Zurfuru"  (50 determinazioni} interne
d) Miniera di "Barega"      (30 determinazioni)
e) Miniera di "Masua"       (20 determinazioni}
f) Miniera di "Mantega"     (20 determinazioni)

-------
                             1696
e 20 microlitri.

Soluzioni standard del metalli
   Sono state adoperate soluzioni acquose del seguenti metal-
li: Pb, Zn, Sb, Cd, Cu, Ag, Ni, Co, della ditta BDH,  tutte al
la concentrazione di 1000 ppm.
   Le membrane utilizzate per i prelievi, su cui si era depo-
sitato il materiale proveniente dall'aria dei cantieri di la-
voro, sono state trattate ripetutamente con aliquote da 5 ml
di HN03 concentrato "Suprapur" della Ditta Merck, onde manda-
re in  soluzione tutto il materiale residue;quindi con acqua
bidistillata abbiamo portato a volume, in palloncini tarati
da 25 ml, le soluzioni ottenute e sulle stesse abbiamo esegui
to i dosaggi dei metalli present!. I metalli sono stati dete£
minati direttamente nelle soluzioni acquose e la loro concen-
trazione 6 stata ricavata da una curva di taratura allestita
preventivamente. Per valutare pid precisamente la concentra-
zione dei metalli, abbiamo operate anche con il metodo delle
aggiunte successive, ed allo stesso modo abbiamo anche con-
trollato le interferenze della presenza contemporanea di di-
versi metalli.
   Ci limitiamo a riportare in tabella dei rilievi eseguiti
nelle varie miniere, i valori  (in mg/m3) massiml e minimi,
nonche, per ciascuna miniera, il numero di prelievi eseguiti.

Conclusioni
   L'uso del fornetto a grafite nella determinazione quantity
tiva dei metalli, present! nell'atmosfera dei vari ambienti
di lavoro, ci ha fornito notevoli vantaggi legati soprattutto
alia alta sensibilita di questa metodica rispetto ad altre.
   Come risulta dalla tabella allegata, con queste tecniche
abbiamo ottenuto sia dei risultati pid precis! nella determi-
nazione di metalli  (come Pb e Zn) present! a livelli piti fa-
cilmente rilevabili, sia nel dosaggio delle tracce minime
 (come Cd e Sb)  non altrimenti evidenziabili con altre metodi-
che. Infatti per la determinazione di queste tracce minime oc
correrebbero dei prelievi massivi e prolungati nel tempo, il
che potrebbe comportare  1'interferenza dei metalli presenti
in concentrazioni  superior!.

-------
                               1697
        LEVELS OF SOME TRACE ELEMENTS IN BABY AND ADULT
                        DIETS IN ITALY

            M, ALLEGRINI+, E, LANZOLA+, C, ROGGI+,
                G, GIANANI++, G, C, SABATINI++
+  Istituto di Igiene, University di Pavia, Italy
++ Ministero della Sanita, Rome, Italy
ABSTRACT

     Cadmium, chromium, copper and zinc have been measured in
nearly 60 samples of various infant foods present on the Italian
market, including powder milk, strained meats, infant cereals,
flours and fruit juices.

     Similar measures have been completed in a pool of food
samples representing an average adult diet for one week.

     The same elements have also been measured in maize, cassava,
rice, peanutef beanst tomatoes, spinach coming from an African
region (Dahomey).

     From the results, the daily intake for infants and adults
has been calculated and compared with data shown in other re-
ports .

-------
                              1698
1.    Introduction
     The problems of food contamination by different elements
is to be approached, keeping in mind that many of these are the
so called "trace elements" whose role in human and animal nutri-
tion is well known since the deficiency of some of them in the
diet may induce particular syndromes of malnutrition.
     At present the nutritional requirements of all the elements
found in the human body, are not yet known.
     Actually the trace elements can be divided into three
groups  (1) the essential ones, the probable essentials, and the
non essentials.
     Ten trace elements considered essential nowadays are: iron,
iodine, copper, zinc, manganese, cobalt, molybdenum, selenium,
chromium, and tin.  Among the trace elements of which there is
no certainty as to  their essentiality, are: nichel,  fluorine,
bromine, arsenic, vanadium, cadmium, barium and strontium.
Furthermore, a certain number of trace elements  (between  20 and
3O)  though  found  in different body tissues, do not meet with
the  criteria established by Cotzias  (2), to be regarded as
essentials.
1.   -  The  criteria are;  Presence in healthy tissues of  all
        living beings;
2.   -  Approximate  constant concentration  in different species;
3.   -  Biochemical, functional  and structural abnormalities
        induced by the deficiency of  an element in  the diet;
4.   -  Reverse of these  abnormalities after the  addition  of  the
        deficient  element to the diet.
     The  non essential  elements are  believed  to  reach  the
human body  trough environmental contamination and,  in  confir-
mation  of  such an hypothesis,  attention  is drawn  to the different
distributing pattern on the concentration  of  essential and  non
essential  trace  elements in the human organs:  for  the  essential
ones normal idstribution has  been  noted, whereas the non  essen-
tial elements  have a log-normal distribution.

-------
Daily intakes  of Cd , Cr. Cu and  Zn
               Figure  1
                               H
                                                                      mg
              10     20
                          50    100    200      500    1000   2
 Cadmium
Chromium
Copper
 Zinc
                                                                  10    20
                                     _L
            '\v\v^v.     .>^^^^->^x^<>x<\SS:^^^\S:..  '  'S^l
                             ///.//////A
                         (  (»)

                           (*)
                                                                             50     tOO   200
= daily intake in infants at the end of first year

=   "      n   «  adults
Results of the

present work



1) Actual intakes



2) Recommended intakes


3) Maximum allowable daily intakes




( values have been calculated from the current litterature  and  in particular

  from oligoelements in human nutrition - QMS Sez. Rapp. Techn.  1973 -  532  )
                                                                          -
                                                                          -
                                                                          -

-------
                             1700
     Cadmium, is among the elements that present this last
behaviour but, for other reasons, it is thought to be essential,
     Finally it should be emphasized that for some elements the
levels which can give rise to accumulation and toxicosis are
not far from the nutritional requirements (3-4-5).

2-    Methods
     Starting from the above consideration four elements have
been measured: three essential, chromium, copper and zinc and
one probable essential, cadmium, in the following items:
a)    infant foods, sold on the Italian market (table I);
b)    samples of infant formula milk produced from 1952 to 1966*
c)    dishes of an average diet for adults, sampled for one week
     and pooled before analysis;
d)    a certain number of foods sampled in an African area
     (Dahomey).
 TABLE I  infant food types analysed.
Food
1 Milk povder
2 Infant flours
3 Strained meats
4 Strained fruits
5 Fruit juices
No producer
7
7
4
5
4
No products
17
14
8
9
7
      The  atomic  absorption  spectrophotometer used  in  this
 research  was  a Jarrel-Ash-Atomsorb.
      50 g.  of  the  sample  was  ashed  at  4OO°C  then  dissolved  in
 concentrated HN03  and left in a muffle furnace  until  the  ashes
 became  completely  white.   The ashes were added  with HC1  (1:1)
 then heated, filtered,  transferred  to  25 ml.  volumetric  flask
 and diluited with  distilled water to the mark.
      All the samples were then measured by the  additioning  me
 method.  The performance  of the analysis was tested by deter-
 mining  the recoveries of  Cd.  - Cr.  - Cu. and zinc.

-------
                              1701
     Therefore known amounts of Cd, Cu, Cr, and zinc were added
to 50 g. of each sample, the samples were then processed by a
procedure already described (6).
     Analyses on fifteen replicated samples indicated recoveries
as follows: 71,O ± 5,7% Cd, 1OO ± 3,2% Cu, 99,1 ± 2,7% Cr,
100,1 ± 3,4% Zn.

     The reproducibility of the analyses has been determined
by analysing each sample three times therefore calculating the
standard deviation.

3.    Results and Discussion
     The results concerning infant foods measured are shown in
tables II, III, IV, V, VI and VII.  Each figure is followed by
twice the standard deviation.
     No evidence has been found of a rise in the levels of Cd,
Cr, Cu, and of zinc in infant milk produced from 1952 onward.
On the basis of the average values in the examined foods the
daily intake of the four elements has been calculated (table VIII
according to the food consumption data in the 1st year of life
(7).
     Table IX shows the daily intakes of Cd, Cr, Cu and Zn in
adults with average adult diet.
     In figure 1 a comparison has been made between the daily
intake concerning infants and adults found in the present
research work and those reported in other surveys; the maximum
non toxic levels, as far as they are known, have also been in-
dicated.  From figure 1 it can be seen that the daily cadmium
intake is very near the maximum allowed level, indicated by the
WHO expert committee in 1973.
     It is also evident that at the end of the 1st year of life
the intake of cadmium and chromium are very near to the adult
level.  The results of the measurements of Cd, cr, Cu and Zn
in some African food samples are shown in table 10.

-------
                           1702
                       TABLE II

Cadmium, Chromium, Copper and Zinc in infant formula millc.
( The values are followed by tvice the standard deviation ).
Samples
n°
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Average
Cd
p.p.in.
0,20 + 0,01
0,23 + 0,02
0,24 + 0,03
0,24 + 0,O2
0,23 + 0,01
O,24 + 0,O2
0,21 + 0,03
0,21 + 0,03
0,22 + 0,02
0,14 + 0,01
0,18 + 0,02
0,23 + 0,02
0,20 + 0,02
0,24 + 0,02
0,23 + 0,04
0,23 j; 0,02
0,21 + 0,05
Cr-
p.p.m.
0,29 + 0,02
0,32 + 0,02
0,27 + 0,03
0,30 + 0,04
0,28 j- 0,02
0,28 + 0,02
0,27 + 0,03
0,28 + 0,05
0,29 + 0,03
0,23 + 0,03
0,23 j- 0,03
0,27 + 0,02
0,31 + 0,04
0,29 + 0,03
0,37 + 0,02
0,37 + 0,04
0,29 + 0,08
Cu,
p. p.m.
0,79 + 0,12
0,77 + 0,06
0,69 ± 0,03
O,47 * 0,06
0,70 + 0,04
0,64 + 0,03
0,53 + 0,04
0,97 + 0,06
0,50 + 0,05
0,26 + 0,02
0,37 + 0,02
0,35 + 0,05
0,53 + 0,03
0,46 + 0,05
0,73 J 0,04
0,67 + 0,10
0,58 j. 0,36
Zn
p. p.m.
18,6 + 4,6
23,3 + 0,5
15,5 + 1,4
23,5 + 0,6
23,6 + 1,2
11,4 + 1,6
27,3 + 3,2
24,4 + 1 ,7
25,6 + 2,1
9,7 + 0,6
19,2 + 2,1
22,2 + 3,8
24,2 + 3,4
20,5 + 1,4
24,2 + 3,4
23,5 + 2,1
21.0 + 10

-------
                           1703
                       TABLE  III

Cadmium, Chromium, Copper and Zinc in infant formula milk produ-
ced from 1952 to 1966. ( The values are followed by twice the
standard deviation )•
Samples

1952
1954
1955
1956
1957
1958
1959
1960
1963
1963
1964
1965
1965
1966
1966
Average
Cd.
p«p.m*
0,20 + 0,01
0,21 + 0,02
0,24 + 0,02
0,21 + 0,01
0,23 + 0,03
0,22 + O,02
0,21 + 0,02
0,22 + 0,02
0,21 + 0,O2
0,21 + 0,02
0,21 + 0,03
0,20 + 0,02
0,21 + 0,03
0,21 ± 0,02
0,21 + 0,02
0,21 £ 0,03
Cr,
p«p*nu
0,34 + 0,03
0,31 + 0,04
0,33 + 0,02
0,32 + 0,02
0,36 * 0,04
0.33 + 0,02
0,36 + 0,04
0,37 * 0,03
0,31 + 0,04
0,33 + 0,02
0,30 + 0,02
0,31 + 0,04
0,30 + 0,07
0,30 + 0,02
0,31 + 0,04
0,32 + 0,04
Cv
p«p.n*
1,0 + 0,14
0,87 i 0,04
0,81 + 0,05
0,73 * 0,2
0,68 + 0,07
0,78 + 0,04
0,80 + 0,02
0,77 * °'°7
0,83 + 0,O8
0,68 + 0,05
0,69 + 0,08
0,67 + °»°3
0,67 + 0,05
0,63 + 0,05
0,64 + 0,10
0,74 t 0,20
Zr.
p*p*m*
21,0 + 0,5
23,2 + 0,7
25,0 ±1,4
19,7 ± 1,0
24,6 +1,6
22,5 ± 1,4
23,2 + 0,7
24,8 + 2,2
22,5 t 1»4
21 J5 + 1,4
22,2 + 0,7
21,6 + 1,4
22,3 + °»5
24,7 ± 0,7
26,2 + 2,0
23,0 + 3,4

-------
                             1704
                      TABLE IV
Cadmium, Chromium, Copper and Zinc in baby foods (strained meats).
( The values are followed by ttrice the standars deviation ).
Samples

1
2
3
4
5
6
7
8
Average
cd-
p.p.m*
0,04 + 0,005
0,04 + 0,005
0,03 -f 0,004
0,05 •«• 0,007
0,07 + 0,008
0,03 + 0,004
0,02 + 0,003
0,05 + 0,006
0,04 + 0,03
Cr-
p.p.m.
0,20 + 0,01
0,23 + 0,01
0,21 + 0,01
0,28 + 0,02
0,22 + 0,02
0,10 + 0,005
0,10 + 0,005
0,10 Hh 0,005
0,13 + 0,14
Cu
p.p*m.
1,2 + 0,07
1,1 j- 0,07
1,3 + 0,08
1,1 + 0,07
0,8 + 0,05
0,4 + 0,02
0,4 * 0,02
0,4 + 0,02
0,8 + 0,7
Zn
p. p.m.
20,0+ 1,3
19,0+ 1,3
35,0+ 2,4
38,0+ 2,5
11 ,0+ 0,7
16,6+ 1,1
16,2+ 1,1
22,8+ 1,5
22 + 18

-------
                            1705
                       TABLE  V - VI

Cadmium, Chromium, Copper and Zinc in baby foods (A - fruit
juices, B - strained fruits ).
( The values are folloved  by tvice the standard deviation ).
Samples

1
2
3
4
5
6
7
Average
Cd.
p. p.m.
< 0,01
<0,01
0,01
0,01
0,01
0,01
0,01
0,01
Cr
p. p.m.
0,04
0,04 + 0,01
0,04 + 0,01
0,04 + 0,01
*. 0,001
•»
0,04 + 0,01
0,07 + 0,01
0,04 + 0,04
Cu
p. p.m.
0,34 + 0,04
0,10 + 0,02
0,30 + 0,03
0,30 + 0,01
0,27 + 0,02
0,29 + 0,03
0,50 + 0,03
0,29 •«• 0,22
w»
Zn.
p. p.m.
0,95 + 0,1
0,76 + 0,1
0,95 -f 0,1
0,57 + 0,2
0,95 + 0,2
0,91 + 0,1
1,33 4- 0,1
Mft
0,92 + 0,44
                              B

1
2
3
4
5
6
7
8
9
Average
0,02
0,02
0,02
0,02
0,02
0,02
0,02
0,01
0,01
0,02
0,05 + 0,01
0,12 + 0,02
0,10 + 0,02
0,12 + 0,03
0,91 + 0,10
< 0,01
0,07 + 0,02
0,04 + 0,01
0,04 + 0,01
0,16 + 0,56
0,6 + 0,2
1,5 + 0,2
0,8? + 0,3
0,91 + 0,2
0,65 + 0,2
0,37 + 0,2
1,05 + 0,3
0,61 -j- 0,3
0,39 + 0,1
0,76 + 0,72
3,3 + 0,5
3,3 + 0,4
3,5 + 0,3
3,3 + 0,4
1,8 + 0,3
1,8 + 0,4
3,0 + 0,5
2,6 + 0,3
1,3 + 0,3
2,7 + 1,5

-------
                            1706
                      TABLE  VII

Cadmium, Chromium, Copper and Zinc in infant flours. (The values
are  folloved by tvice the standard deviation).
Samples
n°
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Average
Cd.
p. p.m.
0,21 + 0,03
0,18 + 0,03
0,12 + 0,02
0,21 •»• 0,03
0,22 + 0,01
0,11 + 0,01
0,18 + 0,05
0,10 + 0,01
0,20 + 0,02
0,18 + 0,03
0,05 + 0,01
0,22 + 0,03
0,18 + 0,02
0,16 + 0,02
0,16 + 0,08
Cr.
p.p.m*
0,22 + 0,0?
0,09 + 0,02
0,16 + 0,02
IP*
0,17 + 0,03
0,42 + 0,07
0,23 + 0,03
0,09 + 0,01
0,05 + 0,01
0,26 f 0,05
0,31 + 0,03
0,03 + 0,01
0,49 + 0,03
1,4 + 0,20
0,13 + 0,04
0,22 + 0,68
Cu.
P.p.D.
4,7 + 0,3
1,7 + 0,1
2,5 + 0,1
1,3 + 0,3
6,8 + 0,5
3,9 + 0,4
5,0 + 0,3
1,2 + 0,1
0,9 + 0,1
2,5 + 0,1
2,5 + 0,3
2,5 + 0,1
1,7 + 0,2
7,4 + 0,4
3,2 + 4,0
Zn
p.p.m*
23,0 + 1,4
11,5 + 2,8
15.3 + 2,2
12,5 * 2,8
25.5 + 3,4
17,3 + 2,4
20,0 * 2,8
11.2 + 0,7
11.7 * 0,7
18.2 + 2,0
12,5 + 1,4
12,5 + 0.28
15,0 + 1,0
15.5 + 1.4
15,4 + 2,2

-------
                           1707
                        TABLE  VIII

Daily intake of Cadmium, Chromium, Copper and Zinc from 4th.
month to12thnonth of life.
gr.
4th. month
milk 640
flours 1 5
meat 30
Total
5th. month
milk 640
flours 25
meat 1 00
Total
6th. mouth
milk 510
flours 50
meat 100
Total
7th. 8th. month
milk 600
flours 50
meat 100
Total
9th. 12th .month
milk 650
flours 70
meat 120
Total
Cd (pg)
21,0 (14-24)
2,4 (0,7-3,3)
1,2 (0,9-2,1)
24,6 (15-30)
21,0 (14-24)
4,0 (1,2-5,5)
4,0 (3-7)
29,0 (18-36,5)
16,5 (11-19)
8,0 (2,5-11,0)
4,0 (3-7)
28,0 (16,5-37)
19,5 (13-22)
8,0 (2,5-11,0)
4,0 (3-7)
31,5 (17,5-40)
22,0 (14-25)
11,2 (3,5-15,4)
4,8 (3,6-8,4)
38,0 (21-49)
Cr (pg)
29,1 (22-39)
3,3 (0,4-21)
4,8 (3-8,4)
37,1 (25-68)
29,1 (22-39)
5,5 (0,7-35)
16,0 (10-28)
50,0 (33-102)
23,2 (17-31)
11,0 (1,4-70)
16,0 (10-28)
50,0 (28-129)
27,3 (21-36)
11,0 (1,4-70)
16,0 (10-28)
54,0 (32-134)
29,5 (22-39)
15,4 (2,1-98)
19,0 (12-33)
64,0 (36-1 7O)
Cu (pg)
58,9 (25-103)
48,0 (13-111)
24,0 (12-39)
131,0 (50-253)
58,9 (25-103)
80,0 (22-185)
80,0 (40-130)
219,0 (87-418)
46,9 (20-82)
160,0 (44-370)
8O,O (4O-130)
287,0 (104-582]
55,2 (24-96)
160,0 (44-370)
80,0 (40-130)
295,0 (108-596;
59,8 (26-104)
224,0 (63-518)
96,0 (48-156)
380,0 (137-778]
Zn. (ing)
2,1 (0,9-2,9)
0,2 (0,2-0,4)
0,6 (0,3-1,0)
3,0 (1,4-4,4)
2,1 (0,9-2,9)
0,4 (0,3-0,6)
2,2 (1,1-3,8)
4,7 (2,3-7,3)
1,7 (0,7-2,3)
0,8 (0,6-1,2)
8,2 (1,1-3,8)
4,7 (2,4-7,3)
2,0 (0,9-2,7)
0,8 (0,6-1,2)
2,2 (1,1-3,8)
5,0 (2,6-7,7)
2,1 (0,9-2,9)
1,0 (0,8-1,8)
2,6 (1,3-4,5)
5,7 (3,0-9,2)

-------
                                1708
                        TABLE IX
  Daily intake of Cadmium, Chromium, Copper and Zinc (average
  diet for adults).

Soup *
Main course *
Dessert *
Wine (1 litre)
Water
Fruit
Different Poods
Total
Cd - mg/d
0,013
0,008
0,006
0,008
K.D.
0,002
0,011
0,048
Cr - mg/d
0,017
0,006
0,007
0,020
N.D.
0,007
0,017
0,074
Cu - mg/d
0,360
0,883
0,099
0,750
N.D.
0,102
0,336
2,530
Zn - rag/d
2,066
3,900
0,366
7,300
0,013
0,262
2,240
16,147
N.D. « Not Detectable
 *   = The values for soup,  main course and  dessert vere obtained
       respectively from pools of various  items.
                          TABLE  X
 Cadmium,  Chromium,  Copper and Zinc  in some  food  samples from
 an African region.  ( The values are folloved  by  tvice  the
 standard  deviation ).
Samples
maize flont
maize paste
cassava
boilet cassava
beans
peanut pastry
rice
rice
tomates
spinach
vegetable sau-
ce
Cd,
p. p.m.
<: 0,05
< 0,05
<0,05
0,05
0,07 + 0,01
0,08 + 0,02
< 0,05
< 0,05
0,05 + 0,01
< 0,05
0,08 •+• 0,02
Cr
p.p*m*
<. 0,1
^0,1
0,10 + O,01
<. 0,10

-------
TABLE  XI
Ratio Cd/Zn in adults diet :

Soup 1 160
Main course 1 500
Dessert 1 63
Wine 1 1000
Fruit 1 1 500
Different foods 1 20O
Total diet 1 30O
Total vine and fruit 1 200














Ratio Cd/Zn in african food
samples :
Beans 1 : 361
Peanut pastry 1 : 361
Tomatoes 1 : 450
Vegetable sauce 1 : 240



Ratio Cd/Zn in flour samples:
14 -4 <• f*t
1 11O
21 £1
1 O J
3 1 120
4 1 60
5 1 115
C. 14 RR














Ratio Cd/Zn in babies diet :

4th, month 1 : 121
5th. " 1 : 162
6th. " 1: 160
7th.-8th. month 1: 158
9th.-12th. " 1 : 150


Ratio Cd/Zn in milk and strained
foods (average);


Milk 1: 100

Strained mead 1 : 550

                                                                            --J
                                                                            o

-------
                              1710
     Although it is not possible to carry out a complete
comparison with similar items produced in Italy the values of
cadmium and chromium seem considerably lower in the African
samples.
     It is known that cadmium is a natural contaminant of zinc
and that the ratio Cd/Zn is 1/35O in the earth surface and
about 1:10O in foods (average).

     Our results  (table  XI) generally confirm such a ratio,
particularly as far as milk and flour are concerned.
     These results also suggest that a limitation of cadmium
in the diet is impossible unless the zinc is also limited.
     In effect considering the ratio of the intake of cadmium
and  zinc is about  1:3OO, the latter element should not exceed
30 mg. in the diet.
     Higher values could bring into the organism an excess of
cadmium.
     Acknowledgments
     The African  foods were sampled by Prof. M. Cresta and Prof.
G.B. Panatta whom the Authors  especially wish to thank.
     References
1.   Underwood E.J.,: Trace elements in human and animal nutrition
     Ac. Press 1971
2.   Cotzias G.C.,: Proc. 1st. Ann. Conf. Trace Substances
     Environ. Health 1967 - D.D. Hemphill Editor, Columbia,
     Missouri
3.   Liebscher K., and Smith H.,  : Arch. Environ. Health 17
     881,  (1968)
4.   Perry H.M. , M'ipton I.H.,  Schoeder H.A., and Cook M.J. ,
     J. lab. clin. Med. 60, 245,  (1962)
5.   Tipton J.H.,  and Cook M.J.,  ; Health Phys. j) 1Q3,  (1963)
6.   Lanzola E., Allegrini M.  and Breuer F.: Lead levels
     in infant food sold on the Italian market. Proc. Intern.
     Symp. Environ. Health Aspects of Lead, Amsterdam Oct. 1972
7.   Burgio G.: Trenta giorni  di Medicina 8, 2-3  (1972)

-------
                              1711
      THE USE OF RADIOISOTOPES 32P AND 35S IN STUDIES ON
              BEHAVIOUR OF PESTICIDES IN THE SOIL
                        Z, SZPERLltiSKI

Institute of Environmental Engineering, Technical University,
Warsaw, Poland
ABSTRACT

     Activated atoms were used to determine the possibility and
extent of contamination of underground waters by pesticides used
in agriculture.   The sorption and dynamics of the migration of
two organo-phosphorous pesticides, fenitrothion and malathion,
were studied.   These two products are currently being proposed
as agricultural pesticides.

     They would replace other substances of the chlorinated hy-
drocarbon group, including DDT, which are more dangerous to
health, but are themselves source of potential contamination of
underground waters.

     Studies were carried out on different types of soil:  sandy
and clay.   Isotherms were determined for the eorption of the
pesticides examined, together with their distribution coeffici-
ents, Kd.   The effects of pH values, exchangeable ions, organic
substances (organic C and N) and the mechanical composition of
the sorption of fenitrothion and malathion were studied.

-------
                              1712
1.  Introduction
     The negative effect  of  pesticides  from the group of chlor-
inated hydrocarbons  on the environment  is commonly known.  This
particularly concerns  such persistent pesticides as DDT and
lindan as well as aldrin, dieldrin,  endrin and heptachlor.  In
Poland, as in many other  countries,  the production and use of
DDT is gradually being limited  with  simultaneous introduction of
such substitutes as  organophosphorus compounds and derivatives
of carbamic acid. Our industry is already producing several
organophosphorus insecticides as  malathion and fenitrothion which
may replace DDT in the combat with insects.  Organophosphorus
pesticides, in comparison with  DDT,  are much easier subjected
to various changes in  soil  and  plants as biodegradation, hydro-
lysis or oxidation and furthermore can  be absorbed by the soil.
However, one should  consider the possibility of the pesticides
being displaced from the  soil  as a result of rainfall.  These
factors favour the migration of pesticides and their metabolities
to subsoil waters.
       This report presents the  results of studies on the so—
 rption of fenitrothion and nialathion  in the soil and the wa-
 shing out of  these pesticides from the soil by rainfall. The
 investigations were  conducted with the use of the radionuolirte
 labelled compound technique, at present often employed  in
 experimental  practice.
 2. Materials  and methods^
       The investigations were conducted on pseudopodsolic SOIIB
 taken from top  layers /O - 25 cm/  from which textural groups
 of soils differing in their mechanical and chemical composition
 cation exchange capacity and organic  matter, were taken
 /table I/
                                                   o e
       The studies were conducted with malathion -   S with

-------
                             1713
                                              Table  I
      Characteristic of soils used for study
Soil - textural groups
1. Silt loam
2. Silt loan
3. Silt loan
4. Silt loan
G. Silt loan
6. Sandy loam
Clay
<0,002 mm
%
4,0
G,0
8,0
7,0
13,0
5,0
Organic
matter
%
0,77
0,80
0,94
1,54
3,55
8,47
PH
"2°
5,4
5,8
7,4
7,6
6,7
6,1
in ICC1
4,4
4,9
7,3
7,4
6,4
5,3
Exchangeable
Cations Ca2* + Mg2*
in m.e per 100 g
1,6
3,O
7,8
9,8
11,6
18,7
                                                  32
specific activity 4,5 mCi/mmol and fenitrothion -   P with
specific activity 1,25 mCi/mraol. The labelled pesticides were
from the Radiocheraical Centre, Amersham, Buckinghamshire, En-
gland. 30 cm long glass columns with inner diameter 2,7 cm
were filled with 100 g samples of soil dried in the air and
then 1 rag of labelled pesticide in 5 ml water solution /mala—
        35
thion -   S/ or 5 ml diluted ethanol solution /fenitrothion —
-32P/
was added drop by drop. The columns of soil were eluted with
distilled water imitating rain water. Through each of the columns
were passed 350 ml of the distilled water, equalling 616 mm of
rainfall* The height of the soil in each of the columns wast
1 - 13,2 cmf 2 - 13,6 cm; 3 - 13,0 cmj 4 - 14,1 cm; 5 - 16,0
cm: 6 - 20,5 cm. After dilution with distilled water the co-
lumns of soil were divided into 1 - 2 cm thick layers. The
radioactivity of individual portions of the eluate /25 - 50 ml/
and layers of soil was determined. The samples of eluate and
soil for radioactivity measurements were prepared in standard
aluminum containers with 24 mm inner diameter and 4 mm depth.
                               qp
The activity of fenitrothion -  "P was determined in 0,5 ml
                            35
of solution and malathion -   S from 1 ml after evaportion to
dryness. The mass of the soil taken for radioactivity measure-
ments was 300 - 500 mg. Additionally a correction was determi-
ned for the self-adsorption of activity emitted by the soil.
The radioactivity of the samples was determined with Geiger
 ii
Muller counter type BOH - 42/0 introduced as part of a measu-

-------
                             1714
 ring  unit produced by ,,Polon". Geometry of calculation  2 "Tl
 time  unit 100,? were employed. The activity of samples was ca-
 lculated  for the concentration of the pesticide  in ug per soil
 mass  unit or volume of solution.
 3. Results of studies and conclusions
        The results of the studies on the penetration of  fenitro-
         o n                  n j*
 thion   -    P and malathion -   S through soils with varied me-
 chanical  and chemical composition are presented  graphically  in
 figures 1 and 2. The axis of ordinates gives the % quantity
 of pesticides eluted from the soil by distilled  water imita-
 ting  the  size of rainfall in ram. Fig 3 presents  the adsoption
                                                    ~r
             100      200      300      400     500      KB
                          WATER  ELUTED . mm
       FIGURE 1.  ELUTION   OF  FENITROTHION   THROUGH  VARIOUS  SOILS

of fenitrothion -32p  and malathion - 35S in individual layers
of a column on the example  of two  soils. The studies indicate
that fenitrothion - 32p  is  adsorbed by soils to a much greater
degree than malathion -  35Se This  is probably due to the dif-
ference in the structure of the molecules of the examined pes-

-------
                               1715
 ticides. The displacment of pesticides from the soil  occurred
 with rainfall up  to  about 350 mm and was for fenitrothion  - ^2p
                                         o tr
 between 10 and  45$ and for malathion -   S between 38 and  76 %
 of the amount introduced. There has been observed a certain co-
 rrelation  bewteen the content of  the clay  fraction/CO»002 mm/,ion
 exchange capacity and content of organic matter in the  soil and
   80-
   70-
   60 -
   50 H
D
LU
3
UJ

g

<
<
40-
                                                   600
                        WATER  ELUTED ,  mm
        FIGURE 2  ELLJTfONI OF  MALATHION  THROUGH  WARlQUS  SOILS
 its capacity  to absorb malathion and fenitrohion /fig  1  and 2/.
 The rate of adsorption of the studies solis in ^ag/g soil is as
 follows? for  malathion - 35S - 1 - 253| 2 - 2,7$ 3
                                                TO
                                                   - 3,45
4 -
4,6^
5 - 5P0| 6 - 6S1$
0 - G,i| 4 - 6,7;
                          for fenitrothion
                                                  P - 1 -
                               - 7,8; 6 - 8fQ0 The sorption  of

-------
                             1716
the studied  pesticides was not similar  for  different layers of
a column  /fig 3/ which may indicate the possibility of the
migration of these compounds in soil as a result of rainfall.
      The conducted studies allow the following conclusions:
1, Soils  have a certain capacity to absorb  fenitrothion and
malathlon and this capacity depends on  the  mechanical composi-
tion, ion exchange capacity and content of  organic matter
2, Fenitrothion and malathion can be displaced from the soil
due to rainfall which indicates the possibility of their
migration to subsoil waters.
                               o FENITROTHION -32P
                                MALATHION - ^>S
                              10
                       COLUMN  LENGTH , cm
            FIGURE  3. ADSORPTION  OF  FENITROTHION -32P AND
                     MALATHION-355  IN SOIL COLUMN

-------
                             1717
3. The use of pesticides labelled with radionuclidos in inve-
stigations on the behaviour of pesticides in the environment
facilitates comparatively accurate measurements of low con-
centrations of these compounds in soil and solutions.

-------
                               1719
    DIE BELASTUNG DER NAHRUNG MIT DEN SCHWERMETALLEN BLEI,
      CADMIUM, QUECKSILBER, ZINN UND ARSEN  IN OSTERREICH

                  H, WOIDICH, W, PFANNHAUSER

Forschungsinstltut der ErnMhrungswirtschaft, Wien, Osterreich
KURZFASSUNG

     Aufgrund der atatiatiach ermittelten Verzehragewohnheiten
dea Oaterreichera und unter Veruendung der analytisehen Da ten
ftir Sohwermetallepuren in Lebenamitteln laird die Aufnahme von
Bleif Cadmium, Queakailberf Zinn und Araen abgesch&tzt.

     Die ermittelten Werte uerden filr ein&n Uberbliok fiber die
durah Lebensmittel aufgenommenen Mengen an potentiell toxisohen
Sahwermetallen in Osterreich und fUr einen Vergleich mit aus
der Literatur bekannten Aufnahmedoaen in anderen LUndevn und
internationalen Toleranzen herangezogen,

     Daruber hinaus haben wir auafilhrlictie Untersuchungen an
verh<nism&asig etark mit Sahvermetallen kontaminierten Lebene-
mitteln durchgeftthrtf wobei aich zeigt, daes ea fur die Elemente
Queakailber^ Cadmium und Araen jeweila ganz beetimmte Gruppen
von Lebenamittel aind, die einen gegenUber den durehschnitt-
lichen Werten anderer Nahrungamittel erhOhten Gehalt aufueiaen.
Zinn wird in erhebliohen Mengen lediglioh in eingedoaten Lebena-
mitteln gefunden* wdhrend Blei unregelmitaaig auf Lebenamittel
verteilt iat.

     Die ermittelte monatliohe Aufnahme an Sahwermetallen betragt
pro Person 2tS mg Blei, 0,6 mg Cadmium, 0,2 mg Quecksilber,
10 mg Araen und 100 mg Zinn.

-------
                               1720
     Gemessen an der Sahwermetallbelastung in anderen Ldndern
liegen die in Ssterreich aufgenommenen Mengen bei Quecksi Iber,
Blei und Arsen verhaltnismassig tief, wdhrend die nahrungsbedingte
Aufnahme von Cadmium an Spitzenwerte industrialisierter Staaten
heranreicht.   Insgesamt liegen alle Werte weit unter den von
FAO/WHO angegebenen vorldufigen annehmbaren Aufnahmedosen.
ABSTRACT

     The intake of lead, cadmium* mercury, tin and arsenic is
estimated on the basis of statistics on Austrian dietary habits
and analytical data on traces of heavy metals in food.

     The values obtained are used in a survey of the intake of
potentially toxic heavy metals via food in Austria, and in a
comparison with intakes (reported in the literature) in other
countries, and international tolerance standards.

     In addition we carried out detailed investigations of foods
which were relatively heavily contaminated with heavy metals.
These showed that there are distinct groups of foods which have
a high content of the elements mercury, cadmium and arsenic, as
compared with the mean levels of these in other foods. Tin is
found in considerable quantities only in tinned foods, while
lead is irregularly distributed in foods.

     The monthly intakes of heavy metals found per person are
2,5 mg lead, 0.6 mg cadmium, 0.2 mg mercury, 10 mg arsenic and
100 mg tin.

     Compared with the incidence of heavy metals in other coun-
tries,  the intakes of mercury, lead and arsenic in Austria are
relatively low, while the cadmium intake via food reaches peak.
levels for industrialized countries.   Taken art a whole, all
levels are well below the FAO/WHO provisional acceptable intakes

-------
                             1721
1.Einleitung
Zur Ermittlung der nahrungabedingten Schwermetallaufnahme
in Osterreich warden die durchschnittlichen Verzehrsge-
wohnheiten nach Ergebnissen der Konsumerhebung  1964  heran-
gezogen. Wir untersuchten alle Nahrungsmittel,  deren monat-
licher Verzehr 0,4 kg Uberstieg,  sowie fallweise  auch in
geringen Mengen konauiaierte Lebensmittel.  Tabellarisch
werden jene Lebensmittel aufgeflihrt, fiir die  eine besondere
Belastung iait einzelnen Schwermetallen festgestellt  werden
konnte.

2.Belastung der Nahrung mit Schwermetallen
2.1. Blei
Blei zahlt wegen seiner hauptsachlichen Verbreitung  Uber
Autoabgase zu den systematisch schwer fafibaren  Schadatoffen,
da der Standort der pflanzlichen  Nahrungs- und  Futtermlttel
von Bedeutung 1st. Es existieren  kaum besondera belastete
Lebensmittel. Viele in Autobahnnahe gedeihenden Pflanzen
weisen erhebliche Mengen Blei auf /"1_7. Die Streuung  der fiir
bestimmte Lebensmittel erhaltenen Werte ist deshalb  auch
sehr groS.
Wir fanden, daB hinsichtlich der  aufgenommenen Mengen die
gewichtsmafiig an der Hahrung stark beteiligten  Lebensmittel
zur Bleiaufnahme den grbSten Beitrag leisten. Die  Tab. I
gibt neben einigen Grundnahrungsmitteln auch  einige  verhalt-
nismaSig hohe Werte in verschiedenen anderen  Lebensmitteln
an, die vielfach nur vereinzelt gefunden wurden.
Die abgeschatzte Gesaintaufnahme von 2,5 mg Pb/Pergon und
Monat liegt unter der fiir hochindustrialisierte Staaten an-
gegebenen Aufnahmedosis von 6-9 mg Pb/Person  und Monat.

2.2. Cadmium
Cadmium hat sich als ubiquitiir verbreitetes Spurenelement
erwiesen. Nach jungsten Berichten [&J hat in  einigen Landern
die Aufnahrne von Cadtaium mit der  Nahrung bereits  die Grenze
der vorlaufigen annehmbaren Dosie der WHO/FAO erreicht.
Einzelne Gruppen von Nahrungsmitteln (Tab. II)  sind  nach un-
seren Untersuchungen £*J tatsachlich vernaltnismaBig stark

-------
                             1722
Tab. I
Bleigehalte ausgewahlter Lebensmittel

Lebensmittel           Proben-      Bereich
verschiedene
Brotsorten
Schweinefleisch
Fische  (konserviert)
Milch (friach)
Milch (konserviert)
Johannisbeersaft
Apfelsaft
Orangensaft
Hafer (Ganzkorn)
Preieelbeeren-
konfitiire
Proben-
zahl
                                                  Gehalt
7
7
18
4
4
1
2
1
11
0,10-0,61
0,05-0,24
0,05-0,47
0,08-0,15
0,20-0,29
-
-
-
0,90-0,95
0,330
0,157
0,190
0,110
0,275
0,74
0,72
0,62
0,93
                                                       0,90
                                                 0,70
Tab.  II
Sorte
            Cadmiumgehalte ausgewahlter Lebensmittel

                     Frobenzahl   Bereich /ppm7  Mittelwert
Hals
Weizen
Hafer (Ganzkorn)
Haferflocken
Haferkleie
Gerste

Muscheln
Thunfisch
Sardinen
sonstige Fische

Schweineniere
Schweinefleisch
3
13
4
3
1
4
39
13
27
65
15
15
0,59 - 1,05
0,006- 1,05
1,1 - 1,93
0,007- 0,090
5,56
0,05 - 0,27
0,05 - 1,74
0,05 - 0,97
0,05 - 1,00
0,05 - 1,46
0,01 - 0,41
0,008- 0,025
-
0,24
1,32
0,045
-
-
0,31
0,14
0,23
0,13
0,15
0,014

-------
                             1723
mit Cadmium verunreinigt. Dazu zahlen die Hullenmaterialien
einzelner Cerealien, Innereien (Niere, Leber) und Muscheln.
In einigen Spezialbrotsorten, die sich durch geringen  Ver-
mahlungagrad auszeichnen, fanden wir beachtliche Cadmium-
gehalte. Aufgrund unserer experimentellen Daten schiitzen
wir die Cadmiuniaufnahme in Osterreich auf 0,6 mg Cd/Person
und Monat.

2.3. Quecksilber
Quecksilber findet sich in wenigen Lebensmitteln in beach-
tenswerter Konzentration. Viele Proben lagen an der Nach-
weisgrenze der verwendeten Hethode &J . Tab. Ill en thai t
eine Aufstellung der mit Quecksilber besonders belasteten
Nahrungainitteln zu denen Thunfische und auch Pilze zahlen
/67. Die Gesamtbelastung der Nahrung belauft sich auf  etwa
0,2 rag/Person und Monat und liegt an der unteren Grenze der
in der Literatur angegebenen Werte $, *£/.

2.4. Zinn
Hauptquelle der nahrungsbedingten Zinnaufnahme sind in ver-
zinnten Behaltern eingedoste Lebensmittel.
Da Zinn unter den betrachteten iSlementen die geringste Toxi
zitat aufweist, wird die Zinnaufnahme aus dem Verzehr von
Gemiise- und Obstkonserven, Obstsaft- und Fischkonserven
allein abgeschatzt (Tab. IV). Basierend auf eigenen Unter-
suchungen /jB,9/ ergibt sich aus diesen Quellen eine monat-
liche Zinnaufnahrae von etwa 100 mg. Zu berlicksichtigen ist,
daQ bei einzelnen Gemiise- und Obatsorten eine Anreicherung
von Zinn in der Prucht festgestellt wurde, die bevorzugt
verzehrt wird
2.5. Arsen
Mit Arsen sind vor allem Muscheln, Innereien und einzelne
Raucherprodukte starker belastet. In einzelnen Fallen weisen
auch Biere merkliche Arsengehalte auf (Tab. V).
Die aufgrund unaerer Untersuchungen geschatzte Belastung der
Nahrung betriigt hbchstens 10 mg/Person und Monat und liegt
dainit deutlich unter Angaben aus der Literatur /10/.

-------
                          1724
Tab.  Ill
                            aus^ew.ihltcT Lebenniai ttel




                    Probensiahl     Hereich      Mittelv/ert
Thunfische
Herir.ge
Sardinen
Muscheln
Scelachs
Hois
Teigwaren
Kler
95
71
55
27
10
7
3
6
0,005-1,57
0,001-0,108
0,001-0,091
0,005-0,077
0,038-0,370
0,001-0,007
0,001-0,018
0,001-0,204
0,330
0,025
0,015
0,027
0,101
0,003
0,007
0,056
Pilze (trocken)
0,002-2,52
Tab. IV
       Zinngehalt  ein^cdoster Lebtnscdttel
1,376
              -97
G-eiiiUKe- und
Bereich
Mittolw
ObH tkonservon Probenzahl /PPii'7 /DDUI?
Spargol
Champignon
An ana a
Krdboeren
Bohnen
Karillen
Pfirsiche
Schvrarzwurzel
Ob si- und GcmiiGeoaf te
Ananasaaf t
Ueiuiisesaf t
Grapcf rui tsaf t
Orangensaft
Karottenaaf t
Tosjatenaaf t
Zi tronensaf t
FiBChkons o ryon
Kuschel
Sard in en
8 117-309
11
5
7
2
2
6
7
1
1
4
4
2
4
2

6
7
61
65
10
- 117
- 244
- 42,5
17,5
77
19
120


63
53
15
16
25

20
15
,5- 97,5
- 77,5
- 408
192
66
- 95
- 216
- 17
- 135
- 80

- 95
- 85
•- * * ^*
211
72
127
25
17,5
87,5
54
225
192
66
76
123
16
32
52

44
25,5

-------
                            1725
Tab. V

       Arsengehalte in ausgewahlten Lebensmitteln

Lebensrnittel       Probenzahl     Bereich       Mittelwert
                                   /PPW

Dorschleber            9        0,05 -  3,54         0,88
Muscheln              34        n.n. -  1,20         0,21
Raucheraal             1           °»38
Tintenfisch            1           0,13
Bier                   4        0,05 -  0,33         0,21
Reis                   4        0,02 -  0,09         0,055
 3. Diskussion der lirgebnisse
 Aus unseren Untersuchungen geht hervor, daB die  Schwerme-
 tallbelastung der Nahrung in Osterreich iin allgemeinen  un-
 ter der stark industrialisierter Staaten  liegt  (Tab.  IVJ.
Tab. VI

 Schwermetallaufnahuie  aus  Lebensmitteln /Ing/Person u.Monat/

    FAO/WHO vorlaufige    Errnittelter      Literatur-
    annehmbare monat-        Wert          angaben
    liche Aufnahme

Pb       12,0                 2,5          6-9      /2,  1&

Cd       1,6 - 2,0           0,6       0,45-0,90   /2,  12, 13.7

Hg       1,2                 0,2       0,2-0,3   /2,6,11,127

As      100  (max.Aufnahme)    10           15-70     /1Q7

Sn       -                  100            100      /"127

-------
                             1726
Die Bleiaufnahme 1st deutlich geringer, als  beispielsweise
in den USA /2/.
l)er Quecksilbergehalt in Lebensroitteln ist sehr gering,
auBgenouiwen bei bestii.nnten Fischarten und bei  Trockenpilzen.
Eine Gefahrdung der Gesundheit  ist  selbst bei  starkem  Fisch-
verzehr nicht zu erwarten.
Zinn als wenig toxisches Ion gelangt nahezu  ausechliefllich
durch eingedoste Lebenswittel in nennenswerter Menge in  die
Nahrung.
Me grofie Eropfindlichkeit von Kindern gegen  Zinn  und Blei
muB aber berucksichtigt werden  und  erfordert laufende
Kontrolle.
Arsen findet sich kumuliert in  touscheln und  Innereien.
Hauptnahrungsmittel weisen dagegen  sehr geringe Arsenspuren
auf.
Lediglich bei Cadmium konnte festgestellt werden,  dafl  die
Belastung der Nahrung im Mittelfeld der Literaturangabe
zu liegen kommt.
Wir erachten es deshalb als notwendig, der Uberpriifung des
Cadmiumgehaltes der Nahrung erhbhtes Augenmerk zu schenken.
                LITiSKATUR  :
[\]   A.Kloke, H.-O.Leh "Verunreinigung von Kulturpflanzen
      mit Blei aue Kraftfahrzeugabgasen".
      Proceedings of the First European Congress on the In-
      fluence of Air Pollution on Plants and Animals.
f2j   WHO Tech. Kept. Series No.505, Genf 1972
      H.Woidich, W.Pfannhauser "Zur Analyse von Cadmium in
      Lebensmitteln." Z.Lebensm.Unters.Forsch. 1974 (im Druck)
      A.Tolan,  G.A.H. Elton  "Total Diet  Studies on Mercury
      and Cadmium in the United Kingdom."
      Vortrag beim Europaischen Kolloquium  Luxemburg  1973

-------
                             1727
      H.Woidich, W.Pfaimhauser  "Zur  quantitativen Be-
      stiuiraung von Queckailber  in  biologischen Materialien.
      H.Untersuchung iiber AufschluB und  Bestiinmung raittels
      Atomabsorption in der  Gasphase."  Z.Lebensm.Unters.
      J?orsch. 1974 (im DruckJ

      R.Schelenz, J.P. Diehl "Quecksilber in  Pilzen".
      Z.Lebensm.Untera.i'orsch.  154,  160 (1974)

      R.Schelenz, J.P. Liehl "Quecksilber in  Lebensmitteln
      Untersuchungen an tiiglicher  Geaamtnahrung."
      Z.Lebensm.Unters.forach.  133,  151  (1973)

      H.Woidich, W.pfannhauaer  "Vergleich dreier  ^ethoden
      zur Beatimmung von Zinn in Konserven"*  Z.Lebensm.Unters,
      Forsch. 151. 114 (1973)
      H.Woidich, W.Pfannhauser  "Untersuchung  Uber die  Ver-
      teilung von Zinn zwischen Prucht und  Aufgufifliissigkeit
      in Gemiiae- und Obstkonserven." Deut.Lebensm.Rdsch.69
      224 (1973)                                         ~~

      Joint PAO/WHO Expert Committee on Pood  Additives
      WHO Tech.Rept. Series No.373,Genf 1967

C\ \]  R.Schelenz, J.jj'. I»iehl  "Quecksilbergehalt von  Lebens-
      mitteln des deutschen Marktes" Z.Lebensm.Untera.i'orsch.
      151. 369  (1973)

fi2}  WHO Tech.Rept.Series No.532 Genf 1973

/137  Workingparty on the Monitoring of foodstuffs for
      Heavy I'letals. Fourth Report: Survey of  Cadmium in  Food.
      Ministery of Agriculture, Fisheries and Food,  UK 1973

-------
                              1729
      ASBESTOS MEASUREMENTS IN THE CALIFORNIA ENVIRONMENT

     J, J, WESOLOWSKI, R, STANLEY/ G, SMITH, L, CARPENTER
                         AND A, BERNER

Air and Industrial Hygiene Laboratory, State of California Dep-
artment of Health, USA
ABSTRACT

     The California population is exposed to asbestos from nat-
ural sources such as wind and water erosion of asbestos-contain-
ing rook formations and from a multitude of anthropogenic sources,
Of particular concern are auoh sources as asbestos mines and
millsj automobile brake linings, asbestos spraying for fire-
proofing of buildings, asbestos cement pipes for transporting
drinking water, talcum, and the widespread use of asbestos fil-
ters in the food, drug and beverage industries.

     Inhalation of asbestos fibers in sufficient quantity is as-
sociated with at least three diseases - asbestosis, lung cancer
and mesothelioma, with latency periods measured in decades,
Thus it is important that long-range monitoring programs be
established now to help assess the health effects of this pol-
lutant.   The mechanism of pathogenicity may be dependent on the
size of the fibers, so monitoring schemes should include size
distribution data as well as identification and counting tech-
niques.   Since most of the particles of asbestos found in en-
vironmental media are too small to be resolved by the optical
microscope, electron microscopy must be used.   This paper
briefly discusses the current status of the sampling methodology
and the electron microscopic techniques used to identify, count
and size asbestos particles in environmental media and presents
data from ambient air sampling near a California asbestos pro-
cessing plant.

-------
                                           1730
1.     Introduction
       This discussion is restricted to the measurement of asbestos in air and water, although
much of what will be  said is  applicable to food,  drugs, beverages and other  media.  The
methodology of asbestos quantification in environmental samples falls into  four  interrelated
areas:  sampling,  preparation  of the  sample  for subsequent analysis, identification,  and
determination of the concentration of asbestos.

2.     Sampling Methods
       Air is  moved  through a membrane filter at a specified flow rate  for a length of time
determined by the  expected concentration of asbestos  in the air. For ambient air monitoring,
the  Air  and  Industrial  Hygiene  Laboratory  (AIHL)  method employs 0.8 pm pore  size
Nuclepore filters of 47  mm diameter, at a flow rate often liters per minute, for 2 to 24 hours.
       For water  samples,  a suitable amount, usually  about  one  liter, is collected in a clean
bottle or flask. The water is then passed through a filter, such as Nuclepore, which collects the
asbestos  particles.  Although  the filter collection efficiency for small asbestos particles remains
to be determined, clearly  the  smaller  the pore  size,  the higher the collection  efficiency.
Unfortunately, filtration of water through a  small pore  filter is  a slow process and  if, as is
usually the case, the water contains many other kinds of particles, the filter may clog before all
the water has gone  through. It has been estimated that errors of the order of 50% may occur as
a result of inefficient filtration through 0.8 micrometer and larger pore size filters.

3.     Sample Preparation
       Two general classes  of  sample  preparation utilized  to obtain  size  distributions by
electron  microscopy are the direct-clearing technique and the  separation technique. The direct
method  presents  the  microscopist with a sample on which  a minimum  of manipulation has
been carried out but which contains  both asbestos particles and all other particles which were
present   in  the environmental  medium when the  sample  was collected.  The  separation
technique presents  a sample which contains a minimum of interfering particles, but which was
prepared with much manipulation.
       In many air samples and in some water samples it is possible to use the direct-clearing
method to size and count asbestos fibers among the other materials present. In this method the
sample filter is coated with a layer of silicon monoxide or carbon to hold the fibers in place.
The  filter itself is then dissolved by vapors of a solvent such as  acetone or chloroform. The film
with imbedded fibers and other particles is then examined by electron microscopy. However,
urban air containing 100 micrograms of particles per  cubic meter will probably contain on the
average only 0.01  microgram asbestos per cubic meter. Thus the sample filter contains 10,000
times as  much non-asbestos as asbestos material by weight. The ratio  of particles depends on
the  densities and sizes,  but for average conditions encountered in urban air will vary from

-------
                                 1731
Fig. 1.   Diffraction pattern obtained from a chrysotile fiber
          collected downwind of an asbestos mill.

-------
                                          1732
around  10,000 to 100,000 non-asbestos particles for every asbestos particle. The ratio is even
larger  in  water  samples.  Under  such  circumstances, qunntitation by  the direct  clearing
technique is difficult.
        In the separation  technique the  filter is ashed to destroy organic  particles and fibers,
carbon, and the filter material  itself. The ashed sample is then mixed with a solution of 0.5%
Parlodion in amyl acetate. An aliquot of the suspension is dropped on a water surface within a
confined  area  of  known  size.  Electron  microscope  grids arc  then drawn through  the
asbestos-containing Parlodion film for subsequent counting and sizing, (Smith |1 ]).

4.      Identification and  Sizing
        After sample preparation, the grid  is scanned at  1000X to determine uniformity of
sample  deposition.  The  search  for fibers  is conducted at 8000X. Each  fiber detected  is
identified at 20,OOOX morphologically and  by electron diffraction if necessary. (Seshan [2]).
The length and diameter arc then  measured.
        Although chrysotilc  may be  presumptively identified  by  recognition of a  central
longitudinal canal in the fibril, positive identification requires electron diffraction. Figure 1 is
an example  of a diffraction pattern obtained from a chrysotilc fiber collected downwind of an
asbestos mill. Crystal lattice parameters arc computed from such spot and ring patterns and
compared with those of standards (Skiknc [3]).
        This technique allows differentiation of asbestos from non-asbestos materials and of
chrysotile from amphibole asbestos. It does not, however, normally allow  the identification of
the specific  type of amphibole  present. Fibrils  smaller than 0.05 pm  in diameter  seldom
produce diffraction patterns suitable for identification.
        Since electron diffraction identification  is slow and expensive, in  most cases, only 5%
of  the  morphologically-identified  asbestos fibers  have  their identifications confirmed by
electron diffraction.
        After examining a minimum of 20 grid holes per each of three grids for each sample
the total number of fibers in eacli size category is calculated, producing the size distribution

5,      Asbestos Size Distribution Data
        An  experiment  was conducted  to  determine  the  feasibility  of measuring the
concentration and size distribution of asbestos fibers in ambient air near an asbestos processing
plant using the Parlodion method developed by AIHL. Sampling was carried out at a chrysotile
asbestos mill located near King City, California. The asbestos ore and spent ore were stored in
unprotected heaps upon the ground within the plant's perimeter. This storage, combined with
strong summer winds, made the operation a possible source of airborne asbestos.
        One sampler was located approximately  500  meters  northwest  of the source and
another 100 meters  southwest. Samples were collected for three hours during the afternoon of

-------
                                    1733
    106 -i
    105 -
a
CD
-t->
G
•H


;X
JH
0)
a>
    10*-
    103_
                                                    SOUTHWEST
                     0.05   0.1
0.5
r
1
I
5
 I
10
                             Fiber  length  —
       Figure 2 — Asbestos fibers in ambient air  near asbestos  mill

                  shown by fiber length.  Points  connected by dashed

                  lines represent the limit of detection in those size

                  intervals where no fibers were  observed.

-------
                                1734
   106-,
£
   104-
(A
^
0)
   10°-
                   NORTHWEST
                                              SOUTHWEST
                   0.05 0.1
             0.5   1

Fiber diameter —yum
I
5
 I
10
         Figure 3 — Asbestos fibers in ambient air near asbestos mill

                    shown by fiber diameter.  Points connected by dashed

                    lines represent the limit of detection in those size

                    intervals where no fibers were observed.

-------
                                           1735
Figure H --

Asbestos fibers in ambient air near asbestos mill
shown by length and diameter.
                                         o

-------
                                          1736
April  26, 1972. During this sampling period the weather was clear, the temperature averaged
65°F  and the wind blew steadily from  the northwest at an average velocity of 600 cm/sec,
occasionally gusting to 1300 cm/sec. Due to the location of the mill at the base of a 100 meter
hill, the wind trajectory may be complicated. Thus both samplers would be expected to collect
source emissions, with the northwest or "upwind" sampler collecting less. Figures 2 and 3 give
the fiber size distributions for length and diameter. As expected the "downwind" sample has a
greater number of  fibers  than the  "upwind", and  the mean fiber diameter is slightly  larger.
Figure 4 is a three-dimensional plot of the number  of asbestos  fibers by  length and diameter
"downwind" of the plant. The number of fibers per liter is shown by the height of the column
plotted over the appropriate cell in the length by diameter matrix. The plot demonstrates that
the large majority of the fibers are  both  thin and short  (less than 0.5 /JLITI in diameter by less
than 2.5 jum long). The "upwind" plot is similar but  with slightly smaller fiber diameter.
       AIHL  has also carried out similar asbestos characterizations  for  other ambient air
locations in California and for  water samples collected from rivers and taps. Although there are
variations in the shapes  of the distributions and in the concentration of fibers present, all
samples demonstrate that the  large  majority of the  fibers arc too small to be detected by light
microscopy.

6.     Conclusions
       We have demonstrated the  applicability  of electron  microscopy  using the Parlodion
method for characterizing the fiber size distribution of asbestos.  Further, our data, as well as
that of others, clearly show that most  fibers are too small to be detected  by light microscopy,
thus  making  electron  microscopy  a necessary  tool  for assessing the  health  impact of
environmental asbestos.

                                   Acknowledgements
       We acknowledge the  contributions of P.K.  Mueller, SuzAnne  Twiss,  and Krishna
Seshan in various aspects of this work. E. Munson and J. Maloney of the Monterey-Santa Cruz
County Unified Air Pollution Control District facilitated the conduct of the field work.

-------
                                       1737
                                     References

\.     SMITH,  C., "Parlodion  Film  Method for the Determination of Asbestos in Ambient
       Atmospheres, AIHL Method 38-A", State Department of Health, Berkeley, CA 94704
       (1973).

2.     SESHAN, K.,  "Morphological Identification of Chrysotile Asbestos by Transmission
       Electron Microscopy, AIHL Report No, 149", State Department of Health, Berkeley,
       CA 94704 (1974).

3.     SKIKNE, M., TALBOT, J., KENDALL, R., "Electron Diffraction Patterns of U.I.C.C.
       Asbestos Samples", Env. Rsch., 4:2, 141-145 (1971).

-------
 EXPOSITIONSMESSUNGEN



  EXPOSURE MONITORING



MESURE DE L'EXPOSITION



MISURA DELL'ESPOSIZIONE



  METING VAN EXPOS IE

-------
                                1741
          SITING  CONSIDERATIONS  FOR AIR  QUALITY  MONITORS
               AS APPLIED  TO  EPIDEMIOLOGICAL  STUDIES

                           P,R,  HARRISON
 Meteorology Research, Inc., Altadena, California, USA
ABSTRACT

     Every  honest researcher  has  the  insecurity  that  his  con-
clusions may be biased  by measurement  technique  or  the  environ-
ments  in which the  tests were  conducted.   A  difficulty we  have
always had  is  "have we  measured what  we  think we  have measured?"

     There  are three factors  in any of the environmental  investi-
gations.    Those factors entail the inner-relationships between
the emission, the transport and the effect on the object  or
organism in question.   Thus,  in  any assessment of  effect of any
environmental assault,  one must consider the  characteristics of
the emission and the effect of its transport  media  before arriving
at the receptor.  This is especially true in "ambient" monitoring.

     Epidemiological studies using air monitoring data,  including
those data provided by  persons whose monitoring purposes are
other than epidemiological,  usually are accepted on face value.
It can be shown that the ambient  air quality  is very sensitive
to the placement of the sensor, relative to the emission source.
Thus, correlations of this data with respect  to health effects
can become overly biased,  according to this placement.   Some
currently important  pollutants are discussed with respect to
these siting considerations  and the relationship between sources
and the ultimate  receptor exposure.

-------
                               1742
     This paper presents some suggestions for siting of air
quality monitors with respect to the pollutant and its source.
It is imperative that perspective air quality data be investigated
to assure that the data is not overly sensitive to the sensor
location.

-------
                                 1743
1.  INTRODUCTION

    It is well known that much of the available air quality data,
historically as well as presently generated, contains many points where
the data becomes sensitive to Interference parameters that are un-
desirable and unknown to the final user.  The three major influences on
data as gathered for air quality measurement purposes are:

     1. Unknown variation in emissions
     2. Sensitivity to various types of transport conditions(meteorology)
     3. Sensitivity of the instrument site to local phenomenon

Unless the above influences are properly described in the statistics,
the conclusions can easily be misleading.

    Most standard techniques use a large sample taken at various
physical locations in order to come to  their conclusions.  Cor-
relations with various population groups are taken as subsequent
analysis.  Usually, those data points lying outside "acceptable" to  the
models are discarded.  There is always  the danger that these exceptions
may be the rule.  In lieu of meaningful documentation one can easily
rationalize that this is the best alternative.

    It is this researcher's opinion that it is better to work at the
"front end" of the problem: either document the existing station so  as
to show  its relationship to its immediate environs, or survey a
potential site in order to place  the sensor at a point which will
optimize the  sensitivity to the function which one is trying to
describe.

2.  EMISSION  CONSIDERATIONS

    When using air quality data one must be fully aware of  the
variations of the emission rates  in both the immediate and  general area
surrounding the sensor.  A case in  point is sulfur dioxide.  Since
sulfur containing fossil fuels are used extensively  in heating of space
 (space heating), as well as industrial  and  power generation uses, there
is a base line on which a seasonal  variation is super-imposed.   This
phenomenon is usually  accentuated by  the use of "cleaner"  fuels  in  a
scheduling called "interruptible  service".   In the northern United
States clean  fuels  are used in the wanner months.  In colder months  the
residential heating units use the cleaner fuel, but  the  industrial  and
electrical  Industries  use fuels containing  more sulfur.

    Suspended particulate also has  seasonal variation.   In Chicago
typical  maximum months would be May  and June as well  as  the winter
months.  The  minimum  is  typically in  July,  caused  by the variation  of
the agricultural activities, July being the maximum  growing month and
minimum  tillage.  This emission  variation,  both from natural  and man-
made  sources, linked  with  the greater frequency of  inversions  during
the latter  part of  the summer,  lead to an interesting variation.

3.  TRANSPORT AND METEOROLOGY

    The  variations  in temperature,  rainfall,  wind  speed and wind

-------
                                    1744
direction play a great part in the measured concentrations of any of
the pollutants.  As for suspended particulate, temperature as well as the
rainfall and growing season make a large difference in the total
measured during various months of the year.  In most areas winter winds
are stronger and more northerly, whereas the summer wind speeds are much
less.  Frequencies of stagnation and/or inversions are greater.  All air
quality values are dependent upon these parameters.  If one considers
re-suspension as a significant source of particulate fl} there is a
point at which increased wind speeds increase the particulate which is
contradictory to most modeling considerations.

    Parameters such as ozone are almost completely dependent upon solar
radiation but are also dependent upon natural sources.   It is well known
that many photo reactive organic materials are emitted by natural pro-
cesses of plant respiration and metabolism.  Similar statements can be
made concerning hydrocarbons.  Location of meteorological monitoring
equipment is extremely important.  Many times it is overlooked to the
point where not only technical skills have been neglected but common
sense also has been ignored.  Thus transport and meteorology are im-
portant controlling factors as to the concentrations of air quality
sensed at any single point.

4.  SITING OF SENSORS

    The technique of properly siting air quality instruments is not at
all well developed.  Each situation exhibits a unique local environment.
Location and deployment of air quality sensors are thus left to the in-
dividual under broad guidelines.  The final user of the information is
at a disadvantage.  He usually has to accept the information without
question and without full confidence that  the air quality data presented
is representative of the organism's exposure.

    Two examples are presented in Figure 1 and Table 1.  Figure 1 rep-
resents a location where the sensor is located six feet above the lip of
a depressed expressway.  Portable carbon monoxide sensors were de-
ployed on 27 occasions under various meteorological conditions.  The
nearest residence was approximately 50 feet from the telemetry station.
The hypothesis was that the telemetry station is not representative of
human exposures.  As one can see from the  results, the nearest household
was exposed to approximately one-half of the concentrations seen at the
telemetry station.  Correlation with health effects with this historical
station would, indeed, be spurious unless  this factor was taken into
consideration.

    Table 1 represents a three-day study at three different stations
throughout the city of Chicago so as to determine the height dependence
relative to total suspended particulate.   It is easily seen that the
lower sensors recorded higher values.  This fact, combined with the in-
formation that most of this added particulate was in a larger "non-
respirable size range, makes the correlation between suspended partic-
ulate and health indeed difficult.  Similarly, Table 2 shows the height
dependence of the highest and lowest station in the Chicago air quality
network.  These stations are located within two city blocks of each
other.  Both sensors are above the city-wide average for total suspended
particulate and are located in the business and commercial center of

-------
                                 1745
                               TABLE 1
                   High/Low 24-Hour Concentration
                                 of
              Suspended Particulate at Selected Stations
Geographical Sensor Height Above
Station Description Ground Level (in feet)
High Low
1. Clay Industrial 57 19


2. Stevenson "Clean" 51 -6
Residential
(Hi-Vol located
in a depression)

3, Austin "Dirty" 61 10
Residential


Concentration
(in ug/m3)
High
32
—
113
56

94

84
100

82
52
Low
81
75
98
67

70

91
152

127
91
Difference
49

-15
11

-24

7
52

45
39
                              ItNNiDY ANO EMNS JUNCTION


                                    .1 IVIISON
                          \ W, MQNI ilOSt  X,
Fig. 1.  Composite  of  27 sets of samples for carbon  monoxide across
         a juncture  of two expressways.   A telemetry  station is
         designated  as TAMN 1.  The nearest occupied residence is at
         "H".   Note  that the nearest residence  is  exposed to 2^%
         less  than  at  the telemetry station and  less than half that of
         the expressway.

-------
                                   1746
                                TABLE 2
                 Loop Area Study of Height Dependence
                                  of
         Suspended Particulate Concentrations (24-Hour Hi-Vols)
1.   CAMP Station:
         12' Height - Severe Canyon Exposure

2.   GSA Building:
        133" Height - Moderate Canyon Effect
         Date
Concentrations (in ug/m^)



(Sun.)


(Sun. )


(Sun.)

(Sun. )


(Sun. )


(Sun. )



Oct. 12/72
15/72
17/72
19/72
29/72
31/72
Nov. 2/72
5/72
7/72
9/72
14/72
16/72
19/72
21/72
23/72
26/72
28/72
30/72
Dec. 3/72
CAMP
218
109
179
239
94
121
127
143
146
299
163
199
94
218
140
63
210
262
170
GSA
72
61
78
150
41
74
87
124
93
128
68
171
65
113
63
-
105
145
33
Difference
+ 146
48
101
89
53
47
40
14
53
171
95
18
29
105
77
-
105
117
137
 metropolitan Chicago.

    It is well  known  that  ozone  is  quickly  converted  by  NO to  oxygen.
 Thus, in  the  areas  of  large automobile  traffic where some NO is emitted
 directly from the  exhaust  pipe,  ozone  concentrations  are relatively
 small.   Maximum ozone  concentrations will occur two to  three hours down-
 wind  of  the  source regions and at higher levels.   At  this point the  ozone
 would be well mixed.

    Hydrocarbons,  like suspended particulate, are  a catch-all  category
 for many compounds and combinations.   Since many hydrocarbons  are
 emitted, both by stationary as well as mobile sources (automobiles), it

-------
                                  1747
is difficult to correlate this parameter with those health effects under
presently available data.

    Finally, sulfur dioxide usually will decrease as the sensor height is
decreased.   Most sensors are located on building roofs slightly below the
horizontal  centerline of the plume.  The further one departs from the
centerline  the less the concentrations become.  (The opposite condition
is true with total suspended particulate due to reflotation phenomenon.)

5.  CONCLUSIONS

     The above comments are a brief description of some of the variables
not always  considered in the statistics and monitoring of air quality
data.   We will probably never rid ourselves of these difficulties.  How-
ever,  some  suggestions can be made.  The most important is that one
should gain a full knowledge of the local environs, both through visita-
tion and photographs and obtain first hand evidence of possible local
influences.  Portable sensor documentation such as presented in Figure 1
is invaluable.  Prevention is always the better quality.  Thus when de-
ploying sensors for such studies one should document the area in question,
placing the sensors so as to best represent the people exposure.  Finally,
any conclusion of the health effects and standards for air quality data
should include some description of the sensors, location and exposure
with respect to the organisms.  A case in point is the current guideline
for siting of carbon monoxide sensors where the sensor can vary from the
curb up to twenty feet from the curb, 15 feet high!  Thus, unless the re-
searcher knows fully the variation in concentrations between these two
extremes as related to the people in that area, the best standards be-
come absurdities relative to the variation in concentration possible
through inadequate siting criteria and guidelines.
 1.  Harrison, P.R., e_t al^, "identification  and  Impact of Chicago's
         Ambient Suspended  Dust1/ unpublished, June  1974.

-------
                                1749
    EVALUATION OF A RAPID INDEX OF ENVIRONMENTAL POLLUTION
       BY TOBACCO SMOKE AND STUDY OF THE COMPOSITION OF
       THE MAIN-STREAM AND SIDE-STREAM CIGARETTE SMOKE

           G, SCASSELLATI SFORZOLINI AND A, SAVING
Institute of Hygiene, Medical Faculty, University of Perugia,
Italy
ABSTRACT

     The Authors refer the results of the researches direct to
study the composition of both the main-stream and side-stream
cigarette smoke.

     The constituents detected are: tar, nicotine, CO, CO,,, NOr,,
                                                         u    A
NH+, and some polycyclic aromatic hydrocarbons: Pyrene, Anthracene^
3,4-Benzpyrene.

     From the results the Authors point out that the quantities
of tar and nicotine found in the side-stream are greater than
those present in the main-stream smoke, with the only exception
for the nicotine in the B) without filter cigarettes.   In all
other cases the ratio between the side-stream and main-stream
smoke is: about 2 for the tar in no-filter cigarettes and from
about 2 to about 6 for filter cigarettes; for the nicotine
about 2 for no-filter and from about 2 to about 4.5 for filter
cigarettes,

     With regard to the gas-phase, the different gases are
differently distributed in the two portions, but, with the
exception of CO, they are also found in greater measure in the
side-stream than in the main-stream smoke.

-------
                              1750
     The Authors note that the polycyclic aromatic hydrocarbons
are also formed during the period of non-inhalation and that
the quantity of 3,4-Benzpyrene found in the side-stream corres-
ponds to more than 2,5 times that present in the main-stream
smoke.

     An index useful for a rapid evaluation of environmental
pollution by cigarette smoke has been worked out and CO was
chosen as a significant index, because: it is always present in
very high quantities in room-air where people smoke; in these
ambients, if they are rationally heated) it is only produced "by
tobacco smoke; it may be evaluated easily, rapidly and exactly
by "Draeger System".

     From CO, using formulas proposed by the Authors, it can
work out other indexes, as the 3,4-Benzpyrene, and calculate
also the equivalent number of cigarettes a non-smoker would in
theory have passively smoked breathing for a some time the
smoky air of social meeting rooms.

-------
                                  1751
1.   Introduction

     The current literature concerning the toxic and carcinogenic
substances present in the inhaled portion of cigarette smoke is very
rich and exhaustive; on the contrary the researches directed to study
the composition of the cigarette smoke dispersed in room-air, in order
to evaluate the environmental pollution by tobacco smoke and to assess
the danger to which no-smokers are exposed in rooms where other people
are smoking, are few and fragmentary.

     The papers by Vynder [1], lotin and Falk [2], Galuskinova  [3]»
Heurath et al. [4], Ferri and Baratta[5]i on the distribution of some
compounds in the different portions of cigarette smoke, should however
be mentioned,

2.   Experimental
     For this purpose, we have carried out different researches, already
published (Scassellati Sforzolini and Saldi [6], Pascasio et al.[7],
Scassellati Sforzolini et al.[8], Savino and Scassellati Sforzolini  [9],
Scassellati Sforzolini and Savino  [10]), that we summarize here*
     In these studies we used apparatus able to smoke cigarettes
reproducing human smoking (according to the standard parameters) and to
collect separately and contemporarily both the main-stream and side-
stream cigarette smoke (Fig • 1 and 2).
2.1  Detection of tarf nicotine and some gases in the main-stream and
     side-stream cigarette smoke.
     The determinations were carried out on different kinds of
cigarettes, with and without filter, and the results shown on Table I.
     From the results it is seen that the amounts of tar and nicotine
found in the ambiental smoke are greater than those present in the smoke
inhaled by the smoker; the only exception is for the nicotine of B)
no-filter cigarettes. In all other cases the ratio between side-stream
and main-stream smoke 1st about 2 for the tar in no-filter cigarettes
and from about 2 to about 6 for filter cigarettes; for the nicotine
equally about 2 for no-filter and from about 2 to about 4.5 for filter
cigarettes. With regard to the gas-phase, the different gases are
differently distributed in the two portions, but, with the exception of
CO, these also are found in greater quantities in the ambiental portion
of smoke.
2.2  Detection of some polycyclic aromatic hydrocarbons (P»A.H.) in the
     main-stream and side-stream cigarette smoke*
     The determinations were made on a single kind of cigarette without
filter.
     On Table II the distribution of some P.A.H., taken as indexes, in
the two portions of smoke is represented.
     From the values shown on the table, it can be seen that these

-------
                               Table  I
Distribution of tar, nicotine and some gases in the mainstream and side-stream
cigarette smoke (mg/cigarette).
rind of
cigarette
A) No-filter
A) Filter
B) No-filter
B) Filter
C) No-filter
D) Double-
filter
Portion
of smoke
Main-stream
Side-stream
Main-stream.
Side-stream
Main-stream
Side-stream
Main-stream
Side-stream
Main-stream
Side-stream
Main-stream
Side-stream
Tar
17.7250
38.4900
8.3650
33.6800
18.8193
38.8447
11.9580
35.3297
26.0481
54.9800
4.3253
27.1000
Nicotine
0.7915
2.0660
0.4564
1.7550
0.9531
0.8695
0.4608
0.7992
1.0337
2.1273
0.1451
0.6524
CO
23.1100
4.9425
22.2200
2.9673
23.4100
3.2950
21.8000
4.9425
18.9000
3.2950
12.0000
2.6360
°°2
55.4200
56.9884
72.5300
98.4344
56.0100
93.2537
72.5000
56.9884
79.2100
88.0729
49.5000
93.2537
NO + N02
detected
as N02
0.0130
0.0550
0.0120
0.0770
0.0150
0.0550
0.0120
0.0770
0.0160
0.0440
0.0080
0.0770
NHj
0.1375
8.0000
0.1000
7.5000
0.1875
7.5000
0.1500
7.0000
0.1687
6.6500
0.0500
5.5000
                                                                                               Oi
                                                                                               t-0

-------
                           1753
                                           1  1

Fig. 1.  Apparatus able to smoke and to collect separately
         both the main-stream and sida-etream cigarette
         smoke:   detection of tar and nicotine.
                     Table II

  Distribution of some P.A.M. in the  main-stream and
  side-st earn of cigarette smoke.

PAH

Anthracene
Pyrene
3,^-Benzpyrene
jug /100 cigarettes

main-stream
^.75
4.56
2.5<+
side-stream
5.^0
1.77
6.51

-------
                                  1754
            Fig. 2.  Apparatus able to smoke and to collect
                     separately both the main-stream and  side-
                     stream cigarette smoke:  detection of
                     gases  and P.A.H.
hydrocarbons are formed during the period of non— inhalation and that,
with regard to 3,4-Benzpyrene, the amount found in the side-stream in
more than 2,5 times that present in the main-stream smoke of the same
king of cigarettes.
2.3. tvaluation of a rapid and significant index of enriyonmental
     pollution by cigarette smote*
     From what is reported in 2.1 and in 2.2, it can be deduced that
the carbon monoxide may be taken as a significant index for a rapid
evaluation of environmental pollution by tobacco smoke, becausei it is
always present in very high quantities in room-air where people smoke;
in these ambients, if they are rationally heated, it is only produced
by tobacco smoke; it may be evaluated easily, rapidly and exactly by
the "Draeger System".
     From the ppm of CO measured  in the room-air, it can calculate
other pollutants present, as  the  3,4-Benzpyrene  (3,4-BP).  In fact

-------
                                   1755
transforming the ppm of CO to mg/m*f making the ratio between the
quantities of 3,4-BF and those of CO present in the side-stream smoke
(see tables) and considering that the rate of air inhaled by a man
under normal conditions is about $00 1/K (Bell et al. [11]), it can
calculate the amount of 3,4-BP taken in 1 hour and also the equivalent
number of cigarettes a non-smoker would in theory have passively smoked
breathing for 1 hour the smoky air of that place, using the following
formulas :
  Atg of 3, 4-BF inhaledA   •   PP* CO • A « B • C              (1)

            CO . to       28 010  .  27V.  m 1 1644
w ere: A -  -------  -  (273^20)""T"22"414  "  '" '
          X  • /ig of 3,4-BP in the side-stream smoke of  1 cigarette
          Y  -  mg of CO      "          "        "       »    »

       £ •  air inhaled under normal conditions •  0,500 m^A

  equivalent n°of cigarettes              ARC
   passively smoked by a      « ppm CO . -w---*---*?-           (2)
   non-smoker                                 ~"

where: D  • yug of 3,4-BP in main-stream smoke of 1 cigarette

 e.g.; if in the room-air there are 10 ppm of CO (an amount very
       frequently found in places where people smoke) f we have:

  tig of 3.4-BP inhaledA   -   10 • 1.1644 • 0.0177 • 0.5  -  0.1030

  equivalent n°of cigarettes
   passively smoked by a non-smoker  •  10 . -«"];»cT"  *
                         References

    VYHDIR, B.L., "The biologic effects of  tobacco".  Little  Brown  and
    Company, Boston  1955,  p.  17.

    IOTIH, P., FALr, H.L., "The role and action of environmental agents
    In the pathogenesis of luag cancer. II. Cigarette smoke", Cancer.
    13, 250 (1960).

    GALUSriBOVA, V., "3,4-benxpyrene determination in the smoky
    atmosphere of social meeting rooms and restaurants. A contribution
    to the problem of the noxiousness of so-called passive smoking",
    Keoplasma. 11, 465 (1964).

-------
                                    1756


 4    NBURATH (von), G., BHKIB, H., und HOBSTMANN, H., "Influence of
      moisture content of cigarettes on the composition of smoke. III."
      Beitrage zur Tabatforschung, 2, 361 (1964).

 5    FBRRI, B.S., BAEATTA, B.J., "Polonium 210 in tobacco, cigarette
      smoke and selected human organs", Public Health Reports. 81, 121
      (1966).

 6    SCASSBLLATI SFOK20LINX, Q., SAIDI, G.t "Ulterior! ricerche sugli
      idrocarburi policiclici del fumo di sigaretta. Confronto fra il
      fumo aspirate e quello raccolto nell'aria ambiente",, Boll.Soc.
      Ital.Biol.Spei.. 37, 769 (1961).

 7    PASCAS10, P., SCASSBLLATI SFORZOLINI, G., SAVINO, A., "Catrame e
      nicotina nella porzione aspirata e nella porzione ambientale del
      fumo di vari tipi dl sigarette", Ann.San.Pubbl.. 27, 971 (1966).

 8    SCASSBLLATI SFORZOLINI, G., PASCASIO, P., MASTSATOREA, V., SAVI
      NO, A., "Attiviti cancerigena del fumo dl sigaretta. Quantifica-
      zione degli idrocarburi aromatici policiclici present! nella por
      zione aspirata", Hiv.Ital.Ig.. 27» 175 (1967).                 ~~
 9    SCASSBLLATI SFORZOLINI, G., SAVING,  A.,  "Valutazione di un indi-
      ce rapido di contaminazione ambientale da fumo di sigaretta, in
      relazione alia composizione della fase gassosa del fumo", Rivf
      Ital.Ig.. 28, 43 (1968).

10    SAVING, A., SCASSBLLATI SFORZOLINI,  G.,  "Determinazione del
      3,4»benzopirene nella porzione ambientale del fumo di sigaretta
      con metodica combinata di gascromatografia, cromatografia su
      carta e spettrofotomctria", Riv.Ital.Ig,« 29-30, 265 (1969/
      1970).
11    BBLL, G.H., DAVIDSON, J.N., SCARBOEOICH, H., "Trattato di Fi-
      siologia e Chimica Fisiologica", Parte II, p. 19, Bd.Scienti-
      fic he Italiane, Napoli 1958

-------
                              1757
           BEEINFLUSSUNG DES KOHLENMONOXIDSPIEGELS
              DURCH LUFTUNGSTECHNISCHE ANLAGEN

            H, RUDEN, K, BOTZENHART UND H, LANGER

Hygiene-Institut der Universitat Bonn, BRD
KURZFASSUNG

     Es -Let bekannt, dass in lilftungsteohnischen Anlagen parti-
kulave Substanzen durch Filter abgesohieden werden ktinnen. Hin-
sichtlich der gasfOrmigen Substanzen ist dies bislang ungekldrt,
uenn man von grossteehn-ischen Anlagen  (z.B. SO^-Abeoheidung in
Rauchga8Da~schevn) absieht.   So wuvden f1lv Kohlenmonoxidj das
in Hauptverkehrastrassen den gesundheitsgeftthrdenden Bereiah
oft evveiohtj indoor-outdoor-Mesaungen vorgenommen.

     Die CO-Bestimmungen wurden vor und hinter  lliftungstechni-
sohen Anlagen duvehgefilhrt, von denen  die erste nux> aua einem
Gebldse ohne Filterung  (Abluftanlage einer Tiefgarage} und die
zveite aus Faser-Filtern, Aktivkohlefilter, Dampfbefeuahtung
und weiteren notwendigen Aggregaten (Klimaanlage in einem Bllro-
gebaude) bestand.

     Die Messungen  haben gezeigt, dasa Kohlenmonoxid in den
Anlagen nioht abgesahieden wird*  so dass in Ubereinstimmung mit
anderen Autoren  die Aussenluftkonzentrationen in gleiahev H&he
aueh im Innenraum erwartet werden mttssen.

     Nenn daher  bei vorgegebener  hoher Immissionsbelastung eine
Klimaanlage  installiert wirdf muse eie so gestaltet werdent dass
der Ansaugstutzen filr die Aussenluft in grOssere HOhen odev in
andere bodennahe Bereiche gelegt  wivdf die weniger Kohlenmonoxid
enthalten.

-------
                              1758
ABSTRACT

     It is well known that particles can be filtered out by aiv
conditioning plants, but with regard to gases the situation is
still uncertain, except in the case of large plants (for example
the extraction of SO,, by flue-gas scrubbers). Accordinglyt indoor
                    &
and outdoor measurements were made of carbon monoxide, which on
main roads often reaches levels dangerous to health. Measurements
were carried out at the intake and exhaust sides of two venti-
lating plants, one of which (the extraction ventilator of an
underground garage) was simply a fan without a filter, while the
other (the air-conditioning plant of an office building) compri.aec
fibre filters, an activated charcoal filter, steam humidification
and other necessary units,

     The measurements showed that carbon monoxide was not removed
by air-conditioning plants, so that the concentration indoors
must be expected to be as high as the concentration outdoors.
This is in line with the views of other authors.

     Hence, if an air-conditioning plant is installed where
there is a high exhaust gas level, the external air intake must
be located at a high point or, if at a low level, at a point
where the carbon monoxide content is lower.

-------
                              1759
1. Einleitung
In den letzten Jahren wurden zahlreiche GebHude erstellt, de-
ren InnenrHume oft tell- oder vollklimatisiert sind. Die Not-
wendigkeit einer Klimatisierung wurde nicht nur unter dem Ge-
sichtspunkt einer ausreichenden Ventilation gesehen, die bei
GroBrSumen durch Offnen der Fenster nicht erraicht werden
kann, sondern sie wird auch gefordert in der Annahme, dafi die
Luft in klimatisierten RSumen eine bessere Oualitat aufweise
als die Aufienluft. Hinsichtlich der Abscheidung von Feststof-
fen durch Faserfilter verschiedener Gtiteklassen trifft diese
Pramisse zu. Ob sich gasformige Substanzen auch in den Aggre-
gaten lliftungstechnischer Anlagen abscheiden, ist bislang
nicht ausreichend geklSrt (1,4,5,6), so auch nicht fiir den
Hauptbestandteil der Kfz.-bedingten Luftverunreinigungen, das
Kohlenmonoxid, das den MIK-Wert in verkehrsbelasteten Regionen
erreicht (2,3). Da klimatisierte GebSude hSufig an Hauptver-
kehrsstrafien mit hoher Inunissionsbelastung liegen, ist es da-
he r von groBem Interesse, den Gehalt der Luft an Kohlenmon-
oxid vor und nach ihrer Passage durch eine Klimaanlage zu
untersuchen. Zu diesem Zwecke wurden Untersuchungen an zwei
Anlagen durchgefiihrt, wobei nur ein Ausschnitt aus den durch-
gefiihrten Messungen dargestellt werden kann.
2. Versuchsmethodik
2.1. Tiefgarage
In einer Tiefgarage mit drei Stockwerken, die von Kraftfahr-
zeugen unterschiedlich frequentiert wurden, wurde die Abluft
durch ein etwa 3O m langes, aus Mauerwerk bestehendes Kanal-
netz tiber Dach eines nahegelegenen Hauses abgeftihrt. Die me-
chanische Entliiftungsanlage bestand nur aus einem GeblSse
ohne zusStzlicher Filterung.
Die MeBpunkte waren alternierend in drei Etagen der Tiefga-
rage sowie im Dereich hinter dem Abluftkanal auf dem Dach in
der frcien Atmosphere angeordnet.
Die CO-Messungen erfolgten mit DrSger-RcShrchen Kohleninonoxid
5/cf deren relative Standardabweichung 15 - 1O% betrSgt. Ent-
sprechend der Gcbrauchsanweisung wurden 1O HUbe gemacht. Die

-------
                              1760
gleichzeitig erhobenen MeBwerte aus den drei Decks der Tiefga-
rage wurden zu einem arithmetischen Mittelwert zusammengefaRt
und den Werten im Abluftschacht gegeniibergestellt.
2.2. Klimatisiertes Gebaude
Die AuBenluft wurde 6 m iiber Niveau einer HauptverkehrsstraBe
entnommen und in einem Kanal aus Zinkblech etwa 1O m abwarts
in ein InstallationsgeschoB geleitet, wo sie neben Erhitzer,
Kiihler und Dampfbefeuchter ein Filter der Klasse B2sowie ein
Aktivkohlefilter passierte. AnschlieBend wurde.die AuBenluft
mit Umluft aus den klimatisierten RSumen gemischt und nach er-
neuter Passage eines B2-Filters den Raumen zugefiihrt. Der
AuBenluftanteil betrug mindestens 1O%.
Die MeBwerte lagen 3 m vor dem Aktivkohlefilter im Zuluftka-
nal, unmittelbar hinter dem Aktivkohlefilter und in einem
klimatisierten Raum. Infolge der geringeren CO-Konzentration
wurden 20 Hiibe durchgefiihrt.
Da alle Messungen im Winter erfolgten, lag die AuBentempera-
tur unter 1O°C, die relative Feuchte betrug 5O bis 70%.
3. Ergebnisse:
Die folgenden Abbildungen zeigen nur einen Ausschnitt aus dem
Me Bprogramm.
Aus Abb.l geht hervor, daB die Schwankungen der CO-Konzentra-
tion in der Tiefgarage relativ groft sind und bis zu 25 ppm
iiber den Werten liegen, die am Ende des Abluftkanals gemessen
wurden. Bei den dort gemessenen Werten muB eine Vermischung
von AuBenluft mit der Tiefgaragenabluft berucksichtigt werden,
Eine Adsorption von CO an die Kanalinnenwande ist im Gegen-
satz zu SO2 (4) unwahrscheinlich.
Die Abb.2 7eigt, daB die CO-Werte vor und hinter dem Aktiv-
kohlefilter fast identisch sind, wenn man in die Beurteilung
die grofie Standardabweichung der Meflmethode mit einbezieht.
Die CO-Raumluft-Konzentrationen liegen haufig unter denen der
ubrigen Meflpunkte, da der Raum von Mischluft versorgt wird,
die zum iiberwiegenden Teil aus den klimatisierten Raumen
stammt.

-------
                                 1761
          Abb. 1 : CO - Messungen in einer Tiefgarage
 ppm CO

120

100

 80

 60

 40

 20
                    Meflpunkte
               	   in der Tiefgarage
                    : nach Ende des  Ab
                          1 uf t kanal s
                    *
           16°°
17C'
                            . 0
18

19°°   Zeit
        Abb. 2  : CO-Messungen in einem klimatisierten  Gebaude
                                          MeRpunkt e

                                     	  : vor  Aktiv kohtef i I te r
                                     	: nach Aktivkohlefilter
                                          : im klimatisierten Raum
                                                       DO

-------
                              1762
4. Schluflbetrachtung
Die Ergebnisse haben gezeigt, dafi Kohlenmonoxid durch GeblMse,
Mauerwerk, Dampfbefeuchtung, Faserfilter oder Aktivkohlefliter
keine wesentliche Venninderung erfMhrt. YOCOM u. Mitarbeiter
(6) konunen aufgrund eigener indoor-outdoor-Messungen von Koh-
lenmonoxid zu dem SchluB, daB die Innenluftkonzentrationen in
klimatisierten Raumen  (ohne die Art der Klimatisierung zu be-
zeichnen) die AuBenluftkonzentrationen von Kohlenmonoxid  ge-
nauer widerspiegeln als in natiirlich beliifteten Raumen.
Mangels einer leistungsfahigen Methode zur CO-Abscheidung
durch herkommliche luftungstechnische Anlagen 1st daher anzu-
raten, den Ansaugschacht klimatisierter Gebaude wegen der mit
zunehmender Hone abnehmenden CO-Konzentration (3) mindestens
4O m iiber StraBenniveau zu legen Oder in Bereiche auszu-
weichen, die weniger CO-belastet sind.

Literatur:
   Benson, F.B., J. J.EIenderson u. E.D.Caldwell: Indoor-Outdoor
                Air Pollution Relationships: Literature
                Review.
                Environmental Protection Agency
                Pat. AP-112 August 1972
   Kettner, H.: Zur Frage der Normierung physikalisch-
                chemischer Luftverunreinigungen in
                Innenraumen.
                Ges.-Ing. 94, 44 - 51  (1973)
   Lahmann, E.: Untersuchungen iiber Luf tverunreinigungen
                durch den Kraftverkehr.
                Schriftenreihe Wasser-,Boden- u.Lufthygiene
                Heft 28, 1-80  (1969)
   Riiden, I!.: Luftverunreinigung und Klimatisierung.
                Umweltmedizin 1, 132 - 136  (1973)
   Thompson, C.R.,E.G.Hensel u. G.Kats: Outdoor-Indoor
                levels of Six Air Pollutants.
                J.Air Poll.Contr.Ass. 23, 881 - 886  (1973)
   Yocom, J.E.,W.C.Clink u. W.A.Cote: Indoor/Outdoor Air-
                Quality Relationships.
                J.Air Poll.Contr.Ass. 21, 251 - 259  (1971)

-------
                              1763
             RECAPITULATION OF EXPOSURES AROUND A
                POINT SOURCE OF AIR POLLUTION

                         M, B, HERTZ

Human Studies Laboratory, National Environmental Research
Center, Research Triangle Park, North Carolina, USA
ABSTRACT

     The Salt Lake Valley offers a unique setting for observing
relationships between the health parameters of the Community
Health and Environmental Surveillance System (CHESS) and a
gradient of human exposure to sulfur dioxide. Four communities
in or near the Salt Lake Valley were chosen to represent an
exposure gradient to sulfur dioxide, emitted primarily from a
large copper smelter.   The study communities, Magna, Kearns,
Salt Lake City, and Ogden, are located approximately St B, 13
and 35 miles from the smelter respectively.

     Air monitoring sites were positioned approximately in the
center of each study population.   Total suspended particulates
and their sulfate and nitrate fractions, sulfur dioxide, and
nitrogen dioxide were monitored daily.   Dustfall concentrations
were measured monthly.

     Arithmetic mean sulfur dioxide and suspended sulfate con-
centrations showed that Magna had the highest levels with Kearns,
Salt Lake City and Ogden following in that order.

     Past sulfur dioxide and suspended sulfate exposures were
estimated from annual smelter emissions and currently measured
levels and indicate that past exposures were considerably higher
than present levels.

-------
                                    1764
1.   Introduction
     Previous epidemiological  studies, in urban areas,  investigating sulfur
dioxide effects have been confounded by mixed exposures to total  suspended
particulates and other pollutants.  In the Salt Lake Valley,  exposures  to
sulfur dioxide and suspended sulfates are primarily influenced by a  large
copper smelter west of the four selected study communities.   The  Utah
CHESS study areas are situated in the north central  part of the State of
Utah and include Ogden, Salt Lake City, Kearns and Magna.   These  communi-
ties were selected on the basis of meteorological  information, emission
inventories and previous air quality measurements  to provide  an exposure
gradient for sulfur dioxide with similar levels of other pollutants  below
the National Ambient Air Quality Standards.
2.   Pollutant Measurement Techniques
     For the Utah CHESS communities, the air monitoring sites  were located
within approximately 1,5-mile  radius of the center of the  human population
under study.
     Twenty-four-hour sulfur dioxide and nitrogen  dioxide  measurements
were obtained with standard gas bubblers using the modified West-Gaeke*- -•
                     P2l
and Jacobs-Hochheiser1 J techniques respectively.   High volume air
samplers collected 24-hour samples of suspended particulate matter.   For
the sulfate and nitrate fractions of suspended particulate,  turbidimetric
and reduction diazo coupling methods were used respectively.   The measure-
ment of atmospheric particulate dustfall collected in a dustfall  bucket
over a month-long period provided an approximate gauge  of  the  amount of
environmental contaminants that precipitated out of the atmosphere on to
the respective CHESS neighborhoods.
3.   Results and Discussion
     Table  I summarizes the Utah CHESS 1971 annual averages for all
pollutants.  The National Ambient Air Quality Standards are shown for
comparison.  Annual average sulfur dioxide and suspended sulfate  con-
centrations show that Magna had the highest levels with Kearns, Salt
Lake City and Ogden, following in that order.  Although Salt  Lake City,
Ogden and Kearns had similar average sulfur dioxide levels, Magna had
the highest daily peak values.  Daily sulfur dioxide peak  values  for
Kearns were considerably lower than Magna, but higher than either Salt
Lake City or Ogden.

-------
Table I.
                       1765

SUMMARY OF UTAH CHESS 1971  ANNUAL AVERAGES
       FOR POLLUTANT CONCENTRATIONS
'ollutant '
Sulfur dioxide,3 yg/rrf5
Suspended sulfate, ^
yg/m
Total suspended ^
particulate, pg/m
Nitrogen dioxide, ug/m
Suspended nitrate, yg/m
Oustfall particulate, a
2
g/m /mo
Ogden
8.0
4.8
66.5
37.4
2.0
4.34
Salt
Lake
City
15.3
6.1
70.9
40.6
2.4
3.56
Kearns
21.7
6.4
38.2
22.3
1.7
2.44
Magna
61.8
9.6
53.9
20.8
1.3
4.61
Standard
80.0
~
75.0
100.0
-
—
 a.   Annual  arithmetic mean values
 b.   National  Primary Ambient Air Quality Standard,
 c.   Annual  geometric mean values.
      The highest total suspended particulate concentration occurred in
 Salt Lake City followed by Oqden, Magna and Kearns in that order.
 Nitrogen dioxide and suspended nitrates exhibited similar gradients.
 Salt Lake City received the highest exposure while Ogden recorded slightly
 lower levels.  Kearns and Magna had similar but considerably lower con-
 centrations that the other two communities.
      It should be noted that annual mean concentrations of total suspended
 particulates and nitrogen dioxide were below the National Ambient Air
 Quality Standards for all communities.

 4.   Past Exposure Estimates
      Measured pollutant concentrations, of course, provide the most
 accurate data for use in epidemiologic studies.  In the absence of
 measured values required for long-term studies; however, it is sometimes
 necessary to estimate past exposure levels.

-------
                                    1766
Table II.  ESTIMATED ANNUAL ARITHMETIC AVERAGE POLLUTANT CONCENTRATIONS
           FOR UTAH CHESS COMMUNITIES, 1940-1971.
Pollutant
Sul fur
dioxide


Suspended
sul fates



Coirmunity
Low (Ogden)
Intermediate I
{Salt Lake City
Intermediate II
(Kearns)
High
(Magna)
Low
Intermediate I
Intermediate II
High
Concentration,3 yg/m
1940-49
—
—
(234)
—
—
—
(27.7)
1950-59
—
(50)
(142)
—
(11-4)
(10.4)
(19.5)
1960-69
(20)
(33)
(95)
( 4.8)
( 7.5)
( 8.8)
(15.3)
1971
8
15
22
62
5.6
7.3
7.8
12.4
 Parentheses indicate estimated values.

 Kearns, the Intermediate II community, was established in 1953.

     Past exposures to sulfur dioxide and suspended sulfates  were estimated
from sulfur dioxide smelter emissions dating back to 1940.  For each year
the ratio of that year's sulfur dioxide emissions to the emissions for the
year 1971 was calculated.  This ratio was then multiplied by the 1971
annual averages for sulfur dioxide and suspended sulfate to give crude
yearly past exposure estimates for sulfur dioxide and suspended sulfate
levels respectively.  These results are shown grouped by decades in
Table II.

-------
                                    1767
References


1.    WEST, P.  W.  and G.  C.  GAEKE,  Fixation  of Sulfur  Dioxide  as
     Sulfitomercurate III  and Subsequent Colorimetric Determination.
     Anal. Chem.  28; 1816-1819 (1956).

2.    JACOBS, M.  B.  and S.  HOCHHEISER.   Continuous  Sampling  and Ultramicro
     Determination  of Nitrogen Dioxide  in Air. Anal.  Chem.  30. 426-428
     (1958).

-------
                              1769
    RECAPITULATION OF METROPOLITAN AIR POLLUTION EXPOSURES
          ARISING FROM MOBILE AND STATIONARY SOURCES

                         D, 0, HINTON

Human Studies Laboratory, National Environmental Research Cen-
ter, Research Triangle Park, NC, USA
ABSTRACT

     Measurements of previous air pollution exposures and con-
tinuing air monitoring surveillance in a study community provide
an exposure basis for epidemiologia studies on the health ef-
fects of ambient pollution.   Sulfur dioxide, total suspended
particulate and dustfall measurements for the oity of Chicago,
Illinois, and for the Chicago-Northwest Indiana Metropolitan
area have been obtained in recent years.   Based on these meas-
ured Values, it was possible to estimate past pollutant exposure
in these areas for the period 1950-19?0(  .

     Limited data were available during the earlier years of the
1950-1970 period.   A consistent correlation was found to exist
between dustfall and total suspended partioulates, and dustfall
and suspended sulfates for those years when these measurements
were available.   Using this relationship total suspended parti-
culate and suspended sulfate exposures were estimated from
dustfall for those years when only dustfall was available.
Sulfur dioxide was not correlatable to dustfallj  however, lim-
ited data provided exposure estimates for 1960-1970.

     The estimates were useful in determining an exposure gra-
dient within the oity of Chicago and across the entire area.

-------
                                   1770
1.   Introduction
     In order to carry out epidemiologic studies on the health effects
of ambient pollution, it is important to have measurements of previous
air pollution exposures for the community under study as well  as con-
tinuing air monitoring surveillance.  An epidemiologic study"-  -" conducted
to determine chronic respiratory disease effects of combined exposure to
particulates and sulfur oxides, required exposure estimates for two dec-
ades prior to 1970.  Data were available from which to determine current
and recent exposure; however, data between 1950 and 1960 were  very limited.
A method for providing exposure estimates during this time period, there-
fore, was required.
2.   Pollution Sources
     Mobile pollution sources include automobiles, trucks and  airplanes.
Stationary pollution sources include primarily power generating plants,
residential, commercial and various industrial sources without dominance
from any one specific type of product.  Combined pollution effluents from
these sources are continually influenced by changing climatological
conditions creating complex exposures to area residents.
3.  Exposure Methodology
     A data search which included city, county, state and Federal
sources was conducted.  Air sampling methods and sampling frequency
were analyzed for consistency.  Graphical trends of individual  pollutants
were determined.  Pollutant concentrations before and after implementation
of control devices were examined.  Meteorological models using available
inputs (wind speed and direction, topography, source emission, source
[stack] height, etc.) were compared with measured data.  One model"- ^
available for the Chicago area was useful in determining sulfur dioxide
distribution in the area from 1960-1970.
     The above procedures were useful in determining estimates of exposure
during the latter decade.  Air monitoring data for the 1950-60 period
were very limited.  Except for annual particulate dustfall in  Chicago,
many data points were missing.  Relationships between pollutants were com-
puted from more recent years to provide reasonable estimates for missing
data points.  The annual arithmetic mean for Total Suspended Particulates
(TSP) and the annual arithmetic mean for Dustfall were examined over time.
Both revealed a similar trend of decreasing concentrations with time.
A least squares fit of TSP (Y) versus Dustfall (X) gave the linear

-------
                                    1771
regression equation:   Y= -49.5 + 14.4X and provided  a  correlation
of r = 0.747.   This approach indicated that Dustfall  is  a  reasonable
basis for estimating TSP; however,  it did not account  for  the  ratios
changing with  time.  A second graph with TSP divided  by  Dustfall  (Y)
plotted against the corresponding year (X) was made  which  provided  the
linear regression equation:   Y = 12.4 - .134X and  r  =  .336.  This ratio
multiplied by  the annual dustfall value was used for estimating  annual
TSP concentrations for 1950-1953.
     Repeating this approach for suspended sulfates  versus dustfall gave
a least squares fit of Y = 3.40 + 1.276X with a correlation  of r =  .738.
A plot of suspended sulfates/dustfall versus time  yielded  Y  =  1.16  +  .008X
with r = .235.  This technique was  used for estimating suspended sulfates
for 1950-1953.
     A similar comparison between sulfur dioxide and  dustfall  data  showed
no consistent  relationship.   Data from a conductometric  instrument  was
used to provide exposure estimates  for 1960-1970.
4.  Conclusions
     Measured  pollution levels provide the most accurate estimate of
pollutant exposure in an area immediately surrounding  the  monitor.
Emission inventories from point sources and traffic  count  of mobile
source offer additional information from which to  base meteorological
models which can be used for larger area coverage.
     In the absence of measured data for a given pollutant,  weighted
consideration  can be given to major source emission  levels,  traffic
changes, meteorological models and  relationships between pollutants to
provide past exposure estimates. Although the limitation  of subjective
analysis and extrapolating to earlier years is recognized, this  provided
a crude estimate of previous exposures in the Chicago  area where measured
data was unavailable.  The exposure estimates indicated  a  definite
pollution exposure gradient highest in east central  Chicago  and  decreased
radially toward the city boundary.   Overall exposure in  the  suburbs was
generally lower than exposure within the city, with  the  exception of
sulfur dioxide.  In some instances, sulfur dioxide exposure  in the
suburbs equaled or exceeded the city exposure, possibly  as a result of
point sources.  Low exposures for non-urban areas, attributed  to fewer
sources per unit area, were verified.

-------
                                    1772
References

1.  HINTON, D. 0., et al.   Human Exposure to Air Pollutants  In the
    Chicago-Northwest Indiana~Retropo1i'tan Region.  1950-19TTTnT
    Health Consequences of Sulfur Oxides:  A Report from CHESS,
    1970-71.  U. S. Environmental Protection Agency.   Research Triangle
    Park, North Carolina.   Publication No. EPA-65011-74-000,  (1974).

2.  FINKLEA, J. F., et al.   Prevalence of Chronic Respiratory  Disease
    Symptoms in Military Recruits":""  Chicago induction'Center,  1969-1970.
    Tn:Health Consequences of Sulfur OxTde~s~i   A Keport from  UHLSS,
    1970-71.  U. S. Environmental Protection Agency.   Research Triangle
    Park, North Carolina.   Publication No. EPA-65011-74-000,(1974).

3.  GOLDEN, J., The Air Quality Display Model Applied  to Cook  County.
    Illinois.   The Mitre Corporation.   Wa'snTngton,  D.  C.  MKl-4148.
    (July 1970).  (Document has not  been approved for  distribution.)

-------
                              1773
         THE PROBLEM OF PASSIVE SMOKING IN MOTOR CARS

                          H.-P, HARKE

Research Institute of the Cigarette Industry, Hamburg, FRG
ABSTRACT

     Interest in measurement of concentrations of tobacco smoke
in closed rooms has been stimulated by a growing concern over
air pollution.   Since 1969 a number of investigators have re-
ported on measurements of tobacco smoke constituentsj especially
of the toxicological most important CO and nicotine.   Measure-
ments of the absorption of these compounds by nonsmokers were
also reported.   The situation in motor cars has not been inten-
sively investigated so far.

     Therefore, we carried out CO measurements in a motor car of
the lower middle-class which was occupied with 3 smokers and one
nonsmoker.    Experiments were carried out in a wind-tunnel.
Doors and windows of the car had been closed and the motor was
off.   At a relative-wind velocity of 0 and 50 km/h ventilation
was varied.   In one series of experiments the 3 smokers smoked
a cigarette simultaneously following intervals, in total 6 or 9
cigarettes;  in another series of experiments one smoker after
another smoked a cigarette, in total either 4 or 6 cigarettes.

     These experiments were completed by CO measurements in
driving motor cars of different types.   During a running time
of 20 min.  on a fixed route through Hamburg 2x2 cigarettes
were smoked under conditions of ventilation as in the wind-tun-
nel experiments.

-------
                               1774

     Results of our investigations may be summarized as follows:

1)    There was an increase of the CO level during smoking of cig-
     arettes in a driving oar.
2)    This CO-level decreased in smokeless periods to ambient CO
     level within a few minutes.
3)    Smoking without interruption led to an equilibrium concen-
     tration but not to a constant cumulation of smoke constitu-
     ents in the air.
4)    Absolute concentrations were depending on ventilation.
5)    Only in a stationaryt unventilated car the CO level constant-
     ly increased when smoking during the experimental period.
     However, this seems to be an unrealistic situation.

-------
                                     1775
 Introduction
 SRCH  (1)  observed that  smoking of  10  cigarettes  to  a  5 aim butt  length  in
 an  enclosed car standing in  a closed  garage  caused  CO levels  up to  90  ppm.
 These cigarettes were smoked by  2  smokers within one  hour. This study  was
 done  to clear  up a murder. SEIFF (2)  published results of investigations
 carried out in a stationary  bus  with  running blower.  Under "worst-case"
 conditions  (all passenger seats  occupied with smokers smoking 50JS of the
 time, i.e.  a 100J? load  factor for  the bus with 1*3 seats  - this  situation
 was simulated  by having one  cigarette burning at each two seating
 positions all  the time)  he found a stable CO level  of 33 ppm  with an
 ambient level  of 7 ppm  about half  an  hour after  the beginning of experiment.
 Under more  realistic conditions  (5 cigarettes burning all the time,
 simulating  that 2Q% of  the bus seats  were occupied  with  smokers smoking
 50% of the  time) the CO-reading  stabilized at 18 ppm  after U2 min.  Ambient
 CO  level was found to be 13  ppm.

 The objectives of our investigations  were to study  the influence of
 smoking and other circumstances, such as ventilation, air stream and
 ambient CO  level, upon  the CO concentration  in motor  cars (3,M.
 Msthod
 Experiments in the wind-tunnel*were  carried out in a U-door  "European"
 motor car (55  PS) at a  relative-wind  velocity of 0  and 50 km/h.  Windows
 and doors had  been closed. Ventilation was varied as  follows: Ventilation 1
 = air jets  closed, blower off. Ventilation 2 = air  jets  opened, blower off.
 Ventilation 3  = air jets opened, blower  on.  The  car was  occupied with  U
 persons - only the driver was a  nonsmoker. Smokers  smoked either
 simultaneously at intervals  or one after another. CO  was measured with non
 dispersive  IR  (Harmann  & Braun,  Germany).

 Experiments in driving  motor cars  were carried out  in 9  different types of
 cars on a route of 15 km through Hamburg. Conditions  of  ventilation were
 the same as in the wind-tunnel experiments.  During  the course of experiment
 air samples were taken  inside the  motor  car  at 13 measuring points.
 Samples were analyzed immediately  after  return to our institute. CO was
 measured with  non dispersive IR  (Maihak, Germany).
*carried out  in  collaboration with  W.  Liedl  and D.  Denker,  Forschungsinsti-
 tut fur Kraffahrwesen  und Fahrzeugmotoren,  University of Stuttgart,  Germany

-------
                                    1776
Table I: Wind-tunnel experiments: Mean CO level in a motor car occupied
with 3 smokers and 1 nonsmoker. The smokers smoked simultaneously at
intervals or one after another.
Speed
on/h



0
0
50
50
50



0
0
50
50
50
Ventilation




1
2
1
2
3



1
2
1
2
3
Ambient
Air
ppm CO


1
2
0
3
2



1.5
2.5
3.5
1.0
3.0
Mean CO level in the car (ppm)
3 cig



Itl
9
25
22
11
pause
20 min


50
2
2
3
2
3 cig



75
8
23
19
8
pause
10 man


80
-
-
—
—
3 cig



96
-
27
-
-
Mean CO level in the car after
smoking
1 cig
15
6
114
6
5
2 cig
33
6
13
8
5
3 cig
U5
1
13
10
5
It cig
57
8
13
8
6
5 cig
67
-
Hi
-
—
6 cig
7^
-
13
-
-
Decrease
of CO
level
to after
ppm min
100 2.5
2 2
5 6
3 1..5
2 1*



62 11.5
5 1.5
5 1*.5
2 2
6 l
 Table  II:  Mean  CO  level  in motor  cars  occupied with 2  smokers  and 2 non-
 smokers.  With smoking  =  9  cars  at ventilation 1  and 2  and  5  cars at
 ventilation  3 were tested. Without smoking  = 23  rides. Between point 2  and
 3 and  between point 8  and  9  cigarettes were lightened. Smoking was stopped
 near to point 6 and between  point 11  and 12.
Measuring
point


1
2
3
It
5
6
7
8
9
10
11
12
13 (1)
Distance
km


0
0.95
1.95
2.1tO
l*.6o
7.10
8.10
9-90
11.50
12.30
13.W)
lit. 80
15.35
Mean speed
km/h


0
33
39
17
77
88
U7
1*2
1*6
33
1*0
35
22
Mean CO level (ppm)
non-
smoking

0.8
3.2
3.8
6.0
2.3
1.1
1.3
3.6
5.6
5.3
U.8
2.5
U.5
smoking
ventilation
1
0.5
1.7
11.3
17-9
21.1
7,8
u.o
^ 5
11.3
2l*.3
21.lt
9.3
7.6
2
0.8
2.5
12.0
15-9
9.9
3.9
2.0
2.6
7-9
18. H
15-7
6.9
5-7
3
1.0
2.3
11.1+
13.6
6.0
3.3
1.5
2.3
8.0
18.9
12.0
5.6
3.9

-------
                                     1777
 The  route  through Hamburg  included  autobahn, main  roads  and by-streets.
 Two  rides  were  carried  out  in  the morning  after  rush-hour. During  the  first
 ride no  cigarettes were smoked, during the  second  ride 2 of the  1*
 passengers smoked simultaneously 2x2 cigarettes. As blowers are  installed
 in some  motor cars only, it was not possible to  carry out measurements at
 ventilation  3,  i.e. air jets opened and blower on.

 Results  and  Discussion
 Results  are  summarized  in table I and II. Experiments in the wind-tunnel
 demonstrated a  quick increase of the CO concentration during the smoking
 period as well  as a quick decrease of CO concentration after smoking 3
 cigarettes simultaneously at intervals. The absolute CO level was
 depending on the ventilation. At a relative-wind velocity of 50 km/h
 ~25 ppm were measured  in an unventilated car and -10 ppm in a well-
 ventilated car.  Only in an unventilated car at a relative-wind velocity of
 0 km/h no sufficient decrease of the CO level was observed in the  smoking
 intervals, and  consequently the CO level increased during the second and
 third smoking period.  Only under these circumstances the Threshold Limit
 Values were exceeded.
 Corresponding results were obtained in experiments when one cigarette was
 smoked after another.  Only in an unventilated car at a relative-wind
 velocity of  0 km/h there was an increase of the CO level after each
 cigarette smoked. In all other series of experiment a stable CO level was
 observed already during smoking the first cigarette.
 Measurements in driving motor cars showed that our experimental results
 from one single motor car may be transferred to other cars. Just as in the
 wind-tunnel  experiments the CO level in the car increased during the
 smoking period  and decreased very quickly after smoking. When smoking the
 first two cigarettes, the car was running relatively fast.  Accordingly,
 there was a  better "natural" ventilation and the absolute CO value was a
 little lower than smoking the other two cigarettes.

   REFERENCES
1  SRCH,  M.:  Dtsch.  Z.  gerichtl.  Med.  60,  80-89  (1967)
2  SEIFF, H.E.:  US Dept. Transport.  Techn.  Rep.  BMCS-IHS-73-1,  April 1973
 3  HARKE, H.-P., LIEDL, W., DENKER,  D.:  Intern.  Arch.  Arbeitsmed. in press

 1+  HARKE, H.-P., PETERS, H. : Intern. Arch.  Arbeitsmed.  in press

-------
                               1779
       ANAESTHETIC VAPOUR POLLUTION IN OPERATING THEATRE
                          ATMOSPHERES

          R, S, BARRATT*, R, BELCHER+, W, I. STEPHEN*
                     AND J, M, THOMPSON4"*"

+  Chemistry Department, Birmingham University, Edgbaston,
   United Kingdom
++ Anaesthetics Department, Queen Elizabeth Hospital, Birmingham,
   United Kingdom
ABSTRACT

     This paper describes part of a comprehensive investigation
to examine the exposure of operating theatre staff to pollution
of the atmosphere by halothane, triohloroethylene and methoxy-
flurane vapours as well as nitrous oxide.

     The extent of atmospheric pollution in the operating theatre
is determined by a continualf  electron capture gas chromatogra-
phic system in which air samples from discrete predetermined
locations in the theatre are analysed automatically at regular
intervals.   The influence of ventilation and scavenging systems
on the pollution levels are important parameters.   The advan-
tages of a diffusion cell for the preparation of standard mix-
tures of anaesthetic vapours in air make this a useful technique
for research in anaesthetics.

     The results of some initial investigations are presented.

-------
                                   1780
1.  Introduction
     There is no doubt that pollution of the atmosphere in an operating
theatre occurs through the discharge of anaesthetic vapours.   It is
important to know the extent of this pollution so that any harmful effects
of chronic exposure can be examined and ways of reducing the  contaimina-
tion can be investigated.  However, there are many ways of measuring the
concentrations of the vapours and of interpreting the results.
     Several workers have taken air samples manually into syringes or
plastic bags, but this approach can cause errors by leakage or adsorption
of samples on surfaces.  Nevertheless, the concept of measuring the
distribution of the pollutants at different points in the theatre by
'snatch' sampling is useful.  It does seem meaningless, however, simply
to produce an average value of these results, particularly when a broad
distribution of concentrations is found.  Trapping the components in a
large volume of air, followed by their analysis, is a more efficient way
of obtaining an average concentration.
     We consider that it is important to examine the distribution of
anaesthetic vapours as well as the exposure levels of selected personnel
in the environment.  For this purpose the analysis of discrete samples
is essential.
2.  Techniques
    2.1  Analysis
     Gas chromatography is very useful for the separation and analysis of
anaesthetics, but major problems concern the sampling technique and the
presentation of the sample to the instrument for analysis.  The system
used in this study incorporates a multi-stream arrangement in which a
pump draws samples from various sites within the operating theatre
through nylon  tubes to a gas sampling valve and then to a gas chromato-
graph  (Hewlett Packard Model 5713).  Each sample point can be selected
automatically in turn for analysis to allow repeated, discrete sampling
over long periods.  This approach gives a reliable estimate of the
distribution of anaesthetic vapours.  For environments where halothane
and trichloroethylene occurred together, a separation was obtained on a
glass column (4.6m x 0.6cm o.d.) packed with silicone fluid,  DC560, on
Universal support  (80-100 mesh).  The carrier gas was Argon/methane (5Z)
flowing at 60 ml min   through the column maintained at 90°C,  For the

-------
                                    1781
ultimate sensitivity to the halogenated agents, a pulsed electron capture
detector is ideal; the linearity and stability of response are completely
compatible with an automatic analysis system.  Sampling through nylon
tubes caused no losses by adsorption and the technique has the advantage
that the tubes can be attached to personnel in order to monitor their
immediate environments.  The whole system is mobile to facilitate
transport to operating theatres for on-site analysis.
     2.2 Calibration
     An accurate method for producing low concentrations of anaesthetic
vapours is essential to give reliable analytical data.  An ideal absolute
calibration was carried out with a diffusion dilution cell as described
by Savitsky and Siggia  [1]  and adapted by Robinson et al [2] .  This
technique relies upon the diffusion of vapour through a precision bore
capillary tube into a gas stream to give a predicted concentration under
established conditions.  The reliability of the analyser system and the
stability of its response were checked by this apparatus.
3.  Discussion
     The techniques described above are well known but hitherto have not
been combined and applied to a study of air pollution in operating
theatres.  The system was used initially to investigate the pollution in
a dental-outpatient, general anaesthetic department where relatively high
levels of anaesthetic in the air were expected from the use of a semi-
open breathing system.  The sampling points were distributed throughout
the surgery in positions of maximal activity and the overall distribution
of positive measurements is shown in Figure 1.  These values were
surprisingly low by comparison with other published data and considering
that air flow measurements  indicated poor ventilation in the room.  The
distribution of the exposure levels of anaesthetists, however, was
considerably wider and halothane levels exceeding 5.10  g.l   were
recorded.  Admittedly, the  anaesthetists' inhalation zone is near to the
expiratory valve of the nose mask, but it must also be significant that
the flow studies indicated  a stagnant zone of air where the anaesthetists
work.
     A complete interpretation of the data is proceeding and studies of
other operating theatres are planned to give a more complete picture of
the extent of air pollution by anaesthetic agents.

-------
                                   1782
  0.3
  0.2
GO
<
CD
O
tr
Q-
  0.1
           IT!
                        0-2
                            HALOTHANE  CONCENTRATION (mg.P1)
                                                                   Q5
   Figure 1      The Range of Concentrations  of Halothane in the Air of
                 an Operating Theatre.   (Sample population 877).
References
     SAVITSKY, A.C., SIGGIA, S.,  "Improved Diffusion Dilution Cell
     for Introducing Known Small  Quantities of  Liquids  into Gases",
     Analyt. Chem., 44^ (1972) 1712


     J. S. ROBINSON et al, work to be published.

-------
                               1783
         THE CONTRIBUTION OF ORGANIC LEAD COMPOUNDS TO
            TOTAL LEAD LEVELS IN URBAN ATMOSPHERES

            R, M. HARRISON, R, PERRY/ D. H, SLATER
Public Health Engineering, Imperial College, London, United
Kingdom
ABSTRACT

     A sensitive adsorption technique for the determination of
organic lead in urban air is described.   Levels of organic lead
as low as 0.01 .ug m~  may be determined with a 30 minute sampling
period.   The validity of the method has been tested, particu-
larly with regard to possible adverse effects of other air pol-
lutants upon the analytical accuracy*   A summary of the levels
of organic lead measured by this method in the London (U.K.) area
is given, together with the levels of particulate lead measured
concurrently.   The highest levels of organic lead and the high-
est proportions of total airborne lead in the organic form were
measured at a petrol station.   A substantial reduction in levels
of organic lead has been found upon moving away from a road or
entering a building.

-------
                                  1784
1.   Introduction
     Organic compounds of lead, used as petrol additives, are exhausted
from motor vehicles largely in art inorganic, particulate form and abundant
data are available on the levels of particulate lead in air.   A very
small proportion of the lead, however, is exhausted in an unchanged organic
form, and this together with crankcase emissions and petrol evaporation
contributes to the organic lead in urban air. Lead alkyl compounds normally
exist in air in the vapour phase, and as they present a more difficult
problem of measurement than the particulate form of lead little informa-
tion regarding the exposure of the urban dweller to such compounds has
been published. Some data have indicated that under certain circumstances,
both for periods of short (Laveskog (l)) and long duration  (Colwill and
Hickman (2)), organic lead in urban atmosphere may account  for in excess
of 50% of total lead, although substantial  variation between the results
of different workers is apparent.   In view of the known toxicological
hazard associated  with organic lead compounds, which is quite different
from that due to particulate lead, and the  uncertainty regarding levels of
organic lead in air, further study appeared appropriate. Consequently a
sensitive analytical procedure has been developed for the determination
of organic  lead in air which allows a rapid assessment of exposure.
2.   Air Sampling  and Analysis
     Few methods have been reported for the determination of organic lead
compounds at the very low levels found in urban air  (Laveskog  (1),
Snyder  (3), Purdue et al.  (It)),  and to date only one procedure has  involved
the  use of  short  sampling periods in  conjunction with modern instrumental
 techniques  of  analysis  (Laveskog(l)). The method developed  in  this  labor-
atory  for  the  measurement of organic  lead compounds in  air,  however,  is
 readily applicable at  levels as  low as 0.01 ng m~5  involving a JO minute
 sampling  period (Harrison _e_t al. (5)).
      In essence,  the air sampling procedure involves drawing approximately
 0.02 m^ of air through an adsorption  tube  packed with  G.C.  column packing
 material  maintained  at the  temperature  of liquid nitrogen (Figure  l),
 after prior collection of particulate lead  by passage  through  a 0.22 pi
 Millipore filter which is subsequently analysed for lead.    A  short heat-
 ing coil immediately preceding the adsorption tube  maintains the enclosed
 tube at slightly above ambient temperature  and ensures that any condensa-
 tion of vapours occurs within the adsorption tube.    After sampling, the

-------
                                   1785
   Heating Coil
               0-22Um    Sample Tube
               Millipore        In
                Filter    Liquid Nitrogen
    Pump
                                                      Gas Meter
                        Figure 1
Air Sampling Apparatus
tube is stored immersed in liquid nitrogen and when required for analysis
is inserted in an oven with its ends sealed by connection to a six-port
switching valve (Figure 2). A nitrogen stream, passed initially through an
identical blank tube, is diverted through the sample tube when the oven
temperature stabilises at 130 . The eluted organic lead compounds are
passed into the combustion air stream of an atomic absorption spectro-
photometer set up for lead measurement. The analytical procedure, calibrat-
ed by the analysis of adsorption tubes dosed with an aliquot of a dilute
toluene solution of a lead alkyl compound, has a detection limit of 0.2 ng
of lead, corresponding to 0.01 ug m   of organic lead in a 0.02 m  sample
of air.
     2.1 Validity of the Method
     The validity and precision of this procedure were tested initially in
a number of ways (Harrison et_ _al_. (5)). A near quantitative collection
efficiency was found for the sampling of known aliquots of tetramethyllead
and tetraethyllead from air and the analytical procedure was found to be
free from errors resulting from background absorption, or matrix effects
due to trapped hydrocarbons and water.
     2.1.1 The Effect of Other Air Pollutants upon the Analysis
     It was subsequently suggested that other air pollutants, such as
oxides of nitrogen, expected to be trapped during air sampling, might
react with trapped organic lead compounds during air sampling or analysis,

-------
                                      1786
                                               Atomic Absorption
                                               Spectrophotonwter
Figure 2
                 Ovw
Organic Lead Analytical Apparatus
KEY
(A)  Sampte Tub*
(5)  Bank-Rib*
©  Six-port VMv»
 thereby converting volatile organic lead compounds into  involatile  lead
 compounds and hence introduce an error into the analytical result.  Were
 this the case, it might have implications  for other procedures involving
 the use of cooled sample tubes for the collection and determination of
 organic lead in air (Laveskog (l), Cantuti and Cartoni (6)), as well as
 for the procedure outlined above.
      To investigate this problem urban air was sampled through two  adsorp-
 tion tubes connected in parallel (Figure 3)« Prior to sampling, one tube
 was dosed with an aliquot of a lead alkyl compound, comparable in lead
 content to the anticipated quantity of organic lead to be collected from
 the urban air sample, whilst the other tube remained blank. Both tetra-
 methyllead and tetraethyllead were used in this procedure.   Air sampling
 was then performed in the normal way, the flows through  the two tubes being
 equalised by use of the needle valves and rotameters (Figure 3). Both
 adsorption tubes were subsequently analysed in the usual manner. Analysis
 of 16 tubes used to sample air at ^ sites of widely differing levels ofair
 pollution and airborne organic lead revealed a quantitative recovery of the
 dosed lead compounds to within the analytical error. Hence other air
 pollutants appear not to effect  the analyses.
 3-    Levels of Organic Lead in Urban Aijr
      Levels of organic lead and particulate lead collected concurrently
 were determined using the above procedure.  Results were calculated both for
 the levels of organic lead in the air sampled,  and the percentages of total

-------
                                    1787
      0*2Um
      Itmipora
      filter
    Figure 3
        SwnptoTubM
           In
        Liquid Mtiogm
   and
RotwiMt*rt
           Pump
Apparatus for Sampling through Two Tubes In Parallel.
lead present in the organic form. The latter statistic  is  valuable as it is
less dependent upon prevailing meteorological conditions and  sampling site
topography anH is more reflective of source characteristics.
     3.1 Kerbside Levels
     In the London (U.K.) area measurements on  streets  of  widely varying
dimensions and traffic flow showed levels of organic  lead  of
0.0** - 0.11 ug m~  with the percentage of lead  in the organic form being
0.3 - 2.6%, the majority of results being in the region of 1  - 2#. Levels
of particulate lead varied from 3«2 - 20.3 ug m . Samples collected in a
busy road tunnel showed particulate lead levels of 23«8 -  32.2 ug m
whilst organic lead levels remained around 0.02 (ig m   ; ca.0.1# of total lead.
     Measurements taken on the forecourt of a busy London  petrol station
revealed organic lead levels in the range 0.21  - 0.59 ug m" , representing
3.9 - 9-7# °f total airborne lead, whilst at a  less busy petrol station out
of the city an organic lead level of 0.0? ug m~^ (k.2&  of  total lead) was
measured. Thus although, as anticipated, elevated levels of organic  lead
were found at petrol stations, particulate lead is still far  the major
form of airborne lead. It is clear that people  working  in  close proximity
to petrol pumps are exposed to significantly higher levels of organic lead
than the population as a whole.

-------
                                        1788
         3.2 Levels at Greater Distance from the Road

         Preliminary studies have indicated that levels of organic  lead  drop

    significantly  with distance  from the road,  and upon entering buildings.

    In moving 20m from the kerbside of a busy London street the  organic  lead

    level was reduced from 0.07 to 0.01 ng ra  i  whilst  particulate  lead

    changed from k.k to 3-3 ug m   • Moving 10m further  and into  a building

    the organic lead level dropped to<0,01 ng m  ,  whilst the  particulate

    lead level was 1.8 ug m  .

         On a less heavily trafficked suburban road the levels of organic and

    particulate lead 2m from the  kerbside were 0.03 and 0.7 ug m   respect-

    ively, whilst moving n further 'tm and into a house  the levels were reduced

    to O.O1 and 0.6 ug m~  respectively. In view of the dilution with ambient

    air upon moving from the source of the lead, and the known instability of

    organic lead in air (Anon (7)) these results are readily explicable.
References:
1.  LAVESKOG, A., "A Method for the Determination of Tetramethyllead (TML)
    and Tetraethyllead (TEL) in Air."   Paper given at the Second International
    Clean Air Congress. Washington D.C., December 1970.                       ~


2.  COLWILL, D.A., HICKMAN, A.J., "Concentrations of Volatile and Particulate Lead
    Compounds in the Atmosphere,   Measurements at Four Road Sites."   Transport
    and Road Research Laboratory, Department of the Environment, TRRL Report LR 51*5

3.  SNYDER, L.J., "Determination of Trace Amounts of Organic Lead in Air.
    Composite Sample Method." Anal. Chem.. 39, 591-595 (196?).


k.  PURDUE, L.J., ENRIONE, R.E., THOMPSON, R.J., BONFIELD, B.A., "determination of
    Organic and Total Lead in the Atmosphere by Atomic Absorption Spectrometry,"
    Anal. Chem.. 45, 527-530 (1973).


5.  HARRISON, H.M., PERRY, R., SLATER, D.H., "An Adsorption Technique for the
    Determination of Organic Lead in Street Air," Atmospheric Environment,  In the
    Press.

6.  CANTUTI, V., CARTONI, G.P., "The Gas Chromatographic Determination of
    Tetraethyllead in Air." J. Chromatog.. 32, 641-647 (1968).


7.  ANON, Survey of Lead in the Atmosphere of Three Urban Communities, Public
    Health Service Publ. No. 999-AP-12.   National Air Pollution Control
    Administration, Public Health Service, U.S. Dept. of Health, Education and
    Welfare; p.10.

-------
                               1789
        ENQUETE PORTANT SUR LA CONSOIflATION FAMILIALE DE
           PESTICIDES DANS LES FAMILIES HABITANT DANS
           DE GRANDS ENSEMBLES DE LA REGION PARISIENNE
                          (METHODOLOGIE)

             C, PIVA, M,L, EFTHYMIOU ET E, FOURNIER
Centre de Toxicologie, HSpital Fernand-Widal, Paris, France
RESUME

     L'£tude epidemiologique de la contamination par pesticides
BB justifie par 1e nombre minime de dosages en France parmi la
population non expoaee. Le propos supplSmentaire de cette etude
est d'eaaayer d'etablir la correlation entre les taux sanguine
lea plus Sieve's et I'origine de la contamination pour proposer
dea mesurea preventives destines a la diminuer.

     Lea diveraea enquStea s'attachent habituellement a preoiaer
I'importance de I'apport alimentairet certes preponderant* mais
non exclusif;  la mesure de la condemnation domeatique d'insec-
ticides managers peut faire apparattre un facteur favorisant de
contamination.
ABSTRACT

     This epidemiological study of pesticide levels is motivated
by the very email number of determinations carried out in France
among the unexposed population.   A further aim of the study is
to attempt to establish correlation between the highest blood
levels and the source of the substances with a view to suggesting
preventive measures to reduce the contamination.

-------
                               1790

     Surveys of various kinds are usually designed to determine
the quantities ingested with the food, this being certainly the
most important, but not the only route of intake.   Determination
of the level of use of household insecticides in the home may
bring to light a factor enhancing contamination.

-------
                                  1791
MATERIEL HUMAIN
Les sujets etudies ont etc selectionnes en fonction de trols criteres :
1.1. - leur habitat urbain dans les immeubles de la banlieue parisien-
       ne, de faqon a exclure 1'exposition semi-professionnelle des ma-
       ralchers et des proprietaires de jardln lors du traitement de
       leurs cultures, de faqon aussl a obtenlr une homogeneite relati-
       ve dans les lieux d'achats des aliments pour laisser son impor-
       tance au facteur d'utilisation de pesticides managers.
1.2* - leur metier, pour eliminer naturellenient tout facteur profession-
       nel d1 expos it ion,,
1.3. - leur cooperation enfin, necessaire pour repondre aux questions de
       1'enquSte et accepter le prelevement sanguin, la visite de 1'en-
       quSteur & leur domicile.
       Aucune selection n'a volontairement port4 sur I1age ou le sexe.
ENQUETE
Divers elements sont precises iors de 1 *interrogatoire :
2.ie - 1'habitat anterieur, pour ne pas meconnaitre un sejour prolong^
       et recent dans une zone d'agriculture intensive riche en pestici-
       des, et la date du dernier deine'nagement.
2.20 - la situation de 1'appartement dans 1'immeuble, car les etages in-
       f^rieurs proches des espaces verts des residences et le dernier
       e'tage poss^dant une terrasse jardin sont plus exposes aux prodults
       de Jardinage { 1'existence de pelouse a 1'entour de la raaison, la
       presence d'arbres dans les cours inte'rieures sont prScls^es*
2.3. - la presence de plantes.
2.4. - 1'existence ou non d1animal doraestique, chien, chat, oiseaux, sou-
       ris, hamster, tortue, herisson, ou autres ; les produits veteri-
       naires contiennent souvent des organo-chlor^s (schampoing insec-
       ticide) ou des organo-phoaphorls (vermifuges), les colliers  insec-
       ticides pour chlens sont a base d'organo-phosphorls, le nettoyage
       et la disinfection des cages sont re'alise's avec des produits mul-
       tiples.

-------
                                   1792

 25   -
        la nature  des  parasites  connus  ou  redoutes,  dans  1'appartement,
        sur les  plantes,  sur  les animaux  ;  selon  qu'll  s'agit  de mousti-
        ques,  de mites, de mouches,  de  puces,  de  pucerons,  I'insecticide
        utilise  differe ; par exemple,  les  produits  centre  les cafards si
        resistants  sont quasi toujours  des  organo-phosphores.
 2.6.  -  la description exacte des produits  menagers  trouves sur place le
        jour de  la  realisation de I'enquete.
        2.6. -1- les produits insecticides, sont  notes  en fonction de
                leur  nom commercial, leur  mode de presentation (bombe,
                plaquette, poudre..e), le  nom du fabricant, le lieu
                d'utilisation (cuisine avec son  risque de pollution ali-
                mentaire, chambre avec contact journalier d'au moins 7
                heures, salle de bains avec temperature plus elevee).
        *}  f\    *}
        •  • ~*~ les produits menagers proprement dits, releves selon les
                memes criteres, correspondant aux detergents, detartrants
                .*.», certains  comma Les desodorisants de lavabos, sont
                a  base d'organo-chlores type  paradichlorobenzene.
        2.6. -3- les produits utilises pour traiter  les piantes d'appar-
                tement0
Pour chacun de ces trois  groupes, des precisions sont demandees  sur les
changements eventuels de  marque par rapport aux mois precedents,  ainsi
que la notion d'incidents imputes par le public a leur manipulation.
Pour cet interrogatoire,  il  est recommande a 1'enque'teur de demander  la
liste sans suggerer lui-mtme les noms commerciaux des produits les plus
courants,  pour ne pas influencer les menageres,

Cette partie de I'enquete suppose le depouiLlament dans un second temps
de la liste de produits deceles sur place et obtenus par interrogatoire  j
I'aide technique du Centre National d1Informations Toxicologiques,
I'A.S.I.T.E.S.T.,  permet  de connaitre la formule des produits utilises»

2.7. les habitudes alimentaires dans le cadre familial sont precisees j
     I'enquete s'attache  a noter surtout deux point :
     2*7.  -1- le lieu d'achat :  super-marche
                                magasin a succursales multiples
                                detaillant individuel,
     car les circuits sommerciaux et la provenance risquent de differer.
     2.7.-2- 1'habitude de consommer cru : les legumes non epiuches
             (salades) les ecorces  de fruits frais.

-------
                                 1793
2.8. la possession ou non d'une residence secondaire qui naturellement
     majore le risque de contamination par 1'emploi de produits pour pro-
     teger les charpentes, detruire les parasites dans la maison, trai-
     ter les plantes du jardin ; la frequence de sejour (tous les week-
     end, les 15 jours, une fois par mois, tous les 6 mois, une fois par
     an) est appreciee.
2.9. le lieu de vacances annuelles comportant comme risques supplementai-
     res : le sejour en milieu rural, 1'installation en zone chaude in-
     festee de moustiques,

Cette enquete relativement detaillee n'a pas fait apparaitre de differen-
ces signlcatives dans le mode d'alimentatioii, ce qul laisse son importan-
ce aux autres facteurs.
PRELEVEMENTS TOXICOLOGIQUES
L1enquete se poursuit par un prelevement sanguin destine au dosage des
pesticides. Lors de la prise de sang sont notes systematiquement :
- le poids et la taille de 1'individu,
- la nature des traitement medicamenteux en cours,
Le volume de sang preleve sur heparine est d'au moins 10 ml. Son uti-
lisation est double.

3.1. - une determination et un dosage des insecticides organo-chlores :
       2 ml de sang sont eleves, apres addition de 2 ml d1 ace"tone et agi-
       tation, sur colonne de chromatographie garnie de sulfate de so-
       dium anhydre et chargee de florisil sur une hauteur de 12 cm ;
       les 16 ml d'eluant ainsi recueillis sont etendus a 50 ml par une
       solution a 2 % de SO,Na2, puis extraits par 2 ml d'hexane, se1-
       ches sur du SO, Na- anhydride et injectes a raison de 5 y*l en
       chromatographie gazeuse.
3.2. - un dosage du taux de cholinesterases sanguines, s£riques et glo-
bulaires. La technique utilisee derive de la methode de Mitchell,
Les erythrocytes, une fois laves, sont maintenus a une temperature de
25° dans une solution riche en phosphates dans laquelle a etc ajoutee
de 1'acetylcholine comme substiat. Les variations de pH, mesurees au
pH metre, exprimees en pH/h/25°, sont comparees au taux consider^ comme
normal dans notre laboratoire, 0,75 pour les cholinesterases erythrocy-
taires (0,70 pour les seriques).

-------
                                 1794
L1analyse peut etre repetee en cas do rcsultat nettement pathologique
puisque le volume necessaire est inferieur a 1 ml ; le lavage preala-
ble des erythrocytes est une precaution supplementaire contre les causes
d'erreur bien que le resultat ne differe guere.
Le dosage direct des organo-phosphores dans le sang n'est pas envisage
dans le premier temps de cette etude,
Les resultats actuels teraoignent d'une contamination relativement homo-
gene de la population dans la banlieue parisienne. L'apparition even-
tuelle de divergences liees a 1'utilisation differente de pesticides
managers suppose :
- soit 1'affinement de 1'enquSte pourtant deja detaillee,
- soit 1'extension de 1'etude a un nombre plus eleve de sujets.
 . Unite de Toxicologie Experimentale U.  26   I.N.S.E.R.M.
   et Faculte Lariboisiere - Saint Louis,  Universite PARIS  VII.

 . Hdpital  Fernand Widal
   200,  rue du Faubourg Saint Denis,  75010 PARIS,  FRANCE.

 . Contrat  de Recherche :  contamination des chatnes  alimentaires.
                          Ministere de 1'Environnement,

-------
 MODELLE





 MODELS





 MODELES





 MODELLI





MODELLEN

-------
                              1797
             APPL1CAZIONE DELL'ANALISI SPETTRALE
             ALLO STUDIO DELIA DISPERSIONE DEGLI
         INQUINANTI ATMOSFERICI NELL'AREA DI VENEZIA

         L, IOVEN ITTI/ M. MENENTI, F, FERRARI-BRAVO,
                   I. GIOVANNINI E A, LEVY
Tecneco SpA, Fano, Italia
RIASSUNTO

     La presente ricerea ha aercato di individuare i meoaanismi
di diaperstone degli inquinanti atmosferici nell'area di Venezia,
partendo da dati rilevati at auolo.

     Si disponeva delle medie oravie delle eerie temporali di
aoncentrazione di SO- rilevate per un mese e mezzo in diverse
etazioni di monitoraggiot eia nell'area induetriale ohe a Venezia
8tes8af e dei dati di velooitd e direzione del vento raocolti in
due atazioni meteorologiche.

     Utilizzando I'analisi di Fourier, sono state evidenziate
le oaratterietiohe apettrali delle serie temporali di concentra-
zione e di velocity e oomponenti orizzontali del vento al auolo.

     Si oonta di ottenere tutta una serie di ulteriori informa-
zioni effettuando I'analiei di ooerenza e faee su  diverse aoppie
di  eevie temporali di oonoentrazione e orossoorrelando i dati
di aonaentrazione aon quelli meteorologiei.

-------
                              1798
ABSTRACT

     This research deals with the evaluation of the atmospheric*
dispersion mechanism of pollutants in the Venice area in connec
tion with wind-speed and   direction patterns at ground level.

     Disposing of six series of continuous records of SO,, con-
                                                        &
centrations collected at different fixed points in the industri
area as well as in Venice itself't the Fourier analysis was
applied to hourly averaged data in order to obtain the spectral
properties of the modulations of SO „ concentration.

-------
                                   1799


1. Introduzione

Se ci si pone 1'obiettivo di indagare sulla dinamica nel tempo e nello spa
zio del campo di concentrazione di un inquinante atmosferico, si incontra
no subito notevoli difficolta quando si voglia entrare nel dettaglio degli ef_
fettivi processi fisici che evolvono nell'atmosfera. Le difficolta diventa-
no praticamente insormontabili quando si vogliano studiare aree caratte-
rizzate da forti disomogeneita  superficiali: in tali condizioni si dimostra
inadeguata gran parte della modellistica oggi disponibile, ne migliori ri_
sultati si ottengono cercando di risolvere le equazioni del moto, date le
notevoli  complessitzl che si incontrano nella loro integrazione.
Una possibile alternativa e quell a di ricavare  informazioni  sulla dinami-
ca dei fenomeni fisici in esame direttamente dai dati sperimentali, me-
diante 1'utilizzazione di opportune tecniche di  analisi. In tal senso va in-
tesa 1'applicazione delle tecniche di analisi spettrale allo studio del tra-
sporto e  della diffusione degli inquinanti atmoaferici (M. Tilley e G.  Me
Bean,  1973). Tali tecniche permettono di scomporre il fenomeno,  per
quanto complesso,  nei processi elementari che lo compongono e di dete£
minarne  1'importanza relativa.
In questo lavoro, 1'analisi spettrale e stata applicata allo studio dell'in-
fluenza dei fenomeni meteorologici  a mesoscala sul trasporto e la diffu-
sione di  SO. sull'area di Venezia.
L'area di Venezia si presenta particolarmente valida per la sperimenta-
zione di  questo  tipo di metodologia. In essa, infatti,  sono present! notev£
li discontinuity  superficiali (terra-mare, zona industriale,  zona urbana,
zona rurale) che danno luogo a fenomeni meteorologici a mesoscala parti_
colarmente complessi,  interessanti sia la circolazione che la struttura
dei bassi strati dell'atmosfera (Lyons, 1973).

2. Dati utilizzati

I dati di  concentrazione utilizzati  sono quelli relativi ai livelli di SO2 riU
vati dalla rete  per  il controllo  della qualita dell'aria nella zona di  Vene -
zia. Tale rete,  realizzata dalla Tecneco SpA per conto dell'Istituto Supe_
riore di  Sanitzk  e dell'ENI,  consta di 24 analizzatori in continua Philips
9700 che campionano a una altezza di 4 m dal  suolo.
Per questo studio,  sono  state utilizzate le medie orarle dei dati raccolti,
nel periodo 1 febbraio -  14 marzo 1973, dalle 6 stazioni di  fig. 1.  Infatti,
a quella  data, delle altre stazioni alcune  non erano state ancora installa
te,  altre avevano avuto del period! relativamente lunghi di non funziona-
mento per manutenzione.
Come dati meteorologici, sono statl utilizzati i dati orari di direzione e
velocitik  del vento,  mediatl sui dieci minuti precedent! 1'ora di attribuzio
ne, rilevati nelle stazioni di  Tessera, del Servizio Meteorologico  dell'A
eronautica Militare,  e dell'Ospedale a Mare (stazioni A e B di fig. 1).
Tali dati sono relativi al periodo 1 febbraio -  14 marzo 1973.

-------
                                  1800

                       Fig.  1.  Map of  the Venice area
3.  Metodo  di analisi

Per le serie temporal! delle concentrazioni di SO^  e del modulo e delle
component! orizzontali della velocita del vento,  sono state calcolate le
funzioni di autocovarianza, a ritardi variabili da 0 a 200 ore,  e le rela
live trasformate di Fourier,  nel campo di frequenza tra 0. 005  e 0. 5 h~ 1 .
Gli algoritmi usati sono in Jenkins  Watts:  "Spectral analysis  and its ap-
plications" (1968).  La finestra spettrale usata e quella di Bartlett; comun
que,  da prove effettuate,  si  e  verificato che la forma degli spettri non
cambia  in modo apprezzabile al variare del tipo di  finestra.  L'effetto di
"aliasing" sugli spettri dei dati della concentrazione di SO., e trascurabi^
le poiche i dati sono mediati sull'ora. FJer i dati di vento,  mediati sui
dieci minuti precedent! 1'ora di attribuzione,  si  e proceduto a filtrare
dalle serie temporal!  le component! con periodo inferiore  a  3 ore.
L'intervallo di confidenza dell'80%  sull'arnpiezza degli spettri  va da
+  90% a - 40% dell'arnpiezza  calcolata.

-------
                                    1801
                                                                 LCCflU t>k
                                                                 jr*i OHC  A
                                                          ^ « . \  ,       , [ i,
                                                           y ''V\  \_  J-/-/W/'*"
•it\- _/^—at—w—±(~t\L/'
  v^xvw:.^-
Fig.  2  a b--Aut°cohovariance (a) and power-specturm (b) of SO  concentra-

             tion time series  at the six monitoring stations indicated in fig.

             1.  In the power-spectrum  the bandwidth is 0.0075  h"1.

-------
                             1802
                                                   • I.I.OH  A
Fig.  3  b

Fig. 3 a,b -
Autocohovariances  (a,b)  and  power-spectra (c,d)  of
time series of wind-speed  (  V  )  and  -horizontal  com-
ponents  (northwards Vyand  eastwards  v  ).   A  and  B
stations refer to  fig.  1.
In the power-spectra the bandwidth is  O.OO75h  -1.

-------
                                      1803
         .  3  d
                                                                            110 k
ig.  3 c,d - Autocohovariances (a, b) and power-spectra (c, d) of time series
              f wind-speed (|v|) and -horizontal components  (northwards  v
            and  eastwards vx).  A and B stations refer to fig.  1. In the   y
            power-spectra the bandwidth is 0.0075 h"1.

-------
                                  1804

4. Risultati

Poiche lo studio e in corso di completamento,  sono stati riportati solo
alcuni risultati  preliminari.  Nella fig,  2,  sono riportate le funzioni di
autocovarianza  e gli spettri di potenza delle serie temporal! della con-
centrazione di SO^ rilevate nelle sei stazioni di misura.
Nella fig.  3, sono riportate le funzioni di autocovarianza e  gli spettri di
potenza del modulo  e delle component! orizzontali della velocita del ven-
to,  rilevate al suolo nelle stazioni A e  B.
Negli spettri di potenza della cpncentrazione,  sono  evidenti le righe diur
ne, semidiurne  e inerziali.  Le righe con periodo attorno a  72 e 48 ore
sono presumibilmente dovute al passaggio di perturbazione sulla zona di
Venezia.  Tale interpretazione e confortata da un  esame qualitative delle
carte meteorologiche relative al periodo in esame.

5. Prospettive

Considerando gli spettri di potenza delle concentrazioni di SO e quelli
di velocita del vento, particolarmente interessante  appare la diversita
nei due casi del pesi delle armoniche di periodo di  circa 48 e 72 ore ri-
spetto all'integrale  dello spettro.  Tale diversita di  peso rende evidente
la necessiti di  cercare  un'interpretazione fenomenologica dell'influenza
del moti atmosferici a piccola energia  (effetti inerziali,  maree atmosfe-
riche, . . . ) sui  fenomeni di inquinamento atmosferico da sostanze in fa-
se gassosa.
Dall'analisi degli spettri di fase delle serie temporal! di concentrazione
di SO  sembrano emergere indicazioni sulla^struttura del regime di brez
za nefl'area considerata. Tale analisi e ancora in corso,  insieme a quel
la sulla coerenza spaziale.
 Bibliography

 . JENKINS, G.M., WATTS, O.G.,  "Spectral analysis and its applica-
   tions", Holden Day (1968).

 - PANOFSKY,  H.A.,  BRIER,  G.W., "Some applications of Statistics
 " to Meteorology", University Park,  Pennsylvania (1968).

   TILLEY,  M.A.,  BEAN,  G.A.,  "An application of spectrum analysis
   to  synoptic pollution data",  Atmospheric Environment,  vol. 7, pag.
   793 (1973).

-------
                              1805
LUMLEY, J.L. ,  PANOFSKY, .H.A. ,  "The  structure of Atmospheric
turbolence",  Interscience Publishers  (1964).

LYONS, W. A. , "Mesoscale transport of pollutants in the Chicago
area as affected by land and lake breezes",  Second Clean Air Con-
gress, Dusseldorf (nov.  1973).

GOSSARD,  E.E.,  "Spectra of Atmospheric  scalars", Journal of Geo_
physical Research, vol.  65, n°  10 (oct. I960).

-------
                               1807
             STUDIO E VALUTAZIONE DEGLI EFFETTI DEL-
   L'lNQUINAMENTO AMBIENTALE SULLA SALUTE DELL'UOMO ATTRAVERSO
       SIMULAZIONE CON MODELLI BIO-INGEGNERISTICO-CHIMICI

      M, DEL BORGHI/ G, FUMAROLA, R, GARDI E G, FERRAIOLO

Istituto di Scienze e Tecnologie dell'Ingegneria Chimica,
University di Geneva, Italia
RIASSUNTO

     Sulla base dei piu veoenti atudi a livello interdiaeipli-
nave ohe si oollegano anche alls appliaazioni biomediche del-
I'ingegneria ohimioa si propone un modello bio-ingegnerietico
ahimico pev la correlaz-Cone f-Ca-ico-matematiea esposizione-effetto
ai fini della valutazione dell'inquinamento ambientale eulla
salute dell'uomo.

     II modello e basato sulla applicazione della ben nota
analia'i. compartimentale g-ia con euooeeao aperimentata nello
studio della diatribuzione, effetti e tollerabilitd dei. farmaoi
nel oovpo umano oonaepito dome uno o piu veattori biochimioi in
     II modello tiene oonto al oontempo dei prinoipali parametri
eignifioativi ai fini degli effetti presunti o oeservati aullo
atato di aalute dell'uomo quali:  la oonoentvasione dell'inqui-
nante, il tempo di eepoaisionet le oavatteriatiohe intvinaeohe
del aoggetto eepoato e per ciaacun tipo di inquinante.

     II modello ooetituiaoe una alternativa di piu rapida infov-
mazione, ad altri a tutt'oggi adottati di tipo epidemiologioo e
etatiatioo e aembva ai poasa ben preetave all'interpreta&ione

-------
                               1808
delle alteration! fisiologiche e del comportamento dell'uomo in
seguito ad esposizione ad inquinanti.

     EBSO inoltre & un'utile risorsa per valutazioni e prevision!
significative sugli effetti:  sinergici, antagonist! e addizio-
nali degli element! inquinanti sia a breve che a lungo termine.
ABSTRACT

     On the "basis of the most recent interdisciplinary studies
concerning the biological and medical applications of chemical
engineering, ideas have been put forward for a mechanical bio-
chemical model designed to provide physical and mathematical
correlations of exposure and effect, to assess the impact of
environmental pollution on human health.

     The model is based on the familiar analytical techniques
already successfully tested in studies on distribution, effects
and  acceptable levels  of drugs in  the human body, and is in the
form of one or more biochemical reactors in parallel series.

     The model is designed to provide simultaneous measurements
of the most important  parameters indicating suspected or observed
effects of pollution on human health, such as pollutant concen-
tration, duration of exposure, and intrinsic characteristics  of
the  exposed subject, relating these  to  the type  of vehicle
 (atmosphere, water, food, noise),  for each type  of pollutant.

     The model represents a faster-working alternative  to the
other  epidemiological  and statistical methods used so far,  and
it could well provide  an excellent means of interpreting physio-
 logical and behavioural changes  in man  as a result of exposure
to pollutants.   Moreover,  it provides  a useful  means of moni-
toring and making significant estimates of  the  synergic, anta-
gonistic and cumulative effects, both short and  long-term,  of
various pollutants.

-------
                              1809
 1 .Introduzione
   Quando un gas tossico e presente nel1'atmosfera  e  viene re-
spirato dall'uomo si stabilisce una certa velocita  di assorbi-
mento attraverso le membrane alveolar!  entro  il  sangue polmona
re prodotta da una differenza di coneentrazione  tra 1'aria e
il sangue.
   Nella generalita dei casi i gas inquinanti  non vengono metj^
bolizzati e escreti attraverso le normal! vie,  ma sono elimina^
ti quasi interamente attraverso i poltnoni da  cui vengono  assor^
biti (1) .
   II trasf er imento netto dagli alveoli  al  sangue e inizialmen.
te rapido e quindi progressivamente diminuisce  fino a che non
si raggiunge un equilibrio.
   Se c  e la concentrazione di inquinante  nel  sangue, c   quel
       s                                                 a    ~~
la alveolare si avra per uguali volumi  di sangue e  aria a co£
tatto:
            dc
                     k.c  - k0c_                    (1)
d                     *x i fc-    "^O
             t        la    2s
e poiche in prima approssimazione  si  suppone  costante  la  concein
trazione alveolare (c =cost) si puo considerare  il  fenomeno  c£
me un assorbimento di ordine zero  ed  una  el iminaz ione  del  1°  or_
dine (2).
   All ' equi1ibrio si puo scrivere:
             kp  = if ^                               f 9 ^
            i ~    " i *••                               \ ^ /
            1 a    2 s
           c      k.
          	s_  =  1  _ s
           Ca     k2
Dove S e la solubilita del gas inquinante  nel  sangue.  Sostitueji
do nella (1) il valore di k. dato  dalla  (3),  il  valore  di  c
dato dalla (2) e chiamato f il rapporto  c  /c   tra  la  concentra_
zione nel  sangue istantanea e quella  al1 ' equilibrio,  la (1)  d^
venta:
           df      ,
          -at- " k2(1-f>                             <4>
cherisoltada:
                      -k t
           1 - f  =  e  L                            (5)
   La Fig. 1 e la espressione grafica della  (5)  e  da  essa  per
vari tipi  di inquinanti caratterizzati da  diverse  S  e  k-,  nota
la concentrazione alveolare si puo ricavare  la  concentrazione

-------
                                 1810
                  i
                  0,2 .
                    0.01
10     t It,
  Fig. 1.  Andamento  del  rapporto o  /c =f in funzione del parametro
                                    5  S

           adimensionale  k?t.
                                              Q  c1
Fig. 2,  Modello a 2 corapartimenti  per  il  sietema aria alveolare


         -frazione acquosa del  corpo  umano.

-------
                              1811
0.2
                           r cmiro
Fig. 3.   Andamento della  concentra
zione alveolare di gas tossico
(rispetto all'equilibrio)  per gas
gacilmente diffusibili k1=8.
                                                              r tmiM
Fig.  k.  Andamento della  concen-
trazione di gas tossico nella
frazione acquosa del corpo umano
(rispetto all'equilibrio) per
gas facilaente diffusibili k =8.
 nel sangue e quindi per  ogni  tipo di inquinante prevedere  gli
 effetti che esso avrS  sul  corpo  umano.
    In un approccio di  seconda approssimazione occorre  tener
 conto che la c  raggiunge  il  valore della concentrazione atnt£
               3
 sferica del1'inquinante  solo  dopo un certo tempo dipendente  e£
 senzialmente dalla solubilita dell'inquinante nel  sangue  in
 particolare, ed in generale  nella frazione acquosa  del corpo
 umano.
  2.Modello a due compartimenti
    Per tale caso piu  generale viene proposto un modello a  due
 compartiaienti che  tiene  conto oltre che del volume  di  sangue o
 frazione acquosa,anche del volume polmonare. La Fig.2  schema-
 tizza il modello (3,A,5,6).

-------
                              1812

   Le equazioni che lo descrivono sono:
      dc.
               1V1C1 + k2V2C2 + ^c^cl)              (6)

                                                      (7)

dove v1 e v9 sono i volumi rispettivamente polmonare  e del  san
gue, c. e c« le concentrazioni di gas tossico nei due comparti
menti, c0 la concentrazione esterna ai compartimenti, k1 e  k
costanti di diffusione attraverso la membrana alveolare, Q  la
portata di aria ai polmoni (It/min).
   AI 1 ' equi 1 ibr io si ha  : c..   = c0 e c0
                           1 CO        £, OO
                                               C2
                                                   s
Le  (6)  e  (7)  si  possono  riscrivere  in  termini  di:
          cl                C2
     y -- - -      e    y=  — -          per  GUI:
dy 9
	—  =   k  v   —  k  v                                    (91
dt      K2yl    K2y2                                   ^>
   Si  tratta di  un  sistema  di  equazioni  differenzia 1 i  lineari
di 1°  grade  che  risolto  con le  condizioni  iniziali  y,=0  e
y~ = 0 per  t=0 conduce  a:

             i     r          ric             r2c~l
yl = 1  - k2(r2-r1} Lr2(k2+rl>e   ' r!^2 + r2)e    J         <10)
           1   I    rlt      r2M
y2 " L  ~ r -r  Lr2e    " rle    J                           (11)
dove  r    =  - j  (ki+k?+-rr-> ± J V(ki+k2+~S~)  -4k.-2-     (12)
                           1                 1         1
   Riportate in grafico  (Figg . 3,4,5,6)  le  (10)  e  (11) mostrano
1'andamento delle  concentrazioni di gas tossici in funzione
del tempo per diversi tipi di gas ovvero per different!  valori
di S e k  e nei due compartimenti scelti ovvero i polmoni  e  la
frazione acquosa del corpo umano.
   Come si puo vedere dalle  curve e ricavare analiticamente
dalle (10) e (11)  per i  gas  poco solubili  dopo  breve  tempo  la
concentrazione negli alveoli  raggiunge  il  valore  di equilibrio

-------
                                1813
o.l
0.6
0.4
0,2
                 10
         20
                      eo
                           I Cffiin)
 Fig. 5.  Andaraento della concen-
 trazione alveolare di  gas tossico
 (rispetto all'equilibrio) per gas
 poco diffusibili k1=0,5.
Fig.  6.  Andamento della concen-
trazione di gas tossico  nella
frazione acquosa del corpo umano
(rispetto all'equilibrio) per
gas poco diffusibili k.,=0,5.
 c0 e  identicamente  la concentrazione nella  frazione acquosa
 raggiunge  il  suo  valore di equilibrio  in  tempi  piuttosto brevi.
    Una  situazione opposta e fornita dai gas  ad  alta solubilita
 per  i quali  1 ' equilibrio,nei due casi  di  alveoli  e frazione a£
 quosa,viene  raggiunto in un tempo piuttosto  lungo.
    Il modello  qui proposto puo costituire una  alternativa di
 piu  rapida  informazione e quanto meno  un  utile  ausilio ad al-
 tri  a tutt'oggi  adottati di tipo epidemiologico e sembra si
 possa prestare  al1 ' interpretazione delle  alterazioni fisiolo-
 giche e del  comportamento dell'uomo in seguito  ad esposizione
 a gas toss ici.
     Gli  Autori  rtngraziano il Prof. F.L.  Petrilli per 1'atten
  zione  rivolta  al  presente lavoro e per  gli  utili suggerimenti
  crit ici ricevuti.

-------
                            1814
                      Bibliograf ia

FISHMAN.A.P., "Dynamycs of the pulmonary circulation",Han-
dbook of Physiology, 2,11 (1967)  (Washington: American Phy_
Biological Society, 1963)
GOLDSTEIN,A., ARONOW,L.,KALMAN,S.M.,"Princjpies of drug
action-The basic of pharmacology",  Harper  International
Edition.N.Y.1963
DEL BORGHI,!!., FERRAIOLO  ,G., GARDI.R.,  "Lo  studio di pr£
blemi ecologici secondo i fondatnenti del 1'ingegneria  chi-
mica", I.C.P. ,Vol.3, Marzo 1973.
RIGGS,D.S., "The mathematical approach to  physiological
problems" Williams and Wilking, Baltimore, 1963
BISCHOFF.K.B., DEDRICK,R.L., J. Theoret  Biel,  29,  pp.63-
-83, 1970
BISCHOFF, K.B., DEDRICK,  R.L., Chem. Eng.  Prog. Sympos..
(Ser. N° 84) ,64,pp.32-44  (1968)

-------
                               1815
       A STATISTICAL ANALYSIS OF THE AIR POLLUTION DATA
       IN CONNECTION WITH THE METEOROLOGICAL CONDITIONS

                          D, YORDANOV

Geophysical Institute, Academy of Sciences, Sofia, Bulgaria
ABSTRACT

     The pollution -in the planetary boundary layer (PBL) with a
height of about 1 "km is investigated.   Under conditions of
stationarity^ and a horizontal homogeneity of the underlying
surface the turbulent state in PBL is fully determined by the
following external to PBL parameters:  Rossby's number* the
external parameters of thermal stratification and two external
parameters of the baroclinicity.  These parameters may easily
be determined by the synoptic information.   In the work it is
looked for the relation between these representative for the
diffusion process parameters and the data on the air pollution
in different cities for a few year period of observation.   It
is shown that there exists a close relation between the external
to PBL parameters and the air pollution.

     These investigations give a possibility by data on the
numerical weather forecast to predict the cases of strong air
pollution, especially important for the industrial city regions
since the necessary preventive measures can be duly undertaken.

-------
                             1816
1.   Introduction
     Meteorological conditions influence considerably the
air pollution. Once studied this relationship enables us
to predict the pollution by means of weather forecasts and
when necessary to take in pood time the required preventive
measures. When statistically evaluating the air pollution
data it is necessary to define the meteorological parame-
ters on which it solely depends. Studies of relations of
arbitrary meteorological conditions is inadequate. We are
going to investigate the pollution in the planetary bount-
dary layer (PEL). That is the layer with height of the
order of 1 km above earth surface and naturally the pollu-
tion in it is of the greatest practical interest.
2.   Representative meteorological parameters
     The air pollution depends on the turbulent state:
wind profile, coefficients of turbulent exchange etc. Let
us assume conditions of stationarity and barathropy. The
contemporary conceptions concerning the structure of PEL
(erf.  [l]  ) show that under these conditions the turbulent
characteristics depend solely on the following dimension-
less parameters
          V
     R  = -"—   surface Rossby number                 (1)
      °   «0
                TFT  * To
     S  = -«—  . —	   external parameter for       (2)
          m     f* V P
           1     •"•  vg      thermal stratification

where V  is the geostrophic wind which could be assumed  to
be equal to the wind of  the top of PBL,  f - Coriolis para-
meter,  Z  - roughness length  of the  order of several meters
for a town, g - acceleration  of gravity, f  - the mean  po-
tential  temperature, TH  - TQ  -  the difference  between  the
potential  temperatures  of height H and ZQ since both exter-
nal parameters  (1)  and  (2)  are  fully  responsible for the
turbulent  state they must solely define  the  diffusion  pro-
cess in the PBL  (erf. [2]).  If we assume  that  the sources

-------
                      1817
J4UU
1300
1200
1100
1000
900
800
700
600
500
400
300
200

100
0

-100
-200
-Wf)

4
)-
-
| -
- s -
•• ••
"T
- • •! J
_ • * i
? *
I Jtf :
• ^ O
O o
o ou oo
•^•o
o
_ —

o
1 1 1 1 1 1 1 1 1
         234S678910
The dependence of air pollution data on the meteorolo-
gical parameters HO, S. • * o ,
slight pollution, respectively.
gical parameters H , S. • » o , strong, average and

-------
                            1818
of pollution around a particular measurement station do not
change and have constant strength

    -b— = F(RQ, S,K)                                  (3)

where b is the concentration measured at the given measure-
ment point, b  is the mean concentration for a given period,
angle K defines the direction of geostrophic wind which is
necessary only when the sources around the measurement
point are disposed unhomogeneously.Usually when measuring
in the centre of a town it could be reasonably assumed that
T-—  does not depend on the direction,
 m
3. Statistical analysis of the air pollution
     Fig.l gives an example for a similar systematization
of date. We have used the air pollution observations in
the towns Krasnoyarsk, Chita, Kursk, Leningrad, Svredlovsk,
Novosibirsk, Krivoy Hog, Irkutsk published in  [3j,J4J and [5] .
As a characteristic of air pollution we use the parameter
P =~£~» where m is the number of observations with a con-
centration of 1.5 above the mean seasonal value for the
period concerned. Since the mean P is approximately equal
to 0.2 for the differents towns, cases with higher concen-
tration of pollution are assumed P > 0.2 and are given ort
Fig.l with black circlets. The cases P < 0.2 are given with
white circlets and those P#» 0.2 with asterisks. It ia
obvious that with great Rossby numbers and little S we have
slight air pollution and vice versa.
4. Conclusion
     The systematization of air pollution data by means of
parameters (1) and (2) is very advantageous since those
parameters are solely responsible for the diffusion process
and could be obtained from the numerical weather forecast.
With a sufficiently reliable statistic it enables ua to
predict air pollution. It is easy to make such a statistic

-------
                              1819
 when  there are  enough experimental air pollution data,

 since parameters  (1)  and (2) could be readily obtained from
 the synoptic maps  or  data of aerology for the corresponding

 past  periods. The  baroclinic effects influence mainly the
 high  sources. In  such cases two more external to PEL baro-

 clinic  parameters  are added as shown in JJ2J . The statistic
 evaluation could  be improved when some other parameters

 are included in (2),  for example the parameter of inertia,
 giving  the extent  of  pollution from the previous day. When

 designing new enterprises we should be able to determine
 the concentration  patterns from them on the ground of syn-
 optic data. A theoretical approach to the solution of this
 problem is studied in [&] and [l].



References

1.    YORDANOV,  D.,  WIPPERMANN, F.,   "The parametrization of  the
     turbulent  fluxex of momentum,  heat and  moisture  at the
     ground in  a  baroclinic planetary boundary layer", Beitr.
     Phys. Atmos.,  45,  (1), (1972).

2.    WIPPERMANN,  F., YORDANOV, D.,   "A perspective  for a routine
     prediction of concentration patterns",  Atmos.  Environment,
     6,  (1972)

3.    SONKIN,  L.R.,  DENISOVA, T.P.,   "Meteorological  conditions
     responsible  for the formation of high concentration of  air
     pollution  in the  towns", Trudi GGO, 238 (1969)

4.    SONKIN,  L.R.,  HRAPATSCHENKO,  W.A., "Contents  of  inclusions
     in the towns in connection with the meteorological conditions
     in summer",  Trudi  GGO, 293, (1973)

5.    GENICHOVITSCH,  E.L.,  GUSCHTIN, W.A.,  SONKIN, L.R.,  "On the
     possibilities for  forecasting the air pollution  in towns
     using the  method of differentiation of  samples", Trudi GGO,
     293, (1973)

6.    YORDANOV,  D.,   "Prediction of concentration patterns  in the
     surface  layer under barotropic and adiabatic conditions",
     Arch. Met. Geoph.  Biokl., A,  22 (1973)

7.    YORKANOV,  D.,   "Prediction of concentration patterns  in the
     surface  layer by synoptical information",  Proc.  3rd Internatio-
     nal Clean Air Congress (IUAPPA) ,  Diisseldorf 1973.

-------
 EXPOSITIONSINDIKATOREN





 INDICATORS  OF  EXPOSURE





INDICATEURS  DEPOSITION





INDICATORI DI  ESPOSIZIONE





 INDICATOREN VAN EXPOS IE

-------
                              1823
        IL CONTENUTO DI PIOMBO NEI CAPELLI COME INDICE
           DI ESPOSIZIONE AL TOSSICO NELL'AMBIENTE

               S,  CACCURI, L. PECORA E A. ROSSI
Istituto di Medlcina del Lavoro dell'Universita di Napoli, Italia
R1ASSUNTO

     Data la oreeeente di.ffuei.one del piombo nell' ambiente vi 8
oggi la possibility di into88icu.zi.one eaturnina non eolo per gli
operai ahe vengono a contatto con il toeeioo per motivi di
lavoro, ma anohe per le popolaxioni oivilif con gravi ooneeguenze
per quanta riguarda i bambini.

     Allo eeopo di oontribuire al oontrollo aia dello etato di
eapoaizione aia dei oaei iniziali di intoeeicazione preolinioa
ei 6 etudiato ee la rioeroa del piombo nei oapelli puo' ooati-
tuire un indiee di espoaizione e di intoeaicazione.

     In un gruppo di aoggetti eaturnini e in un gruppo di aoggetti
eoelti a oaso fra la popolasione ai e determinate il taaeo del
piombo nel aangue, nelle urine e nei oapelli. E' a tat a anohe
pratioata la rioeroa delle protoporfirine libere eritrooitarief
delle aoproporfirine urinarie e dell'ALA urinaria.

     Si d inoltre determinate il piombo nei oapelli di gruppi
di aoggetti in quartieri diverai della oitta di Sapoli.

     E' etato trovato ohe vi 6 una differenza aignifioativa fra
la quantity di piombo nei oapelli dei aoggetti normali e dei
aoggetti eaturnini. Inoltre il oomportamento di queeto parametro
rieulta molto eimile a quello degli altri parametri etudiati, e

-------
                              1824

che vengono oomunemente impiegati per la diagnosi biologica di
saturnismo.

     Non sono state trovate difference significative nel conte-
nuto di piombo nei capelli fra gli abitanti di quartieri diverai
della cittd.

     In conclusione, da questo  studio risulta ohe la ricerca
del piombo nei capelli puo ' essere un mezzo molto utile per le
indagini sul saturnismo e riguaranti grandi masse di popolazione.
ABSTRACT

     In view of the •increasing spread of lead in the environment
it is now possible for lead poisoning to affect not only workers
who come into contact with the poison in the course of their
work, but also the general population, with serious consequences
for children.

     To assist the investigation of the exposure situation and
initial cases of pre-clinical poisoning^ a study was made of
whether the  lead content of hair could be used as an index of
exposure and poisoning.

     The level of lead in the bloodj urine and hair was determined
in a group of subjects suffering from lead poisoning and in
another group chosen at random from the population. The free
protoporphyrins in red blood cellss the urinary  ooproporphyrins
and  the urinary ALA. were also measured.

     In addition^ the lead content was measured in the hair of
groups of subjects in different parts of the city of Naples.

     It was found that there was a significant difference betwe&n
the  level of lead in the hair of normal subjects and in that of

-------
                               1825
subjects suffering from lead poisoning.    This parameter was
Very similar to the other parameters investigated,  which were
those generally used in the biological diagnosis of lead poisoning,

     No significant differences in the lead content of the hair
were found between the inhabitants of different parts of the
city.

     To conclude3 this study showed that determination of the
lead content in hair might be a very useful method for studies
of lead poisoning involving large sections of the population.

-------
                             1826
     Tntroduzione
     I metalli pesanti costituiscono una fonte importante e
seropre piu pericolosa di inquinamento ambientale. Per quanto
riguarda il piombo e atato calcolato che negli ultimi 100 anni
ne aono atati immeaai nella bioafera oltre 100 railioni di ton-
nellate.
     Gib porta ad vans, creacente eapoaizione e ad tin aempre piti
forte aaaorbimento del tossico da parte degli organistni viven—
ti, per cui la poaaibilita di intoaaicazione non riguarda piu
aolo coloro che vengono a contatto per motivi profeaaionali,
ma tutta la popolazione civile e in particolar modo i bambini,
piu aenaibili al toaaico apecie per quanto riguarda il siatema
nervoso.
     Oltre le prowidenze per ridurre le fonti di inquinamento
e limitare I'iramiaaione nell'ambiente, fc necesaario non aolo
oontrollare il grado di eapoaizione delle popolazioni civili
ma anche il taaao di aaaorbimento nei aingoli individui e gli
eventuali atati di intoaaicazione.
     Sia per il pran numero di aoggetti da eaaminare, aia per
le difficolta pratiche del prelievo del sangue e della raccol-
ta delle urine, apecie nei bambini, aarebbe piu agevole e piu
pratico un controllo eaeguito attraverao la ricerca del piom-
bo nei oapelli.
     Perche cib aia poaaibile fe neceaaario accertare un. effet-
tivo e coatante rapporto fra la quantita di piorabo nell'ambien
te, il grado di aaaorbimento, la piombemia, la piombiuria e
gli altri parametri biologici impiegati per la diagnoai di sa-
turnismo, con la quantita di piombo nei capelli.
     Sinora pochi autori ai aono occupati di queato argomento.
Pra gli altri Kopito ha riportato 5 caai di bambini nei quali
I1intoaaicazione saturnina fu aospettata per l*elevato tasso
di piombo trovato nei capelli. Hammer ha trovato che la con-
centrazione di piombo ed altri metalli peaanti nei capelli di
5 citta in USA riflette fedelmente i diversi livelli di eapo-
aizione ambientale.
 Ricerche
     Tenendo conto di queate premeaae, abbiamo atudiato il con
tenuto di piombo nei aangue, nelle urine e nei capelli di due"
gruppi coatituiti da 42 aoggetti normali e 41 soggetti satur-
nini. Sono atati anche rinercati i taaai delle protoporfirine
libere eritrocitarie(PLE), delle coproporfirine urinarie(CPU)t
e dell'acido 'Is^.ta-aminolevulinico urinario(ALAU).
     E' stato anche determinate il contenuto di piombo nei ca—
pelli, -n gruppi diverni di vari quartieri della  c^tt^ di Napo-
li.
     I riaultati della prima inda/rine 3ono riportati nella ta-
bella I. Si ^ trovato che i dati medi per ciaaoun parametro

-------
                                              TABELLA I
Soggetti
normal! (42)

Sogretti
saturnini(41)
                  Sargue       Urine         Capelli
                    Pb           Pb            Pb
               microg/100ml  microg/24 h  microg/g peso
                                                   secco
                                              PLE
 37+15


118±24
29±18


132+27
42+15


114+23
globuli rosai

   24±19


  201+38
                                               CPU        ALAU

                                           microg/24 h   mg/?4
82+15
3,2+1*2
172+20     7,4+2,3
                                                                                                             00
                                                                                                             Is)
               Quartiere
                                               TABEI.IA II

                                Contenuto dl. Pb nei capelli(microg/g peso secco)
                           in gruppi di vari quartieri della citta di Napoli
               S.Lorenzo
               Chiaia
               S.Ferdinando
               Ponticelli
               Barra
               Secondigliano
               S.Giovanni
           N°sog,e;etti

               11
               15
               10
               12
               11
               10
               12
                          Pb

                         41 ±7
                         38±2
                         37*8
                         44±s
                         43+P
                         4U10

-------
                               1828
nel gruppo dei sog^etti saturnini sono stati sempre e signifi-
CPtivamente superior! ai dati cmrispondenti nel gruppo disog-
petti normal.!. Per quanto riguarda il contenuto di piombo nei
capelli la media nei r.aturnini fe risultata superiors di circa
tre volte quella dei soggetti normali. Questo rapporto e quasi
uguale a quello trovato per il piombo nel sangue e nelle urine
( circa 4:1), per cui si pub ritenere che la validity del ri-
lievo e quasi la stessa per i tre pararaentri.
II tasso delle PLE nei saturnini e risultato rnolto elevato ris-
petto ai normali ( 8:1 ), mentre per quanto riguarda le CPU e
1'ALAU il rapporto e circa uguale a quello trovato per i ca-
pelli.
     I risultati della seconda indagine sono riportati nella
tabelln II, e non mostrano difference significative fra i
gruppi dei vari quartieri cittadini, per quanto riguarda il
contenuto di pjombo nei capelli,
Conclusion!
     I risultati da noi ottenuti prova.no che la ricerca del
piombo nei capelli pub essere considerate un test valido dell*
assorbimento  organico  del piombo« Inoltre la facilitk del pre-
lievo, la possibllita  della conservazione e della ripetizione
lo rendono adatto come primo esame su grandi masse di popola-
 zione. Esso rende possibile di individuare  facilraente i sog-
 getti  che possono essere coniderati normali e quelli che do-
 vranno essere sottoposti ad ulterior! accertamenti per quanto
 riguarda 1'assobimento di piombo e 1'eventuale intossicazio-
ne.  Inoltre la omogeneit^ dei dati di base, provata dal fatto
 che  non sono  state  osservate differenze fra i gruppi nella po-
 polazione di  una grande citta, rende piu agevole la individua-
 zione  dei oasi di aumentato assorbimento. Questo dato aumenta
 1'utilitk dell'impiego della ricerca del piombo nei capelli
 come indagine di massa.
 Bibliografia
 1.-   KOPITO L.,BRILEY A.M. ,SH\VACHIilAN H., "Chronic Plumbism in
       Children"  JAMA, 209, 243 (1969)
 2.-   HAMMER D.T.,PTNKLEA J.F., HENDRICKS  R.H.,SHY C.M." Amer.
       J. Epidemiol." 93, 84 (1971)

-------
                                 1829
               APPLICATION OF LEAD MOBILIZATION
                   FOR EXPOSURE EVALUATION
              LJ, GRAOVAC-LEPOSAVI6, D, DJURl6,
                 S, MILIC AND V, VALJAREVIC
Institute of Occupational and Radiological Health, Belgrade,
Yugoslavia
ABSTRACT

     Various parameters -indicating lead exposure are followed.
The investigation has been performed on following groups of
persons showing various levels and duration of exposure:

I Group:   workers exposed to high lead concentration for some
           months with acute clinical signs.
II Group:  workers from the same place but without severe cli-
           nical symptoms,
III Group: workers with chronic lead poisoning.
IV Group:  workers suspected  of lead poisoning.
V Group:   inhabitants living around a lead smelting plant.

     For studied groups the following analyses are presented:

     Hematologic determination in blood:  Hbt erythrooytes, SRC,
ETC.   Toxicologio determination in urine:  ALA, coproporphyrin,
lead.   Lead mobilization by  1 injection  (for children h amp)
of Ca Na^EDTA.   Lead in urine was determined before injection
and in 24 hours sample after.

     Analyzing obtained data  it is quite  obvious that mobiliza-
tion of lead represents a very useful method, importance of

-------
                                1830
which depends on the level and duration of exposure.    In acute
and chronic poisoning all parameters are positive, including
mobilization yield.   On the other hand in the case of low but
long exposure (inhabitants and workers in group IV) mobilisation
yield represent the most characteristic parameter pointing out
the integral lead exposure.

-------
                              1831
     For evaluation of the lead exposure many "exposure tests"
are available.  Generally, such tests are based on the direct
determination of lead in biological samples (blood, urine) or
on determination of the effects of toxic action of lead on
porphyrin synthesis or on erythropoiesis.  In environmental and
occupational health, however, very often it is necessary to
evaluate the integral exposure during the life.  For such in-
vestigation all existing exposure tests are not useful.  Recently
some papers appeared suggesting evaluation of the lead body
burden in the skeleton using "mobilization" by CaEDTA injection
and determination of the quantity of excreted lead  (mobilization
yield)  (1,2,3,4,5,6,7,8,9).
     Here is presenting a study on various parameters, character-
istic for lead exposure and mobilization yield in some groups
of workers and a group of population with different level and
duration of exposure. I group: 22 monters hospitilized in the
Institute due to severe signs of poisoning (acute abdomen, lead
colics, adinamy or digestive disturbances).  They worked some
months, exposed to high lead concentration 0,5-2Omgr/m .
     II Group: 13 monters from the same place with slight clinical
signs reached the Institute 2 months after the end of work with
lead exposure III Group: 31 workers exposed to lead 5-10 y. in
battery factories treated into the Institute due to chronic lead
poisoning. IV Group: 30 workers lead welders, who were examined
as suspected on lead poisoning.
     V  Group: 124 inhabitants of a valley living 1-6 km from
lead smelting plant emitting more than 20O t. of lead aerosols
Concentration of lead in air was registered from 1,3-85 micrograms/
m  .  In these persons ALA excretion was  from 1,4-42 mgr/1 varying
in the  course of years.  In this group 2O males, 37 children and
67 females are included.  In all persons mentioned  analysis has
been performed except ALA for groups I and II.

Results
     Obtained results are presented in a series of  graphs as
percentage of cases with values out of normal limits.  Per-
 centage of cases with  decreased Hb values  under 80%  (for women

-------
                             1832
and children 75%  is presented on graph  I.  The lowest values
of Hb were registered in group I (64% under 75% Hb).
                         GRAPH  I

              PERCENTAGE OF CASES WITH  DECREASED
                    Hb VALUES UNDER 80%
       wo
        90
        80.
        70.
        SO
        50
        40
        30
        20
        W
                  II
        III
examined
IV
                                   m
                                   HI
                           \A/
                           VY
                                   ~..-Jk:	
                               croups
         FOR WOMEN AND CHILDREN UNDER 75'/.Hb.
    Percentage of persons with decreased number  of Er  under
3.900.000 (for women and children  3.6OO.OOO) is  presented on
graph II.   In the group I also the  lowest absolute numbers are
evident,  in 45% of examined less than  3 mil.  In group iv and
inhabitants decrease of Er is  practically negligible.

-------
                             1833


                          GRAPH II


            PERCENTAGE OF CASES WITH DECREASED
               NUMBER OF Er.  UNDER 3.90O.OOO
100-
90
80
70
60
50
40
30
20
10

'•

















I














II





III



r-i-J-tn-,
m \ c\ w
IV V
examined groups
      FOR WOMEN AND CHILDREN UNDER 3.500.000 Er.
         On graph III the percentage of cases  with  the increased
number of stipled red cells (SRC)  over 300 on  million Er.(for
population over  100) is presented.  Comparing  the absolute number
group I is showing 32% cases with over 2.OOO RSC and group III
only 19%.  In  population in 65% of males,  73%  children and 55%
of women we found increased number of RSC.  To add  that  in 54%
of children number of RSC was over 300.

-------
                            1834


                          GRAPH III
           PERCENTAGE OF CASES HAVING INCREASED
              NUMBER OF SRC OVER  300  (MILLION Er.)
       o/
       /e
     ea
     70.
     so
     40
     30-
     20-
     10-
                II
III
             examined
IV
™ jg;
        groups
     FOR POPULATION OVER 100 SRC MILLION Er.
           Next graph is presenting the percentage of cases with
increased number of RTC over 1,6%.  Group I and III are showing a
moderate percentage of increased values.  In other groups per-
centage of cases is negligible.

-------
                          1835
                       GRAPH IV
      PERCENTAGE OF CASES HAVING INCREASED
      VOLUESOF RTC OVER  1.6%
       60.
       50.
       40.
       30.
       20
        10
                  II
III
IV
                examined
m jc  \w\
   V
        groups
          Graph  Vis presenting the percentage of persons with
an increase of coproporphyrin excretion over 500 yg/1  (for pop-
ulation over 20O yg/1).  The highest values are found  in group
It in 27% examined results are from 2OOO-5OOO Hg/1.

-------
                          1836
                        GRAPH V
       PERCENTAGE OF CASES WITH INCREASED COP-
       ROPORPHYRIN EXCRETION OVER 500mq/l
       90.
       80.
       70
       50
       40
       30
       20
       10.
                 II
            examined
III
IV
m ' c \ w
    groups
      FOR POPULATION OVER 200mq/t
    Graph VI is presenting percentage of cases with increased
excretion of ALA over 2O mgr/1 (for population 1O mgr/1).  Results
for  group I and II are missing.  In group III all examined had
values from 20-100mgr.

-------
                         1837
                      GRAPH VI
      PERCEN TAGE OF CASES WITH INCREASED
      ALA  EXCRETION OVER  20mq/l
     100-
       90'
       80-
       70
      50-
      40.
      30.
      20-
       10-
                 II
III
IV
            >    i
      m  « c ! w !
                                    V
               examined
         groups
          FOR POPULATION OVER  Wmq/l
    Graph VII is presenting the percentage of cases with in-
creased values over 12O ug/1 of spontaneously excreted lead and
the "mobilization yield".   In group I excretion is  increased in
78% of examined and 32% of them showed quantities over 3OO pg/1,
In group III the increase of excretion is registered in similar
percent  (71%) but only 3,2% of examined had the increase over
300 yg/1.

-------
                                 1838
    a/
    /o
100
 90
 BO
 70
 60
 50
 40
 30
 20
 10
        PERCENTAGE OF CASES WITH SPONTANEOUS EYCRETED LEAD OVER
          120mg/l AND MOBILISATION YIELD OVER 10OO mg/1
                       r   ] spont. pb
                II
Iff
IV
           e  x a m i n e d    groups
                                              m
                                   w
             FOR  POPULATION MOBILISATION YIELD OVER 5OOmg/l
         In inhabitants 6% showed increased values of lead  excretion.
    As normal mobilization yield is accepted for workers  l.OOO
    of lead and  for population 500 yg/1.   In all groups high per-
    certage of cases with increased mobilization yield is evident
    but a significant difference between  groups exist in  the quantity
    of mobilized lead.  For example, in group I in 86% of cases
    mobilization yield is 2.0OO - 4.OOO Mg/l, and in  9% even more
    than 4.000 yg/1.  In group III values over 4.000  yg/i are regis-
    tered in 54% of examined.  In population 86% of men, 96% of

-------
                              1839
children  and 78% of women excreted more than 5OO yg/1 of lead.
The yield of l.OOO - 2.OOO yg/1 is registered in 41/3% men,
61% children and 18% women.

Discussion
     Comparing parameters for each group some interesting obser-
vation could be made.
     In group I with clinical signs of acute poisoning/ all
parameters are changed in high percentage of examined.  Also
the absolute value of each parameter is showing the greatest
deflection from the normal value.
     In group II with slight clinical  symptoms the absolute
values and percentage of deflection are smaller in relation to
the first group.  Here the spontaneous excretion of lead  re-
presents the most characteristic  finding*

     In group III  (chronic lead poisoning) ALA excretion  repre-
sents the most important findings; increase  is established in
all examined.  In this group is lacking the  parallel  increase
of coproporphyrin excretion.  Comparing results obtained  for
group I and III it  is possible to conclude that the percentage
of cases with increase SRC and RTC in  both groups are similar.
In Hb, Er and coproporphyrin expressed difference in  percentage
is evident but it is not statistically significant.
     In group IV  (workers  exposed to lower lead level)  only
decrease of Hb and  increase of ALA represent characteristic
parameters.
     Exposure of population of lead is characterized  by two
parameters: SRC  and ALA.   ALA  increase in population  is more
expressed than in workers  of IV group. This could be explained
probably by the  fact that  inhabitants  of valley are exposed to
lead 24 hours per day from birth.
     We  followed  all parameters  in inhabitants of valley  for
5 years  consecutively and  found  great  variation each  year,
according to  technologic  and meteorologic conditions.  But
determination of  mobilization  yield gave us  real apprehension of
the size  of  the  integral  exposure from birth.

-------
                               1840
References


1.   TEISINGER J., SRBOVA J., : Diagnosticka mobilizace olova
     pomoci CaEDTA, Pracov.  Lek. 10  (1958) 402.

2.   TEISINGER J., SRBOVA J. , : Mobisization of lead by CaEDTA  in
     the Diagnosis of Lead Poisoning, Brit. J, industr. Med.
     .16  (1959) 148.

3.   TEISINGER J.,: Zakus'enosti s ambulantnim vys"etrovanim
     diagnosticke mobilizace olova, Pracov. 16  (1964) 145.

4.   PREROVSKA I.,: Vztah spontane vyludovaneho olova v moci a
     olova vylutfovaneho po CaNa-EDTA, Pracov. Lek. 21  (1969)361


5.   TEISINGER J., PREROVSKA I., ^EDIVAC V.,: Stanoveni depozita
     olova v tele u experimentalni otravy olovome, Pracov. Lek.
     19_  (1967) 251.

6.   TEISINGER J., &EDIVED V., PREROVSKA I., FLEK J., ROTH z.,:
     Attempt on Determination of Biologically Active Lead in
     Organism in Experimental Poisoning, Int. Arch. Gewerbepath.
     Gewerbehyg 25. (1969) 240.

7.   PREROVSKA I., TEISINGER J., : Excretion of lead and its
     Biological Activity after Termination of Exposure,
     Brit. J. Industr. Med.  27 (1970) 352.

8.   STANKOVIC M., PETROVIC  Lj . , POLETI J., : A Contribution
     to the Laboratory Diagnostics of Early Saturnismus, Arh.
     Hig. Rada .13  (1962 189.

9.   GRAOVAC-LEPOSAVlfi Lj.,  DJURIC D., VALJAREVl6 V., SENICAR  H. ,
     SENIOR Lj., MILIC S. ,  DELI<5 V.,: Environmental Lead
     Contamination of Mez*a-valley.  Study on Lead Exposure of
     Population,  Proc. Int.  Symposium "Environmental Health
     Aspects of Lead, Amsterdam, October 1972, p. 685.

-------
                              1841
          BEZIEHUNGEN ZWISCHEN RAUMLUFTKONZENTRATION,
            BLUTSPIEGEL UND METABOLITENAUSSCHEIDUNG
                   AROMATISCHER LOSUNGSMITTEL

                         D, SZADKOWSKI

Zentralinstitut fiir Arbeitsmedizin der Gesundheitsbehorde der
Freien und Hansestadt Hamburg, BED
KURZFASSUNG

     Urn den Beziehungen zwischen dem Genalt von -Toluol der
Raumluft, seiner Konzentration in Blut und der Metabolitenaus-
eoheidung im Earn naehzugehen, wurden bei 94 BesohUftigten einer
Druokerei Bestimmungen dieses Litsungsmittela im Blut und seines
Metdboliten Hippursaure in Harn durohgefilhrt und den gle-ifohzei-
tig erfolgten Raumluftmesaungen gegenubergestellt, Bei einen
Veraleiah der aeneseenen Toluolkonzentrationen in der Raumluft
mit dem MAK-Wert filr Toluol ereohien arbeitsnedizinisoh-toxiko-
loaisoh die Belastung relativ gering.   Trotzdem Daren die mitt-
leren Toluolkonzentrationen im Blut gegenttber den Werten einer
Kontrollgruppe stark erhtiht.

     Es zeigt sich, dass eine Beurteilung der Gefdhrdung am
Arbeitsplatz auf Grund von Raumluftmessungen mit erhebliohen
Vnsioherheiten behaftet ist.   Die Hippursdureaussoheidung im
Harn is ausserberufliohen Einfltlssen unterworfen.   Diese Bin-
eohrtlnkungen gelten fttr eine L0 sung emit telbestimmung  im Blut im
wesentliahen nicht.   Bin vorlttufiger Grenzwert fUr einen
arbeitsmedizinisah noah akzeptablen Bereioh der Toluolkonzentra-
tion wird auf Grund der Untersuohungsergebnisse vorgelegt.

-------
                               1842
ABSTRACT

     In order to investigate the relationships between the
toluene content of air in an enclosed space, toluene concentra-
tion in the blood and excretion of toluene metabolites in the
urine, the levels of this solvent in blood and of its metabolite
hippuvio acid in urine were determined in 94 employees of a
printing works and compared with toluene levels in the  air in
the room measured at the same time.   Compared with the MAC
value for toluene* the atmospheric toluene concentration measured1
in the room seemed to indicate relatively slight exposure from
an industrial medical and toxicological viewpoint. Despite this,
the average toluene levels in the blood were very much higher
than the figures for a control group.

     It is clear that considerable uncertainty attaches to the
assessment of health hazards at places of work from measurements
of the room air concentrations.   The excretion of hippuric acid
in the urine io affected by non~occupational factors.   In the
main, these reservations do not apply to the determination of
the solvent in the blood.   On the basis of the experimental
results, a provisional limit value is proposed for a range of
toluene concentrations which is acceptable from the standpoint
of industrial medicine.

-------
                             1843
FUr eine Uberwachung des Arbeitsplatzes 1st es relativ
einfach, die Konzentration aromatischer Losungsmittel in
der Raumluft zu bestimmen. Bin Vergleich dieser Raumluft-
analysen mit den zugehbrigen MAK-Werten ermoglicht in
vielen Fallen eine Beurteilung des gesundheitlichen Risikos.
Man mufl sich Jedoch dariiber im klaren sein, dafi diesem Vor-
gehen nicht mehr als eine orientierende Bedeutung zukommen
kann. Individuelle Faktoren, etwa das Atemminutenvolumen,
lassen sich nSmlich hSufig nicht gebtihrend beriicksichtigen.
Bestimmungen des jeweiligen Schadstoffes Oder seiner
Metaboliten im biologischen Material geben daher eine
bessere AussagemSglichkeit zur individuellen Belastung.
Auf diese Veise kbnnen n^mlich nicht nur inhalative Re-
sorptionsquoten mit erfaflt werden, persSnliche Faktoren
oder spezielle Arbeitsplatzsituationen gehen ebenfalls
integrierend in den Endwert ein.

Urn den Beziehungen zwischen dem Gehalt von Toluol der
Raumluft, seiner Konzentration im Blut und der Metaboliten-
ausscheidung im Ham nachzugehen, wurden bei 94 BeschSftig-
ten eines Tiefdruckbetriebes Bestimmungen dieses LSsungs-
mittels im Blut und seines Metaboliten HippursSure im Ham
durchgefUhrt und den gleichzeitig durchgeflihrten Raumluft-
messungen gegenUbergestellt. Die Toluolbestimmungen in Luft
und Blut erfolgten gaschromatographisch nach einer anderen-
orts von uns beschriebenen Methode. Die Hippursaurebestim-
mungen erfolgten nach dem Verfahren von Lieberman.

Die Toluolkonzentrationen in der Raumluft schwankten Je
nach Arbeitsplatz zwischen 318 mg/m^ und 935 mg/m^. Die
mittlere Toluolkonzentration im Blut von 68 Beschaftigten
im Tiefdrucksaal lag bei 159,1 tig/100 ml (s « 107,0) und
von 11 Tiefdruckern im Andruck bei 161,5 ^ig/100 ml
(s = 118,4), wShrend die mittlere Toluolkonzentration
im Blut bei einem Kontrollkollektiv (n = 30) von BUroange-
stellten 5,3 |ag/100 ml (s = 4,2) betrug. FUr die Hippur-

-------
                              1844
saureausscheidung im Harn ergaben sich in der gleichen
Reihenfolge 3,7 g/1 (s = 2,2), 2,8 g/1 (s = 1,8) und
0,82 g/1 (s = 0,4).

Bei einem Vergleich der gemessenen Toluolkonzentrationen
in der Raumluft mit dem MAK-¥ert fiir Toluol erschien
arbeitsmedizinisch-toxikologisch die Belastung relativ
gering. Trotzdem waren die mittleren Toluolkonzentrationen
im Blut gegeniiber den Werten der Kontrollgruppe stark
erhoht, die Unterschiede lieBen sich varianzanalytisch
mit einer Irrturnswahrscheinlichkeit von p < 0,01 sichern.
Inwieweit hier temporare Erhb'hungen der Toluolkonzentra-
tion in der Raumluft (gemessene Spitzenwerte lagen bei
1872 mg/m^) oder die Tatsache eine Rolle spielte, da3 die
Drucker ihre stark verschmutzten Hande regelmaflig in
Toluol wuschen, mag dahingestellt bleiben. Unverkennbar
ist jedoch, daB eine Abschatzung der Gefahrdung am Arbeits-
platz allein auf Grund von Raumluftmessungen einen groBen
Unsicherheitsfaktor impliziert. Eine Beurteilung auf
Grund der Hippursaureausscheidung im Harn ist jedoch auch
nicht ohne Probleme. Es ist bekannt, daB die Hippursaure
nicht nur ein Abbauprodukt von Toluol ist, sondern eine
vermehrte Ausscheidung auch nach dem Genufl bestimmter
Nahrungsmittel, etwa Backpflaumen oder Preiselbeeren, zu
beobachten ist. Auch die zu Konservierungszwecken manchen
Nahrungsmitteln zugesetzte Benzoesaure bewirkt eine er-
hohte Hippursaureausscheidung. So kann es geschehen, dafl
die Hippursaureauscheidung zu Beginn einer Arbeitsschicht
hoher liegt als bei Schichtende, wie wir es in einem kiirz-
lich durchgefUhrten weiteren Untersuchungsprogramm an
Tiefdruckern in Einzelfalien sahen.

Die stark erhohten Toluolkonzentrationen im Blut lassen
zunachst nur die Aussage zu, daB es sich hierbei urn eine
vermehrte Belastung der Exponierten handelt. Um nun dariiber

-------
                            1845
hinaus auch konkrete Beurteilungskriterien iiber eine
beginnende Vergiftung in die Hand zu bekommen, haben wir
neben den quantitativen Toluolbestimmungen im Blut
samtliche fur eine Beurteilung der Leberfunktion rele-
vanten Parameter untersucht, wobei aus verstandlichen
Griinden bei einer derartigen epidemiologischen Studie
von Leberbiopsien Abstand genommen werden muflte. Vorbe-
haltlich der noch nicht vbllig abgeschlossenen Auswertung
kann schon Jetzt gesagt werden, daB es bis zu Toluolkon-
zentrationen im Blut von 400 [ig/100 ml nicht zu klinisch
oder labordiagnostisch objektivierbaren Leberschaden
kommt. Man wird diesen Wert zunachst als einen vorlaufigen
Grenzwert ansehen kb'nnen, der den arbeitsraedizinisch
akzeptablen Bereich charakterisiert. Darliber hinaus hoffen
wir, durch eine Verkniipfung der Toluolkonzentration im
Blut mit der zugehorigen Hippursaureausscheidung im Harn
zu noch differenzierteren Aussagen zu kommen. Dazu sind
jedoch noch weiterfiihrende Untersuchungen und statistische
Analysen notwendig.

-------
             GEWEBSMESSUNGEN





           TISSUE MEASUREMENTS





MESURES RELATIVES AUX TISSUS B10LOGIQUES





      MISURE NEI TESSUTI BIOLOGICI





      METING VAN BIOLOGISCH WEEFSEL

-------
                               1849
         HOW USEFUL ARE THE BULK OF ENVIRONMENTAL DATA?

                         JON C,  VAN LOON

Institute of Environmental Studies and the Departments of Geology
and Chemistry, University of Toronto, Canada
ABSTRACT

     It ia crucial -in the setting of environmental or health
standards that a reliable data bank be available.    Unfortunately
a review of the literature on most pollutants reveals that there
ia frequently no way to Judge the validity of the  numbers which
are given.   Reasons are poor documentation and control of samp-
ling, sample preparation and the subsequent analysis.

     The inadequacies of data have been demonstrated frequently
but never more clearly than in the case of mercury pollution prob-
lems around the world.   Mercury analyses have been done routinely
for most of this century.   Much of the work was undoubtedly car-
ried out using so-called "standard methods"t commonly available
from health and pollution control organizations.   However* it
was only subsequent to the 195$ Minamata incident  that the true
chaotic nature of mercury data was revealed.   In  fact the liter-
ature shows that as recently as 1971 little agreement could be
found even on the choice of the analytical method.

     In this presentation evidence of deficiencies in other health
related data banks e.g. blood lead values, will also be discussed.
Suggestions will be made regarding a more rational approach in the
future.

-------
                                 1850
1.  INTRODUCTION
    Reviewing the literature for environmental chemical  data  reveals
    that there is frequently no way to judge validity of the  numbers.
    Reasons are poor documentation and control of some,  most,  or all
    aspects of the analytical process including sampling,  chemical
    analysis procedures, and data dissemination.  These inadequacies
    cause serious problems when government standards for environmental
    quality have to be decided upon and enforced.
2.  DOESJJJ5EOF A "STANDARD METHOD" GUARANTEE USEFUL DATA?
    It is commonly believed that use of a "Standard Method" (one
    published by a number of health and pollution control organizations)
    is sufficient to validate data produced.  In published environ-
    mental studies it is common place to find that the description of
    analytical methods and  assessment of the validity of the results
    consists only of a reference to the use of a "standard method".
         A number of authors (e.g. ref.3,4) have provided data to show
    that the use of standard methods does not necessarily result in
    accurate results.  In addition performance of a laboratory must be
    continually be evaluated through use of interlaboratory control
    samples, "standards" and participation in round robin studies.
3.  MERCURY - A GOOD EXAMPLE OF INADEQUACIES IN DATA
    What happened when high levels of mercury were found in fish is a
    good example of the problems to be faced in setting and enforcing
    environmental standards.  When the dangers of mercury were realized,
    a limit was put on the amount allowed in fish flesh offered for
    sale in Canada and the United States.  Because of the almost
    complete lack of valid data, however, the selection of an 0.5-ppm
    level, was, at best, an educated guess.  Fortunately, as the vast
    quantities of useful data now accumulating are evaluated, this
    arbitrary, choice is being justified.

-------
                              1851
     In establishing permissible levels for a substance such as
mercury in fish, many questions must be asked.  What are the normal
background levels for mercury in various fish species? What levels
of mercury can be safely tolerated by the human population in
question?  Is there analytical data of suitable validity to allow
sensible answers to either of these questions?  When officials were
faced with the task of establishing the permissible mercury content
of fish, the answer to this last question was an emphatic no !
     The inadequacy of mercury data is well documented in the
literature.  The Niigata report (2) in 1967 shows variations
between neutron activation (NA)  and dithazone mercury values, for
the same sample, of up to 100%.   In 1970, 13 laboratories analyzed
a mercury spiked, flour sample by the same method (MA) (5). A mean
value of 4.6 ppm was obtained with a range of values for 10 of the
13 laboratories of 1.4 ppm.  The other laboratories were outside
this range/ one giving a value deviating from the mean by 50 timesI
In 1971 two homogenized fish samples were analyzed by (NA) and
flameless atomic absorption spectroscopy (FAAS) (6)   Results showed
that although the mean values for each technique were close,
1,37 ppm (FAAS), 1.36 ppm (NA) for sample 1 and 0.11 ppm  (FAAS),
                  *
0.10 ppm (NA) for sample 2 there was a wide range of values sub-
mitted by the laboratories, 0.9 to 1.8 and 0.03 to 0.21 respectively.
     In light of the variability of this data one might ask how it
is possible to enforce a limit of 0.5 ppm mercury for fish flesh.
A recent evaluation of the performance of laboratories engaged
in mercury analysis has shown that with a careful selection of
good laboratories, data in good agreement for standard samples can
be obtained  (1).  Gelatin samples spiked with known amounts of
mercury .were sent to 18 laboratories.  Results collected from all
these laboratories on a sample spiked with 2 ppro mercury showed a
mean of 2.02 ppm and a maximum range of from  188 to 2.18 ppm - a
variation of 15% at this level of concentration.   This is a vast
improvement over performances at earlier times recorded above.

-------
                               1852
     In spite of this, it could still be asked whether the same
laboratories would perform with the same reliability after doing their
own sampling and then analysing a natural sample such as a fish.
Fortunately samples more like those chemists actually test, such as
leaves, liver, and fish flesh, are now being prepared and tested
as standards for mercury analysis.  How laboratories will perform
on these samples is yet to be seen.  Because of this, it still
remains a question in this writer's mind how closely the 0.5 ppm
limit set for mercury in fish flesh can be reasonably enforced.
  4.  WHAT ABOUT THE FUTURE?
     Few environmental pollutants have been the subject of such an
intensive appraisal of data validity as mercury.  Fortunately, in
this case the arbitrarily chosen limit of 0.5 ppm seems to be
standing the bombardment of valid data.  One must, however, worry
about sudden needs to set limits in the future for other substances
in the environment.  Will we have the same luck as we did with
mercury?  Will skilled analytical chemists help us approach the
problem in a more rational way?  Or will inadequacies in chemical
data force us to guess about questions vital to health?
REFERENCES
(1)   Anderson, D.H,Murphy, J.J.White, W.W. Anal.Chem., 1972 44,2099
(2)   Kawasaka,!., "Report on the Cases of Mercury Poisoning in
     Niigata, Part 11,"1967, Ministry of Health and Welfare,
     Tokyo,  Japan.
(3)   Les,  G.F.,  Analytical chemistry of plant nutrients,  in
     Eutropication Causes, Consequences, Correctives." 1969,
     National Academy of Sciences.
(4)   Stanley, G.H.Richardson,G.R.,"Automation of the Single Reagent
     Method  for  Total Phosphorus, Advances in Automation Analysis,
     1972, Technician International Congress Future Publishing Co.,
     Mount Kisco, New York, N.Y.
(5)   Tugsuval, A.,  Merten, D., Suschny,O., "The Reliability of Low
     Level Radio Chemical Analyses," MARCH 1970 IAEC Vienna,Austria.
(6)   Uthe, J.F.,  Armstrong, F.A.J., Tam,K.C. "Determination of Trace
     Amounts of  Mercury in Fish Tissue:  Results of a North American
     Check,"  J.Ass.Offic.Anal.Chem., 1971,866.

-------
                              1853
           AUSVIRKHNfiEN DEP BLEI-UMWELTKONTWINATION
             AUF LEBENSMITTEL VON SCHLACHTTIEREN;
          EIN BEITRA6 ZUR PROBENAHME UND BEURTEILUNG

           w, KREUZER+, p, WISSMATH"*"/ B, SANSONI++,
                  W,  KRACKE"1"1" UND F, DIETL++

+  Bereich fur Hygiene und Technologic der Lebensmittel
   tierischen Ursprungs der Universitat Munchen, BRD
++ Gesellschaft fiir Strahlen- und Umweltforschung mbH, Institut
   fiir Strahlenschutz, Gruppe Radiochemie und Analytik,
   Munchen, BRD
KORZFASSUNG

     Von 139 Rindern aus e-inem vorwiegend landwirtechaftlioh
etrukturierten Gebiet dee sehwttbisch-bayerisQhen Voralpenlandea
und 22 Tieren aua etttrker bleiexponierten Arealen der BRD wurden
in Fleiechf Leber und Niere von jedem Tier naeh nasser Minerali-
sation mit H^O^/Fe   der Pb-Gehalt flammenlos  atomabsorptions-
            u G
epektrometriech beatimmt.

     Mit nur 2 Auenahmen enthielten die Nieren der  Tiere aus
dem Voralpenland eteta die hOeheten Pb-Gehaltet gefolgt von
Leber und  Fleiaah.   In den Nieren variierten  die Pb-Konzentra-
tionen zwieohen 0S10 und Qt80 ppmf in den Lebern zwiachen  Ot05
und Of40 ppmj  im Fleisch  lagen  sie faat immer unter 0,05  ppm.
Die Variationskoeffisienten betrugen bei Lebern  54,2%,  bei
Nieren S9,0%.   Sie dilrfen primar individuell  veruraacht  aein.
Eine Abhtingigkeit der  Pb-Konsentration  in Nieren und Leber von
Alter, Geechlecht und  Jahreazeit war daner  etatietiaah  niaht
abzueichern.   Dagegen beatand  zwischen den Nieren- und Leber-
Pb-Gehalten der Kinder ein hoaheignifikanter Zuaammenhang.

-------
                               1854
Das Fleiach starker bleiexponierter Binder enthielt selbst dann,
wenn die Tiere mit Symptomen einer Pb-Intoxikation notgeschlach-
tet werd&n musaten, kaum erhOhte Pb-Gehalte, Nieren und Lebern
yedoch ein mehr ala 100-faahes der Konzentrationen wie in
Nieren und Lebern der Kinder aus dem Voralpenraum.   Eine ein-
deutige Abhangigkeit von der Expositioneinteneitdt konnte dabei
jedoah nioht featgestellt werden.

     Fleiech kann ale ein Pb-avmes, Nieren ale ein Pb-reiahes
Lebensmittel gelten, Lebern als durchaehnittlich Pb-haltig,
Der Beitvag von Fleiach, Lebern und Kieren zum GeBamt-Pb-Gehalt
der Nahrung iet gering.
ABSTRACT

     The lead content of the fleah, liver and kidneys of eaeh
of 139 heads of cattle from a predominantly agricultural aub-
alpine area of Swabia and Bavaria* and of 22 animals from areas
of the German Federal Republic which are more severely exposed
to lead, Has determined by atom absorption apectrometry under
flame-free conditions after wet mineralization with fl^Og / Fe

     With only 2 exceptions, the kidneys of animals from the
subalpine area always had the highest lead contents, followed
by the liver and flesh.   In the kidneys the lead concentrations
were in the range 0.10 - 0.80 ppm, in the livers they were
0.05 - 0.40 ppm, and in the flesh they were almost always below
0.05 ppm.   The coefficients of variation were 54.2% in livers
and 39.0% in kidneys.   They may depend primarily on individuate.
Hence, a dependence of lead concentration in the kidneys and
liver on age, sex, and season could not be established statisti-
cally.   On the other hand, there was a highly significant aoz*r«-
lation between the lead contents of the kidneys and livers of
the cattle.   The flesh of cattle more heavily exposed to lead
showed hardly any increase of lead content even when the animals

-------
                               1855
had to be emergency slaughtered with symptoms of lead poisoning*
but the kidneys and livers showed concentrations more than 100
times as high as those in the kidneys and livers of the cattle
from the subalpine region.   However* it was not possible to
establish a definite dependence on intensity of exposure.

     Flesh oan be considered to be a food with a low lead
content* kidneys as one with a high lead content* and liver as
one with an average lead content.   The contribution of flesh*
liver and kidneys to the total lead content of the diet is
small,

-------
                            1856
1. Einleitung
   Uber die Auswirkungen der Aufnahme geringer Mengen Pb fiir
langere Zeit auf den menschlichen Organiamus 1st relativ we-
nig bekannt. Unklarheit besteht vor allem iiber seine Lang-
zeittoxizitat. Pb reichert sich in den Knochen, aber auch
in Leber und Nieren des Kbrpers an und ubt offenbar schon in
schwachen Konzentrationen bestimmte biologische Wirkungen
aus.
Die Hauptzufuhrquelle stellt dabei, bestinunte Extremsitua-
tionen ausgenommen (Eraissionen von Pb-Hutten, Pb-verarbei-
tender Industrie, GroBstadt), die Nahrung dar. Mit ihr wer-
den dem menschlichen Korper mehr als die Halfte seiner Ge-
samt-Pb-Aufnahme zugefiihrt (10, 44, 47). Die WHO veran-
schlagt diesen Anteil auf 57 - 66 $ (47).
Allein schon deshalb beansprucht Pb als potentieller Ruck-
standsbildner in Lebensmitteln lebensmittelhygienische Be-
achtung, Tierischen Lebensmitteln kommt dabei besondere Be-
deutung zu, da Pb in ihnen iiberwiegend organisch gebunden
vorliegt. Die leichter loslichen organischen Pb-Verbindun-
gen sind aber einer Resorption im Verdauungstrakt leichter
zuganglich als die im allgemeinen weniger gut loslichen an-
organischen.
Wenn auch bis jetzt beim Menschen noch keine augenscheinli-
chen Pb-Langzeitintoxikationssymptome beobachtet werden
konnten, halt man aus Unkenntnis der Pb-Langzeitintoxika-
tionswirkung eine weitere Zunahme der Pb-Belastung in Um-
welt und Bevolkerung nicht allein aus Grunden der Vorsicht
fiir kaum mehr vertretbar,
Im Interesse des Konsumentenschutzes haben deshalb bereits
verschiedene Staaten flir Lebensmittel zulassige Pb-Hbchst-
werte festgelegt, andere, unter ihnen die BRD, erwagen ge-
genwartig deren Pestsetzung.
Dies gab Veranlassung, ab Fruhjahr 1972 Fleisch- und Organ-
proben von Schlachtrindern auf ihren Pb-Gehalt zu unter-
suchen.
Diese Untersuchungen diirften aufierdem Hinweise auf das Ver-
teilungsmuster des Pb im Rinderorganismuo und Uber den An-

-------
Tabelle 1: Pb-Konzentrationen (ppm/Fg)  in Pleisch,  Leber und  Nieren von Schlachtrindern
           Muskulatur   n  Leber
n  Nieren
n Land Unters.-Jahr  Literatur





JgBulle/
Parse




Fruhling
Herbs t





0,16
0,29-1,65 8
<0 , 1 10
0,092 127
(0,03-0,35)
0,96 30
0,1

0,01-0,50
<1,0

-------
                             1858
tell des Rindfleisches am Pb-Gehalt der Gesamtnahrung, bei
Fleiacherzeugnissen auchKUckachlUsse auf eine:  bei der Her-
at ellung und Verpackung erfolgte Kontamination geben. Mb'g-
lichkeiten dazu beatehen beim Kontakt mit Pb-haltigen Ma-
schinenteilen oder dem Zusatz von Gewlirzen und Premdstoffen.
   Einen Uberblick Uber bisher in Fleisch, Lebern und Nieren
von Rindern erraittelte Pb-Konzentrationen gibt Tabelle 1.
Aus den Daten iat jedoch mit Ausnahme der Ergebnisse von
HECHT, MIRNA und SGHRAML (17) sowie GRAHWIT (13) nicht er-
sichtlich, welche Fleischstucke unteraucht wurden und ob die
analysierten Fleisch-, Leber- und Nierenproben jeweila von
einem oder von verschiedenen Schlachtkorpern entnommen wor-
den sind. AuQerdem fehlt es i.d.R. auch an Hinweisen zur
Probenahme, insbeaondere an Angaben Uber Alter, Geschlecht,
Rasae, Haltung und Futterung.
2. Untersuchungen
   2.1.: Untersuchungsmaterial
Von 77 am 12.5., 5.7., 26.7. und 13.12.1972 sowie 62 am 14.
3., 19,6., 18.7. und 25.9.1973 geschlachteten Rindern unter-
achiedlichen Altera (0,8 - 15 Jahre), Geachlechts und ver-
schiedener Rasse wurden pro Tier Jewells 100 - 200 g schwere
Proben aus den Mm. adductorea, dem Lobua caudatua der Leber
und dem apikalen Pol der rechten Niere achlachtfriach ent-
nommen und bia zur Analyse tiefgekiihlt aufbewahrt.
Pleisch war fiir die Untersuchungen wegen aeiner Bedeutung
in der menachlichen Ernahrung ausgewahlt worden, Leber und
Nieren als Target- und vermutliche Speicherorgane des Pb im
tierischen Organismua,
Pur die  Bntnahme der Fleischproben empfahlen sich bei der
grofien Konatanz ihrer grobgeweblichen Zusammensetzung die
Mm. adductorea (26).
Me sehr inhomogene Vorteilung dea Pb in Lebern (25; 26) und
Nieren (25) veranlaBte, die Proben immer gleichen Stellen
dieser Organe zu entnehmen, um die Reproduzierbarkeit der
Ergebnisse zu sichern.
Die Rinder atammten aus einem vorwiegend landwirtschaftlich
strukturierten Gebiet des schwabisch-bayerischen Voralpen-
landes mit Weidehaltung im Sommerhalbjahr. Daa Probenein-

-------
                            1859
zugsgebiet liegt zwischen den Orten Augsburg - Memmingen -
Ettal - Weilheim und FUrstenfeldbruck.
In dieaem Areal achwanken die langjahrigen Niederschlags-
mittel zwischen 700 ram in Raume Augsburg und 1200 mm Nieder-
schlag im Ammergebirge (Ettal). Dieses Gebiet war ausgewahlt
worden, weil ea als reprasentativ fiir andere, ahnlich struk-
turierte landwirtschaftliche Gebiete Mitteleuropaa gelten
kann. Verschiedentlich konnten dabei wiederholt Proben von
Tieren aua gleichen Ortschaften bzw. gleichen Betrieben ent-
noramen werden.
Auawahl der Tiere und Schlachttermine sollten die Priifung
eines eventuellen Einflussea von Alter, Geschlecht, Herkunft,
Jahreszeit und PUtterung auf die Pb-Konzentrationen in
Fleisch, Leber und Nieren ermb'glichen.
Zum Vergleich waren noch Proben von Rindern aus versohie-
denen starker Pb-exponierten Gebieten der BRD untersucht
worden. Sie stammten von
a) klinisch gesunden Rindern aus Betrieben in unmittelbarer
Autobahnnahe und aus der Wesermarsch, wo im Friihjahr 1972
mehr als 60 Kinder wegen einer schadhaften Pilteranlage des
Preussag-HUttenbetriebes verendet waren und
b) von mit Verdacht auf Pb-Vergiftung notgeschlachteten Tie-
ren aus der Weaermarsch, aus 2 weiteren Gebieten der BRD, wo
immer wieder Rinder an den Symptomen  einer Pb-Intoxlkation
sterben und
c) von Tieren, die nach Aufnahme von Mennige notgeschlach-
tet werden muQten,
2.2  : Methodik
Die flestimmung des Pb in Pleisch-, Leber- und Merenproben
erfolgt flainmenloa atomabsorptionsspektrometrisch im Gra-
phitrohr (Perkin-Elmer HGA 70 und 290 G). Das Untersuchungs-
material, bei Ploinch, Leber und Nieren je 100 g, wird zu
diesem Zweck mit UNO, nuprapur und HpOp (DAB 7) naS verascht,
vorhandenes Fett abfiltriert, das Pb  aus einem Aliquot der
Loaung durch Pallung mit NH.QH angereichert,  abzentrifugiert
und zur Injektion in das Graphitrohr  in 20 ml salpetersaurer
Lbsung aufgenommen. Die Ausbeuten werden mittels tragerfrei-
                212
em radioaktiven    Pb kontrolliert und schwanken zwischen

-------
Tabelle  2: Pb-Konzentrationen  (ppa/?g) in Pleisch. Leber und Nieren 1972 geschlachteter
           Hinder  (n =  77)


Datum  ii Alter-s(a)    Leber±s      Y  (#)   r*  Niere*s     V (%) r*  Pb-Konz.      Fleiscn
_ _ __ ___ _ _ _ _ Nj e r en/Leber _

12.4.  11 4,5*3,2  -   0,21±0,09x   42,8   0,39  0,43±0,23  53,5 -Q,26     2,5        0,09
          (1,0-9,0)   (0,08-0,39)               (olia-olso)             (0,8-6,2)   (^,65-
                                                                                     0,24)

 5.7.  17 4,3^3,1      0,13±0,06    40,2   0,00  0,32±0,14  43,8  0,13     2,5      -£0,05
          (0,8-10,0)  (0,05-0,2)                (0,10-0,51)             (1,0-4,5)

26.7.  22 5,3^,0      0,15^0,05                 0,35^0,12  34,3  0,09     2,5
          (2,6-11,0)  (3,06-0,23)   33,3   0,20  (0,14-0,57)            (1,3-4,5)    <0,05



27.9.  15 7,4^3,5      0,l6io,05N   31,3  -0,14   0,37±0,15  42,9 -0,05     2,4      <0,05
          (2,0-15,0)  (0,06-0,23)                (0,16-0,80)             (1,7-6,3)

13.12.  12  6,4*1,2     0,17*0,06    39,2  -0,17   0,27-0,05  13,5 -0,11     1,7      <0,05
          (5,0-8,5)    °09°
1972   77   5,3        0,16*0,07    43,8          0,34*0,15   44,1           2,4       ^0,05
           (0,8-15,0) (0,05-0,39)                 (0,10-0,80)              (0,8-6,3)

 *
   = Schwankungsbreite der Einzelwerte

-------
Tabelle 3: Bleikonzentrationen  (ppm/Fg) in Fleisch, Leber und Mieren 1973 geschlachteter
           Hinder  (n = 62)
Datum  n  Alters(a)  Leber*s      V  (?5)  r   Niere*s      V ($)  r*  Pb-Konz.     Fleisch
_  Njeren/Leber _

14.3. 12  5,3*1,9     0,17*0,06    ±39,2 0,07 0,27*0,10    ±37,0 0,18     1,6       CO, 05
          (2,0-9,0)  (0,10-0,24)              (0,10-0,22)               (1,2-2,7)


19.6. 17  4,0*2,3  K   0,13*0,04    *30,8 0,07 0,39*0,46    *117,9 0,13    2,2       <0,05
         (1,5-10,0)  (0,07-0,23)              (0,18-2,75)               (1,5-4,3)


18.7. 17  5,5*2,7     0,12*0,03    *25,0 0,27 0,28*0,11     ±39, 3 0,28    2,5       <0,05
         (1,5-11,0)  (0,06-0,40)              0,14-0,52)               (1,2-3,8)


25.9. 16  4,8*2,4     0,15*0,03    *20,0 0,13 0,38*0,09     *23,7 0,10    2,7       <0,05
1973: 62    4,9       0,14*0,04    *28,6      0,30*0,10     ±33,3         2,7       <0,05
         (1,5-11,0)  (0,06-0,40)              (0,10-2,75)               (1,2-4,3)
   #
  r  = Korrelationskoeffizient
  V  = Variationskoeffizient

-------
                  1862
A
Q9
0.8
0.7
0.6
as
0.4
0.3

0.2

of

O.L
o.i
«n.
04

ppm/Fg




If
i



-W4
«
• v
_





i
I
i


I

V 72.4.72
A
0.6 i
0.5
0.4
03"
—
Q2 \
\
oj


0.1

°'2 \
0.3__


5pm/Fg

',
M-r- •

i ' ; '" -; • * .
jr. •-•;?!
i i ""i . '; J
: t . ; ' *

< ; ; : • • ]
' i • ' i \
'i'^J
*;

ij1 '
-74.3. 7J .-






III
_, . .




|i!
«j
!j
1,






Tolerant -
!j wert* Til

1
'i
i*
1
T!r
^. .





A
I)'




5. 7. 72



|
k
*
:,;
*v . .
::^??i.
1 ••', •• • •>.• «
.' •• ! I
:.:Vvi|
'•'* ''''*''!!!
f : ' • ; ;! :: ^ !i •' » »




! !
i
i
i
i
i
i


[ID


j

"Lp
P



I).

Hi.
iM 5'
,,1,
' '" r





Illu k
j < t Ji ~ ~~
1 j } j 1 Uj
* i ill Cfc-:
I ' • 1 5 j j j ^y
i M : ' { hi ^
i 1: , | ^ i '•] • j
lj >;-lj !l fiiii f-
1 iiijlj il ill rri
' ' *li in I ^

~T wert* -*-• • —

! 25.7.72 27.9.72 /3./2.72



!!!!


II II,
, i! »"?jj
fSn. 1 i '!i'
1 1 	 Toleranz-
It t
Ms-1'.'litPli I" ; '• '• •: «;§ UJ
w J ' l j ' - • •* j * '• '536 /^^*
J* . T * :i » ^ * ; " I "* f ' ! r i * i t «^^
t ' li ^ * i n 1 ' ', •• ' * '. i -I '• •< J> *

i ' 1. .LJ.'.l | ...'._..,•.".! . '.'.L.,.
1 '.:
' 0:-'0'-!'
i
s
i r
I f ii''1'"
! '
\ 	 73.6. 73-..

'" i < J
i - "• • M i ^
^di^Tji^j
3! ; > ;.;
itli » • ::;

!r ' iiii
i ^ i?' ' |i»| I'll
?", ^

"^
*
r^^
i^^

1 * i^i
' .~7ft 7 7J 	 1 	 25.9. 73 	 ^'fCH^ 	
' *^ werP
*s. Tabelle 6
, , ABBILDUNG /; Bleikonzentrationen in Nieren and Lebern von Schtacht-
rindern  aus dem oberbayerisch-schwabischen  Vor-
alpengcbict

-------
                           1863

85 - 95 #. Die Nachweisgrenze li-3gt fUr 100 g Einwaage bei
20 ppb, der Gesamtfehler der Pb-Bestimmung bei<10 #.
3. Ergebnlsse
Die Durchschnittswerte, Standardabweichungen, Streubereiche
und Variationskoeffizienten der Pb-Konzentrationen in Lebern
und Nieren gibt fUr die 1972 geschlachteten Rinderkollektive
Tabelle 2 und fUr das Jahr 1973 Tabelle 3 wieder. Beide Ta-
bellen enthalten noch jeweils das Jahresdurchschnittsmittel,
deaaen Standardabweichung sowie Variationskoeffizienten und
geben Auskunft Uber den Umfang der Probenkollektivey das Al-
ter der Tiere, den Korrelationskoeffizienten zwischen der Pb-
Konzentration von Leber und Niere und dem Alter der Tiere so-
wie den Quotienten der Pb-Konzentrationen von Nieren und Le—
bern.
Die Einzelwerte der Leber- und Nierenkonzentrationen sind
fUr alle untersuchten Kinder aus Abbildung 1 ersichtlich*
Ihre Anordnung erfolgte in Schlachtkollektiven nach fallen-
der Nieren- Pb-Konzentration. Gleichzeitig veranschaulicht
die Tabelle auch die inter- und intraindividuellen Unter-
schiede im Pb-Gehalt beider Organe und deren eventuelle jah-
reszeitliche Abhangigkeit.
Tabelle 5 faBt auQerdem noch die Nieren- und Lebereinzelwer-
te zu Konzentrationsklassen zusammen, wobei ein Klasseninter-
vall Jewells 0,1 ppm betragt.
Tabelle 4 inforraiert Uber die Pb-Konzentration in Pleisch,
Leber und Nieren der starker Pb-exponierten Kinder. Die Da-
ten sind durch entsprechende Angaben aus der Literatur er-
gSnzt. Die Tabelle weist auch ftir Kinder in Leber und Nieren
vermutllch toxische, offensichtlich toxische und todllche
Pb-Konzentrationen aus.
In Tabelle 6 sind verschiedene zur Zeit  diskutierte, empfoh-
lene oder fUr Pleisch und Organe bereits gliltige Pb-Toleranz-
werte zuaammengestellt.
4. Besprechung und Diakussion der Ergebnisse
Bei 137 von 139 aus dem bayerischen Voralpenraum untersuch-
ten Rindern enthielten die Nieren stets die hochsten Pb-Kon-
zentrationen. Es folgen Leber und Pleisch. Lediglich in 2
Fallen wiesen die Lebern die hb*heren Konzentrationen auf. Ein

-------
Tabelle 4: Pb-Gehalte  (ppm/Fg)  in Fleisch, Lebern und Nieren starker Pb-exponierter Rinder

Klin. ges. Rinder
1 . GroBstadtnahe
2 . Autobahnnahe
(massiv Expos.)
3. Wesermarsch
Notgeschl. Rinder
(Verd.Pb-Vergifta.)
1 . Wesermarsch
2. Wesermarsch
3. Anderer Herkunft
4 . Anderer Herkunft
Aufnahme v.Mennige.
Alter
a


3,5
1-6


5,5

2-6

3-6
Fleisch

0,42±0,71
0,08
0,02-0,03


0,08
0,15.0,98
O,11-O, 38
< 0,1
0,04;0,07
0,O9;0,18
Leber

0,23±0,48
0,66
0,23-0,85


2,89
3,87-11,73
3,80-12,20
18,0-22,8
5,7;20,2
3 , 1 ; 4 , 4
Niere

0,42±0,59
1 ,44
0,64-1 ,91


2,60

4,9-22,8
23,2-62,8
14,0;59,0
7,6;5,9
n

20

7



8
7
5
4
Jahr
a

1971/72
1972
1972


1972
1972
1972/73
1971/72
1971
Literatur

13





15/16/17

13
24
                                                                                                 00
 Vermutl.tox.Konz.
 Offens.tox.Konz.
 todliche Konz.
2-  20
4-  2O
> 20
10-  25
1O-  50
   25
Lit. L.14;35;38;39;
    N.1 ;2;6;14;21;27;35
    L.14,-30;35;42;43;
    N.6;11;14;21;27;3O;35
      43;

    L. 1 ; 2 ; 31 ;
    N.1; 31;42;43;

-------
Tabelle 5: Zusamiaenstellung der Einzelwerte  (Leber, Nieren) nach Konzentrationsklassen

           (ppm Pb/Fg)
                 Lebern
Nieren
Klassen
<0,1
0,1-0,2
0,2-0,3
0,3-0,4
0,4-0,5
0,5-0,6
0,6-0,7
0,7-0,8
0,8-0,9
Probenzahl
0
14
44
36
21
11
0
1
2
Anteil in %
0
10,9
34,1
27,9
16,3
8,5
0
0,8
1,5
Probenzahl
24
79
24
2





Anteil in
18,6
61,2
18,7
1,5





*









                                                                                                oo
Tabelle 6: Verschiedene fiir Fleisch und Organe von Schlachttieren diskutierte  und vorge-

           schlagene  zulassige  Pb-H6chstwerte (ppm/g Pg)

WHO/PAO Expert Committee: Vorlaufiger wbchentlicher Toleranzwert fur Lebensmittel: 3,0 mg
Land
BRD:
VO-Entwurf
Empfehlung
Pleisch
0,05
0,2
Innereien
0,3
*
0,5
2,0**
Pleischerzeugnisse
0,1

46
12
               •
 ohne Nieren,   Leber

-------
                            1866
Grund hierfiir verraag nicht genannt zu werden. Moglicherweise
batten beide Tiere kurz vor der Schlachtung mehr Pb aufgenom-
men als die anderen Hinder.
Mit Ausnahme des am 12.4.72 geschlachteten Tierkollektivs la-
gen die Pb-Konzentrationen im Pleiach stets unter 0,05 ppm/Fg.
Die Tiere dieses Kollektivs waren vermutlich mit Putter, ins-
besondere Heu und Grummet aus dem Jahr 1971 ernahrt worden,
1971 fielen in Mitteleuropa extrem wenig Niederschlage. Trok-
kenes Wetter aber beglinstigt im Griinfutter hohe Pb-Kontarai—
natidnen.
Lebern enthielten durchschnittlich 0,16 ppm, Nieren 0,33 ppra
Pb/Pg. Die Minimalwerte betrugen bei Lebern 0,09 ppm und bei
Nieren 0,10 ppm, die Maximalwerte 0,40 ppm bzw. 2,75 ppm.
Die Niere mit dem Pb-Gehalt von 2,75 ppm stammt moglicher-
weise von einem Rind, das entweder langere Zeit bleireiches
Putter aufgenommen oder frtiher einmal Zutritt zu stark Pb-
haltigen Gegenstanden besessen hatte.
Die Nieren enthielten, wie die Quotienten aus der Nieren-
und Leber-Konzentration ausweisen mehr als doppelt so viel
Pb wie die Lebern (Minimalwerte: 0,8, Maximalwerte: 6,3).
   Besonders fallt in Lebern und Nieren die grofle Streuung
der Pb-Einzelwerte urn die Mittelwerte auf (V = 54,2 % bzw.
39 fO #). Selbst bei Tieren aus gleichen Bestanden oder dem
gleichen Ort waren sie kaum geringer. Die Variationskoeffi-
zienten schwankten bei Lebern innerhalb der Einzelkollektive
zwischen 20 und 46 #, bei Nieren zwischen 18,5 und 53,5 #.
   Dennoch bestand zwischen der Pb-Konzentration in Leber
und Nieren des Einzeltieres, innerhalb der Kollektive und
der Gesamtheit der untersuchten Tiere ein hochaignifikanter
Zusammenhang fcc= 0,001). Hinder mit hohen Leber-Pb-Werten
weisen demnach auch hohe Pb-Konzentrationen in den Nieren
auf.
Bine statistisch gesicherte AbhSngigkeit der Pb-Konzentra-
tionen in Nieren und Lebern vom Alter der Tiere besteht, wie
ein Blick auf die Korrelationskoeffizienten in Tabelle 2 und
3 beweist offenbar nicht. Deshalb kann auch die Prage, ob
die Pb-Konzentrationen in Organen von Rind :rn ahnlich wie

-------
                             1867

beim Menschen rait steigendem Alter zunehmen (10) oder der
Binbau des Pb bereits nach einer beatimmten Zeit im wesent-
lichen abgeschlosaen 1st (4) anhand des vorliegenden Unter-
suchungsmaterials nicht schlilssig beantwortet werden. BARRY
(4) beobachtete bei Menschen in Welchgeweben nach dem ersten
Lebenajahrzehnt kaum mehr eine Pb-Zunahme.
Auch ein jahreazeitlicher EinfluB auf die Pb-Konzentrationen
von Leber und Nieren war statiatisch nicht absicherbar,Aller-
dings ergab der t-Teat beim Vergleich der Pb-Gehalte von Le-
bern und Nieren 1972 und 1973 geachlachteter Tiere nur 1972
fUr Lebern signifikant hbhere Pb-Werte (t = 1£9; n = 77;
«(= 5 #). Moglicherweiae begtinatigten die 1972 nur geringen
Niederachlagsmengen diesen Umatand.
Die weitere AufechlUsselung des Untersuchungsmateriala nach
Raaae, Geachlecht, Herkunft, Haltung und Flitterung lieferte
bia jetzt noch keine zusatzlichen Anhaltapunkte fUr eine Er-
klarung der gefundenen groflen Schwankungabreiten im Pb-Ge-
halt von Lebern und Nieren,
Die Ursache dafUr dUrfte primar individuell bedingt sein und
mufi vermutlich in Unterachieden bei der Abaorption, Vertei-
lung und Ausscheidung dea Pb in verschiedenen Tierkbrpern ge-
sucht werden. Sie wtirden auch erklaren, warum aelbst bei
gleichalten, gleichgeschlechtlichen Tieren eines Bestandea
ahnlich groSe Schwankungen gefunden wurden wie innerhalb der
Tierkollektive.
Ein Blick auf Tabelle 4 laflt erkennen, dafi starker Pb-expo-
nierte Hinder in Leber und Nieren zumeist erheblich hbhere,
in Pleiaoh dagegen nur geringfUgiger erhbhte Pb-Konzentra-
tionen aufweisen ala die von una in  weniger stark umweltbe-
lasteten Arealen untersucht en Hinder. Wenn sie mit den Symto-
men einer Pb-Vergiftung notgeachlachtet worden waren, ent-
hielten ihre Lebern und Nieren Pb-Konzentrationen, die den
Verdacht einer lebensbedrohlichen Pb-Intoxikation recht-
fertigten. Die Zunahme dea Pb in den Fleieohproben war auch
bei diesen Tieren nur gering. Diese Befunde decken sich mit
einer Mitteilung HOGGERS (19), der bei drei 36 Tage mit Heu
von einem Autobahnmittelatreifen (99 mg Pb/kg TS) gefUtter-

-------
                               1868
ten Kiihen in Leberri unJ uieren uber eine utarke, in der Mua-
kulatur dagegen nur schwache Zunahme der Pb-Konzentrationen
berichtete, Auch unaeren Untersuchungen zufolge, bedarf es
ftir eine erhebliche Erhbhung des Pb-Gehaltea in Leber und
Nieren von Rindern des Zusammentreffens einer Reihe die Akku-
raulierung des Pb in Putter und Tieren optimal beglinatigender
Faktoren. Bei Rindern aua autobahnnahen Betrieben konnte bis
Jetzt erst bei einer Kuh in Fleisch  ein signifikanter An-
stieg der Pb-Konzentration heobachtet werden. Das Tier stanun-
te aus einem 10m tiefer, unmittelbar neben der Autobahn in
Richtung der Hauptwinddrift gelegenen Betrieb, wo zum Tran-
ken der Hinder Wasaer aua einem TUmpel verwendet wird, der
Sickerwasser und auch von der Autobahn herabrinnendes Regen-
wasser aufnimmt, DaQ Rinder aus autobahnnahen Betrieben bis-
her in Nieren und Lebern keine starker erhdhten Pb-Konzentra-
tionen aufwiesen, diirfte mit darauf zuriickzufiihren aein, dais
die Tierbesitzer entweder von sich aus auf die Verflitterung
von unmittelbar neben der Autobahn gewachsenem Putter ver-
zichten oder den Tieren in erheblichem Umfang weniger stark
Pb-kontaminierte Puttermittel zufiittern.
Die in den eigenen LJntersuchungen bei Fleiach, Lebern und
Nieren von weniger und starker Pb-exponierten Rindern er-
mittelten Pb-Konzentrationen entaprechen im wesentlichen den
in der Literatur mitgeteilten Werten (Tabelle 1 und 4).
Allerdings iat ein Vergleich nur bedingt mbglich, da es zu-
meist an detaillierten Angaben zur Probenahme fehlt,
Der durchschnittliche Pb-Gehalt der Nahrung betragt in Eng-
land 0,13 ppm (44» 45). Vergleicht man damit die Pb-Werte
in Fleisch, Lebern und Nieren der von uns untersuchten, we-
niger stark Pb-exponierten Rinder , so zeigt sich, dafl
Fleisch relativ geririge, Lebern durchschnittliche und Wieren
iiber dem Durchachnitt gelegene Pb-Konzentrationen aufweisen.
Auch Fleisch von starker Pb-exponierten oder sogar mit Symp-
tomen einer Pb-Vergiftung notgeschlachteten Rindern enthielt
fast stets Pb-Konzentrationen unter 0,13 ppm/Pg.
Fleisch zahlt deranach zu den bleiarmen, die Nieren zu den
bleireicheren Lebensmitteln. Lebern enthalten etwa Durch-

-------
                                1869
schnittswerte.
Der Anteil dea Pleisches am  Gesamt-Pb-Gehalt der  Nahrung
kann daher als gering veranschlagt werden. Das  gleiche trifft
trotz  hb'herer Pb-Konzentrationen auch  fur Lebern  und Nieren
zu, wenn man deren geringen  Beitrag an der Gesamtnahrung  in
Rechnung stellt.
Allerdings sollte hier auf bestimmte KontrollmaBnahmen nicht
verzichtet werden, da Hieren auch bei  starker  Pb-exponierten
Rindern bereits  beachtliche  Pb-Konzentrationen enthalten.
Legt man den im  Entwurf der  "VO Uber Umweltkontaminanten"
der BRD fUr Innereien diskutierteri  zulassigen  Hochstwert  von
0,3 ppm Pb/Fg  zugrunde, hatten 55 $ der von  uns untersuchten
Nieren und 1,5 % der Lebern  Pb-Konzentrationen Uber diesem
Toleranzwert besessen.
Auch  bei Zugrundelegung einer Toleranz von 0,5 ppm Pb/Pg
Ubersohreiten  immer noon  10,8 96 der untersuchten Nieren  die-
sen Wert.
Es wird deshalb  vorgeschlagen, fUr  Nieren eigene Pb-Toleranz-
werte zu erstellen.

                           REFERENZEN
  (l)  ALLCROFT, R. "Lead Poisoning in cattle and sheep,"Vet.Rec.,63 (l95l)t
       583;

  (2)  ALLCROFT, R, and BLAXTER, K.L. "Lead as a nutritional hazard to farm
       livestock. V: The toxicity of lead to cattle and sheep  and an evalua-
       tion of the lead hazard under farm conditions", J.comp.Path.Ther.60
       (1950), 209;

  (3)  ALLCROFT, R. ""Distribution of lead in the tissues of bovines after
       ingestion of various lead compounds", J.comp.Path.Ther._60 (1950)140;

  (4)  BARRY, P.S.I. "A comparison of lead concentrations in human hones
       and in soft tissues",Fnvironmental health aspects, EUR 5004 d-e-f,
       Commission of the European Communities, Centre for Information and
       Documentation (CID), Luxembourg 1973;

  (5)  BERGNER, K.O. und HALLER H.F.. "Ueber das naturliche Vorkommen von
       Arsen, Blei, Cadmium, Selen, Zink und Zinn in Lebensmitteln",Kitt.
       Bl.GDCh, Fachgruppe  Lehensmittelchemie und gerichtl, Cheraie 18
       (1964) 113}                                            "•

  (6)  BUCK, B.W. "Lead ond organic pesticide poisonings in cattle",
       J.Am.Vet.Ked.Ass. 156 (1970), 1460;

  (7)  CHF.FTFX, H. "Le plomb de,ns les aliments"»Ann. Falsificat  Traudes 43
       (1950), 210j

-------
                                     1870
 (8)  DANKWORTT,  P.W.  and HOFLL,  K.  "Chemische Untersuchungen tiber das
      normale Vorkommen von Blei  in Orgnnen und Knochen ge sunder Haustiere" ,
      DTW.  Jl (19M) 586;

 (9)  DE TREVILLE,  R.T.  "Natural  occurence lead" , Arch. Flnvironment. Health 8
      (1964), 2125

(10)  DINN'AN, B.D.  "Airborne lead in perspective.  Environmental  health
      aspect s", FUR 5004 d-e-f.  Commission of the  European Communities,
      Centre for Information and  Documentation (CID), Luxembourg 1973;

(11)  DONAWICK,  W.J. "Chronic lead poisoning in a cow", J.Am.Vet.tfed.Ase.
      148 (1966), 655;

(12)  GORENSTEIN, B. und SPROULE, J.D.  "Pesticide and heavy metal resi-
      dues in Canadian meats",  Proc. 18th Meeting of Veat Research Workers,
      Guelph, Ontario, Canada 1972;

(13)  GRAHWIT, G. "Untersuchungen tiber den Gehrlt an toxischen Schwermetal-
      len in Organen normal geschlachteter Haustiere", Vorlaufige Kitteilung:
      Blei in Rinderorganen, Arch.Lebensmittelhyg. £3  (1972), 213;

(14)  HATCH, R.C. and FUNNEL, H.S. "Bleigehalte in Geweben und Kageninhalt
      vergifteter Kinder" ,Uebersicht 15 Jahre, Cand.Vet.J.JLO (1969),  258;

  (15)  HF.CHT, H. "Toxische Elemente im Fleisch",  Ref.  Symp.  "Toxische
        Spurenelemente in Lebensmitteln",  Oktober 1973,  Karlsruhe

  (16)  HECHT, H, "Untersuchungen tiber Spurenelemente im Fleisch",
        Arch. Lebensmittelhyg.  2£ (1973),  256;

  (17)  HECHT, H.,  KIRNA, A. und  SCHRAKEL, P.  h Untereuchungen uber den
        Geha.lt an Spurenelementen in Schweine-, Rind- und Kalbfleisch",
        Fleischwirtschaft ^ (1973), 237;

  (18)  HERNBF.RG, S. "Biological  effects of low lead doses",  Environmental
        health aspects - EUR 5004 d-e-f, Commission  of  the European
        Communities (CID), Luxembourg 1973;

  (19)  HOEGGER-,  I). "Die hygienische Bedeutung der Benzinverbleiung",
        Environmental  health aspects - TUR 5004 d-e-f,  Commission of  the
        European Communities, Centre for Information and Documentation (CID)
        Luxembourg 1973;

  (20)  HOEGL, H.A., TARDENT P. und  SULSF.R,  T.O.  "Bleigehalt  in  Lebens-
        mitteln  und Wasser", Mitt. Gebiete Lebenemittelunters. Hyg. 53
       (1961), 188;

  (21)  HUTCHINS, K. "Lead poisoning in cattle ",  Aust.Vet.J. Jl2 (1955), 327;
  (22)   INTONTI,  R. ,  WCCIIETTI G.  et  STACCHIN,  A.  "Determinp.zione  di
        oligoelementi per piezzo dells  spettografin  di  emissione ottica  nei
        prodotti  alimentari.  Nota 1.  Picerce, e  dorBagio del  plombo nelle
        carni  fresche e conservate",  Ann.  1st.  Super Sanit*  ^ (1967),  571;

-------
                                  1871
(23)   KAURANEN,  P21QJAAKKOLA T.  end NIETTINEN, J.K. "Stable lead and
      radiolead  (   Pb)  in  the "arctic food chain", Kongresspapier aus
      dem "Department  of Radiochemistry, University of Helsinki", 1969j

(24)   KREUZER, W.,  KRACKE, -W. und  SANSONI, B. "Zur fleischbeschaulichen
      Beurteilung akut bleivergifteter Schlachttiere", SVZ ^3  (1973), 369;

(25)   KREUZER, W.  "Unveroffentlichte  Untersuchungen des  Institute fiir
      Nahrungsmittelkunde der Universitat Kunchen"(l97l);

(26)   KHEUZER, W.,  SCHROEDER, K. und  RING, Ch. "Zum K-Gehslt in Fleisch-
      proben vom Rind  und Schwein", Pleischwirtschaft 48 (1968) 802;

(27)   LEARY, S.L.,  BUCK, W.R., LLOYD, W.E. and OSWELLES, G.D.  "Epidemiolo-
      gy of lead poisoning  in  cattle", Jowa State University Veterinarian
      Issue Mr.  3 (1970) 112;

(28)   LEHNFRT,  G.,  STADELNANN, G., SCHALLER,  K.H. und SZADKOWSKI, D.
      "Usuelle Bleibelastung durch Nahrungsmittel und Getranke", Arch.
      Hyg. J£3jl968)  403;

 (29)  LOB, K.,  GUILLKMIN,  H.  und VALLATOK, W.N. "Interest des  parametree
       modernes pour I1Evaluation de I1action biologique du plomb  dus des
       sujets exposes et non-exposes proffessionellement"t  Environmental
       health aspects - FUR 5004 d-e-f,  Commission of the European
       Communities, Centre for  Information and Documentation (CID),
       Luxembourg 1973;

 (30)  Me. EWEN, I.E. "The seasonal  incidence of lead poisoning in the
       north of Scotland"., Vet.  Rec. ]$ (1963), 515?

 (31)  Kc.  INTOSH, I.G., ref. TODD,  I.R. "A knackery survey of lead
       poisoning in cattle in Northern Ireland", Vet.Rec. J£ (1962), 116;

 (32)  Kc.  LF.AN, K.S.,  BYERS, D.L. and BROWN, K.N. "Spectrographic deter-
       mination of lead in agricultural and related materials", J. Ass.
       Off.  Anal.  Chero.^0  (1967)  , 1366;

  (33)  NELSON,  R.  und KOLAR, K. "The content of  lead in the liver of pigs
       and cattle", Proceedings 16th European meat research workers,
       Varna 1971;

  (34)  PEDEN, J.D. "A survey of the  arsenic, copper  and lead  contents
        of pig and  other animal  livers", J. Assoc. Publ.  Analyses 8  (1970)
        14;

  (35)   ROSENBERGER, G.  "Krankheiten  des Rindes", Verlag P. Paray, Berlin
        1970;

  (36)  RUESSEL, H. "Ueber  die Bleiverteilung in Leber bei  Vergiftung und
        die Probenentnahme  zur chemischen  Analyse", D.T.W.  68  (1969), 96}

-------
                                   1872
 (37)  SANSONI, B., KRACKE, W., DIETL, P. und FISCHER, J.   "Kikrospuren-
      bestimmung  von Blei in verschiedenartigen Umweltproben durch
      fl^mmenloBe Atomabsorption nach externer Nassveraschung mit
      Environmental health aspects - F,UH 5004 d-e-f, Commission of
      the European Communities, Centre for Information and Documentation
      (CID) Luxembourg 1973;

 (38)  SCHOFJ3ERL,  A. und RUESSEL, H. "Bleivergiftungen bei Tieren und in
      Futter- und Lebensmitteln". D.T.W. J8 (l97l)f 341;

 (39)  SCHOEBERL,  A. "Erfahrungen und Ergebnisse beim Nachweis von Tier-
      vergiftungen", D.T. W. _62 (1955), 3;

 (40)  SCHROEDER,  H.A. und BALASSA, J.J. "Abnormal trace metals in man:
      lead", J. Chron. Diseases 1£ (l96l), 408;

 (41)  TKR HAAR, G. "The sources and pathways of lead in the environment"
      Environmental health aspects - CUR 5004 d-e—f, Commission of the
      European Communities, Centre for Information end Documentation (CID)
      Luxembourg  1973;

(42)  TODD,  J.R. "The  incidence of lead poisoning  in cattle in Northern
      Ireland", Vet.Rec. J[6  (1964), 845;

(43)  TODD,  J.R. "A knackery  survey of lead poisoning in  cattle in Northern
      Ireland", Vet.Rec. J4.  (1962), 116;

(44)  TOLAN, A. und ELTON, G.AH. "Lead intake  from food", Environmental
      health aspects  - EUR 5004  d-e-f, Commission  of the Europeen
      Communities  ,  Centre for Information and Documentation  (CID),
      Luxembourg 1973;

(45)  N.N.  "Ministry  of  Agriculture,  Fisheries and Pood  (1972)",
      Survey of lead  in  food,  HMSO, London

(46)  N.N.  "Entwurf  zu einer Verordnung iiber hochstzulassige  Gehalte
      an Arsen,  Cadmium,  Blei und  Quecksilber  in Fleisch und  Fleischer-
      zeugnissen des  Bundesministeriums  fur Jugend,  Familie und  Gesund-
      heit  (1972);

(47)  N.N.  "Evaluation of Hg,  Pb,  Cd  and the  food  pdditives amaranth,
      diethylpyrocarbonnte  and octyl  gallate1', WHO-Food  Additives
      Series No.4, WHO-Organization Genf 1972;

-------
VERZEICHNIS






   INDEX






   INDEX






  INDICE






   INDEX

-------
                           LXXXV
AASETH, J., 913
ADLER, M.W., 72
ALBERT, R., 1167, 206I, 2O68,
  2069
ALESSIO,  L., 1123, 1129, 1130,
  1167
ALLART-DEMUL, C., 1303
ALLEGRINI, M., 1697
ALLEN, J.R.,  385, 397
ALTMAN, D.G., 289
ALTSHULER, B., 2061
ALTSHULLER, A.P., 21O9
ANDERSON,  J., 1449,  1461, 1468,
   1469
ANGERER, J.,  1317, 1327,  1328
ARHIRII, M.,  339
ARONOW,  R.,  1177
ARSAC,  P., 6O3.
AUBERT,  M.,  1613

 BABCOCK,  L.R.,  2083
 BACKHAUS,  F., 2231
 BAKER, F.D., 879
 BARHAD, B., 333
 BARQUET, A., 695
 BARRATT, R.S., 1397, 1779
 BARSAN, E.T., 1O73
 BARTH, D.S., 1875, 1877, 1921,
   1922, 1925,~192~6,  1928, 1930,
   1931, 1933, 1934,  1935, 1939,
   1940, 1950, 1959,  1960, 1962,
   1963, 2073,
 BASTENIER, H.,  1303
 BATES, D.V., 1007, 1967, 1978,
   1979, 2001, 2003,"2TT2
BATTI, R.( 1531
BEACONSFIELD, P., 2397
BECK, E.G., 1031, 1040, 1O41
BEITZ, L., 1417, 1430
BELCHER, R., 1779
BELL, A., 11O1
BENARIO, M., 2169
BENINSON, D., 845,  1878,  1924,
  1927,  1934, 1935,  1941,  1949,
  2297
BENSON,  F.B., 423
BERGLUND, B., 119
BERGLUND, U., 119,  142
BERLIN,  A.,  552,  611,  629, 693,
  859,  1087, 1100,  1238,  1959,
  2003,  2185, 2254, 2257
BERLIN,  M. ,  .156,  491,  895, 1259
BERNA,  M., 231
BERNER,  A.,  1729
BERNSTEIN, A.D.,  105,  116, 117,
  1029, 2298
BERNSTEIN, D.,  431
BIANCO, A.,  1O39
BIERSTEKER,  K., 1881,  1924, 1926,
  1933, 1934,  1943, 1957, 1958,
  1959
BIGNON, J.,  1189, 1196, 1197
BINDER, R.E.,  669
BITTEL, R.,  714,  1441, 1449,
   1469
 BLACKBURN, C.R.B., 63,
De BOECK, R., 1131
 Den BOER, M.C., 1247
 BOGDANOVIC, E., 2271
 BONNAUD, G., 1189
 BONNEFOUS, M., 51O
  Underlined numerals denote  authors
  Arabic numerals denote  participants  in discussions

-------
                           LXXXVI
BORDAS, E., 1145
del BORGHI, M.f 1807
BORLAUG, N., 2397
de BORTOLI, M., 1287
BOTZENHART, K., 1757
BOUDENE, C., 603, 612
le BOUFFANT, L., 1645, 1651
BOUHUYS, A.r 669, 675
BOUQUIAUX, J., 1239, 1298
BOURBON, P., 269
BOURDEAU, Ph., 263 363, 366,
  510
BOUVILLE, A.f 1531
BRADEN, M., 485
BRAETTER, P., 2255, 2301
BRAMAN, R.S., 117, 258, 1328,
  1363, 1370, 1397, 1954
BRAVO A, H., 468, 2O91, 2160,
  2161
BREIDENBACH, A.W., 751
BRILLE, D., 269, 356, 667
BROCKHAUS, A., 781
BROMBERG, P.A., 1989
BROOKS, A.G.F., 77
BRUAUX, P., 1131, 1143
                                CACCURI,  S., 1823
                                CAGNETTI, P., 1451
                                CALANDRA, J.C., 772
                                CANTON, J.H., 1479, 1489
                                CAPURRO,  P.U., 1579
                                CARNOW, B.W., 45, 313, 353, 266,
                                  368, 588,  1129, 1167
                                CARPENTER, L., 1729
                                CARPI di  RISMINI, A., 2397
                                CARSTENS, L.A., 385
                                CARSTENSEN,  J., 969
                                CARTER, M.H., 1399
                                del CASTILHO, P., 2185
                                CASULA, D.,  1693
                                CERNIK, A.A., 1207, 1221, 1237,
                                  2254
                                CERQUIGLINI-MONTERIOLO, S., 1301
                                  1383                         "
                                CHAMBERS, P., 1O4, 259, 397, 51O,
                                  702, 2045
                                CHANTEUR, J., 2386
                                CHAPMAN,  R., 193, 2O7, 208, 210,
                                  645, 658
                                CHARLTON, J., 112O
                                CHATTOPADHYAY, A., 1685
                                CIALELLA, N., 2225
BRUCH, J., 781, 791, 1040, 1206 CICOLELLA/
de BRUIN, A., 259, 911, 2O59
BUCKET, J.P., 631, 887, 2185
BULCRAIG, W.R., 209
BURGER, E.J.jr.,  145,  155,
   156,  157
BUSH, B., 879, 885, 911, 1339
BUSTUEVA, K., 1O09
BUTLER, G.C., 1431, 1884, 1930
   1937, 1948, 1955, 1956, 1978
BUXTON, R. St.J., 1113
CIGNA ROSSI, L., 1451
CLAEYS-THOREAU, F., 1131
CLAUDE, J.R., 397
CLAYTON, J.W.Jr., 371, 383, 383
          G.F., 260, 1O28, 1451
                                COFFIELD,  T.H.,  858
                                COIN, L.,  7O3
                                COLMAN, R., 485
BYRNE, A.R., 245, 258, 259, 26ofCOWMBIMIf M., 159
  1370

-------
                             LXXXVII
COLUCCI, A.V., 1043
COOPER, W.C., 555, 568, 569,
  1196
COTE, R.W. , 4_7
CROCKER, K., 2068
CROSSMANN, G., 14O7
CRUZ, R., 1685
CUCU, M., 333

DAHL, R., 2231
DAMS, E., 513, 1430
DAMS, R., 4O9
DANIEL, H., 1645, 1651
DANIEL, J.W., 877, 9O2, 1041
DANIELSON, L., 116, 885
DAVID, 0., 588, 1549
DAVIDOW, B., 545
DAVIES, J.E., 695
DAVOUST, P., 2384
DEAN, G., 190, 643
DEHNEN, W., 781
DELEANU,M., 1583
DELCARTE, E., 1675
DELVES, H.T., 2215, 2258
DEMUYNCK, M., 409
DENNIS, C.A.R., 1029, 1543,
  1953                	
DEPAUS, R., 1341
von DEPKA, J., 5O6
DERWENT, R.G., 1669
DESBORDES, J., 1199
DEVOTO, G., 1693
DIEHL, J.F., 958, 1121
DIETERICH, B.H., 2451
DIETL, F., 1853
DI FERRANTE, E., 1956
DJURIC, D., 1829
DOBIN, D.D., 1223
DOBRYSZYCKA, W., 685
DOLGNER, R., 279
DONALDSON, W.T., 1399, 1405,
  14O6
DONNIER, B., 1613
DONZELLI, A., 231
DUBOIS, L., 1331
DUGANDZIC, M. 2285
DUMONT, M., 2231
DUNCAN, K.P., 2175
DUNCAN, L.J., 1241
DUPUIS, P.J., 491, 1196, 1314

EDWARDS. H.W., 1277, 1285, 1286
EFTHYMIOU, M.L., 1789
EGELS, W., 1407
EISENBUD, M., 431
ENGLISH, T.D., 4O1, 470, 471,
  472          	
EPSTEIN, S.S., 552, 569, 749,
  814, 1120, 2367, 2382, 2383,
  2384, 2385, 2386
van ESCH, G.J., 1O17
ESPINOSA, M.E., 2091
EVENDIJK, J.E., 1351

FACCHETTI, S., 1287, 1298, 1299
FAGNIART, E., 1675
FAIRWEATHER, F.A., 1113
FALK, H.L., 2331, 2348, 2349,
  2350
FAVRETTO, L., 1511
FAVRETTO GABRIELLI,  L., 1511
FERRAIOLO, E.G., 18O7

-------
                            LXXXVIII
FERRARI-BRAVO, F., 1797
FINE, P.R., 1223
FINKLEA, J., 193, 645
FISCHER, A.B., 1031
FISHBEIN, L., 725, 749
FLOREY, C., 289
FONDIMARE, A., 1199, 1206
FOURNIER, E., 1433, 1439, 1789
FREDERIKSON, M., 1959
FREEMAN, G., 685, 833, 844,
  1007
FRENCH, J., 193, 645
FRIBERG, L., 2307, 2315, 2316
FRIEDMANN, J., 5O7, 51O, 511
FRIEDRICHS, K.H., 715, 723,
  724
FUMAROLA, G., 1807
FURIOSI, N.J., 833

GADDO,  P.P.,  1287
GAFFEY, W.R., 555
GAGE, J.C., 895
GAGLIONE, P., 1287
GARDI,  R.,  1807
GARDNER, D.E., 705, 713, 714,
GARIBALDI, P., 1287
GARNIER, A., 1441
GENT, M., 1263
GHELBERG, N.W.,  1145,  1583
GHETTI, P.F., 1957
GIANANI, G., 1697
GIBBS,  G.W., 1197, 1206, 2271,
  2296, 2297, 2301, 2302
GIBSON, R.J.W.,  289
von GIERKE, H.,  1249
GIOVANNINI i  I.,  1797
CLAUDE, P.,  1341
GLOBUS, G., 507
GODIN, J., 6O3
GOERKE, W., 37
GOLDBERG, A.M., 793, 8O3
GOLDBERG, H., 193
GOLDSMITH, J.R., 62, 155, 189,
  382, 585, 675, 1889, 1923,
  1935, 1936, 1942, 1945, 195O,
  1956, 2156, 2165, 2179,
GOLDSTEIN, I., 1275
GONO, E., 591
GOOTJIS, P., 1316, 1951
GRAB, B., 339
GRADISKI, D., 1631, 1661
de GRAEVE, J., 523
GRAOVAC-LEPOSAVIC, L.,  1829
GRASSO, C., 861, 878
GREENLAND, R.D., 805
GREVE, P.A., 1479
GRIECO, A., 231
GRIFFIN, H.E., 23O5
GROLL-KNAPP,E., 989
GROS, R., 2225
GRUENER, N., 1067
GRUNSPAN, M.J.,  773, 2003
GUINEE, V.F., 545, 551,  552,
  553, 1166

HAAG, A., 1317
HAASE, J.,  1417
HADDAD, R., 451
HAGEDORN-GOETZ, H.,  2231
HAIDER, M., 475, 989, 999,  2393
HARDWICK, D.F., 961
HARKE, H.P., 1327, 1773
HARRISON, P.R., 1741
HARRISON, R.M., 1111, 1285, 1349,
  1370,  1405,  1783

-------
                           LXXXIX
HAYES, C., 645
HAZUCHA, M., 1979, 2001
HEBBELINCK,  D., 1303
HEM, B., 1189
HENDERSON, P.Th., 2047
HENIN, J.P., 1651
HERNBERG, S., 568, 692, 1129,
  1142, 2395
HERTZ, M.B., 1763
HICKEY, N.,  658
HILLERY,  P.J., 7_
HILPERT,  K., 2231
HINE, C.H.,  207,  260,  1O4O,
  1130,  1937, 2029,  2298
HINTON,  D.O., 1769
HISLOP,  J.,  959
HO, M.T., 1519
HOFMAN,  B.,  1O17
HOGGER,  D.,  1247, 2157
 ROLL, K. , 612
 HOLLAND, W.W.,  39_,  45
 HOLM, S., 895,  902
 HOLMQVIST, It  613, 629
 HORIE, Y., 2143
 HOSEIN,  H.R.,  669
 HOUCK, C.L., 879
 HOWER, J., 591, 601
 HUETER, F.G., 261, 352, 363,
    364
 HUTCHINSON, T.C., 1685
 HUTH, F., 715
 HUUNAN-SEPPALA, A., 2263

 IMPENS,  R., 1675
 IOVENITTI,  L.,  1797
 IRWIG,  L.,  289,  360
 IWANKIEWICZ, S.,  685
IZMEROV, N.F., 2409
JACKSON, D.L., 161, 176, 177,
  178, 1956
JACOBSEN, M., 89, 211, 228, 229,
  1478, 1502, 2365
JACYSZYN, K., 685
JAMIN, P., 523
JANSEN, G.,  513, 999,  2O44
JEANMAIRE, L., 2225,  2252,  2256
JERVIS, R.E., 1685
JOHNSON, D.L., 1363
JOOSTING, P.E.,  76,  2005,  2029
JOST,  D., 2115
JUHOS, L.,  833

KAMINSKI, E.J.,  551, 803,  1168
KARCHER,  W., 1341,  1349
KARHAUSEN,  L.,  367
O'KEEFFE, A.E.,  21O9, 2164, 2166
   2168, 2169, 2172
KEITZ, E.L., 1241, 1247, 1248
 KELLER, M.D., 4_2
 KEVANY, J., 89
 KILPIO, J.O., 2263
 KIRCHMANN, R., 1675
 KJELLSTR6M, T., 62,  1221,  1939,
   2197, 2252, 2254,  2257,  2299,
   1353, 2328, 2365,  2383
 KLAHRE-, p"  2231
 KLEINMAN, M., 431
 KNAUTH, P.,  2O31, 2O43, 2044
 KNEIP, T.J., 431, 466,  473
 KNELSON, J.H.,  181,  189,  190,
   361* 221
 KOPPLE, J.D., 847
 KORICANAC,  Z.f  2285
 KOSTA,  L.,  245
 KOTLAREK-HAUS,  S.,  685, 692, 693

-------
                               xc
  KRACKE,  W.,  1853
  KREBS/  H.,  2397
                                             f
KREUZER, W., 601, 1853, 2045,
  2299
    LOON,
LORKE, D.,
LOSER, E., 817
                                                '  Q3°' 831
 RUMMER,  J.,  1303,  1314,  1315,    LOVE,  G.J.,  3O1,  645
    1316,  1339
                                  LOWE,  A.C.,  2O91
 KUMPF M.J.,  339,  363,  365,  1957  LYNAM,  D.R
 LACOURLY, G., 1441
 LAFONTAINE, A., 382, 1131, 1942,
   2437
 LAMBO, T.A., 11
 LANESE, R.R., £7
 LANG, R., 659
 LANGER, H., 1757
 LANZOLA, E.,  1697
               .., 1339,  1340,
 LAUER, G.,  423, 2156,  2163
 LAUWERYS, R.,  542,  568, 631,
   831, 887, 1238,  2185  	
 LEEDER,  S.R.,  6_3, 76,  1111
 LEFEVRE,  M.J., 988,  2O29
 LEGRAND,  M.,  1131
 LEHNERT,  G.,  1317
 LEHTO, V.P.,  1007, 2315
 LELLOUCH, J.,  269
 LENGHEL,  I., 1583
 LEUNG, S.,  1248
 LEVERE, T.E.,  493, 506
 LEVY, D., 1263, 1275, 1276
 LEVY, E.A., 1797
 LINDr B., 2197
LINDVALL, T.,  119,  1954
LINNMAN,  L., 2197
                                MAGADUR, J.L., 1631, 1661
                                MAGE, D., 176, 207, 1275, 2068
                                  2156, 2158, 2162, 2164, 2165,
                                  2168, 2170, 2174, 2176, 2178
                                  2179, 2181, 2315, 2364
                                MAGE, D.T. ,  2097
                                MAGI, F., 857
                                MAGNAVAL, R. , 1531
                                MAHEU,  R.,  17
                                MALONE, D.W. , 1569
                                MANOJLOVIC, N.,  1Q31
                                MARCUS, A.H., 1505
                                MARQUARDT, H. ,  1607

                                                    1120'
                               MARTIN, J., 973
                               MARTIN, J.C., 1645,  1651
                               MATERNE, D. , 631
                               McCABE, E.B., 1168
                               MCDONALD, G.C., 1491
                               McGUIRE, J.M., 1399
                                           ' ^' sat' 586-
                               McNESBY, J.R. , 1371, 1383
                               MEININGER, J.f 603
                               van  MEIRHAEGE, A. , 2O44
                               MENENTI, M. , 1797
                               MERIAN, E., 929, 1396, 2175

-------
                             XCI
                               NORBACK,  D.H.,  385
                               NORSETH,  T.,  913
                               NOTTEN,  W.R.F., 2047, 2059, 206O
                               NURNBERG, H.W., 2231

                               OLEKSYK,  E.,  685
                               OLOFFS,  P.C., 702, 723, 75O,
                                 961, 1065,  1953, 235O
                               OMENN, G.S.,  1563, 1952
                               OREL, J.V., 2177
                               ORIOL, P., 269
                               OTT, W.R., 2097
                               OUW, K.H., 11O1

                               PACIGA,  J.,  1685
                               PACKHAM,  R.F.,  1468
                               PATTI, F., 2225
                               PEAT, J.K.,  63
                               PECORA,  L.,  1823
                               PELECH,  L.,  279
                               PERROTEY, J.,  1199
                               PERRY, R., 1285,  1328, 1385,
                                  1405,  1783
                               PETERS,  R.,  2397
                               PETERSON, R.W. , 3_
                               PFANNHAUSER, W.,  258, 858, 1328,
                                  1719,  2253
                               PHAM, Q.T.,  208,  470, 973, 988
NAGDA, N.L., 2O83              PHILP,  J.M., 2350
KANGNIOT, P.,1^5             PICHE,  L., 1314,  1468
NEEDLEMAN, H.L. , 584, 1155, 1166PI^f OR. M. ,^830,  951, 958,
  1167, 1168, 1169, 1936, 2433
NEUBERGER, M., 989
NEWHOUSE, M., 383, 1263, 1953
NEWILL, V.A., 161
NIEUWSTRATEN, N.,  1351
MERLUZZI, F., 231
MICHAELSON, I.A., 805
MILIC, S., 1829, 2285
MILOVANOVIC, Lf., 2265
MIRE, B., 973
MITCHELL, C.A., 669
MITCHELL, R.I., £7, 62
MOKEMATKENGUEMBA, G., 178,
  1941,  1951
MOLDOVAN, N., 1583
MOLLARET, P.,   1938
MONCELON, B., 1661
MONCHAUX, G., 1189
MONKMAN,  J.L.,  1015,  1331
MOONEY,  T.F., 1637
MOORE,  M.R.,  535,  1171,  2252
MOORE,  W.,  751
MOKEAU,  M.,  1315
MORGADE, C.,  695
MORGAN,  G.B.,  2073
MORRESI, N.,  231
MORRIS, S.C.,  677,  683
MOSE, J.R., 1617
MOULE,  Y.,  967
MRAK, E.M., 1965
MURAYAMA, H. , 9_1
MUSSENDEN, R.,  833
                                 PITTWELL, L.R., 2303
                                 PIVA' C"
                                 POTT' F"
                                 POWELL, W., 833

-------
                             XCII
PRINZ, B., 591, 1471, 1489
PTASNIK, J.A., 571

QUAGLIARDI, A., 1287
QUINOT, E., 1519

RABINOWITZ, M., 847
RAFFONELLI, A., 695
RAINSBURY, R. , 2397
RALL, D., 37
RAMACIOTTI, D., 659, 2316
REAY, J.S.S.,  1669
RECHT, P., 157, 1894, 1928,
  1938, 1946,  1961, 2641
REEVES, A., 724, 1637, 2385
RETHFELD, H.,  1407, 1416
REUSMANN, G.,  591
REUTER, L., 1O66
REY, P., 659,  667
ROBERTS, T.M., 1685
ROELS, H., 631, 887, 1237,
  2185
ROGGI, C., 1697
RONDIA, D., 523, 535, 1339
ROOSKEN, A.A., 1351
ROOTS, L.M.,  1113
van ROOYEN, G.I., 2382
ROSCA, Gh., 1073
ROSCA, S., 1073
ROSIVAL, L.,  877, 1187, 1237
ROSSANO, A.T., 2083
ROSSI, A,,  1823
RUBIN, R.J.,  903, 911, 912
RUDEN, H., 1757
RUDOLF, W., 2115
RUTENFRANZ, J., 2031
RUTZEL, H., 2231
RYLANDER, R., 137, 477, 484,
  667
SABATINI, G.C., 1697
SADOUL, P., 973
SALAZAR, S., 2O91
SANOTSKY, I.V., 1009, 2349,
  2409, 2433
SANSONI, B., 1853
SANTARONI, G.P.f 1030, 1451
SANTOLUCITO, J.A., 702, 1051,
  2387, 2394
SAUERHOFF, M.W., 8O5
SAVING, A., 931, 1749
SAYERS, M.H.P., 12O7
SCASSELLATI-SFORZOLINI G., 931,
  1749
SCHALLER, K.H., 1087
SCHNEIDER, T., 1954, 2071, 2155,
  2158, 2162, 2166, 2168, 2172,
  2174, 2175, 2178, 2182
SCHILLING, R.S.F., 669
SCHLATTER, C., 723
SCHLIPK5TER, H.W., 19O, 228,
  369, 521, 667, 723, 771,
  2328
SCHMIDT, P., 279, 354
SCHOENBERG, J.B., 669
SCHUCK, E.A., 2073, 2159, 2161,
  2164, 2171, 2173, 2176, 2177,
  2179, 2180, 2181, 2182
SCHULLER, P.L., 1017
SCHULZ, C.O., 903
SCHWING, R.C., 1491, 1502, 1503,
  2178, 2181
SCOPPA, M., 958
SEBASTIEN, P., 1189
SECCHI, G.C., 1123

-------
                             XCIII
SEIFERT, D., 14O7
SERWER, D., 2383
EL-SEWEFY, A.Z., 1589
SHAHEEN, H., 1589
SHAMS El-DEEN, A., 1589
SHAPIRO, I.M., 1155
SHAPIRO, M.A., 677
SHERWOOD, R.J., 471, 1340, 1954
SHUVAL, H.I., 1067
SILBERGELD, E.K., 793, 814, 1051
SILVERMAN, A.P., 1601
SKJAERASEN, J., 1233
SLATER, D.H., 1783
SHEETS, J., 1Q87, 2185
SMIDT, U., 1557
SMITH, R.G., 1637
SMITH, G., 1729
SORENSEN, S., 137, 477
STANKOVIC, B., 2285
STANKOVIC, M., 2285
STANKOWSKA, K., 685
STANLEY, R., 1729
STARA, J.F., 714, 751, 771, 772,
  813             	
STEELE, T.D., 1954, 1961, 2170
STEENSBERG, J., 472, 1006, 1960
  2432
STEPANEK, V., 1473, 1478, 1503
STEPHAN, W.I., 1779
STERN, A.C., 2143
STEVENS, L., 1001
STEWART, H.N.M., 1669
STIDL, H.G., 989
STOEPPLER, M., 2231, 2259
STOOFF, W., 1479
STOPPS, G.J., 229
STUIK,  E.J., 537
STUPFEL, M., 89, 77O, 844,
  1275, 1625, 1963, 2043
SUESS, M.J., 339
SULAIMAN, A.B., 657, 683, 1952
SUTHERLAND, L.C., 484, 485, 491
SWYNGEDOUW, I., 1131
SYROTA, M.J., 399, 457, 459,
  462
SZADKOWSKI, D., 1841, 2183, 2249,
  2256, 2259
SZPERLINSKI, Z., 1711

TAKABATAKE, E., 2197
TATI, M., 2251
TAYOT, J., 1199
TER HAAR, G., 803, 1177
THOMAS, R.S., 1331
THOMAS, T.J., 1569
THOMPSON, J.M. , 1779
TOMATIS, L., 1053, 2317, 2328,
  2329
TOMPKINS, E., 2197
den TONKELAAR, E.M., 1O17, 1O28,
  1029, 1030
TRAKOWSKI, A.C., 2455
TREMOLIERES, J., 19O2, 1921,
  1927, 1944, 1950
TRUFFERT, L., 2255, 23OO
TSUCHIYA, K.r 2197, 2351, 2364,
  2365              	
TULLIEZ, J., 921, 929
TUMASONIS, C.E., 879
TWIBELL, J.D., 1385, 1397, 1398
TYTUN, A., 545
VALENTA, P., 1416, 2231
VALJAREVIC, V., 1829

-------
                              XCIV
VANINI, G., 1289                YORDANOV, D., 1815
VEIL, S., 1, 2jj
VERBERK, M.M., 2005
               	             ZAPHIROPOULOS, M.,  1961
VOINIER, B., 659
             	                de ZEEUW, M., 1947
VOUK, V.B.,  467, 469, 473,
  1912, 1925, 1930, 1931, 1947  ZELENKO, V., 2£7
VUORI, E., 2263, 2296           ZIELHUIS, R.L., 175,  364,  537,
           	                   542, 543,  692,  1O40, 1100,
                                  1168, 2002, 2060, 2348,  2384
WALDBOTT, G.L., 1575            ZUNIC, R., 2386
WALLER, R.E., 77, 89, 90        ZWIERS, J.H.L., 1017
WANNAG, A., 1233, 1237, 1238
WARNER, P.O., 1001, 1006, 1O07,
  1008
WASSERMANN, D., 1053
WASSERMANN, M., 103, 156, 396,
  683, 702, 713, 877, 1053,
  1065, 1066, 1439, 1955
WATANABE, H., 91, 104
WEAVER, N.R. , 369
WEBER, 0., 260, 2261, 2295, 2297,
  2299, 2301, 2302, 2303
WEICHERT, N., 1417
WEIR, F.W., 1989, 2001, 2002,
  20O3, 20O4
WEISS, B., 2415, 2432, 2433
WESOLOWSKI, J.J., 471, 1729
WETHERILL, G.W., 847, 857, 858,
  859, 1298
WILCOX, S.L., 1241
WILLIAMS, H., 1601
WILLIAMS, M.K., 551
WILLIAMS, R. , 4_2
WISSMATH/ P., 1853
WOIDICH, H., 1719
WOOLCOCK, A.J., 6_3
WORTH, G., 1557
YANIV, S.L., 1249, 1260

-------
                                               SALES  OFFICES
 Belglque - BelgIS

 Moniteur be/ge - Belgisch Staatsblad
 Rue de Louvain 40-42 - Leuvenseweg 40-42
 1000 Bruxelles - 1000 Brussel
 Tel.5120026
 CCP 000-2005502-27 -
 Postrekening 000-2005502-27
 Sous-dap6t — Agentschap:

 Librairie europaenrte - Europese Boekhandei
 Rue de la Loi 244 - Wetstraat 244
 1040Bruxetles- 1040 Brussel
 OanmarK
  Ireland
  Stationery Office

  Beggar's Bush
  Dublin 4
  Tel. 668433
 Italia

 Libreria dello State
 Piazza G.Verdi 10
 00198 Roma - Tel. (8)8506
 CCP 1/2640
 Telex 62 006
 Agenzie:
 00187 Roma - Via XX Settembre
             (Palazzo Minislero
             del Tesoro)
 20121 M llano -Galleria
             Vitlorio Emanuele
             3 - Tel. 806406
  United States of America

  European Community Information Service

  2100 M Street, N.W.
  Suite 707
  Washington, D.C. 20037

  Tel. 296 5131
 Schweiz - Suisse - Svizzera

 Ltbrairfe Payer

 6, rue Grenus
 1211 Geneve
 Tel. 3169 SO
 CCP 12-236 Geneve
 J.H. Schu/tz - BoghanHel
 M«i1ergade19
 1116KebenhavnK
 Giro konto 1195
 Tel. 141195
Deutschland (BR)

Verlag Bundosanzeiger
SKoln 1 - BreiteStraBe - Postfach 106006
Tel, (0221)210348
(Fernschreiber; Anzeiger Bonn
08662595)
Postecheckkonto 83400 Koln
France

Service d« vente en France des publications
des Communautes europeennes
Journal officiel
26, rue Oesaix
75732 Paris - Cedex 15
TO. (1) 5786) 39 - CCP Paris 23-96
 Grand-Duch6
 de Luxembourg

 Office das publications officiates des Com-
 munautes europtiennos

 5, rue du Commerce
 Boftepostaie 1003 - Luxembourg
 T6I.490081 -CCP 191-90
 Compte courant bancaire:
 BIL 8-109/6003/300
 Nederland
Staatsdrukkerij- en uitgeveriibedrijf

Christoffel Plantijnstraat,
 s-Gravenhage
Tel.{070(81 4511
Postgiro425300
United Kingdom

H.M. Stationery Office

P.O. Box 569
London SE1 9NH
Tel. 01-9286977,ext. 365
 Sverfge

 LibrairieC.E. Fritze

 2, Fredsgatan
 Stockholm 16
 Post Giro 193, Bank Giro 73/4015
                                        Espafia
 Libreria Mundi-Prensa

 Castell637
 Madrid 1
 Tal, 2754655
Other countries


Office  for  Official  Publications  of  the
European Communities

5. rue du Commerce
Bolte postals 1003 - Luxembourg
Tel. 490081 -CCP 191-90
Compte courant bancaire:
BIL 8-109/6003/300

-------