September 1977
Environmental Health  Effects Research Series
                                                 CARBON  MONOXIDE
                                                    Health Effects Research Laboratory
                                                   Office of Research and Development
                                                  U.S. Environmental Protection Agency
                                           Research Triangle Park, North Carolina  27711


Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination of traditional grouping  was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health Effects Research
     2.  Environmental Protection Technology
     3.  Ecological Research
     4.  Environmental Monitoring
     5.  Socioeconomic Environmental Studies
     6.  Scientific and Technical Assessment Reports (STAR)
     7.  Interagency Energy-Environment Research and Development
     8.  "Special" Reports
     9.  Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series. This series describes projects and studies relating to the toler-
ances of man  for unhealthful substances or conditions. This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies. In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques  utilizing ani-
mals — but always with  intended'application to human health measures.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

                                  September 1977

      Subcommittee on Carbon Monoxide
Committee on Medical and Biologic Effects of
          Environmental Pollutants
         National Research Council
        National Academy of Sciences
              Washington, B.C.
          Contract No.  68-02-1226
              Project Officer

              Orin Stopinski
    Criteria and Special Studies Office
    Health Effects Research Laboratory
    Research Triangle Park, N.C. 27711


     This report has been reviewed by the Health Effects Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication.  Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.

     The project that is the subject of this report was approved by the
Governing Board of the National Research Council, whose members are
drawn from the Councils of the National Academy of Sciences, the National
Academy of Engineering, and the Institute of Medicine.  The members of
the Committee responsible for the report were chosed for their special
competences and with regard for apropriate balance.

     This report has been reviewed by a group other than the authors
according to procedures approved by a Report Review Committee consisting
of members of the National Academy of Sciences, the National Academy of
Engineering, and 'the Institute of Medicine.


     The many  benefits  of  our modern, developing,  industrial  society  are
accompanied  by certain  hazards.   Careful  assessment  of  the  relative risk
of existing  and new man-made environmental  hazards is necessary  for the
establishment  of sound  regulatory policy.   These  regulations  serve  to
enhance the  quality of  our environment  in order to promote  the public
health and welfare and  the productive capacity of our Nation's population.

     The Health Effects Research Laboratory,  Research Triangle Park,
conducts a coordinated  environmental health research program  in  toxicology,
epidemiology,  and clinical studies using  human volunteer  subjects.  These
studies address problems in air pollution,  non-ionizing radiation,
environmental  carcinogenesis and the toxicology of pesticides as well as
other chemical pollutants.  The Laboratory  develops  and revises  air quality
criteria documents on pollutants for which  national  ambient air  quality
standards exist or are  proposed, provides the data for  registration of new
pesticides or  proposed  suspension of those  already in use,  conducts research
on hazardous and toxic  materials, and is  preparing the  health basis for
non-ionizing radiation  standards.  Direct support to the  regulatory function
of the Agency  is provided  in the form of  expert testimony and preparation of
affidavits as  well as expert advice to  the  Administialor  to assure  the
adequacy of  health care and surveillance  of persons  having  suffered imminent
and substantial endangerment of their health.

     To aid  the Health  Effects  Research Laboratory to fulfill the functions
listed above,  the National Academy of Sciences (NAS) under  EPA Contract
No. 68-02-1226 prepares evaluative reports  of current knowledge  of selected
atmospheric  pollutants. These  documents  serve as background  material for
the preparation or revision of  criteria documents, scientific and technical
assessment reports, partial bases for EPA decisions  and recommendations
for research needs.  "Carbon Monoxide"  is one of  these  reports.
                                        John  H.  Knelson,  M.D.
                                  Health Effects Research Laboratory

                                                                                    COMMITTEE ON MEDICAL AND BIOLOGIC EFFKCTS OF ENVIRONMENTAL POLLUTANTS
RONALD F. COBURN, University of Pennsylvania School of Medicine,

     Philadelphia, Pennsylvania, Chairman

ERIC R. ALLEN, State University of Hew York Atmospheric Sciences

     Research Center, Albany, New York

STEPHEN M. AYRES, St. Louis University School of Medicine, St. Louis,


DONALD BARTLETT, JR., Dartmouth Medical School, Hanover, New


EDWARD F. FERRAND, New York  City Department of Air Resources,

     New York, Now York

A. CLYDE HILL, University of Utah, Salt Lake City, Utah

STEVEN M. HORVATH, University of California Institute of Environ-

     mental Stress, Saitta Barbara, California

LEWIS H. KULLER, University of Pittsburgh Graduate School of Public

     Health, Pittsburgh, Pennsylvania

VICTOR G. LATIES, University of Rochester School of Medicine and

     Dentistry, Rochester, New York

LAWRENCE D. LONGO, Loraa Linda University  School of Medicine,  Santa

     Barbara,  California

EDWARD P. RADFORD, JR., The Johns .Hopkins University School of

     Hygiene and Public Health, Baltimore, Maryland

JAMES A. FRAZIER, National Research Council, Washington, D.C.,

     Statf Officer
 HERSC1IEL E.  GRIFFIN,  University of Pittsburgh, Pitrsburgh, Pennsylvania,


 RONALD F. COBURN, University, of Pennsylvania School of Medicine, Philadelphia


 T. TIMOTHY CROCKER, University of California College of Medicine, Irvine,


 CLEMENT  A. FINCH, University of Washington School of Medicine, Seattle,


 SHELDON  K. FRIEDLANDER, California Institute of Technology, Pasadena,


 ROBERT I. HENKIN, Georgetown University Medical Center, Washington,  n.C.

 IAN T. T. niGCINS, University of Michigan, Ann Arbor, Michigan

 JOE W. HIGirrOWER, Rice University, Houston, Texas

 HENRY KAMIN, Duke University Medical Center, Durhnm, North Carolina

ORVILLE  A. I.EVANDER, Agricultural Research Center, Beltsvlllc, Maryland

DWIGKT F. METZLER, Kansas State Department of Health and Environment, Topeka,


I. HERBERT SCHEINBERG, Albert Einstein College of Medicine, Bronx, New York

RALPH G. SMITH, University of Michigann, Ann Arbor,  Michigan

ROGER P. SMITH, Dartmouth Medical School, Hanover, New Hampshire
T. D. BOA7., JR., National  Research Council, Washington,  U.C.,  Executive  Director


     Members of the Subcommittee on Carbon Monoxide, under the Chairmanship of

Dr. Ronald F. Coburn, wrote this report.  It is appropriate to mention here

that the same people also wrote the section on carbon monoxide in the report

for the U.S. Senate Committee on Public Works.*

     Drs. Coburn and Eric R. Allen wrote the introduction.  Dr. Allen also

wrote Chapter 2, on properties and reactions, and Chapter 3, on sources,

occurrence, and fate of atmospheric carbon monoxide.  Dr. Edward F. Ferrand

wrote Chapter 4, on environmental analysis and monitoring, and Appendix A,

on methods of monitoring.

     For Chapter 5, Dr. Edward P. Radford, Jr., wrote the material on uptake;

Drs. Coburn and Donald Bartlett, Jr., on physiologic effects; Dr. Lawrence D.

Longo, on effects on the pregnant woman, the developing embryo, the fetus, and

the newborn infant; Dr. Lewis H. Kuller, on cardiovascular effects; Dr. Victor G.

Laties, on behavioral effects; Dr. Steven M. Horvath, on effects during exercise

and effects on populations especially susceptible to carbon monoxide exposure

owing to reduced oxygenation at altitudes above sea level; Dr.  Bartlett, on the

effects of chronic or repeated exposure; and Dr.  Coburn, on dose-response character-

istics in man.

     Dr. A. Clyde Hill wrote Chapter 6,  on the effects of carbon monoxide on

bacteria and plants.
*National Academy of Sciences,  National Academy of Engineering.   Coordinating
 Committee on Air Quality Studies.   Air Quality and Automobile Emission Control.
 Vol.  2.   Health Effects of Air Pollutants.   U.S.  Senate Committee Print Serial
 No. 93-24.  Washington, D. C.:  U.S.  Government Printing Office, 1974.  511 pp.

     Chapters 7 and 8, the statements in the summary, and recommendations, were

written by the members of the Subcommittee and assembled by Dr. Coburn.

     Dr. Radford wrote Appendix B, on measurement in biologic samples.

     The Environmental Protection Agency's Air Pollution Technical Information

Center supplied information from its computer data base.  Dr. Robert J. M.

Horton of theEPA obtained various technical and scientific documents and other

resource information.  The staff of the NRC Assembly of Life Sciences Advisory

Center on Toxicology gave assistance in obtaining resource information and also

reviewed the report.

     The report was reviewed by the Academy's Report Review Committee with the

assistance of anonymous reviewers selected by that Committee.  The members of

the MBEEP Committee reviewed the report in depth.  It was also reviewed by the

Associate Editor, Dr. Henry Kamin, and five anonymous reviewers selected by him.

     The staff officer for the Subcommittee on Carbon Monoxide was Mr. James A.

Frazier.  We also acknowledge the staff support of Mrs. Renee Ford for editing

the report, Ms. Joan Stokes for preparation and verification of the references,

and Mr. Norman Grbssblatt for his editorial assistance.

1.     Introduction

2.     Properties and Reactions of Carbon Monoxide

3.     Sources, Occurrence, and Fate of Atmospheric
          Carbon Monoxide

4.     Environmental Analysis and Monitoring

5.     Effects on Man and Animals

6.     Effects on Bacteria and Plants

7.     Summary and Conclusions

8.     Recommendations

       Appendix A:  Methods of Monitoring Carbon Monoxide

       Appendix B:  Measurement of Carbon Monoxide in
                       Biological Samples


                                   CHAPTER 1.


     Man has experienced the effects of carbon monoxide poisoning at least since

that period in prehistory when he first discovered the art of making and utilizing

fire.  Numerous accounts of tragic events, circumstances, and phenomena that can

be directly or indirectly attributed to the toxic properties of carbon monoxide

have been related in folklore and mythology.  Lewin,  who traced the early history

of carbon monoxide, was led to the conclusion that this form of poisoning is

unique in its close association with the history of civilization.  For example,

carbon monoxide poisonings drastically increased in the fifteenth century when the

use of coal for domestic heating increased.  These poisonings were attributed both

to inhalation of carbon monoxide formed by incomplete combustion in the heating of

homes and to the exposure of coal miners to the deadly "white damp," encountered

after underground explosions and mine fires.

     The introduction of illuminating gas (a mixture of hydrogen, carbon monoxide,

methane, and other hydrocarbons, also known as carburetted water gas) for domestic

heating purposes further increased the hazard of carbon monoxide poisoning.

Although this fuel is still used extensively in Europe, it has largely been

replaced in the United States with natural gas.   More recently,  the introduction

of the internal-combustion engine and the development of numerous technological

processes in which carbon monoxide is produced  have increased still further the

hazard of exposure to this toxic gas.  Despite awareness for many centuries  that

human exposure to combustion fumes was hazardous, it was not until 1919 that

industrial production of carbon monoxide was recognized to be an environmental

health problem of national importance.  At the First International Congress  of

Labor, the rapidly increasing use of the internal-combustion engine as a source of

industrial power was cited as a major contributor of the carbon monoxide being


inhaled by workers.  The exhaust from the internal-combustion engine is the

principal contemporary anthropogenic source of carbon monoxide.  Over the

centuries, the problem of dealing with carbon monoxide exposure has expanded from

dwellings to work environments and now includes    the ambient air in cities.  This

report reflects our concern with the adverse effects of exposure to carbon monoxide

at the concentrations found  in our urban and  industrial air.

     At the time that the Air Quality Criteria for Carbon Monoxide appeared in 1970,

data was produced       which suggested that, when man received an acute carbon

monoxide exposure that produced carboxyhemoglobinemia as low as 3 percent satura-

tion, there were adverse effects on complex mental functions such as vigilance.

There was also epidemiologic evidence relating the incidence of myocardial

infarction and the concentration of carbon monoxide in air.  The possibility was

raised that a significant fraction of the urban population might be experiencing

adverse health effects due to carbon monoxide.  A major criticism was that it was

not known whether the data could be extrapolated to urban population groups.

     Studies on the biologic effects of carbon monoxide on man since 1971

apparently support the conclusion that carboxyhemoglobin levels as low as 3-5%

may have effects on vigilance and aerobic metabolism under conditions of

exercise at maximal oxygen uptake.  It is uncertain whether these experimental

results can be extrapolated to urban populations, but it is strongly suspected

that they can.  In addition, an awareness of a spectrum of carbon monoxide

susceptible populations is growing, particularly* .in patients with  coronary  vascular

diseases and the fetus.

     This document summarizes the carbon monoxide literature related

to effect's on man and his environment for the consideration of the Environmental

Protection Agency in  updating the information in the Air Quality Criteria on


Carbon Monoxide.  It emphasizes recent major advances in our knowledge of carbon

monoxide:  chemical reactions in air; biologic effects on man; problems in

monitoring urban concentrations and relating such data to the exposure of popu-

lations; data concerning the identification of susceptible populations; and

evidence implicating carbon monoxide as a causal factor in disease.  We have not

tried to review all published articles but only those deemed to be the important

studies related to carbon monoxide air quality criteria.  There is a large

literature on adverse effects of cigarette smoking and some of these effects may

be related to carbon monoxide.

                                   CHAPTER  2


     Carbon monoxide  (CO) is an imperceptible poisonous gas, the most common

source of which is the incomplete combustion of carbonaceous materials.  It is

probably the most publicized and the best known of all air quality criteria

pollutants* because of the frequency of accidental deaths attributed to its

inhalation over the years.  Its toxic and sometimes	?

lethal properties, however, are due to acute effects resulting from exposures

to very high concentrations in confined spaces, generally exceeding 500 ppm for

several hours.  In this report we are mainly concerned with the deleterious ef-

fects resulting from human exposures to much lower concentrations over consider-

ably longer periods of time.  In the latter context we are also concerned with

the role this oxide of carbon plays as a chemically reactive environmental

pollutant and atmospheric trace constituent.  This requires a detailed quantita-

tive understanding of major physical and chemical factors governing production,

control, transformation,  and removal of carbon monoxide prior to and during its

atmospheric life cycle.  Recently,  there has been a renewal of interest in reac-

tions involving carbon oxides because of their fundamental importance in the

genesis and evolution of  planetary atmospheres as well as in pollutant-forming

combustion, flame and explosion processes and in terrestrial atmospheric chemistry.

     In this report the fundamental scientific and technical knowledge concern-

ing carbon monoxide in its role as an air pollutant affecting public health and

welfare is updated and assessed.  Special emphasis is  placed on the more recent

advances in our understanding and relevant discoveries made during the last five years.
*Those pollutants for which an air quality.criteria document has been published
 as required by the Clean Air Act.


     For a comprehensive review of pertinent literature published prior to 1972,

there are two excellent reports, one commissioned by the National Air Pollution

Control Administration, U.S. Public Health Service^   and the other by the North

Atlantic Treaty Organization, Committee on Challenges of Modern Society,298 pub-

lished in March 1970 and June 1972, respectively.  They treat the topic of air quality

criteria for carbon monoxide in considerable detail and provide a basis for this

updated review.  In addition, Cooper" has compiled an extensive carbon monoxide

bibliography with abstracts of the literature published prior to 1966, that pro-

vides a further source of information.

General Physical and Chemical Properties

     Carbon monoxide, a stable compound, is a heteropolar diatomic molecule.

It absorbs radiation in the infrared region corresponding to the vibrational ex-

citation of the electronic ground state of the molecule,    CO (X £ ).  Because

thermal population of excited vibrational levels is extremely inefficient, even

at 1000 C less than 1% of all the molecules present will reside in the first

vibrationally excited (v'~l) state above the ground vibrational state (v"=0).

Radiation in the visible and near ultraviolet regions of the electromagnetic

spectrum is not absorbed by carbon monoxide, but in the vacuum ultraviolet

region, a structured relatively weak absorption band extends from 155 to 125 nm.

In this  so called  spectroscopic fourth positive band system of carbon monoxide

the electronic transition COCA^ •<- X1^ ) occurs.167'172  The characteristics of

                             frequency ^dependent

the absorption spectrum and the/extinction coefficients, as well as the photo-

chemistry of carbon monoxide in the vacuum ultraviolet region, have been reviewed

and discussed elsewhere.  '     Interest in carbon monoxide photochemistry has

recently been stimulated by the discovery of this gas at about 0.1% in the Martian

           26 197                                             98 344
atmosphere,  '    and by detection in the atmosphere of Venus.  '     These


discoveries suggest that it plays a significant role in the development  of

primitive planetary atmospheres in our solar system.

     Carbon monoxide has a low electric dipole moment  (0.10 Debye) , short

interatomic distance (1.13 A), and high heat of formation from atoms, or  bond

strength (1,072 kj/mole) suggesting that .this diamagnetic molecule is a

resonance hybrid of the three structures ;
a)  :C: 0:              b) :C::0:               c)
                                                          —   +
These structures correspond to forms with single, double;and triple covalent

             + _          _ +
bonds, i.e., C-0, C=0 and C=0.  They all apparently contribute about equally

to the normal state of the molecule, thus counterbalancing the opposing effects

of the number of covalent bonds and charge separation.  Carbon monoxide is iso-

electronic with molecular nitrogen  (N'2) , the nitrosyl cation (NO"*"), and the cyanide

anion (CN ).  The similarity to nitrogen causes difficulties in its physical

separation and identification in air.  It is a colorless, odorless and taste-

less gas  that is slightly lighter than air and difficult to liquefy.  Although

an anhydride of formic acid (HCOOH) , it is unreactive with water and is only

slightly soluble.  General physical properties of carbon monoxide are presented

in Table 1.

Production and Preparation282 » 308

     Carbon monoxide is produced when carbon or combustible carbonaceous ma-

terial is burned in a limited supply of air or oxygen, i.e., under fuel -rich


     It is manufactured on a large scale by reducing carbon dioxide with carbon

at high temperatures.   Below 800 C the reduction is slow but above 1000 C the

conversion, corresponding to the endothermic reaction (1) ,  is quite fast and

efficient .


                            TABLE 2-1

          Physical Properties of Carbon Monoxide  —
 Molecular weight

 Critical point

 Melting point

 Boiling point

 Density, at 0 C,  1  atm

          at 25 C,  1 atm

 Specific gravity  relative to air

 Solubility in water3-,  at 0 C

                       at 20 C
                       at 25 C
Explosive  limits  in air
Fundamental  vibrational  transition
               1 -e v" =  0)
Conversion  factors:
   at  0 C,  1  atm
   at 25 C,  1 atm.

-140 C at 34.5 atm.

-199 C

-191.5 C

1.250 g/liter

1.145 g/liter


3.54 ml/ 100 ml
 2.32 ml/100 ml
(29.0 ppmm)

2.14 ml/100 ml
(26.8 ppmm)

12.5 - 74.2%

2,143.3 cm"1
(4.67 urn)
1 mg/m3 = 0.800 ppm £•
1 ppm  = 1.250 mg/m3
1 mg/m  = 0.873 ppm
1 ppnr  = 1.145 mg/m3
—Volume of carbon monoxide is at  0 C,  1  atm (atmospheric  pressure
 at sea level =760 torr)

—Farts per million by mass (ppmm  = mg/1)

•^•Parts per million by volume

C(s) + C02(g) -> 2CO(g) - 163 kJ/mole
     The monoxide is a major constituent of the synthetic fuels,  "producer

gas" (25% CO) and "water gas"  (40% CO).  The former, used mainly  for  heating

purposes, consists primarily of nitrogen and carbon monoxide and  is prepared

by passing air through a bed of incandescent coke, the residue of coal  remain-

ing after destructive distillation.  If coal is used in place of  coke,  coal

gas will also be present.  This fuel mixture, commonly used for domestic

heating in Europe, has been almost completely replaced by natural gas in the

United States.  "Water gas" is made when steam is blown through incandescent

coke at 1100 C.  It is a mixture of hydrogen (49%) and carbon monoxide  (44%)

                        H20(g) + C(s) -»• C0(g) + H2(g)                   (2)

with traces of nitrogen (4%), carbon dioxide (2.7%) and methane (0.3%).

     "Semi-water gas" is produced by blowing a mixture of steam and air through

incandescent coke.  Carbureted (enriched) water gas,  which burns with a lumin-

ous flame, is prepared by mixing water gas with partially unsaturated hydro-

carbons.  "Water gas" burns with a blue, nonluminous  flame,  produces  consider-

able heat -In combustion, and may be used with Welsback mantles for illuminating

purposes.  The calorific values (available fuel energy) for combustion of pro-

                                                         3          6    ^
ducer and semi-water gas are low, of the order 125 Btu/ft  (4.4 x 10   J/m ) ,

compared with 350 Btu/ft3 (12.3 x 106 J/m3)  for water gas and 600 Btu/ft3

        6    3
(21 x 10  J/m ) for coal gas.  Carbon monoxide can also be produced by several

other methods , including :

     •  Reduction of carbon dioxide with zinc dust or iron filings at red heat —

        for example, with zinc dust.

                        C02(g) + Zn(s) ->• C0(g) + ZnO(s)                 (3)


     •  Heating charcoal with either zinc, iron or manganese oxides—

        for example, with zinc.

                        C(s) + ZnO(s) -»• C0(g) + Zn(s)                    (4)

     •  Heating carbon with certain alkaline earth carbonates,  such

        as chalk or barium carbonate—for example, with barium  carbonate.

                        BaC03(s) + C(s) -> BaO(s) + 2CO(g)                (5)

On the laboratory scale, carbon monoxide for analytical and other purposes can

be conveniently prepared by the following methods:

     •  The reaction of concentrated sulfuric acid (H2S04) with formic

        acid at 100 C (reaction 6), or oxalic acid at 50 C  (reaction 7),

        or sodium formate (reaction 8) or potassium ferrocyanide at room

        temperature (reaction 9):

                           HCOOH   -»•   H20 + C0(g)                        (6)
                                 100 C

                           '   H2S04
                       (COOH)0  ->      H90 + C0(g) + C09(g)               (7)
                             ^ 50 C                    Z

               2HCOONa.+ H-SO,     Na0SO. + 2H« + 2CO(g)                  (8)
                                             25 C   *

                                                     >,SO,                (9)

In the reaction with oxalic acid  (reaction  7),  note  that  equal volumes of

carbon dioxide and carbon monoxide are produced.   Some  sulfur  dioxide (S02)

is generated in all the reactions (reactions 6-9)  by the  reduction of sulfuric

acid as shown in reaction (10):

                 H2S04 + C0(g) + H20 + S02(g) + C02(g)                  (10)

     •  The reaction of chlorosulfonic acid with formic acid also

        produces carbon monoxide  (reaction  11):

                   HCOOH + CS03C1 -»• HjSC^  + HC1 + C0(g)                UD
In the above procedures the carbon monoxide produced can be purified by bubbling

through caustic soda and then drying over phosphorus pentoxide.

     •  Thermal decomposition of nickel carbonyl at 200 C yields a high

        purity product:

                        Ni(CO)4(g) -»• Ni(s) + 4CO(g)                    (12)

     Carbon dioxide absorbs solar ultraviolet radiation in its first two

absorption bands between 170 and 120 nm.  At wavelengths less than 165 nm,

sufficient energy is available to dissociate the dioxide into the monoxide

and electronically excited atomic oxygen^O according to the spin-conserving

reaction (13).

             G00(X1zt) + hv ->- CO(X1£j') + 0(1D)                         (13)
               L    8               *

Here C02(X Eg) is the electronic ground state of carbon dioxide, CO(X zf)

is the electronic ground state of carbon monoxide and (0 D) is an electronically

excited state of atomic oxygen.


The quantum efficiency (number of molecules of product per absorbed photon of

radiation) of this process is almost unity near 150 nm.

     Carbon monoxide is also a product of the pyrolysis or photolysis of many

oxygenated organic compounds.  For example, at wavelengths below 340 nm ali-

phatic aldehydes/which are constituents of photochemical smog, may photolyze

directly into hydrocarbon and carbon monoxide as shown in reaction (I4a):

                      RCHO + hv •> RH + CO                              (14a)

In the case of formaldehyde  (HCHO), the quantum efficiency of this process is

about 0.5 at 313 nm and ambient temperatures, but the efficiency is much less

for higher aliphatic aldehydes.  The photolysis of aldehydes also produces

acyl radicals (RCO) according to reaction (14b)|

                     RCHO + hv + RCO + H                               (14b)

In the absence of molecular oxygen, acyl radicals can/decompose into hydrocarbon

radicals (R) and carbon monoxide.  However, in air this reaction is suppressed
by the dominant formation of acylperoxy  radical adducts (RCOo) with oxygen.


Chemical Reactions of Carbon Monoxide


     Carbon monoxide is quite stable and chemically inert under normal condi-

tions (25 C and 1 atm) despite a carbon valency of two;  At higher tempera-

tures it becomes reactive, behaves as though unsaturated and can act as a

powerful reducing agent—a property used in many metallurgical processes, such

as in blast furnaces.

     At high  temperature,  in  the presence  of  a catalyst (palladium, iron,

nickel),  carbon monoxide undergoes  the  reversible  disproportionation reaction


                        2CO + C + C02 +162 kJ/mole'1                   (15)

As the temperature is increased the equilibrium fraction of  carbon dioxide

decreases, for example, the equilibrium percentages by  volume of dioxide

are   90  at 550 C, 50 at 675  C, 5 at 900 C and less than 1 at 1000 C.   The

reverse reaction is utilized  industrially  to  manufacture the monoxide  (see

reaction  1).

     Photolysis of carbon monoxide  at 129.5 nm ftenon radiation)  results in the

disproportionation into carbon dioxide  and carbon  suboxide (C-C^).      Since

the absorption of radiation at wavelengths <£lll nm is  required  to  photodissociate

carbon monoxide into its component  atoms it has been postulated  that the photo-

chemical  reaction involves electronically excited molecules,  CO  (A  ir),  and

the    following sequence of reactions has been suggested:

                             CO + hv -»• CO(A1Tr)                          (16)

                        CO (A1 IT) + CO ->• C02 + C                          (17)

                              C + CO -v C20                              (18)

                            C20 + CO -> C302                             (19)

At this wavelength (129.5 nm) the quantum efficiency of carbon monoxide  re-

moval is 0.8 ± 0.4 but at 147 or 123.6 nm it is almost zero.15*  similar

products are formed by electrical and high frequency discharges  in  carbon


     Combustion. "°  Carbon monoxide burns in air or oxygen with a bright blue

flame but does not itself support combustion (see reaction 20).

                        2CO + 02 -* 2C02 + 565 kj/mole                   (20)

Although heat is evolved during combustion, the fuel value is low (320 Btu/ft^
            6    o
or 11.3 x 10  J/m ).  The blue flames seen on top of a clear fire consist of

burning carbon monoxide.  It is presumably produced in fires due to the reduc-

tion of carbon dioxide  that is formed in the lower regions of glowing fuel

near the entering air draught as it rises through an incandescent mass of

carbon (see reaction 21).  The carbon monoxide burns on top of the fire where

an excess of air is available.  An interesting feature of these processes is

that the reaction of carbon with oxygen at temperatures producing carbon

dioxide generates heat (exothermic process—see reaction  21), whereas the

                       C(s) + 02(g)->- C02(g) + 1109 kJ/mole              (21)

 reaction of carbon dioxide with carbon at the same temperature absorbs heat

 (endothermic process—see reaction  11).  Thus, it is possible to adjust the
 ratio of air* (or oxygen) to carbon dioxide,  so that a desired temperature

 can be maintained continuously.   In industry,  large quantities of carbon

 monoxide are formed during the incomplete combustion of charcoal or coke,

 and to a lesser extent by other carbonaceous fuels, in a limited supply of
 air,i.e., .fuel-rich conditions.   The presence of carbon monoxide in furnace

 gases is used as an indicator of improper air supply and its estimation in

 flue gases is used as a check on the operating efficiency of the furnace.

      Stoichiometric mixtures (2:1 by volume) of carbon monoxide and oxygen

 explode upon ignition in the presence of trace amounts of moisture or hydrogen-

  ous compounds, such as methane (CH^) or hydrogen sulfide (f^S).  These neces-

 sary impurities apparently act as a source of oxyhydrogen free radicals


 (hydroxyl and hydroperoxyl), which provide low energy reaction paths  leading

 to an explosive chain branching mechanism.  Explosive limits in oxygen  range

 from 15.5 to 93.9% by volume of carbon monoxide.  These proportions may be

 compared with the limits in air given in Table 2-1.

     Heterogeneous Reactions.  Carbon monoxide is the anhydride of formic

 acid but it does not react with water (liquid or vapor) at room temperature.

 However, the forward reaction in the equilibrium with formic acid vapor shown

 below (see reaction 22) may be achieved by the application of an electrical

 or high frequency discharge in stoichiometric mixtures (1:1) of carbon  monoxide

                        C0(g) + H20(g) * HCOOH(g)                       (22)

 and water vapor.  The reverse dehydration process is accomplished by the cata-

 lytic action of metallic rhodium.   Carbon monoxide at 120 C and 3 to 4  atmospheres

 pressure is rapidly and completely absorbed by a concentrated solution  of caustic

 soda forming sodium formate.  This salt is also produced when carbon monoxide is

 passed over caustic soda or soda lime heated to 200 C.  (See reaction 23.)

                        NaOH(s) + C0(g)  -»• HCOONa(s)                      (23)

Anhydrous formic acid is manufactured economically in quantity from sodium

 formate by distillation with concentrated sulfuric acid.

     In the presence of a suitable catalyst,  such as freshly-reduced metallic

nickel,  carbon monoxide can be hydrogenated to methanol (CHoOH),  methane (CH,)

 or other organic compounds depending upon the conditions  selected.   Its conver-

 sion to methane is the basis for the flame ionization gas chromatographic separa-

 tion procedure used for the detection and estimation of carbon monoxide in ambient


     When carbon monoxide is passed over heated metal oxides such as lead

oxide  (PbO) in reaction  (24), oxygen is extracted leaving the metal.

                        PbO(s) + C0(g) + C02(g) + Pb(s)                 (24)

Rapid and quantitative reaction with various oxides at high temperatures has

been used for the estimation of carbon monoxide; some examples are:

     •  with cuprous oxide at room temperature or copper-cupric oxide

        at 270-300 C t carbon dioxide produced is quantitatively absorbed

        by caustic soda solution and the volumetric loss on absorption  is

        measured in the Orsat-Lunge apparatus.

     •  with iodine pentoxide at 90 C,iodine vapor is formed stoichiometri-

        cally by reaction (25) and the iodine is collected and estimated by


                        I205(s) + 5CO(g) •> I2(g) + 5C02(g)              (25)

     •  with mercuric oxide at 150 C,mercury vapor is released which is

        measured spectrophotometrically at 254 nm.

                         HgO(s) + C0(g) -»• Hg(g) + C02(s)                (26)

     The rate of oxidation of carbon monoxide in oxygen, although insignificant

in homogeneous mixtures, is known to be enhanced by metallic catalysts  such as

palladium on silica gel or a mixture of manganese and copper oxides (Hopcalite).

It has also been shown that carbon monoxide is made catalytically active by ad-

                                  I f\ O

sorption on hot metallic surfaces.     The reactions of nitrous oxide (N-O)

with carbon monoxide chemisorbed on copper, charcoal, Pyrex glass or quartz

surfaces at high temperatures (above 300 C) have been studied extensively214,255,369,

and found to be quite efficient in oxidizing carbon monoxide (see reaction 27) .


              C0abs + N2°(g) "" C°2(g) + N2(g)                           (27)

     Carbonyl and Coordination Compounds.282'362  In the presence  of  light,

equal volumes of carbon monoxide and individual halogens (F2,Cl2,Br2,I2)  or

cyanogen (C2N2) react to form the corresponding volatile and highly reactive

carbonyl halides or cyanide .  In reaction  (28):

                             CO + X2 -»• COX2                             (28)

X is a univalent element.  Carbonyl chloride (phosgene), formed by reaction

with chlorine (C12), is the best known of these compounds and is highly

poisonous at a concentration which has been suggested to be considerably  less

than that for carbon monoxide.  It is used frequently as the solvent  in non-

aqueous systems of acids and bases.  Phosgene undergoes ammonolysis when  passed

into a solution of ammonia in toluene, forming urea (CO(NH2)2) an^ tne salt

              COC12 + 4NH3 •»• CO(NH2)2 + 2NH4C1                         (29)

ammonium chloride, which provides an effective means of removing the  phosgene.

It is also used in the manufacture of urea.  Carbonyl fluoride (COF2) is  ex-

tremely reactive attacking glass and many metals.

     When carbon monoxide is passed over heated sulfur or selenium, carbonyl

sulfide (COS) and selenide (COSe),respectively, are produced.  All of the

carbonyl compounds mentioned above are used widely and frequently in  industrial

organic syntheses and organic chemical production.

     Carbon monoxide forms volatile metallic carbonyl compounds in which  the

carbon monoxide molecule exclusively is attached to a single metal atom or

group of metal atoms.  These include the carbonyls of chromium, molybdenum

and tungsten in Group VI of the periodic table of the elements; rhenium in


Group VII; iron, ruthenium and osmium in Group VIIIA; cobalt, rhodium and

iridium in Group VIIIB; and nickel in Group VIIIC.  Although these are coordina-

tion compounds, their unusual composition is determined more by their tendency

to form closed  (filled) electronic shells rather than by the valence of the

central metal.  For example, in simple binary carbonyls, containing one metal

atom (M) in the complex, i.e., (M[CO] ), the carbon monoxide donates two electrons

to the shellj thus, the effective atomic number  (EAN) of the metal M becomes the

atomic number of the next higher inert gas element for a stable complex.

(Atomic number' of element + 2y.)  Thus, the EAN's for some of the metals mentioned

above are:  36 in chromium hexacarbonyl, Cr(CO)/-, iron pentacarbonyl f Fe(CO)ijj

and nickel tetracarbonyl>Ni(CO)^; 54 in molybdenum hexacarbonyl,MO(CO)g, ruthenium

pentacarbonyl Ru(CO)c*and 86 in tungsten hexacarbonyl,W(CO)g,and osmium penta-

carbonyljOs(CO) .  Similar considerations apply to the complex carbonyls with

several metal atoms or other co-coordinated groups.

     The addition of carbon monoxide to metallic elements, as shown in .reaction

(30) for the case of nickel tetracarbonyl, occurs with a large decrease in

volume and the formation is promoted by employing high pressures (750 atms).

                          Ni(s) + 4CO(g) •*• Ni(CO)4(l)                  (30)

In the large scale commercial production of iron pentacarbonyl, reduced iron

is heated at 180 to 200 C under 50 to 200 atmospheres of carbon monoxide.

Cobalt, molybdenum and tungsten carbonyls are prepared on a commercial scale

under similar conditions with sulfur as a promoter.  These carbonyl compounds

are used in the separation of metallic elements  as a source of highly purified

metals^ and in the manufacture of plastics.  Nickel carbonyl is as highly  poisonous

as phosgene at concentrations comparable (1 ppm) j these toxicity levels are well

below that

for carbon monoxide, and these compounds  should,  therefore,  be handled with great care.

At elevated temperatures (>2QO C) the  gaseous  binary metallic carbonyls decompose

into carbon monoxide and deposit a high-purity metallic  element.   Similarly,

iron and nickel carbonyls in the gas phase or in solution can  dissociate  readily

at room temperature when irradiated at 366 nm.

     Carbon monoxide can penetrate heated iron and escape through  the  iron  flues

of stoves or furnaces that are operated with an insufficient  supply of air.  At

high pressures carbon monoxide reacts with solid iron and invariably iron penta-

carbonyl is formed in gas cylinders containing compressed carbon monoxide or

gases such as commercial hydrogen that contain carbon monoxide.  This  impurity

can lead to erroneous results in reaction studies and in other experiments  where

cylinder gases are used.^°a

     Other complex carbonyl derivatives that are prepared by partial substitu-

tion of metallic carbonyls are:   carbonyl halides such as with iron

(Fe[CO]i ha^); amines of the carbonyls and carbonyl halides, such as with

rhenium (Re[CO]~py2, Re[CO]_pyphal);  carbonyl hydrides such as iron carbonyl

hydrides  (HgFetCO]^ or Fe[C02][CO.H]2) and their metallic derivatives; mixed

carbonyl-nitrosyl compounds such as  iron nitrosyl carbonyl  (FefCO^tNOjp) and

compounds with only one carbon monoxide group on the central metal such as

potassium nickel cyanocarbonyl.  (K2[Ni(CN)3CO]).   Mixed carbonyl halides are

easily prepared by Hiebeh's method,  that involves  heating metallic halides with

carbon monoxide at high pressure.

     The aqueous solubility of carbon monoxide is  enhanced by the formation of

coordination compounds with metal atoms,  e.g.,  copper (Cu),  silver (Ag), gold

(Au)  and mercury (Hg).   Owing to this property, the gas can be quantitatively

absorbed by the following solutions."

     •  Hydrochloric acid, aqueous ammonia or a potassium chloride

        solution of cuprous chloride that produces colorless crystals

        with the composition CuCl.CO.H20 and the probable structure:
                             °\  A
                                 Cu      Cu

                               X  \  /  \
                            H20      Cl      CO
        Silver sulfate solution in concentrated (fuming) sulfuric acid.

        Dry aurous chloride, forming benzene  soluble Cl-Au-«-CO.

        Mercuric acetate in methanolj the product is convertible to the

        chloride complex by adding KC1:
                     CHoCOO.Hg            KCL    Cl.Hg                 (31)

                               ^ OCH3   — »-         "^ OCR

These coordination compounds can be used to estimate carbon monoxide se-

lefrtivetyor to separate it from other gaseous mixtures.

     The toxicity of carbon monoxide is due to its strong coordination bond

formed with the iron atom in heme which is 200 times stronger than that for

molecular oxygen.   The molecule (C^E^^^OiFe) , is a ferrous ion complex of

protoporphyrin IX which when combined with the protein, glob in, constitutes

hemoglobin.  Hemoglobin consists of a tetramer each with a globin chain and

a single heme group, corresponding to 95% by weight globin.  The planar,

macro ring structure of heme is shown below.

                                                   The purple  compound  carboxy-

hemoglobin formed by absorption of carbon monoxide in blood has a character-

istic visible and near ultraviolet absorption spectrum, which  has been  used

in the determination of the amount of carbon monoxide uptake in humans  and


     Reactions with Atmospheric Constituents and Trace Contaminants.  The

efficiencies of gaseous atmospheric reactions capable of oxidizing carbon

monoxide to carbon dioxide were reviewed by Bates and Witherspoon   in  1952.

     •  Reactions with Molecular Species.  The homogeneous gas phase reaction

        with molecular oxygen (02) freaction (32), is far too slow to be signifi-

        cant, even in urban atmospheres where carbon monoxide concentrations

                        2CO + 02 -»• 2C02 + 565 kj/mole                   (32)

        are relatively high (10-100 ppm).  Experimental confirmation has been

        provided by the observed long-term stability (7 yrs)  of carbon monoxide

        in either dry or moist mixtures with oxygen (02)  when exposed to sun-

        light.  /0  Other possible homogeneous reactions with atmospheric con-

        stituents and trace contaminants include those with water vapor  (H20),

        ozone (0-j) and nitrogen dioxide (NO ).

                 CO  +  02   -*  C02  +  0    +33.5  kJ/mole              (33)

                 CO  +  H20  ->  C02  +  H2   +41.9  kJ/mole             (34)

                 CO  +  0    •>  C02  +  02  +  423 kJ/mole              (35)

                 CO  +  N02  -»•  C02  +  NO   +  226 kJ/mole              (36)

The reactions with molecular oxygen  (reaction 33), and water

vapor (reaction 34), may occur in the lower atmosphere but are

very slow and have high energy barriers  (activation energies) to

the reactions, 213 and 234 kJ/mole, respectively.  Also, at ambient

temperatures (25  C) they exhibit very low molecular reaction colli-

sion efficiencies123'149*403 (<10~15).  Reactions 35 and 36, with

ozone and nitrogen dioxide also have been shown.to have high activa-

tion energies^0'133'162'321'457 84 and 117 kJ/mole, respectively.

Therefore, the rates of reactions 33 through 36 become significant

only at substantially high temperatures  (above 500 C) and consequently

are unimportant under atmospheric conditions and at ambient concen-

trations at altitudes below 100 km.  Above this level in the thermo-

sphere, molecular kinetic temperatures increase rapidly with altitude7

from 200 K (-73 C) at 100 km to 1403 K (1130 C) at 200 km and, thus,

reaction (33) can become significant.  The increased thermodynamic

probability that  this reaction take place above 100 km is at least

partially offset by decreasing atmospheric densities (-^3 x 10   atms)-and

lower molecular collision frequencies (<2 x 103 sec"1).  At these high

altitudes competing reactions with atmospheric ions and electrons plus the

solar photodissociation of carbon monoxide are possible.

Gas Phase Reactions with Unstable Intermediates.  There is a possi-

bility that rapid reactions can occur between carbon monoxide and

certain reactive  intermediates, such as atoms or free radicals,

generated by chemical processes in the natural and polluted atmosphere.

For example^  oxygen atoms are produced by the photolyses of nitrogen

dioxide (A<436 nm) and/or ozone (X<1140 nm) in sunlight near the ground,

         and in the upper stratosphere by the solar photodissociation of

         molecular oxygen at wavelengths in    the range of 246 to 176 nm

         (corresponding to the Herzberg and Schumann-Runge absorption bands)•

         Note that significant concentrations (ca. 0.02 ppm) of nitrogen dioxide

         and ozone are produced and play important roles in photochemical smog

         resulting from atmospheric chemical transformation in automobile

         exhaust products.

      The probability of reaction between carbon monoxide and atmospheric atomic

 oxygen has been discussed by Leighton.      Estimates of the rate of reaction

                                3                                      i  +
 (37), in which oxygen atoms, 0( p) and carbon monoxide molecules, COQif-Z^) - both

 in their electronic ground state and accompanied by a chaperone molecule M

(to remove the excess energy), react to form carbon dioxide,  C02(X £• )

 electronic ground state show that this  spin-forbidden process  is insignificant

                o         l +             i  +
              0(JP)  + CO(X EO + M + CO^X1! )  + M + 532 kj/mole        (37)
                            8               g

 in air.   The predominance of molecular  oxygen in the atmosphere and  the spin-

 conservation rules (which govern the probability of reaction between species

 in different spin states) favor instead the formation of  ozone by reaction

 (38), involving reaction of the electronic ground states  of atomic oxygen,

   o                            3
 0( p) and molecular oxygen, 02( £•*). in  the presence of a  chaperone molecule,


                        0(3P) + 02(3£r)  + M -»• 03(1A)  + M                (38)

 The relative efficiencies* of reactions (37) and (38) at  ambient temperatures in
 *The  rate  (R)  of  any elementary reaction is  given by  the  product  of  the rate

  coefficient  (k),  at the temperature  desired,  and the concentrations of the

  reactantSj     in the same units (dimensions)  as used in  the rate coefficient,

  e.g.,  for the reaction A + B -»• C

                           R_A+ R_B - RC - k[A][Bj

  where  R_^ and R_B refer to the rates of removal of A and B, respectively,

  and  Rg refers  to  the rates  of  formation of C.


air are given by the ratio of the rates  (Ro7, ROD) of reactions  (37) and  (38),

respectively.  Thus, where ko-, and k~0 are the rate coefficients of reaction for
                            J /      JO

                           *„   k  [CO]         _6
                            37.37     = u x 1Q °
reactions (37 and (38) respectively, [CO] = 10 ppm, [02J = 21%.  Reaction  (37)

is fundamentally important at high temperatures in the combustion  (flames and

explosions) of carbonaceous materials and in the photochemistry of planetary


     The kinetics and mechanisms of the reactions of oxygen atoms with carbon

monoxide are reviewed and discussed in considerable detail elsewhere.28,89,160a,2.

Although there have been many studies of this reaction at room temperature and

at the very high temperatures (2300-3600 K) in shock tubes, there still are

large unresolved discrepancies in the reported kinetic measurements.  The rate

                                                    -7           -5    -9    -1
coefficients at room temperature range from 1.8 x 10   to <5 x 10   ppm   min

and activation energies vary from -24 kcal/mole to +4.5 kcal/mole.  Some studies

found the reaction rate was second order dependent on reactant concentrations,but

             order  dependence,
others  showed a third/     suggesting a termolecular process involving chaperone

molecules (M).   Additional studies are required to resolve these difficulties.

The discrepancies may be due in part to: reaction conditions, wall effects,

hydrogenous  impurities generating hydroxyl radicals which react rapidly with

carbon monoxide, or iron carbonyl impurities in the carbon monoxide used.

     Simonaitis and Heicklen ^"^ have made a recent study of this reaction

employing mercury photosensitization of nitrous oxide and competition for the

oxygen atoms produced by carbon monoxide and 2-trifluoromethylpropene.  They

suggested the reaction proceeds via intermediate electronically excited states

of carbon monoxide, and proposed a mechanism to  explain  their  observation that

the reaction rate was intermediate between second and  third  order in reactants.

                        0(3P) + CO + M 2 CO  (3B  ) + M                   (39)

                              C02(3B2) *- CO^Bj)                       (40)

                        COj^Bj) + M   t CO (1Eg) + M                   (41)

The reaction was found to be pressure dependent  approaching  second order

kinetics as the temperature was increased at any pressure.   They  derived  ex-

pressions for limiting low and high pressure rate constants, k and k, re-
spectively; kQ = 5.9 x 109exp[-ItlOO/RT]M~2 sec"    (with nitrous oxides as  the

chaperone molecule, M) and k   = 1.6 x 107exp(-2900/RT) M"1 sec"  where kQ = k^-

and k   = koQk/n/k ^o (kon and k.  are the rate constants for the forward  reac-
     oo    3? 40   J»   39      40

tions in equations (39) and (40) and k._39 is that for the reverse reaction in

                                                         —7     ")    _1
equation (39).  These values corrrespond to kQ = 5.7 x 10   ppm   min"1

                                        and k   = 2.9 x 10   ppm   min

at ambient temperatures (25 C).

     The bimolecular reactions (42) and (43) also have been studied.

                           C02  +  0(3P)  -»•  CO  + hv                  (42)

                           CO   +  0(3P)  •*  C02                       (43)

The rate constants evaluated for reactions (42) and  (43) are:

k42 = 8.3 x 102exp(-2590/RT) IT1 sec"1 = 2.5 x 10~5 ppm"1 min"1 at 25 C


k43 = 1.8 x 107exp(-2530/RT) tT1 sec"1 = 6.1 x 10"1 ppm-1 min"1 at 25 C

Note the similarity between k,~ and k^ in the previous processes (reaction 39 to 41),


     The reaction of carbon monoxide with electronically excited oxygen atoms,

O(-'-D), is permitted by the spin conservation rules and reactions (44) and  (45)

would therefore be expected to be considerably faster than the corresponding

                                         o                     77 7fi
reactions with the ground state atoms, 0(JP).  Clerc and Barat, ">'° using the

               COCX^"1") = 0(1D) + M -> CO-CX^*) + M + 720 kJ/mole          (44)
                     g                  2    g

                         CO + 0(XD) ->- C02                                  (45)

far ultraviolet flash photolysis of carbon dioxide,estimate that k// = 1 ppm

min~  and k^ = 1+0.5 x 10^ ppm   min~* at about 300 K.  However, the ambient

concentrations of the excited Specie,  0( D), in the lower natural or polluted

atmosphere are considerably less than those of the species, 0( P), owing to

a less efficient source of the former species (0  solar photolysis at X<310 nm),

and to efficient removal of 0( D) by processes  involving collisional deexcitation wi

molecules of oxygen, nitrogen, water, and argon in air, and by chemical reaction

                                                       3         1
with molecular oxygen and water vapor.  Oxygen atoms, 0( P) and 0( D), are pro-

duced more efficiently at higher altitudes and it is therefore probable that

atomic oxygen can oxidize carbon monoxide at altitudes in the regions of the upper

stratosphere and in the mesosphere where carbon dioxide itself can be photolysed.

In urban air and the lower stratosphere, the reaction with atomic oxygen is

insignificant, as are the reactions of carbon monoxide with atomic hydrogen and

organic free radicals because of the strong affinity of these transient species

for atmospheric oxygen.

     Hydroxyl radicals (OH)are generated by several recognized processes in the

atmosphere, particularly in heavily polluted air resulting from automobile ex-

hausts. These include:  the solar photolysis of nitrous acid vapor (HONO) at

X<400 nm, reaction of electronically excited oxygen atoms, 0( D), with water

vapor by reaction  (46); abstraction  of  hydrogen from hydrocarbons by ground

                   O^D) + H 0 -» 20H +  117  kJ/mole                     (46)

state atomic oxygen, 0( P), as in reaction  (47);  and indirectly by the solar

                   RH + 0(3P) -*  R. + .OH +  (59-84) kJ/mole             (47)

photolysis of aldehydes (A<350 nm) followed by  the reactions  of the product

species with atmospheric constituents and trace contaminants,  oxygen and

nitric oxide, as shown by the sequence  of reactions (48  through 51).

                             RCHO h£ R.+ .CHO                           (48)

                              .CHO -»• .H + CO -  100 kJ/mole              (49)

                       H.+ 02 + M •»• H02-+ M +  197  kJ/mole              (50)

                         HO . + NO -»• N02 +.OH + 38  kJ/mole              (51)

   Photolysis of  aldehyde (reaction 49)  is not a significant source  of

atmospheric carbon monoxide compared to its production by the  internal  combus-

tion of gasoline-air mixtures in  the automobile.   In polluted  atmospheres,  the

solar photolyses of aldehydes, which are partially  oxidized hydrocarbons, by

reaction (48) and specifically that of  formaldehyde by reaction  (52), are im-

portant sources of transient reactive intermediates such as hydrogen atoms  (H),


                          HCHO  *.H +.CHO                                   (52)

hydrocarbon radicals (R, RO, R02, RC-0),  hydroperoxyl (H02) and hydroxyl  (OH)

radicals.   The hydroxyl radical reacts rapidly with carbon monoxide at both

                                                                                 109  152
ambient and sub-ambient temperatures forming carbon dioxide and atomic  hydrogen.    '

                   OH + CO -»• CO, + .H +  105 kJ/mole                        (53)

From an evaluation   of reported rate constants, the most reliable kinetic

expression for reaction (53) is k53 = 5.6 x 108exp(-1080/RT) M"1 sec~l.

Thus, large rate coefficients at ambient and sub-ambient temperatures may

be derivedji.e., kco - 2.2 x 10^ ppm"1 min"* at 300 K, because of the low

activation energy (4.5 kJ/mole) of this reaction.  This fast reaction between the

hydroxyl radical and carbon monoxide is believed to be important in polluted

atmospheres as well as in the upper atmosphere, where hydroxyl radicals are

produced in quantity through reaction (46).  In polluted atmospheres, however,

trace contaminants such as hydrocarbons (particularly olefins), sulfur oxides

and nitrogen oxides can compete successfully for available hydroxyl radicals

and thus reduce significantly the probability for reaction with carbon monoxide.

     It has been suggested    that the rapid conversion (half-lifesl hr) of

nitric oxide to nitrogen dioxide in photochemical smog can be explained by

the cyclic chain of reactions (50), (51) and (53).  In this sequence, hydroxyl

radicals consumed by reaction (53) are subsequently regenerated from the hydrogen

atoms produced in reaction (53) through the consecutive steps, of reactions

(50) and (51).  In this way repeated cycles regenerate and maintain ambient

hydroxyl concentrations while simultaneously converting carbon monoxide and

nitric oxide to carbon dioxide and nitrogen dioxide, respectively.  The rate

of conversion of nitric oxide to nitrogen dioxide would not be greatly affected

by variations in ambient carbon monoxide concentrations if the carbon monoxide

concentrations were always much higher than the nitric oxide concentra-

tion.  This condition is usually met in urban pollution and elsewhere, when the

carbon monoxide concentrations exceed those of nitric oxide by at least two

orders of magnitude.  The ambient carbon monoxide concentrations would be

affected very slightly (<1%) and immeasurably by involvement in these processes

since that amount reacted would be less than the inherent error in standard

monitoring systems, such as nondispersive infrared spectrophotometry.

      Reaction (53) is a chemical sink for carbon monoxide in the stratosphere.

 Despite the low temperatures near the tropopause (213 K or -60 C),  the high

 rate of this reaction explains the relatively rapid decrease in the  mixing ratio

 of carbon monoxide above the tropopause (11 km at mid latitudes).  Estimates

 of the atmospheric concentrations of hydroxyl radicals required  to  react with

 carbon monoxide by reaction (53)  have been made.*226'228 Until  recently concen-

                                                       —7   —8
 trations for these chemical species  in the range of  10-10   ppm were below

 the limits of detection by the  available  methods.   In 1976,  it was reported

 that hydroxyl radical concentrations were measured in the upper  stratosphere

 ranging from 4.5 x 106 c.m   at  30 km to 2.8 x 107 c;m~3 at 43 km,  by  a molecular

 resonance fluorescence emission detection      instrument.   Before mechanisms

 such as those cited above can be  postulated with any  degree  of certainty,

 the concentrations of these and other reactive intermediates (atoms  and free

 radicals),  such as hydroperoxyl (H02)  and nitrate  (NO-)  have to  be determined in

 situ in ambient urban atmospheres.   Some  progress  is  being made  at the  present time


 in developing techniques  and instrumentation  for this  purpose,    but we are

 still a long way from being able  to  measure reproducibly and accurately all

 important reactants and products  of  either  photochemical or  other types of

 pollutants  in urban environments.  The  formidable task of performing similar

 measurements at much  lower  concentrations to determine the background concen-

 tration of  the relatively  clean,  natural atmosphere appears,  to be impracticable

 at  this time.
 G. J. Doyle, Stanford Research Institute, Menlo Park, California.  Unpublished

 data.  1968.

     Westenberg and deHaas^° used electron spin resonance detection in a dis-

charge flow system to investigate in the laboratory  the competition between

carbon monoxide and hydrogen atoms for hydroperoxyl radicals, by reactions (54)

and (55), respectively.

                   CO + H02*   ->   CO  + -OH + 264 kJ/mole             (54)

                   •H + H02'   -»•   2-OH + 159 kJ/mole                  (55)

From analysis of their data, they claimed that reaction (54) involving oxida-

tion of carbon monoxide by the hydroperoxyl radical could be faster than oxi-

dation by the hydroxyl radical, reaction (53).  Gorse and Volmany*6 on the

other hand, us?d  the static photolysis of hydrogen peroxide in the presence

of carbon monoxide estimate that the room temperature rate coefficient for

reaction (54) is at least ten orders of magnitude less than predicted by

Westenberg and deHaas.  This large discrepancy must be resolved before it

can be stated with confidence that a significant reaction exists between

carbon monoxide and hydroxyl, which can effectively compete with the conver-

sion of nitric oxide to nitrogen dioxide,**^7 as shown in reaction (51).

     There is some evidence that halogen atoms, from the photolysis of fluorine,

chlorine and bromine, may catalyze the oxidation of carbon monoxide^&3'^

in oxygen at ambient temperatures (25 C).  At this time, the extent to which

these reactions would be of importance in the atmosphere is uncertain.

Participating halogenated species are produced and exist in the marine

atmosphere as well as in the upper atmosphere, where they are produced by

the decomposition of halocarbon aerosol propellants and refrigerants as well

as natural materials, such as methyl chloride.

•  Chemical Modelling of Carbon Monoxide Reactions.   In the previous

   section, in which reactions of carbon monoxide were discussed, the

   processes described were isolated and studied by  suitably selecting

   and controlling experimental conditions in order  to determine the

   kinetic characteristics of the reaction system.   In the atmosphere,

   however, the situation is complicated by the  enormous  variety of

   competing and consecutive processes  and interactions that occur.

   As a result, simple assumptions cannot provide a  reliable picture

   of the transformations occurring;  nor can the influence be reliably

   assessed of the effects of variations in one  contaminant on the  rates

   of conversion and levels of concentration achieved  by  the other  con-

   taminants.   In order to assess  the influence  of varying one or more

   parameters  in the atmospheric reaction system it has been necessary

   to develop  theoretical mathematical models that  incorporate the

   most recent kinetic information on all the significant  reactions.

   Mathematical models have  come to prominence with the advent  of high

   speed and large capacity  computers, which make handling masses of

   data efficient  and manageable. The initial efforts to develop reliable
                          the analysis of   data obtained by
   and realistic models have been  in/smog chamber/ simulating  synthetic

   photochemical pollution reactions  at near ambient concentrations of

   reactants.   As  these models are  tested and refined, it should be

   possible eventually to  achieve a reliable predictive model combining

   chemistry,  physics  and  meteorology which can be used to assess

   the  design   and   impact  of various pollution control strategies.

   Present  models, however,  are generally capable of describing reactions

   only in  simple mixtures under carefully controlled conditions and are

        of limited atmospheric application due to the sensitivity to chemical

        input rather than because of the model design.  These models make their

        greatest contributions by the identification of the limitations of

        smog chamber experiments and by indicating the lack of knowledge about

        many of the elementary chemical processes involved.

     Theoretical considerations suggest  that carbon monoxide as well as

hydrocarbons can be involved in the conversion of nitric oxide to nitrogen

dioxide in polluted atmospheres where reactions  (50), (51), and (53) can be

represented by reaction (56):

                             CO + NO + 02  -»•  C02 + N02                (56)

     Modelling computations indicate that carbon monoxide may play a role

in ozone production. •**  Ozone concentrations were determined to be 50% higher

in the presence of 100 ppm carbon monoxide than calculated for the same

hydrocarbon-nitrogen-oxide-air mixture with no carbon monoxide present.  This

suggests that substantial reductions in carbon monoxide emissions could reduce

ozone production in photochemical smog, but this has been questioned    at least for


carbon monoxide concentrations <35 ppm.  Also, because these simulations were

performed using a single hydrocarbon reactant, isobutylene, in a controlled

environment, this conclusion cannot  be extrapolated with certainty to the

complex blend of organic compounds found in automobile exhausts and the

conditions existing in urban atmospheres.

     Calvert  et al.   have developed a model to simulate a simple analogue

of the sunlight-irradiated auto-exhaust polluted atmosphere.  They concluded

that although reactions (50), (51) and (53) constitute a major regeneration

step for hydroperoxyl radicals in the chain oxidation of nitrogen dioxide,

 it is by no means the main source of hydroxyl to hydroperoxyl  conversion in

 their model.  They suggest that reaction of alkoxy radicals with oxygen  pro-

 vides a more efficient source.  In a computer simulation study of the effect

 of carbon monoxide on the chemistry of photochemical smog the same authors^0

 conclude that the presence of small levels of carbon monoxide in a  nitric

 oxide-containing atmosphere can enhance the photooxidation of nitric oxide

 to nitrogen dioxide ultimately forming significant amounts of ozone.  Never-

 the less, there are still many discrepancies between the effect predicted by

 chemical mathematical simulation models and the results from experimental

 analyses of reacting mixtures in smog chambers.   These discrepancies will be

 resolved as our knowledge about these elementary chemical processes increases,

 the analytical techniques improve, and as the models are subjected to more

 rigorous testing and refinement.

     In the last few years several photochemical models of the natural tropo-

 sphere have been developed^!»185,226,227,228,422  Using plausible reaction

             22fi 99R
 schemes,,Levy    *    has demonstrated that  radical reactions are important in

 this region of the atmosphere.  He has  shown specifically that hydroxyl radicals

 achieve significant concentrations in the sunlit atmosphere and that their

 subsequent reactions with trace gaseous constituents,  including carbon monoxide,

 could be important.   The hydroxyl radical is essentially unreactive with the

major atmospheric components, nitrogen, oxygen,  argon and carbon dioxide »this

 accounts for the apparent preference for reactions with minor constituents.

 McConnell et al. ,   using a model, similar to that of Levy,  estimated that

 at noon time the hjcdboxyljcoricentration  in '

 the lower legions of the troposphere was  8 x 10   ppm-  with a daily average

 of about 2 x 10~8 ppm.  The latter value decreases with increasing altitude  to

 about 1/3 this concentration in the vicinity of the tropopause.

                       226 2?R
     According to Levy,   »  ° the major source of hydroxyl radicals in the

lower troposphere is the reaction of electronically excited oxygen, 0(^0),

with water vapor, shown in reaction (46).  The oxygen specif 0( D), is

produced by the efficient solar photolysis of ozone at the long wavelength

at the end of the Hartley absorption band  (X=290-350nm).  Although most of

                      1                                               o

the excited oxygen, 0(^0), atoms are converted to the ground state, 0( P),

by collisions with atmospheric nitrogen and oxygen molecules, a few percent

still are available to react    with water vapor.  The concentrations of

hydroxyl radicals are maintained by the processes discussed previously

(reactions 46, 47 and 50).  An important consequence of .these observations

is that hydroxyl radicals react rapidly with methane (CH ),102,185,228,259

as well as with carbon monoxide, and at a  sufficiently high rate which provides

the most significant natural source of carbon monoxide identified thus far.

The primary step in the initiation of atmospheric oxidation of methane involves

hydrogen abstraction by the hydroxyl radical, by reaction (57).

                   CH  + OH  -»•  CH3 + H20 + 71 kJ/mole                 (57)

The rate coefficient determined   for reaction (57) is kc^ = 12 ppm~l min~l

at 25 C.  In conjunction with a methane concentration of 1.4 ppm and the

average hydroxyl concentration given previously, we find the upper limit for

carbon monoxide production, R   = 3.4 x 10~' ppm min~ , assuming that all CH*
                             C#O                                             "

molecules reacted are converted to carbon monoxide.  To produce carbon monoxide,

reaction (57) is followed by reaction of the methyl radicals (CH,) produced

with molecular oxygen to produce formaldehyde (CHJD) by reactions  (58) and (59a)

and formyl radicals (CHO) by reaction (59b).  -The solar photolysis of formaldety

                             CH3 + °2   *   CH3°2
                             CH 02.     +   CH20 +.OH                   (59a)
                             CH302.     ->  .CHO  + H20                  (59b)
                             CH20       +v  R2 + CO                     (60a)
                             CH0       -»•  .H+.CHO                     (60b)
produced by reaction (59a) is an important natural source of carbon monoxide
by reaction (60a) and of hydrogen atoms and formyl radicals by reaction  (60b) .
Formaldehyde is a reactive and photo chemically unstable intermediate which
may, efficiently produce carbon monoxide in the absence of oxygen.  Several
authors102 .228,259,428 have estimated that the oxidation of biologically pro-
duced methane leads to a global production rate of carbon monoxide at least
ten times greater than that obtained from anthropogenic sources.  Recently,
however, Warneck    has proposed that Levy's model should be modified to take
into account the possible scavenging of hydroperoxyl radicals (an intermediate
in the hydroxyl recycling process) by atmospheric aerosols.
     Numerous photochemical models also have been used to describe stratospheric
processes.  These models use sets of sequential reactions similar  to those' used
to describe photochemical smog but with appropriate modifications for different
temperatures,  pressures, concentrations and solar spectral intensity distributions.
In the stratosphere the dominant scavenging process for removal of carbon
monoxide is the reaction with the hydroxyl radical.

                                    CHAPTER 3


     Carbon monoxide is released into the atmosphere from both natural and

                                                    102 228 259 428
anthropogenic sources.  Recent theoretical estimates   '   '   '    suggest

that on a global basis natural sources contribute at least ten times more

carbon monoxide than manmade sources to the total atmospheric burden.  The

                                                     228 259
most important natural source that has been suggested   '    is that result-

ing from the oxidation of atmospheric methane with lesser contributions from

forest fires, terpene oxidation, and the oceans.  The incomplete combustion of

fossil fuels is the principal anthropogenic source of carbon monoxide.   At the

present time, carbonaceous fuels are widely used, primarily in transportation

and to a lesser extent in space heating and industrial processing.   In con-

trast to natural sources, which are globally widespread,  anthropogenic sources

are mainly located in urban and metropolitan areas and concentrated in the

Northern Hemisphere.   The influence of naturally produced carbon monoxide on

the carbon monoxide concentration in urban air is believed    to be negligible,

but recent discoveries indicate that it could be important in determining

the "background" concentrations of carbon monoxide and the mean residence

time (atmospheric lifetime)  in the terrestrial atmosphere.

     Anthropogenic carbon monoxide production is directly related to man's

technological growth and productivity, as well as to his  economic and social

well-being.  It is well within our present technologic capability to reduce

the emission of this pollutant, and maintain it at a low  level without  economic

hardship, despite its being an unwelcome by-product accompanying national

economic growth along with increasing  energy  requirements.  There

appears to be no simple panacea, however, for controlling pollutant emissions,

as shown by the new pollution problems that appear when specific controls are

implemented.  Much more attention and research have to be devoted in the future

to the indirect as well as the obvious consequences of applying control tech-

niques to both stationary and mobile sources.

     It is estimated that the total anthropogenic emission of carbon monoxide

exceeds that of all other man-made pollutants combined.  This fact, coupled

with the purported excessive natural emission of carbon monoxide and its long

lifetime (residence time) owing to its stability and apparent lack of chemical

and biologic reactivity, supports the conclusion that carbon monoxide is the
most abundant and most commonly occurring of all the atmospheric pollutants

and minor constituents.

Sources of Carbon Monoxide
                                       OQQ A f)£
     Contrary to what has been believed   '    it is now considered possible that

atmospheric carbon monoxide could be principally of natural origin.228,259

This gas was first discovered as a trace constituent of the terrestrial

atmosphere by Migeotte^?? in 1949.  In a study of the solar spectrum, he

attributed specific absorption lines to ambient carbon monoxide originating
*Although carbon dioxide (CC^) is produced and released to the atmosphere
 in much greater quantities than carbon monoxide, it is not usually classi-
 fied or defined as an air pollutant.  Little can be done at this time to
 control carbon dioxide emissions despite a regularly observed and signifi-
 cant (0.7-1.0 ppur/yr)  annual increase in the already large concentrations
 of global ambient carbon dioxide ( 330 ppm )  or the associated global cli-
 matic implications should the annual-increase trend continue, This gas is a
 natural end-product of all combustion processes involving carbonaceous ma-
 terials.  Thus, not until it is economically and practically feasible to replace
 conventional fossil-fuel burning energy sources with non-combustive systems
 such as nuclear, solar or geothermal power and provide non-carbonaceous fuels
 for transportation will it be possible to markedly reduce carbon dioxide emissii

  at a wavelength of about 4.7 micrometers (ym) in the infrared region.  Similar

  observations  '   »   '     were made during the next decade in the United

  States,  Canada and Europe which demonstrated that carbon monoxide concentration

  levels  average about 0.1 mg/m3 (0.1 ppm)   in clean air.   From these early in-

  vestigations,  it was concluded by Junge^2 in 1953 that  background atmospheric

  carbon  monoxide concentrations were highly variable and  erratic in nature but

  generally ranged in concentration from 0.01 to 0.2 ppm.

       In 1968,  Robinson and Robbins340 estimated that about  2.8 x 108™fonsC

  (2.8 x  1011  kg) of carbon monoxide were discharged into  the atmosphere

  worldwide from anthropogenic sources during the year of  1966.   Somewhat more

than half of the / 1.5 x lO^jJtons (l.5 x 10*1 kg)  was estimated  to be generated

  in the  continental United States.   Transportation is by  far the largest single

  man-made  source of carbon monoxide in this country,  accounting for more than

  two-thirds of  the emissions from all anthropogenic sources  in  the 1960's.

  It has been  estimated that the total U.S.  emissions  of carbon  monoxide showed

  almost  a  two-fold increase during the period 1940 through 1968.   This  dramatic

  increase  was due almost  exclusively to the increasing use of the motor vehicle

  during  that  period.   After 1968 there was  an initial decline of about  10%  in

  the carbon monoxide emission followed by a levelling off in mobile source

  emission  inventories owing to the required installation  of  emission control

  devices in new vehicles.   In 1974,  there were about  125 million vehicles

  registered in  the United  States,405 which  produced annually about 1 x  108™fonsc

  (1 x 10    kg)  of carbon monoxide.

       The  magnitude of the problem in controlling internal combustion engine

  emissions may  be seen   from the fact that  approximately 3  lb  (1.4 kg  )  of

carbon monoxide, along with other pollutants is  produced in the combustion

of 1 U.S. gal  (3.8 liter) of gasoline.  Assuming an average density for  gasoline

(octane,  C0H.0)  of 0.7 g/cm  ,  then,  on the basis of mass, approximately
          o  lo

25% of the carbon in gasoline  is converted to  carbon monoxide in the internal

combustion engine.  This corresponds to an oxidation efficiency of 92%,  assuming

carbon dioxide and water are the only  other products.  If the average auto-

mobile (without emission control equipment) travels 15 miles (24 km) per

gallon of fuel consumed, then  about 0.2 Ib   ( 91 g) of carbon monoxide per

mile (52 g/km) would be released into  the surrounding atmosphere, which  was


the situation prior to 1968.   This average emission is two and/half times greater

than the 1970-1971 automobile  emission standard set for carbon monoxide  (34 g/miL

20 g/km) but more significantly it is  25 times larger than the pro-

posed standard for 1975-1976 model automobiles (3.4 g/mile, 2.0 g/km), as

prescribed by the 1970 Clean Air Amendments.  It should be pointed out here

that the required installation of catalytic converters in 1975 model auto-

mobile exhaust systems in conjunction  with the use of unleaded gasoline  has

greatly reduced carbon monoxide emissions; but the reduced emissions of

criteria pollutants were achieved at the expense of generating potentially

harmful sulfuric acid aerosol.  Apparently there is no simple solution to

the problems of automobile emission control.

     Technological Sources.  Until recently, anthropogenic sources producing

large quantities of carbon monoxide were considered to be primarily responsi-

bile for the observed   '    global concentrations.  Although  carbon monoxide

produced in urban areas by man's activities far surpasses natural contributions,

it now appears that anthropogenic emissions only approximate natural contribution

in the Northern Ifemisphere and on a global scale man's activities  contribute

only about 10% to the total worldwide production.  Estimates    '    of  global

tonnages of carbon monoxide produced by anthropogenic sources show a marked

increase in annual emissions which correlates with the worldwide increase

                                                                      I Qg

in the consumption of fossil fuels.  A comparison of a recent estimate    of

global carbon monoxide emissions from man-made sources with an earlier  esti-

mate^O shows that annual worldwide anthropogenic carbon monoxide  emissions

have increased by 28% during the period 1966-1970.   Jaffe    has estimated

that in 1970 the global carbon monoxide emissions from combustion  sources

                                            8                       11
were about 359 million metric tons (4.0 x 10  short tons or 3.6 x 10   kg).

This estimate is based on an analysis of worldwide fossil fuel usage,

agricultral practice, mining activities and waste disposal in conjunction

with recently revised (1972) carbon monoxide emission factors for various

sources.   Table 3-1 gives the estimated worldwide contributions from major

sources plus the corresponding fuel consumption figures.  These data show

that as in previous estimates, the internal combustion engine  in motor

vehicles  still represents the largest single source of carbon monoxide;  providing

55% of all anthropogenic emissions.  The second most important anthropogenic

sources and industrial processes and certain miscellaneous activities in the

stationery source category, which together contribute almost 257, of the total.

In 1970 the U.S. contribution was 37% of total global production.

     During the period 1940-1968, there has been a dramatic increase in carbon

monoxide  emissions in the United States due almost  exclusively to the increased

use of automobiles.  Since 1968, automobile emissions-have leveled off through

1970 and  declined from 1970  - 1975 because of the installation of emission

control devices.  Trends in the emissions from major sources for the period

                                  TABLE 3-1
     Estimated Global Anthropogenic Carbon Monoxide Sources for 1970
World Fuel
10° metric tons/yr
CIO9 kg/yr)	
Motor vehicles, gasoline ")
                diesel   J)

Aircraft (aviation gaso-
   line, jet fuel)



Other (non-highway) motor
   vehicles, construction
   equipment, farm tractors,
   utility engines, etc.)
World Carbon Monoxide
10° metric tons/yr
(109 kg/yr)	



Coal and lignite;

Residual fuel oil


Distillate fuel oil

Liquefied petroleum gas

Industrial processes
   (petroleum refineries,
    steel mills, etc.)
               .  «J.
Solid waste disposal (urban
   and industrial)

Miscellaneous (agricultural
   burning, coal bank refuse,
   structural fires)





                       Total anthropogenic carbon monoxide


 1940-1975 are shown in Table 3-2.  During the 35-year period covered  in Table

 3-2, transportation in the United States was clearly the dominant  contributor

 to the total carbon monoxide emissions, as shown by the increasing percentage

 from this source, from 41% in 1940 to 77% in 1975.  The carbon monoxide emissions

 from industrial sources also have been greatly reduced during the  period 1940

 through 1970, although there has been a rapid industrial expansion during  this

 period.  Agricultural burning and industrial process losses, representing  9.3%

 and 7.6%, respectively, of the total carbon monoxide emissions, are two  im-

 portant sources that remained relatively constant until 1970.  A considerable

 reduction (77Z) in emissions from stationary sources occurred during the

 period from 1940 through 1975, probably owing both to a changeover  from solid

 fossil fuels to liquid and gaseous fuels and to increased furnace efficiency.

All of the sources listed can be controlled with the exception of those in

 the miscellaneous category.  These include such essentially uncontrollable

 sources as forest fires, structural fires and burning in coal refuse banks.

     Table 3-3 gives a more detailed breakdown of emission sources with

estimates of the amounts of carbon monoxide emitted  »     during calendar

year 1975.  These data show that in 1975 the carbon monoxide emissions in

the U.S.  from technological sources (all categories except forest fires)

was more than 85 x 106 metric tons (85 x 109 kg,  93 x 10  short tons).  The

major source of carbon monoxide in the U.S.  in 1975 was the combustion of

 fossil fuels in vehicles producing about 67 million metric tons (67 x 10  kg,

 74 x 10  short tons) annually, with gasoline-burning internal combustion

 engines accounting for 65% of the total.

                                  TABLE 3-2

         Nationwide Estimates of Carbon Monoxide Emissions,


                    in 106 metric tons*/yr    (1Q9 kg/yr)
Source Category    1940    1950    1960    1968    1969    1970    1972     1975

Industrial pro-
  cess losses


Fuel combustion
  in stationary

Solid waste


31.7    50.2    75.7    102.5   101.6   100.6   70.4    66.6
13.1    17.1    16.1
 8.3     9.4    11.2
 1.6     2.4     4.6
                 7.7    10.9    10.3   15.8    13.3
                12.6    12.5    12.5    1.5
1.6     0.7    1.1
                 7.3     7.2
                         6.6    4.5
5.7     4.1    4.2
77.5    93.3   115.8    136.8   139.5   134.8   97.5    87.0
*To convert to U.S. short tons, multiply by 1.1 (1 short-ton = 2000 Ib).

                                    TABLE 3-3

Detailed Summary of  1975 Carbon Monoxide Emission Estimates in the United  States
Source Category
Fuel combustion in stationary sources

     Commercial and institutional

         Total fuel


     Gasoline vehicles
     Diesel vehicles

         Total road vehicles

Other non-highway use

         Total transportation

Solid waste disposal

     Municipal incineration
     On-site incineration
     Open burning

         Total solid waste

Industrial process losses
Agricultural burning

         Total controllable


     Forest fires
     Structural fires

         Total miscellaneous

Total all categories
Estimated Carbon Monoxide Emissions
(103 metric tons*/yr    106 kg/yr)








 Source - U. S. Environmental Protection Agency Data for 1975 National Emissions
Report, National Emissions Data System (NEDS) of the Aerometric and Emissions
Reporting System (AEROS), Research Triangle Park, N.C.  U. S. Environmental Pro-
tection Agency, Office of Air and Waste Management.  Unpublished, 1976.
"k                                                                o
 To obtain U.S. short tons, multiply by 1.1 (1 short ton = 2 x 10  Ib).

At the present time,the remaining sources in order of decreasing contribution are
industrial processes, solid waste combustion, miscellaneous fires and agricultural

burning.  Fuel combustion in stationary sources for space heating and power generation
provide less than 2% of the total carbon monoxide emissions and, thus, is of minor

importance.  Carbon monoxide is also produced in high concentration by burning

cigarettes, explosions and the firing of weapons, but these point sources are
insignificant in terms of total annual production.
     Natural Sources.  Prior to the 1970's, the known natural sources of carbon

 monoxide were considered to be of minor importance in comparison with technologi-

 cal sources, and  the principal natural source was thought to be forest fires
 resulting from natural causes, such as lightning.340  Other natural emission

 sources identified were volcanic activity, natural gases from marshes and coal
 mines*2^ and electrical storms.^40  Secondary sources of natural origin included
 photochemical degradation of naturally occurring organic compounds, such as
                                                                    4 5 224
 aldehydes, which are alsp involved in photochemical smog formation, * *    and the
 solar photodissociation of carbon dioxide which becomes feasible in the upper
 atmosphere at altitudes above 70 km.

     In the last few years, several potentially large natural sources of geo-

 physical or biological origin have been identified.  Estimates of the magnitude

 of these sources range from 3 to 25 times that of anthropogenic sources, de-

 pending upon the data base selected for comparison.
                            OT C
     In 1972, Stevens et^ al. '  reported a comprehensive study made of the

 isotopic composition of atmospheric carbon monoxide at numerous locations

 at different times of year, which included a comparison with the carbon

 monoxide originating from automobile exhausts and from selected natural

sources.  They found five major iso topic types of carbon monoxide, two  con-

taining light oxygen (^O enriched) and three containing heavy oxygen

(*80 enriched).  The light oxygen varieties^ which were present throughout

the year, were a predominant fraction of atmospheric carbon monoxide, with

constant concentrations of 0.10-0.15 ppm.  The heavy oxygen species were

minor constituents whose production was apparently seasonal.  On the basis

that the light oxygen species originate in the atmosphere rather than the

biosphere, they estimated an atmospheric production rate in the Northern

                              9                    19
Hemisphere of more than 3 x 10  metric tons (3 x 10iz kg) .  From Robinson

and Moser's*2 estimate that 95% of the global anthropogenic carbon monoxide

emissions originate in the Northern Hemisphere and the total global man-made

carbon monoxide emitted in 1970 given in Table 3-1,  the man-made emissions

in the Northern Hemisphere in 1970 are calculated to be about 3.4 x 10**

metric tons, which is ten times lower than the estimate from the isotopic

studies .

                       an(j formaldehyde^^ have been suggested also as natural
atmospheric sources of carbon monoxide.   McConnell et^ al .     have estimated

that atmospheric oxidation of biologically produced methane can provide a

global source of about 2500 million metric tons (2.5 x 10*2 kg) annually.

This is ten times greater than the previously reported    worldwide anthro-

pogenic emission rates and 7 times larger than the estimates for 1970 given

in Table 3-1.  In 1972, Weinstock and Niki^2° derived a carbon monoxide produc-

tion rate via methane oxidation based on calculations of hydroxyl radical

concentrations in tropospheric air.   They concluded that this mechanism could

provide a source strength about 25 times greater than man-made sources.

    In some regions, the oceans appear to be a significant source of carbon

monoxide.  Independent studies3-^, 388 Q^ Atlantic Ocean surface waters have

shown that the surface layers are supersaturated with carbon monoxide ranging

from about 10 to 40 times the equilibrium water-air ratio.  Similar conclusions

have been drawn from subsequent marine air-water interface studies in the

Atlantic221'387'444 and South Pacific Oceans.385  In 1971, Junge, Seiler and

Warneckl94 calculated that the oceans may contribute the equivalent of about

0.3% of the total anthropogenic carbon monoxide production.  More recently,

Linnenbom, Swinnerton and Lamontagne241 have revised their previous estimates

of the oceanic carbon monoxide production to a Northern Hemisphere flux of

9 x 10'  metric tons per year or about 25% of the man-made output in 1970.

In 1974, Liss and Slater242 developed a mathematical model describing the flux

of various gases across the air-sea interface.  Taking a mean atmospheric

carbon monoxide concentration of 0.13 ppm221 and a surface water concentra-

tion of 6 x 10~° cm3 C0/cm%20 in their model, they estimate a total oceanic

flux of 4.3 x 10'  metric .tons (4.3 x 10^0 kg) carbon monoxide per year.  All

of these recent estimates suggest that the oceans are a source rather than,

as previously thought, a sink for atmospheric carbon monoxide.  Biological

organisms including marine algae, siphonophores and microorganisms apparently

are responsible for the large quantities of carbon monoxide in the surface

layers of the oceans.

    In 1960,  Went434 proposed that atmospheric photochemical reactions in-

volving naturally produced terpene hydrocarbons could be an important source

of atmospheric trace constituents.  Revised estimates of natural terpene

production*-" were used by Robinson and Moser338 to calculate that approximately

54 x 106 metric tons (5.4 x 1010 kg) of carbon monoxide are produced annually

by atmospheric photochemical oxidation.  An identical figure (5.4 x 1010 kg/yr)

for carbon monoxide generation from the degradation of chlorophyll was arrived

at by Crespi, e_t _al.100a  This source strength was used by these investigators

together with an estimate of production by the bilin biosynthesis, which takes

place in blue-green algae, to show that plants could produce a total of 90

million metric tons (9 x !Cr° kg) of carbon monoxide per year on a global


    Charged particle deposition mechanisms and atmospheric electrical dis-

charge phenomena, including lightning in the troposphere, have been investi-

gated* ^ as potential sources of atmospheric carbon monoxide.  However, these

sources appear to be small compared with anthropogenic sources.

    Total global emissions of carbon monoxide can be estimated from anthro-

pogenic and natural sources given in Table 3-1 and from the data presented

above.  Annual production rates for the major sources are given in Table 3-4.

                                                                         I O
These estimates show that approximately 3.2 billion metric tons (3.2 x 10   kg)

of carbon monoxide can be released annually into the atmosphere by all processes,

Occurrence of Carbon Monoxide

    Community Atmospheres.  Carbon monoxide concentrations in metropolitan

areas vary considerably both temporally and spatially.  Analysis of aerometric

data collected at continuous air monitoring program (CAMP) stations in selected

cities has revealed distinct temporal patterns in ambient carbon monoxide levels

Diurnal, weekly and seasonal trends have been observed which correlate with

traffic volume, vehicle speed and meteorological conditions.

                              TABLE 3-4

           Estimated Carbon Monoxide Production Rates from

                Natural and Anthropogenic Sources, 1970
                       CO Emission rate
Source	    10^ metric tons/yr (10* kg/yr)    Reference
Methane oxidation
Forest fires
Terpene oxidation                     54                    338

Plant synthesis and
   degradation                        90                    186

Oceans                               220                    241

   Total, all carbon
        monoxide sources           3,233

     There is a great variation in the carbon monoxide concentrations  in

urban metropolitan areas, ranging from 1 to more than 140 ppm, with the

higher concentrations being observed as brief peaks in dense traffic.  As  a

result, persons in moving vehicles subject to heavy traffic conditions can

be exposed to carbon monoxide concentrations greater than 50 ppm for sustained

periods.  Larsen and Burke^" have developed a mathematical model to statisti-

cally analyze and review the extensive aerometric data collected at numerous

sampling locations in many large cities.   They estimated that maximum annual

8-hour average carbon monoxide concentrations were approximately  115 ppm

in vehicles in heavy traffic downtown, 75 ppm in vehicles operating on express-

ways or arterial routes, 40 ppm in central commercial and mixed industrial

areas, and about 23 ppm in residential areas.   These estimates indicate that

carbon monoxide can achieve levels in heavy traffic on city streets almost

3 times that found in central urban areas and 5 times that found in residential

areas.   Mathematical urban diffusion models for carbon monoxide have been

developed at Stanford Research Institute-^04 and elsewhere in order to describe

better the observed spatial and temporal  variations of this pollutant.

     In special situations, such as in underground garages, tunnels,  and

loading platforms, carbon monoxide levels have been found to exceed 100 ppm

for extended periods.    To prevent excessive exposure to carbon monoxide,

alarm systems have been installed in newly constructed tunnels  which auto-

matically activate auxiliary ventilating  units when predetermined undesirable

carbon monoxide levels are reached.

     Background Levels and Distribution of Carbon Monoxide.  Ambient carbon

monoxide concentrations measured in relatively clean air  (remote from strong

sources) are quite low but variable.  Some of the early measurements of the

infrared solar spectrum absorption by atmospheric carbon monoxide at locations

in Canada,244 Switzerland41.277,278 and in the United States244'359'360 showed

a range of concentrationjfrom 0.03 to 0.22 ppm carbon monoxide, and an average

concentration359 of 0.11 ppm.

     Junge^92'!94 reported that background concentrations of carbon monoxide

range from 0.01 to 0.2 ppm.  The most extensive measurements of carbon monoxide

background levels at various locations have been made by Robinson, Robbins and

their colleagues.  They have found concentrations ranging as low as 0.025 ppm

and up to 0.8 ppm in North Pacific marine air;33"  0.04 to 0.8 ppm carbon

monoxide in non-urban air over California;336 0.06 to 0.26 with an average of

0.09 ppm at Point Barrow, Alaska;64 0.05 to 0.7 with an average of 0.11 ppm

at Inge Lehmann Station in Greenland339 and an average of 0.06 ppm in the Sou then


Pacific. Jy  They have concluded from these investigations that the observed

variability of carbon monoxide in unpolluted areas is a characteristic of the

air mass in transit and reflects the prior history of the air mass.  Background

levels as high as 1.0 ppm  are observed when the air mass has recently traversed

densely populated areas.  On the basis of data collected on five cruises in the

Pacific Ocean,339 they measured and determined the average background concen-

tration distributed latitudinally over the Pacific.  The highest concentrations,

about 0.2 ppm, were found between 30° and 50° N.  These were associated with

the high population density at mid-latitudes in the Northern Hemisphere.  The

Northern Hemisphere mid-latitude value decreases to about 0.07 ppm  in the

Arctic and to 0.09 ppm at the equator.  In going south from the equator, the

average concentrations fall to a minimum of about 0.04 ppm at 50   S  then rise

to about 0.08 ppm in the Antarctic.  Based on this work, they suggest average

carbon monoxide concentrations of 0.14 ppm in the Northern Hemisphere,  0.06  ppm

in the Southern Hemisphere and a global average of 0.1 ppm.

     Seiler and Junge^55 have measured similar average values over the  Northern

(0.18 ppm) and Southern (0.05 ppm) Atlantic Ocean.  Generally higher (>30%)

background concentrations have been found over the Atlantic Ocean than  over  the

Pacific Ocean by several investigators.  In general, because of its remoteness

from major polluting sources, the Pacific Ocean may be considered to be more

closely representative of "background" clean air.

     The altitude and vertical distributions of atmospheric carbon monoxide

have been reported by Seiler and Junge.339,355  They have found consistent

carbon monoxide concentrations averaging about 0.13 ppm at 10 km altitude in

both the Northern and Southern Hemispheres, in contrast to the large differences

observed near the surface.  During polar flights they observed^39 upper

tropospheric concentrations averaging  0.10  ppm, whereas the stratosphere con-

centrations were markedly lower, falling in the range of 0.03 to 0.05 ppm

carbon monoxide.  Subsequently, Seiler and  Warneck-"" found a decrease in

concentrations from M).15 ppm below the tropopause to ^0.05 ppm above this

region of discontinuity.   A 1972 investigation    of the infrared solar spectrum

made with a balloon-borne spectrometer has  shown a gradual decrease in carbon

monoxide concencration with increasing altitude,  from ^0.08 ppm at 4 km to

0.04 ppm at 15 km.  These observations coupled with those made near the earth's

surface indicate that important contributions to surface carbon monoxide result

from the concentration from Northern Eemisphere industrialization at mid-latitudes

     Residence times (T = atmospheric mean life) for carbon monoxide in the

atmosphere have been estimated by many investigators using the available data.

An approximate estimate of this temporal turnover parameter can be made using

the global average carbon monoxide concentration to calculate the total amount

in the atmosphere (M  ) and the total rate of production from natural and anthro-

pogenic sources (R™) .  Thus, TQQ = MQQ/RCQ.  Using the global annual production

rate of 3.2 x 10  metric tons, (from Table 3-4) and an average global concen-

tration of 0.1 ppm, corresponding to 580 x 10  metric tons of carbon monoxide


in the atmosphere, then T = 5.8 x 10  ton = Q.18-/yr.

                            3.2 x 109 ton/yr


     These rough estimates are /to the variability in and interpretation

of atmospheric concentration measurements and to the current knowledge of source

concentrations.  They are however, a useful guide in judging the effectiveness

of natural processes in cleansing the atmosphere of this man-made pollutant

and natural constituent.  The early estimates of residence times for atmospheric

carbon monoxide ranged from 2.7 to 5 yrs. 192, 336, 340, 386  However, in 1969,

Weinstock    proposed that a residence time could be derived from radiocarbon
 data because "hot" carbon-14  (C) nuclei produced  from the nitrogen-14  (*N)

 (n, p) reaction with cosmic ray neutrons are  fixed  primarily as carbon monoxide-

   C  ( ^CO) prior to conversion into carbon dioxide- ^C (  CC^) .  From an analysis

 of available data on carbon monoxide-^^C (^CO) concentrations in the atmosphere

 and estimates of its rate of  formation, Weinstock derived a residence time of

 0.1 year which is much shorter than the earlier estimates.  Such a short atmospherij

 mean lifetime suggested that  an efficient mechanism for carbon monoxide removal

 must be operative . and Weinstock proposed reaction with the hydroxyl radical.
 In 1971,  Levy   derived plausible  theoretical  estimates of tropospheric

hydroxyl radical concentrations and their reaction with atmospheric carbon

monoxide and calculated a corresponding residence time of 0.2 year.  The

reaction of tropospheric hydroxyl radicals with atmospheric methane as the

principal natural source of carbon monoxide was used by McConnell et al.
to estimate a residence time of 0.3 year.  Subsequently, Weinstock, et al

confirmed their earlier estimate of 0.1 year from an analysis of both radio-


active and stable carbon monoxide in/ troposphere.  There now is sufficient

evidence to suggest that the residence time of carbon monoxide in the tropo-

sphere is about 0.2 + 0.1 year .

Fate of Atmospheric Carbon Monoxide

      On a global basis the average concentration of carbon monoxide in the

atmosphere is about 0.1 ppm (0.12 mg/m^) and it does not appear to have been

increasing substantially in recent time. 336  If efficient natural removal

processes (sinks) were not operative to transform or scavenge this gaseous

atmospheric pollutant, then based on the 1970 estimated total inputs of

carbon monoxide (Table 3-4) , the global average concentration would be in-

creasing by about 0.5 ppm annually.  Similarly, using current anthropogenic

production estimates alone . atmospheric carbon monoxide would be expected to

increase at the rate of 0.06 ppm/yr.  As this does not appear to be the case,

efficient sinks for carbon monoxide must be present in order to maintain the

present atmospheric concentrations.  Several possible natural sinks have been

identified in the upper atmosphere, the biosphere and the chemosphere.  These

will be discussed in more detail in the following sections.

      Upper Atmosphere Sink.  Carbon monoxide produced at the earth's surface

or in the lower troposphere, could conceivably migrate  vertically to the

upper regions of the atmosphere by atmospheric transport, turbulent mixing aand

diffusion.  A rapid decrease in carbon monoxide mixing ratio* above the polar

tropopause was first observed by Seller and Junge356 in 1969.  They attributed

the reduction in carbon monoxide across the tropopause to the reaction between

carbon monoxide and ambient hydroxyl radicals in the lower stratosphere.  In

1970, theoretical estimates of the hydroxyl radical concentrations necessary

to oxidize the flux of carbon monoxide from the troposphere into the strato-
sphere were reported to be  10   ppm  . lo:*»J^u  Subsequently, flights were

made in the winter of 1970-1971 to measure the gradient of the carbon monoxide

mixing ratio above the tropopause.   '  This sink  appears to be controlled by

the vertical eddy-type diffusion of  carbon monoxide, and accounted for about

13 + 2% of the total annual carbon monoxide based on 1968 estimates.  This

sink does not, therefore, appear to  be adequate to compensate for carbon

monoxide production at the present time.

      Soil as a Sink.  Certain microorganisms in the soil> as well as

some terrestrial plant species appear to be the major biological sinks for

carbon monoxide.  Soils both produce and absorb carbon monoxide simultaneously,

but the net result is they act as a  sink.  Ingersoll et al.    measured soil

                                                            -11        9
carbon monoxide uptake rates ranging from 21.1 to 319.4 x 10    g CO/cm /s at

 100 ppm carbon monoxide.  This study, which was based on investigations in

the field  at 59 locations in North America ^0 showed that desert soils took
 *The mixing  ratio  is defined  as  the  fractional volume  (mass) of gas in

  unit volume (mass) of  air  under the same  conditions of temperature and


up carbon monoxide at the lowest rates and tropical soils at the highest

rates.  Agricultural soils had  a lower carbon monoxide uptake than un-

cultivated soils presumably because they had less organic matter in the

surface layer.  There was also some evidence that soils exposed to higher

carbon monoxide concentrations (such as near a freeway interchange) had a

greater carbon monoxide uptake rate.  This was attributed to the development

of a larger or more effective population of carbon monoxide converting micro-

organisms .

      Data concerning the effect of temperature on the sink capacity of soils

are limited and contradictory.  Ingersoll et^ al.    and Seiler    agree that

the net uptake rate is drastically reduced as the temperature is increased

from 30 C to 50 C.  At temperatures in the range of 10 C to 30 C
measured high uptake rates iwhereas Ingersoll ^t al. °  found the maximum

uptake rate to be at 30 C with greatly decreased rates at both 10 C and 20 C.

      Seiler (1974)    calculated the average soil carbon monoxide uptake rate to

be 1.5 x 10"11 g/cm2/sat0.2 ppm carbon monoxide,  since the carbon monoxide

concentration used in this study is typical of the concentrations of the gas

found in the Northern Hemisphere, this value appears to be one of the most

reliable estimates of the carbon monoxide uptake rate by soil.  The measure-

                         1 QO           .  —11     9

ment made by Inman et^ al.    of 23.4 x 10    g/cm /s at a carbon monoxide concen-

tration of 100 ppm can not be safely extrapolated to typical ambient concen-

trations because the relationship between carbon monoxide uptake rate and con-

centration is not known over this range.   In highly contaminated areas, where

ambient concentrations are particularly high, however, soil carbon monoxide

uptake rates could approach this level.  The relationship between uptake rate

and concentration needs to be determined before uptake in the contaminated

areas can be evaluated.

                and Smith et^ al.    have shown that production of carbon

monoxide also occurs in soils.  It is not clear whether this results  from

nonbiological processes or from a combination of biological and nonbiological

processes.  Apparently the net rate of carbon monoxide removal by soil  is

determined by the effects of uptake and production combined.  Additional

studies, perhaps using carbon-14  (^C) are needed to clarify  this point.

      To discover which types of soil microorganisms were most active in

carbon monoxide uptake, Inman and Ingersoll^2 isolated over 200 different

species of fungi, yeast and bacteria from three soil samples.  They found

14 species of fungi capable of removing carbon monoxide from an artificial
atmosphere.  These included four strains .of Penicillium digitatum. Penicillium
                       each             ^
restrieturn; four species.of Aspergillus, Mucor hiemalis; and two strains each

of Haplosporangium parvum and Mortierella vesiculata.

      Bacteria also have been isolated  from soils which utilize carbon

monoxide  in metabolism or fermentation.  Kluyver and Schnellen'iu reported

that under anaerobic  conditions the species Methanosarcina barkerii produce

methane utilizing carbon monoxide as the only source of carbon.  They presented

evidence  that this fermentation proceeds in the following two steps,  reactions

 (61) and  (62):

                        4 CO + 4 H20 ->• 4 C02 + 4 H2                      (61)

                          C02 + 4 H2 ->• City + 2 H20                       (62)

Combining these  two equations gives the following net reaction  (63):

                         4  CO + 2 H20  -»• 3 C02 + CHA                      (63)

      Evidence for the existence of an autotropic aerobic carbon monoxide

utilizing bacteria is not conclusive, ' although several isolated species

have been reported to have the capacity to oxidize carbon monoxide to

carbon dioxide.  Since many of these organisms also could oxidize hydrogen

or assimilate simple organic compounds^ their true aerobic and autotrophic

nature has been questioned. ^

      Vegetation.  Some plant species apparently have the ability to remove

carbon monoxide from the atmosphere but the available data are conflicting.

In 1957, Krall and Tolbert^lS exposed excised barley leaves to an artificial

atmosphere containing 60% carbon monoxide-  C (CO) and determined the rate

of conversion to serine and other compounds.  The conversion at 28 C was

0.038 ymoles/g/h.  In 1972, Bidwell and Fraser^ investigated the incorpora-

tion of carbon-14 (  C) from an artificial carbon monoxide atmosphere into

plant carbon compounds (mainly sucrose and serine).   The uptake of carbon

monoxide by excised leaves was found in seven out of the nine species studied.

Six species showed a measurable uptake rate at concentrations of 2 ppm or

lower.  If we assume that the uptake rate varies linearly with the concentra-

tion in the range 0.2 to 2 ppm, the data for these six species indicate an

average uptake rate of 0.003 ymole/dm /h at 0.2 ppm.  This would be equivalent

to 0.23 x 10~H g/cm^/s if we assume a leaf area that is 10 times the ground

                                                                  O C/
surface area.   This is 15% of the average value reported by Seller    for

soils at this concentration, indicating that soils are probably more important

than plants as a sink for carbon monoxide.  In 1973, Kortschak and Nickell^l2

using methods similar to those of Bidwell found the uptake of carbon monoxide

by the leaves of sugar cane to be 1 x 10~^ mg/cm^/h at 2 ppm carbon monoxide.

In contrast, Inman and Ingersoll^-"^ were unable to measure any carbon monoxide


removal from an artificial atmosphere of 100 ppm by any of the 15 species

of higher plants they tested, although their methods were more than adequate

to establish carbon monoxide uptake rates by different soils.  The assessment

of the vegetation as a sink for carbon monoxide will have to wait until this

discrepancy is resolved.

     Several workers such as Delwiche,^^^a demonstrated the production of

carbon monoxide by higher plants.  The relative importance of plants as a

source and as a sink for carbon monoxide can not be determined from the

available data.

     The uptake rate of carbon monoxide needs to be determined for a number

of plant species over the range of atmospheric carbon monoxide concentrations

found in ambient and polluted atmospheres.  Also the rates of evolution of

carbon monoxide by plants, if any, need  to be known before the overall sink

properties can be ascertained.  In addition, the relationship between soil

carbon monoxide uptake and carbon monoxide concentration should be determined,

especially in the range between 0.2 ppm and 50 ppm.  The effects of tempera-

ture on the evolution of carbon monoxide by the soil also still remains to

be learned.  Knowledge of these parameters will improve the determination of

the carbon monoxide sink properties of soils and vegetation.

     Biochemical Removal.  The binding of carbon monoxide by porphyrin-type

compounds, found in plants and animals, is analogous to carbon monoxide uptake

by hemoglobin in blood and is a potential sink for carbon monoxide.  However,

the carbon monoxide is reversibly bound by the heme compounds found in man

and animals and thus is eventually discharged from the blood, and only a

small  fraction of carbon monoxide is retained.  *

     Absorption in the Oceans.  Unlike carbon dioxide, the oceans can no

longer be considered a sink for atmospheric carbon monoxide.     In fact,

in the geographical areas studied, the carbon monoxide supersaturation and

the high diurnal variations of carbon monoxide observed in the upper layer

of the ocean provide evidence that the oceans are a significant source of

carbon monoxide that comes from the photobiologic processes of marine algae.

Inland expanses of fresh water have not been adequately assessed at present

but it is doubtful that they can serve as a sink for the large amount of

carbon monoxide injected  into the atmosphere.

     Removal at Surfaces.  Kummler et al.    have extrapolated the reported

heterogeneous reaction rates of nitrous oxide with carbon monoxide to ambient

temperatures (300 K) and suggest that atmospheric reaction between these

gases in the presence of such common materials as charcoal, carbon black or

glass is a feasible scavenging process for atmospheric carbon monoxide.   This

                                                                    l "\")
mechanism is derived  from observations made by Gardner and Petruccilj  of the

chemisorption of carbon monoxide at room temperature on the oxide films  of

copper, cobalt and nickel using infrared spectroscopy.  Also,  Liberti^l has

observed from analyses of collected urban particulate material that quantities

of carbon monoxide ranging from 10 to 30 ug/g were associated  with the aerosol.

He suggested that absorption by dust may be an important removal mechanism

for ambient carbon monoxide and, through codeposition, it is made available

to soil microorganisms for oxidation.  These studies are inconclusive and

additional investigations are necessary.  Quantitative evaluation of the

catalytic efficiency and absorbing capacity of common surfaces and atmospheric

particulate materials under ambient conditions is required to determine whether

or not they provide adequate sinks for atmospheric carbon monoxide.

                                   CHAPTER 4


     There are major problems in correlating atmospheric data to the health

effects data for carbon monoxide pollution despite both the financial invest-

ments in monitoring and legal reliance placed on these data.  Some of the

problems related to both the nature of the pollutant and the monitoring de-

vices used will be considered in this chapter.

     There are varying amounts of a large number of trace substances in the

atmosphere.  When the concentration of certain of these increases above a

threshold amount, there is a harmful effect on human health or welfare.  When

this happens the atmosphere is considered polluted and the causal agents are

called air pollutants.  Most substances considered to be air pollutants

originate from both natural and man-made sources, therefore, origin cannot be

used as a criterion for distinguishing between pollutants and nonpollutants.

     The determination of the threshold concentration for each pollutant is

difficult and very often controversial.  Therefore, specification of a polluted

atmosphere is based upon concentration standards established by a consensus among

panels of experts designated by air pollution control officials.  A distinction

is made between primary standards which are based on health effects and secondary

standards which relate to welfare.

     The amount of carbon monoxide emitted into the atmosphere is about ten

times greater from natural than from anthropogenic sources.  However, the

natural sources are so widely dispersed that their contribution to atmospheric

pollution can be neglected..  Analyzing pollutant exposure is complicated both

because people's activity patterns are unpredictable and because there is a

wide concentration variation with respect to location and time for carbon

monoxide from anthropogenic sources.

     Carbon monoxide, when taken into the body, is converted into carboxy-

hemoglobin, and then quantitatively eliminated.  In a uniform environment,

its concentration in the body reaches a steady state and then remains constant.

Exposure to an environment with a higher carbon monoxide concentration in-

creases the body burden; while exposure to an environment with a lower concen-

tration causes the elimination of carbon monoxide thus decreasing the total

body burden.  Such changes characteristically take place over several hours.

     Another consideration is the choice of the best way to express the large

number of measurements.  Conventionally, the "hourly mean" values and "eight-

hour means" have been expressed as arithmetic means.  The "hourly mean" and

the "eight-hour mean" were calculated on the assumption that a significant

number of people will be exposed either for 1 or  8 h.   In principle,

the measure of central tendency used should be such that those periods having

the same mean value should also have the same physiologic manifestations.

For any averaging process/however, there are limiting cases in which this

condition cannot be met; for example, exposure to a low concentration for

 8 h is not  equivalent  to a 5 min exposure  to a lethal concentration fol-

lowed by 7 h and 55 min at zero concentration.  But even in less extreme

cases, exposure periods with the same arithmetic average concentration do

not all stress the receptor equally.  Additional study is needed to arrive

at the optimal expression of central tendency.  When a large number of hourly

arithmetic means have been collected it is frequently found that they have a

lognormal distribution which permits the expression of the entire data distribu-

tion as a geometric mean and a geometric standard deviation, i.e., the entire


ensemble of data can be represented by two numbers.  This compactness  of  ex-

pression is very convenient, and is not intrinsically incorrect.  Neverthe-

less the limitations of the data must be kept in mind so that the researcher

does not do further statistical analyses that ignore the character of  the

data.  For example, when using these statistical data to compute the optimum

number of sampling stations  within a monitoring network and the optimum

sampling frequency the individual sample data taken at sequential times and

adjacent locations should  be checked for correlation.   It is possible to reduce

the number of sampling stations if good correlation can be shown.  Furthermore,

the data on each pollutant must be evaluated separately owing to differences

in source-receptor relationships that affect the correlations for pollutants

other than carbon monoxide.

     To summarize,  it is convenient to use statistical measures of central

tendency and dispersion to represent aerometric data;  these measurements are

so closely correlated in space and time that it is disadvantageous to use

additional statistical manipulations that  depend on the randomness or inde-

pendence of successive measurements.  As yet, the ideal spacing of monitors

has not been achieved.

Emission of Carbon Monoxide

     Motor vehicles are the largest anthropogenic source of atmospheric carbon

monoxide.   Diesel-powered  vehicles emit a  much lower amount than those using gasoline

as a fuel.   A 1972 gasoline-powered light-duty vehicle emits 59,0 g/km

(36.9 g/m)   as compared to a pre-1973 diesel-powered light-duty vehicle that

emits 2.7 g/km (1.7 g/m) .-^   A compilation of the estimated man-made emissions

of the major pollutants for the United States and for  three major cities is

given in Table 4-1.  The breakdown of carbon monoxide emissions by source,
both  nationally and for New York City, given in Table 4-2, shove the pre-
ponderance in the mobile sources.
     In individual locations,however, stationary sources can be equally
important.  For example, proximity to a poorly-controlled petroleum refinery
could, under certain circumstances, result in exposures to significant concen-
trations of carbon monoxide.
     Emission factors for a number of sources are listed in Table 4-3.
The industrial factors are multiplied by appropriate measures of industry
size to obtain the total emissions.     To obtain the total emissions for
motor vehicles, the emission factor adjusted by the appropriate speed correc-
tion factor is multiplied by the total vehicle miles travelled.  Figure  4-1
has been calculated from data in the reference and gives the speed correction
factor for 1975 with the additional assumption that 88% were autos and 12%
were light-duty trucks.  "
     Vehicle speed has long been recognized as a critical variable in pre-
dicting motor vehicle emissions.  It is only recently with the introduction
of sophisticated emission control systems such as catalysts and stratified
charge systems that the importance of ambient temperature and hot/cold weighting
 (a measure of the relative mileage contribution of warmed-up vehicles) has been
recognized.  This has led to the quantification of the corresponding emission
adjustment factors.  Calculations based upon USEPA's emission factor report
indicate that carbon monoxide emissions from conventional vehicles not equipped
with catalytic exhaust control devices are three times greater during cold
starts at -7C (20F) than at 27C (80F).1  Over this same ambient temperature
range the carbon monoxide emissions of catalyst-equipped vehicles during cold
starts differ tenfold.  As these improved pollution-control systems become more
prevalent, it will be particularly necessary to take adjustment factors  into

                             TABLE 4-1
      Estimated Man-Made Emissions for Year 1973, 10  Tons/Yr
U.S. TOTAL (1972 )X
  U. S. Environmental Protection Agency

2 New York City Department of Air Resources 74b

3 Los Angeles County Air Pollution Control District 250

^ City of Chicago Department of Environmental Control 74

                             TABLE 4-2
               Carbon Monoxide Emissions, 10  Tons/Yr

Fuel Combustion
Fuel Oil
Natural Gas
Industrial Processes
Solid Waste Disposal
Motor Vehicles
Other (Off-Highway)
Percent of
Percent of
1.  Environmental Protection Agency

2.  New York City Department of Air Resources

                                        TABLE 4-3
Bituminous Coal Combustion




Fuel Oil  Combustion

Carbon Black Manufacturing

Charcoal Manufacture

Meat Smokehouses

Sugar Cane Processing

Coke Manufacture

Steel Mills


Petroleum Refineries


Utility and large industrial

Large commercial and general
 industrial boilers

Commercial and domestic

Hand-fired unit
Power plants
Industrial, commercial, and

Channel process
Thermal process
Furnace process

Pyrolysis of wood
Field burning

Without controls


Gray iron cupola

Uncontrolled fluid
catalytic cracking

Uncontrolled moving
bed cracking
     1 Ib/ton coal  (0.5 kg/t)
     2 Ib/ton coal  (1.0 kg/t)
    10 Ib/ton coal  (5.0 kg/t)
    90 Ib/ton coal  (45 kg/t)

     3 lb/103 gal (0.36 kg/103l)
   4-5 lb/103 gal (0.47-0.6 kg/103l)
33,500 Ib/ton (16,750 kg/t)
 5,000 Ib/ton (2500 kg/t)

   320 Ib/ton (160 kg/t)

   0.6 Ib/ton meat (0.3 kg/t)

   225 Ib/acre burned (253 kg/ha)

   0.6 Ib/ton (0.3 kg/t)

 1,750 Ib/ton (875 kg/t)

   145 Ib/ton (72.5 kg/t)

13,700 lb/103 bbl (38.63 kg/103l)

 3,800 lb/103 bbl (10.7 kg/103l)
Approximately 20 mph(32 km/h)
 for each of the years
g/mi (55.3 g/km)
 11   (48.5 g/km)
 11   (47.6 g/km)
 "   (44.4 g/km)
     (40.6 g/km)
 "   (38.0 g/km)
     (19.3 g/km)
 "   (7.02 g/km)
   *Average emission factors for all vehicles on the road in given year
                                *                          ,- -N


   o m
   P.  *





                                                                                                 ...— HONOLULU

                                                          TABLE 4-4

                                            CARBON MONOXIDE  (PPM) HOURLY READINGS
                                                    STREET LEVEL STATIONS*
                                                      Calendar Year, 1974
                                                                                   8 HOURS 7 AM - 3 PM EST
   STATION            NUMBER OF    ARITHMETIC   PEAK         NUMBER OF    NUMBER OF    AVERAGE      PEAK         NO. > 9
                      READINGS     AVERAGE                   HOURS > 35    8 ». AVGS

   	1973  1974   1973  1974   1973  1974   1973  1974   1973  1974   1973  1974   1973  1974   1973  1974

                                                   •t ,.

   00-121 Street      6018  5473    4.5   4.1     35    26      00    213   192    4.5   4.5   13.0  15.6      8     5

   94  59th Street    7140  8463   18.5  16.2     70    57   384     89    277   335   21.7  18.5   49.6  37.4    276   334
•f      Bridge
   96 Canal Street    7635  7698    8.8   9.3     36    45     1      1    302   309   10.4  11.0   21.5  20.5    184   217
   98  45th Street    8503  8698   11.3  12.6     61    43    29      5    349   355   13.6  14.6   25.3  25.8    264   307
         *City of New York, Bureau of Technical Services, Dapatrtaaent of Air Resources

                                                      TABLE 4-5

                                        CARBON MONOXIDE, HOURLY READINGS, PPM
                                                   ROOFTOP STATIONS*
                                                  Calendar Year, 1974

1 -Bronx HS of Science
,L 14- Queens College
30- Springfield Gardens
5 -Central Park Arsenal
10-Mabel Dean Bacon
18 -Brooklyn Public
26 Sheepshead Bay HS
3 4- Sea view Hospital

NO. >
*City of New York, Bureau of Technical Services, Department of Air Resources^3

      Temporal Variations.  A lognormal plot (normalized logarithmic probability

plot) of data separated both by season and by day of the week    for a station

in New York City is shown in Figure 4-3.  Seasonal differences are small com-

pared with the difference between Sundays and weekdays.  This is probably not

surprising since there is little seasonal difference in New York City's traffic

density.  Sunday traffic however, is significantly lighter than weekday traffic,

and this correlation shows up clearly.

      The correlation between carbon monoxide concentration and traffic is still

more apparent in Figure 4-4.^  These plots of the diurnal course of both traffic

and carbon monoxide reveal their typical patterns and  show their similarity.  Cities

with a marked rush-hour traffic peak in the morning and afternoon tend to show

similar carbon monoxide patterns.  Cities such as New  York that experience saturation

traffic throughout business hours tend to show a plateau during the day rather than

a mid-day minimum.

      Spatial Variations.  In narrow canyon-like streets, most of the carbon

monoxide is  emitted at ground level.  Dilution is largely through mechanical

turbulence induced by the movement of the traffic and  possibly by convection

from the excess heat produced by the vehicles.  The resulting vertical distribution

 (Figure 4-5) shows smoothed vertical profiles up the sides of two high-rise buildings

in New  York  City.     The data  are given by  season to  distinguish between the heating

and nonheating  seasons.   The  differences shown are probably not significant.  One

conclusion that can be drawn  is that  the concentrations measured at a given station

are very  sensitive to the height of  the intake tube.   This point is discussed further


         FIGURE 4-3.  1967 Cumulative Distributions of Hourly Average Concentrations of Carbon Monoxide
                      (110 East 45th Street, Manhattan)



                                                                                            WEEKDAY 1/6 - 5/17
                                                                                              .WEEKDAY 7/30 - 9/14
       SUNDAY 1/6 - 5/17

SUNDAY 7/30 - 9/14
                                                                             January 6- May  17:   1936 weekday hours
                                                                                                   360 Sunday hours

                                                                             July 30- Sept.  14:    731 weekday hours
                                                                                                   144 Sunday hours
                                                                             1 ppm = 1.145 mg  CO  at  25°  C
                           10     20   30   40  50  60  70    80     90    95
                                        Percent of values £  ordinate value
                                                                                    98    99




                             FIGURE  4-4
                        DIURNAL VARIATION  OF
                     CARBON MONOXIDE AND TRAFFIC
                   CARBON VONOXIOE
                 I — I — I _ I _ i  i   »  i   i _ I
                                               L_l — I _ i
                                                  1  1  I
                                10    N
                                                8    10
    35 i—
    30 r-


     Meteorological Effects.  An additional complication occurs when a pre-

vailing wind blows across a deep street-canyon.  Studies have shown that this

induces a downward air motion on the side of the street facing the wind, a

flow across the street in the opposite direction to the prevailing wind, and

an upflow on the upwind side of the street.  This reverse eddy causes low

carbon monoxide concentrations on the downwind side of the street and high

concentrations on the upwind side.

     Local wind behavior combined with the configurations of nearby structures

can be responsible for significant differences in the measurements reported

by closely adjacent monitors.  For example, differences as great as twofold

were found at different corners of the same intersection in Tokyo. Ola

     Indoor-Outdoor Relationships.  Because carbon monoxide is relatively

inert, there is little adsorption  on surfaces.                   Therefore,

with the possible exception of a  brief time lag, a close relationship would

be expected between concentrations in the open air and those inside buildings.

Figure 4-6,which gives indoor and outdoor concentrations and adjacent traffic

count for a third-floor apartment in a high-rise building in New York City,

shows that this is true in many cases.13^  The daily course of carbon monoxide

for a detached suburban home  (Figure 4-7), where leakage from an attached

garage and the effects of gas-cooking and smoking largely obscure the outdoor

influences^,is in marked contrast.^2  The effects of indoor sources including

tobacco  smoking predominate over  the influence of the outside atmosphere.

Indoor-outdoor combined effects will be increased if windows and doors  are

open and decreased if the house is tightly closed^as is the case during the

heating  season.




             25 r

                         	Inside 3rd Floor
                               -Outside 3rd Floor
                        Heating Weekdays
                                                                     I	I	I	I
200      400     600     800     1000     1200    1400

                                   TIME OF DAY
                 Figure  4-6  -   Diurnal  Carbon Monoxide &  Traffic  -  Site 1  -  Heating Season  - 3rd Floor - Weekdays.


                                                            I       I        I
                                                CAR BEING TAKEN FROM GARAGE

                                           	. FAMILY ROOM

                                           	• — OUTSIDE
     1200     1700    2200 I   1300

                                        1300     1800

                                        MAR 6

                                     TIME, hours
 900     1400

   Figure  4-7   -   Indoor-Outdoor Diurnal Variation  in Carbon Monoxide.

     Summary .   Urban carbon monoxide concentrations are at least ten times

higher than background concentrations; seasonal differences are small; concen-

trations on Sundays are almost invariably less than those on weekdays; diurnal

concentration patterns follow diurnal traffic patterns, with a tendency to

peak during morning and evening rush hours  (the intervening period may or

may not show a decrease depending on the nature of nearby traffic) ; concen-

trations decrease steeply with increasing height  and are also affected, even

when averaged over the long-term, by persistent air circulation patterns.

Because carbon monoxide emission decreases and mechanical turbulence increases

with increasing vehicle speed; high speed roads tend to yield lower ambient

carbon monoxide concentrations in their vicinity even though the traffic

density is increased.   In addition, because high speed roads tend to be built

in open spaces and on elevated roadbeds they are less subject to canyon effects.

Measurement of Carbon Monoxide

     Several of the techniques used for monitoring carbon monoxide are dis-

cussed in detail in the Appendix.  The previous discussion has pointed out a

number of measurement problems common to all monitoring techniques.   There

are however, additional problems specific to the measurement of carbon monoxide.

The Environmental Protection Agency has set ambient air quality standards for

carbon monoxide based on the resulting concentrations of carboxyhemoglobin in

human blood.  These were supposed to remain below 2% in nonsmokers,  which

corresponds to carbon monoxide concentrations of 40 mg/m  (35 ppm)  for 1 hour,
or 10 mg/m3 (9 ppm)  for 8 hours. ^   Since there is a finite probability of

exceeding any given  concentration the statistical status of these standards

was further defined  by stating that  they could  be exceeded by experiencing

one higher  concentration  per year without constituting a violation.   That is

to say,  these standards refer to  the second highest concentration measured in


any year.  It has recently been argued that using three fixed periods per day

(0000-0800, 0800-1600, and 1600-2400) for the 8-hour averages leads to under-

estimating concentrations.  Therefore, 8-hour moving averages of one-hour

average concentrations should be used instead.  "  The number of violations

which result if running averages are used should be limited by counting only

nonoverlapping 8-hour averages as violations.

     These standards have had a profound effect on both the measurement and

reporting of carbon monoxide concentrations.  Whether or not the selection of

concentrations has been completely accurate it can be seen from the foregoing

discussion that a typical mobile individual may experience a history of carbon

monoxide exposure very different from that recorded by either one or several

stationary monitoring devices.  A situation can be pictured in which a monitor

at the second-floor level would measure one concentration, while a person

on the curb directly beneath would experience ten times that concentration,

and another person on an upper floor would be exposed to significantly less

than the measured concentration.  A sampling probe can always be located so

that much higher concentrations are measured than a person in the vicinity could

reasonably experience for any extended period.

     The statistical problem of the optimal theoretical design for sampling

systems has not yet been solved.  Consequently* to obtain reliable data first

the number and locations of the sampling stations  must be decided and then

many trials must be made to minimize any distortion of the data because of one

or two anomalous locations.

     Assuming optimal locations have been selected and tested the quality

control of the individual monitoring instruments remains a problem.  All

carbon monoxide monitoring instruments have features in common; they contain coopl

electronic circuits; they are operated near the sensitivity limits set by


inherent noise; and most significant of all their calibration is arbitrary.

None of the available instruments can be calibrated solely from consideration of

its operating principle.  For this reason the instruments must be calibrated

using standard gas mixtures.  Their accuracy is thus limited by the accuracy

of available standards.  Additional potential sources of error are the

electronic circuits that add the possibility of malfunction or calibration

drift not necessarily obvious from inspection of the output data and unless a

program of frequent calibration, preferably daily, is instituted.  Much of the

carbon monoxide monitoring in the past has been performed by nondispersive

infrared instruments,  wfaich are very susceptible to water vapor, and which

require highly-trained technicians.  Thus, some of the data generated have

been unreliable.

 Simulation Modelling

      The preceding discussion has  pointed out  the inherent difficulties  in

 obtaining carbon  monoxide  concentration data in order  to calculate  accurately

 human exposures.   Ideally,  the  individuals being  tested should  carry a personnel

 monitor that  would continuously record the concentrations to which  they  are

 exposed.   Alternately,  a detailed  record of their movements combined with moni-

 toring data representative  of each location could  be  used to calculate  ex-

 posures and rate  of carbon  monoxide uptake.  For  technical and  economic  reasons

 neither of these  alternatives is feasible at present.

      A major  step towards  the ability to characterize  the concentration  in

 a complex area has been the development of simulation  models*   Once a successful

 model has been achieved, estimates  of concentrations at a particular location can

 be made, given the intensity.of  emissions and their spatial and  temporal  patterns,

 the micrometeorology,  and the structural configurations bounding the area.  In

 combination with  a relatively small number of  strategically located monitoring


stations, such models can be used to depict the concentrations at many locations

in a large area.  Unavailability of such sophistication in the past necessitated

using oversimplified models.  The pressures created by the cost of pollution

control require investing in the development of more effective models.  Modelling

has also become important in the preparation of environmental impact statements.

Therefore, there is an increased effort  to develop successful air quality models.

The models for pollutants such as sulfur dioxide  are usually long-term.  Be-

cause of the difficulty  in  depicting meteorological conditions over extended

periods., these can be in error by as much as 100% or more.

     Short-term models for  urban carbon monoxide  conditions are currently

under development in order  to meet  state implementation plan requirements and

to prepare environmental impact statements.  These short-term models are
capable of calculating concentrations within 25%  of measured values.

Because of the short-term relation  between carbon monoxide exposure and health

effects, concentration models have primarily emphasized hourly averages.


     Carbon monoxide is not significantly altered metabolically when taken

into the body.  It binds reversibly with heme pigments, principally hemoglo-

bin, the red pigment of the blood.  A direct indication of the amount of

human exposure can therefore be obtained by hemoglobin measurements.  From

the standpoint of health these are more significant than ambient air measure-

ments, since the concentration of carboxyhemoglobin (HbCO) in the blood is

related to the physiologic effects of carbon monoxide on people.

     There are  practical limits to the value of biologic sampling for

carbon monoxide uptake.  Carbon monoxide is taken up relatively slowly by

the body from ambient sources.  Since the rate of uptake is dependent on

several physiologic factors, the interpretation of measurements in terms of

ambient sources may not be simple.  Because cigarette smoking is a major

source of carbon monoxide exposure, widely prevalent in the urban population,

ambient air sources can be evaluated only in nonsmokers.  Occupational sources

of carbon monoxide exposure, for example in garages, may also contribute to

its presence in urban dwellers.

     Routine sampling of blood has two additional practical limitations; the

first is that the discomfort of taking a blood sample by any method would

make random sampling in the general population unacceptable, and the second,

is that analytical techniques for determining low concentrations of carboxy-

hemoglobin are as yet not sufficiently reliable.  At the time of the previous

National Academy of Sciences report (1969) rapid spectrophotometric methods

were not considered accurate for measuring very low carboxyhemoglobin concen-

trations. 293  Accuracy is crucial if the blood carboxyhemoglobin is used to

indicate exposure to the low concentrations in urban air.  The methods currently

used to measure carboxyhemoglobin are described in Appendix A.

     Carboxyhemoglobin can also be estimated indirectly by measuring alveolar

gas, which is the gas present in the deep regions of the lungs.  Using subjects

with normal lungs and properly applying this technique, good results are

achieved with a minimum of discomfort.  The subject's cooperation and well-

trained personnel are required.  Therefore, the wide applicability of this

technique for routine sampling of carbon monoxide exposure is unlikely.  This

technique is described in the Appendix.  Because a small amount of carbon

monoxide is generated within the body, there is about 0.4% carboxyhemoglobin

present even when no carbon monoxide is in the inspired air, which adds a

further complication to biologic sampling for carbon monoxide exposure.  The

production rate of carboxyhemoglobin can be increased by certain diseases and


physiologic conditions such as hemolytic anemias.    These can increase the

carboxyhemoglobin 2 to 3% above normal endogenous production.  It is also

increased by ingesting drugs that induce the hepatic oxidizing enzyme, cyto-

chrome P-450.    For example, barbiturates slightly increase carbon monoxide


     We can add the effects of carbon monoxide produced in the body to that

of  the carbon monoxide from the ambient air.^2  The expected carboxyhemoglobin

concentration at equilibrium,calculated using an endogenous carbon monoxide

production rate of 0.4 ml/h, is shown in Table 4-6.  For comparison there is

a calculation based on the empirical equation, % carboxyhemoglobin = 0.4 + (p/7),

where p is the inspired carbon monoxide concentration in ppm.  Table 4-6 shows

that if equilibrium is achieved (for comment see the section on uptake) 2%

carboxyhemoglobin should result from inspired carbon monoxide of about 12 ppm,

and 3% carboxyhemoglobin from about 18 ppm.

                                        TABLE 4-6

        Calculated Carboxyhemoglobin at Equilibrium with Inspired Carbon Monoxide

                      Concentration, Pressure in Parts Per Million

                             (Applicable to nonsmokers only)

                                                    % Carboxyhemoglobin

Inspired Carbon Monoxide    % Carboxyhemoglobin     from empirical equation       pn

ppm	    from Coburn, et al.     ~ ~  .   -----	    « *  . _ CO,.,**
_ - u
*Assumptions used in the equation of Coburn,  Forster  & Kane:     the carbon monoxide pro-

 duction = 0.4 ml/h STPD;  the diffusion capacity,  D_CO = 20 ml/min/torr;  the barometric

 pressure, 760 torr;  the alveolar ventilation,  3,500  ml/min STPD;  the Haldane constant =

 220;     the mean pulmonary capillary oxygen  pressure,  C00 = 100  torr  and the fraction

 of unbound hemoglobin is  constant at 3%.

**PtCO  is the inspired carbon monoxide concentration  in ppm.   This equation is applicable

  up to 50 ppm.


     The most extensive study of carboxyhemoglobin in the general public has

been carried out by Stewart and his associates, who sampled blood drawn from

about 31,000 individuals by blood donor mobile units in 17 urban areas and

also in some small towns in New Hampshire and Vermont.     Their results showed

that smokers' blood contained a much higher percent of carboxyhemoglobin than

nonsmokers.  They found however, that in all regions studied a large percentage


of the nonsmokers had over 1.5% carboxyhemoglobin, indicating significant ex-

posure.  Frequency distributions for each of these regions indicated that a

small percentage (1-2%) of the nonsmokers^ despite claiming to be exsmokers, were

actually still smoking.  Their inclusion as nonsmokers does not change the

results significantly.  In addition, some of the high values found in non-

smokers in four of the cities  may have been due to special occupations

in which unusual exposures to high carbon monoxide concentrations occurred in

enclosed areas.

     The effect of occupationally related carbon monoxide sources was emphasized

by Kahn and associates who did a similar analytical study of more than 10,000

blood samples from nonsmokers and 6,000 samples from smokers in the St. Louis

area.1^  For the nonsmokers classified as "industrial workers" the mean car-

boxyhemoglobin concentration was 1.4% and for those classified as "other than

industrial" the mean was 0.8%.  The overall mean for both employed and un-

employed was 0.9%.  The conclusions drawn from the results of Kahn et_ al.

                                            377 378
differ markedly from those of Stewart e£ al.   '     Whereas Stewart and his

coworkers concluded that 35% of nonsmokers in St. Louis were exposed to ambient

carbon monoxide causing their carboxyhemoglobin to be greater than 1.5%, Kahn

et  al.Lyv concluded that only a small percentage of nonsmokers (many of whom

were industrial workers), had values greater than 2%.  Of the nonindustrial

workers, 5.7% had values above 2%, but some of these were unquestionably smokers.

     Kahn et al.    did show that a small urban-rural gradient  for  carboxy-

hemoglobin existed in the St. Louis area.  Three areas, "highly urban,"

"urban," and "rural" were designated according to the 1970 census data.   The

mean carboxyhemoglobin for nonsmokers who were not industrial workers was

0.8%, 0.7% and 0.6% for the highly urban, urban and rural areas respectively.

These very small differences do not support the conclusion that carbon monoxide

uptake from ambient sources in the urban air is physiologically significant.

     The discrepancy in the conclusions of these two studies is important

with respect to current air quality control strategies.  Both studies empha-

size the significance of smoking and occupational exposures.  Stewart's group,

however, infers that in cities with such different meteorological conditions

and population densities as New York, Los Angeles, Denver, New Orleans,

Phoenix, Anchorage, St. Louis and Washington, one-fourth to three-fourths of

the nonsmokers are consistently exposed to carbon monoxide concentrations

above the current standard of 8.7 ppm (which would result in carboxyhemoglobin

above 1.5%, see Table 4-6).  Kahn et^ al_.    conclude^,  contrarily,  that the

carbon monoxide in the urban environment of St. Louis  contributes  only negligibly

to carbon monoxide uptake by the general population.

     Other measurements have been made that apply to the questions raised by

these studies.  Aronow ^ ^1.  '  '   found carboxyhemoglobin values around

1.0%  in 25 nonsmoking subjects in the Los Angeles control area which rose

to 5.1% when these subjects were driven in a car on freeways for one and

one-half hours.  Ambient carbon monoxide  was reported to be 1 to  3 ppm in

the laboratory and about 50 ppm in the car.   Radford et. ^i.- (unpublished

observations) found values of 0.9% carboxyhemoglobin in the blood  of 19

laboratory workers and other personnel in Baltimore, who had never smoked,

and values of 0.6% carboxyhemoglobin In 32 nonsmoking residents of Hagerstown,

Md., a small rural city.  The carbon monoxide concentrations observed in

Baltimore are comparable to those reported in St. Louis, and these results

resemble those of the St. Louis study.  Horvath  (unpublished observations),

in Santa Barbara, California>has found 0.6% carboxyhemoglobin in the blood

of approximately 150 nonsmokers, indicating low  carbon monoxide exposures

for this city, even though high ambient air values have been recorded in

Santa Barbara.  Ayres et al.2* reported that 26  nonsmoking hospitalized

patients in New York City had a mean value of 1.0% carboxyhemoglobin.  The

carbon monoxide concentrations in the hospital were similar but slightly below

those found outdoors at the same time.  These last two studies are particularly

important because the carboxyhemoglobin was measured by gas chromatography

rather than by the spectrophotometric techniques used by the other invest!-

     Carboxyhemoglobin  concentration in smokers  depend*on the number of cigarettes

smoked, degree of inhalation and other factors.  The  carbon monoxide content  of
cigarette smoke is up to 5% by volume    and about 80% of carbon monoxide

produced by smoking cigarettes is retained.  Carboxyhemoglobin levels

as  high as 15% have been reported in chain smokers, but values are usually in the

3-8% range  in most smokers.

                                   CHAPTER 5

                           EFFECTS ON MAN AND ANIMALS


     Carbon monoxide In the body comes from two sources; endogenous, from the

breakdown of hemoglobin and other heme-containing pigments; and exogenous, from

inhalation.  The catabolism of pyrrole rings is the source of the endogenous

production of carbon monoxide, which in adults normally leads to carbon monoxide

production of about 0.4 ml/h (STP).     This can be increased by hemolytic

       Q ^                                                            T/\

anemias   and the induction of hepatic cytochromes from taking drugs.    Dihalo-

methanes may have increased endogenous carbon monoxide production and markedly

elevated carboxyhemoglobin.

      Inhalation is the first  step in the process  of  exogenous  carbon monoxide

 uptake,followed by an increase in carbon monoxide concentration in  the alveolar

 gas with diffusion from the gas phase through the pulmonary membrane and  into

 the blood.  The rate of uptake into  the body  is limited by the rate of diffusion

 from the alveoli and the combination with  the blood.  When the concentration of

 carbon monoxide is very high  the rate of uptake may  be partially limited by  the

 amount that can be inhaled with each breath.

      Using modern concepts of the physiologic factors that determine carbon

 monoxide uptake and elimination,  Coburn, Forster, and Kane  developed an

 equation for calculating the  blood carboxyhemoglobin as a  function  of  time.

 The basic differential equation was:

      d(CO) = '      [HbCO] ^ PC°2 v        1        .  Pi00	

       dt      CO   [HbO?] A  M      1  + PD - 47     1  + P -  47
                        *•           ——    o	    7T-     »

                                     ^  \         *   —

 where d(CO) is  the rate of change of  carbon monoxide in the body

     ,   dt


          V^Q       is the carbon monoxide production rate

          [HbCO]    is the concentration of carbon monoxide in the blood

          [HbC^J    is the concentration of oxyhemoglobin

          *C 2      is the mean pulmonary capillary oxygen pressure

          M         is the Haldane constant (220 for pH 7.4)

          Dr         is the diffusion capacity of the lungs

          Pfi        is the barometric pressure
          V^        is the alveolar ventilation rate

          p-j-CO      is the inspired carbon monoxide pressure.

     This equation was designed to investigate the measurement of blood

carboxyhemoglobin as an indicator of the rate of carbon monoxide production.

Its solution could therefore be based on the assumption  that the mean pul-

monary capillary oxygen pressure, Pr®2> an(* t*ie concentrati°n °f oxyhemo-

globin, [HbC^^were constant and independent of the concentration of carboxy-

hemoglobin, [HbCO].  With these assumptions a solution of the equation be-
came possible.  Researchers who have used the Coburn, Forster, Kane solution

have accepted these assumptions.

     In the general case however, the oxyhemoglobin concentration depends

on the carboxyhemoglobin concentration in a complex way and a solution is

only possible using special computer methods.  A second approximation solution

of the Coburn, Forster, and Kane equation permits the evaluation of the

kinetics of the washing in and washing out of carbon monoxide over a fairly

wide range of carboxyhemoglobin concentrations.*  The basic assumptions on

which the differential equation and its solutions  are developed are described
  e are indebted to Dr. Alan Marcus and Mr. Philip Becker of the University
  of Maryland at Baltimore County for providing both this solution and the
  computer printouts.


in the original paper by Coburn et^ al.^2  These assumptions are not re-

strictive however, and therefore the solutions are applicable generally.

The inspired gas is assumed to be ambient air to which carbon monoxide is


     Besides the carbon monoxide production rate,    the principal factors

that determine the rate of uptake or release of carbon monoxide from the

body are:  the concentration of carbon monoxide inspired; the diffusion

capacity, DL, a function of body size and to some extent the level of

exercise; the alveolar ventilation, v^,    also dependent on the amount of

exercise; the mean pulmonary capillary oxygen pressure, P,^, a function

both of the barometric pressure and the health of the lungs; and the blood

volume, determined by the body size.  Thus,  the principal factors relating

the change in carboxyhemoglobin  concentration, after exposure are: concentration

of carbon monoxide inspired;    endogenous carbon monoxide production; amount

of exercise;  body size; lung health (including diffusion capacity, DA;  and

barometric pressure.

     In addition to exposure from ambient sources, tobacco smoking, particu-

larly of cigarettes,  is an important special case of carbon monoxide exposure.

The theory predicts that a smoker with a normal concentration of carboxyhemo-

globin during the day frpm ambient carbon monoxide sources  will have an

additive amount cf carboxyhemoglobin from smoking.  Recent measurements  by

Smith (unpublished data) in Calgary, Alberta,of ambient exposures and alveolar

carbon monoxide concentrations  have shbwn that the theory predicts end-of-day

carboxyhemoglobin with reasonable accuracy as long as the ambient carbon

monoxide does not fluctuate rapidly during the day as it does with certain

occupational exposures.

     As illustrated in the figures, the theory can also be applied to the

influence of various factors such as cigarette smoking, lung health, duration

of exposure and altitude on the rate of change of carboxyhemoglobin during the

daily activity cycle and to its concentration at night during sleep.

     In the following graphs the variables used in the Coburn, Forster,

Kane equation are for adult subjects at sea level.  Any modifications are

indicated in the legends to the figures.

     The values assumed for the general case are:

           Haldane constant = 220  (for pH 7.4)

           = 0-  (mean pulmonary capillary oxygen, partial pressure) = 95 torr

            c  !

           D   (diffusion capacity) = 20 ml«min   torr"1

           Sleeping V.  (alveolar ventilation) * 3 liters»min
                      4                                 -1
           Light work V   (light exercise) - 5 liters*min

           V   (blood volume) - 5 liters



, 7
1 1 1 f

i i i i
P. __. _ __
r r i i
— »
1 i i i
                                     TIME (H)
   FIGURE  5-1.
Upper graph:  carboxyhemoglobin in a nonsmoker  exposed  to

carbon monoxide during the period 8 AM to 6 PM  and not  exposed

at other times; sleep from 10 PM to 6 AM, light exercise when

awake.  Zero time is midnight on the first day.  The initial

carboxyhemoglobin is an arbitrarily chosen value.

Solid line:  Ambient carbon monoxide 10 ppm during day4

Dashed line:  Ambient carbon monoxide 20 ppm during day.

Lower graph:  this shows the carbon monoxide exposure pattern

used to calculate above carboxyhemoglobin -


     The peak values reached are less than the equilibrium values (Table 4-6).

For 10 ppm the peak percent of carboxyhemoglobin reached on the second day

(when the initial conditions no longer are important) is 1.5% compared to

1.85% predicted for equilibrium.  At 20 ppm the peak value reached at the

second day is 2.86% compared to 3.29% for equilibrium.  These results show

that if exposure to carbon monoxide occurs only during the working day from

ambient or occupational sources, the carboxyhemoglobin reached will be well

below equilibrium for persons not doing heavy work.

% Cone.
                                     TIME  (H.)



f>-*i '"V
• /*i*^*
i i

it* •''!•'


t §^^*" ]
&"'-•/?••* i


T i i i

i' ••We .' ... ^
». -r-\ .


0 25 50 75 100 125 150 17^
                                     TIME (E)
 FIGURE 5-2.   Upper graph:  Carboxyhemoglobin in a smoker during one week,

              smoking 48 cigarettes per day,  but with relatively light in-

              halation.   Sleep and awake conditions as in Figure 5-1.  At

              zero time  (midnight), the initial carboxyhemoglobin is assumed

              to be 0.4%,  which is consistent with endogenous production only.

              Lower graph:  The pattern of exposure to carbon monoxide appli-

              cable to the upper graph.  Each single vertical line represents

              a cigarette  smoked for 5 min, with a mean alveolar pCO of 100
                            *                       X

              ppm during that  five minute interval (light inhalation)*

     By the second day, a consistent daily pattern of carboxyhemoglobin

has been achieved.  The peak carboxyhemoglobin reached in the evening in

this light inhaler is 4.5%.  The carboxyhemoglobin remains well above that

from endogenous production even when smoking is stopped during sleep.

10  ,

 4 '
 0 .
   i  i  iii  i  r
20          30



     The respiratory and cardiovascular systems working together transport

oxygen from the ambient air to the various tissues of the body at a rate

sufficient to maintain tissue metabolism.   In one step in this overall

process, oxygen is carried by the blood from the lungs to extrapulmonary

tissues.  Nearly all the oxygen in the blood is reversibly bound to the

hemoglobin contained in the red blood cells.  The most important chemical

characteristic of carbon monoxide is that  it too, is reversibly bound by

hemoglobin, competing with oxygen for the  binding sites on the hemoglobin

molecule.  Hemoglobin's affinity for carbon monoxide is more than 200 times

greater than for oxygen.  Therefore, carbon monoxide can seriously impair the

transport of oxygen even when present at very low partial pressures.

     The proportion of hemoglobin combined with carbon monoxide at any time

is determined not only by the partial pressure of carbon monoxide, but also

by that of oxygen.  The approximate relation stated by Haldane and his associ-

ates,    in 1912, showed that the ratio of the concentrations of

carboxyhemoglobin (HbCO) and oxyhemoglobin (Itt^) is proportional to the ratio

of the partial pressures of carbon monoxide and oxygen:
                               [HbCO]  = M (pCO)
                                    ]      (Po2r

The constant M is about 210 for human blood.  The accuracy of this expression

has been questioned, particularly when a large fraction of the hemoglobin present

is not combined either with oxygen or with carbon monoxide.188  The equation

approximates the actual relationship closely, however, and has proven very useful

for quantitatively  analyzing the influence of carbon monoxide on oxygen trans-

port  by the blood.

      The oxyhemoglobin dissociation curve (Figure 5»^F) describes

oxygen  transport.   Under  normal conditions when the arterial blood has been

equilibrated in the lungs with a° oxygen partial pressure  (p02) of 90-100 mm Hg,

it contains slightly less than 20 vol % of oxygen bound to hemoglobin (point a).

As the  blood flows  through the various tissues of the body, oxygen diffuses out

of the  capillaries  to meet metabolic needs.  The  p02 falls as the oxygen content

of the  blood is reduced by an  amount determined by the metabolic rate of

the tissue and the  capillary blood  flow.  For a typical tissue or for

the entire body,  the venous oxygen  content is about 5 vol % lower than the

arterial,and the  venous pOn is about 40 mm Hg  (point V).

      The normal oxygen transport by the blood may be impaired by a variety of

factors.  In Figure 5-5 the effects of carbon monoxide on venous p02 are shown

and compared to the normal state and to the   —?

effects of a decrease in blood hemoglobin content (anemia).  Arterial p02

and oxygen content   A indicated by  the symbols ^  and a/ and the symbols v,

v|, and v£ indicate venous p02 and  oxygen content assuming the arteriovenous

 (A-V) oxygen content difference remains constant  at 5 ml/100 ml.  This figure

shows that there  are two consequences of  increasing carboxyhemoglobini an affect

like  anemia that  results  in decrease of p02  in tissue capillary blood, and a shift to

the left of the oxyhemoglobin  dissociation curve, which further decreases the p00


                              /    (0, Hb Cipicltv - 10 ml/100 ml)
Figure  5-5.  Oxyhemoglobin dissociation curves of normal  human

blood,  of  blood containing 50% carboxyhemoglobin, and  of  blood

with a  50% normal hemoglobin concentration due .tq anemia.
                            O'gAg                          34-4-3
(Derived from Rahn and Fenn     and Roughton and Darling    )

in the capillary blood in peripheral tissues.  For the sake of clarity an

example of severe carbon monoxide poisoning has been used in this discussion.

Under conditions of mild carbon monoxide poisoning as shown in Figure 5-6,

the effects are qualitatively similar but less severe.

     The ultimate indicator 
        g   14
Figure  5-6.  Oxyhemoglobin dissociation curve of normal human
blood containing  10% carboxyhemoglobin and blood with 10% de-
crease  in hemoglobin concentration (10% anemia).(Symbols v, V]
and v*2  feave  same  meaning as in Figure 5-5).

                        PERCENT OF Hb COMBINED WITH CO
                     6       10       IS         20


                        BO         100        160
                           CO CONCENTRATION (ppm)
Figure  5-7.   Effect of carbon monoxide on venous p02*  The  arterio-
venous  oxygen content  difference with normal blood flow and oxygen
consumption is assumed to be 5 vol %.  (Modified slightly from
Figure  3 of reference  27).

     This analysis assumes that alveolar ventilation, metabolic rate, and

blood flow all remain constant and that little or no mixed venous blood is

shunted through the lungs without being equilibrated with alveolar air.  It

is well known, however, that 1-2% of the cardiac output is shunted past the

alveolar capillaries in normal subjects, and much larger right-to-left shunts

exist under disease conditions.  Mixed venous blood flowing through these

right-to-left shunts combines with blood that has undergone gas-exchange in

the pulmonary capillaries to form the mixed arterial blood.  Because the

oxygen content of the shunted blood is lower than that of the end-capillary

blood with which it mixes, the resulting oxygen content of the arterial blood

is lower than that of the end-capillary blood, and arterial pC   is somewhat
lower than mean alveolar pC^.   Brody and Coburn,  * have pointed out that if

the oxygen content of the mixed venous blood is abnormally low, as in anemia

or carbon monoxide poisoning,  the effect of the shunted blood in lowering the

arterial pOn will be greater than normal, resulting in a small increase in the

alveolar-arterial oxygen pressure difference (A-a 002).  If the mixed venous

pC^ and the right-to-left shunt remain constant,  the change in the shape of

the oxyhemoglobin curve due to the presence of carbon monoxide also increases

the A-a DC^.  A similar phenomenon occurs  when some lung regions have non-

uniform ventilation perfusion  ratios,  the  case in many types of cardiopulmonary

disease.  Figure 5-$^ taken from the work  of Brody and Coburn, " shows that

slight increases in carboxyhemoglobin  concentration have little or no influence

on the alveolar-arterial oxygen pressure difference (A-a DC^) in normal sub-

jects but in patients with large intracardiac right-to-left shunts or with

chronic lung disease and mismatching of ventilation and perfusion of carbon

monoxide increases the A-a D0«.  This  phenomenon adds a component of arterial

hypoxia to the effect of carbon monoxide on oxygen transport in such patients.











                       4       6       12       16

                             HbCO (%)
Figure 5-8.,  Effect  of carbon monoxide administration on the
alveolar arterial^pO^ difference in normal subjects  (•)» in
patients with VA/Q abnormalities due to chronic lung disease (o),
and in patients with intracardiac right-to-left shunts (x).
(Reprinted  with permission from Brody and

     It has been suggested that cytochrome P~^50is important in oxygen trans-

port in cells and that one effect of increased carbon monoxide tension is to

block transport facilitated via this mechanism.  At present, however , there

is no convincing evidence to support this postulate.

Intracellular Effects of Carbon Monoxide

     Tissue Carbon Monoxide Tension.  Intracellular effects of carbon monoxide

depend on the carbon monoxide partial pressures (pCO) in the tissues.  Calcula-

tions have been made of tissue pCO from blood carboxyhemoglobin and an assumed

mean capillary oxygen partial pressure pO^, using the Haldane equation.  At a

blood carboxyhemoglobin concentration of 5%, pCO is 2 x 10   mm Hg (assuming

the mean capillary pO^ is 40-50 mm Hg) , about 5 times greater than when the

carboxyhemoglobin is normal.  This is about 60-70% of the inspired pCO for a

steady state condition.

     Gothert and coworkers^^ have recently (1970) estimated tissue pCO in

the rabbit peritoneum from measurements of pCO in an air pocket in the peritoneal

cavity.  Measurements were made with rats, guinea pigs or rabbits breathing

carbon monoxide at 86-1,000 ppm.  The pCO in the gas pocket was 42-69% of

the partial carbon monoxide pressure in the alveolar air.  This figure probably

underestimates tissue pCO in tissues where the ratio of oxygen extraction to

blood flow is less than in peritoneum.  Campbell •*• made similar measurements

of pCO in a gas bubble in the peritoneum of mice.

Intracellular pQ
     The pOo in proximity to intracellular compounds which bind carbon monoxide

is a critical factor in possible intracellular effects of carbon monoxide,

since oxygen and carbon monoxide binding are competitive.  In general, recent

research is pointing to the presence of a lower intracellular p02 than previously

thought.  Studies using polarographic microelectrodes in liver (Kessler

demonstrated a tissue p02 of < 10 mm Hg in 10-20% of their penetrations.

Whalen439 found that intracellular p02 in skeletal and cardiac muscle averaged

5-6 mm Hg.  Computing mean myoglobin p02 from carbon monoxide binding to myo-

globin also give a normal p02 value of 4-7 mm Hg (Coburn85).  Tissue p02

levels in brain are somewhat higher.  Since we have no idea about either p02

gradients or compartmentalization in cells it is possible that intracellular

p02 at the  site of carbon monoxide binding compounds are considerably below

these levels.

     Chance at al.   indicate that the p02 in mitochondria under conditions

of tight respiratory coupling and presence of ADP  (state 3) is inhibited 50%

 at 0.01-0.05 mm Hg.  Since  it is possible that the p02 in mitochondrial cristae

 is as low  as 0.01 mm Hg even though mean cytoplasmic p02 is 4-6, later in

this chapter we used      this low value in calculations of possible effects

of carbon monoxide on cytochrome ag function.

      Zorn    studied  the effects of carbon monoxide inhalation on  brain and

 liver p02>  using Lubbers'  platinum electrode and surface electrodes.   It was

 found in both tissues that tissue pO« fell,  even at carboxyhemoglobin of 2%

 saturation, and that the fall was almost directly related to the increase  in

 carboxyhemoglobin.  For a 1% fall in oxyhemoglobln saturation due  to an increase

 in carboxyhemoglobin, pO. decreased 0.2 to 1,8 mm Hg.   This is a particularly

 nice approach since,  if carbon monoxide had only an intracellular  effect,  tissue

 p02 would be expected to increase.   When the experimental animal breathed  air

 not containing carbon monoxide,  tissue pO« returned toward normal.

     Weiss and Cohen431 have reported similar effects of breathing 80 and 160 ppm

carbon monoxide                              saturation           g.^
   for 20 minutes (resulting in carboxyhemoglobin of less than J.J/6;     on
  A                                              /*
rat brain cortex p02 and rat biceps brachii muscle, as measured with a bare

platinum electrode.

     Direct Carbon Monoxide Effects on Intracellular Processes _-  Cytochrome

a^ and Mitochondrial Electron Chain Transport.  Carbon monoxide affinity to

intracellular compounds has customarily been given in terms of the Warburg

partition coefficient:

                             K = [n/l-n] [C0/02]

where n is the fraction bound to carbon monoxide and CO/02 -*-s t*ie ratio of

carbon monoxide to 02 .   The data are usually given where n=0.5/ which gives

the ratio of carbon monoxide to oxygen for 50% saturation with carbon monoxide.

There is apparently no new information about K for aj which has been widely

quoted to be 2.2-28.  In the normal aerobic steady state in the rapidly

respiring State 3, the concentration of reduced cytochrome 33 is very low,

probably less than 0.1%.  Because only reduced cytochrome 33 binds carbon

monoxide, the affinity of carbon monoxide for cytochrome 33 under this condition

is small.  In coupled mitochrondia at low ADP concentration (State 4), the

steady-state concentration of reduced cytochrome 33 may be even lower.

     Earlier studies of Chance^ have not been appreciated in previous publi-

cations concerning possible effects of carbon monoxide on mitochondria.  He

studied the transient from anoxia to normoxia in pigeon heart mitochondria in

both the absence and presence of carbon monoxide.  It was found in un-

coupled mitochondria that CO/02 ratios of 0.2 caused a marked delay in this

transient.  Thus these mitochondria were markedly more sensitive to the

effects of carbon monoxide, when studied in this state.  Chance also points

out that the dissociation of carbon monoxide from reduced a$ is so slow that

it should take 3-4 minutes for one-half unloading to occur; thus after a

hypoxic episode, cytochrome a3 function can be influenced by much lower tissue


     Recent data have  suggested that mitochondrial respiration may be more

sensitive to carbon monoxide under some conditions than previously indicated,

however, there is no solid evidence for implicating this system at low blood

carboxyhemoglobin levels.  Even using C0/02 ratios of 0.2 for 50% binding to

33 which Chance found  in uncoupled mitochondria during transients, and a pC>2

of 0.01 mm Hg, computed  tissue pCO at 5% carboxyhemoglobin of 2 x 10 * mm Hg

is slightly below that necessary  to cause  50% binding to a^, however at a

carboxyhemoglobin of 10%,  tissue  pCO is high enough.  There is a probability

that binding of carbon monoxide to a~ is physiologically significant during

tissue hypoxia at very low blood  carboxyhemoglobin levels.  Previous reviews

of possible effects  of carbon monoxide on  a3 have failed to point out that

the chemical constants and investigations  are never performed at 37 C but at

25 C or  less.

     A recent  line of  investigation               has revealed that heart

mitochondria isolated  after  chronic arterial hypoxemia have a higher State

3 mitochondrial  oxygen uptake,  per gram protein,  than mitochondria  isolated

from an animal which was not chronically hypoemic.  These data may be pertinent

to  acclimatization  to  carbon monoxide,which has been demonstrated in experimental


     Cytochrome P-450.  The  Warburg coefficient for6 Cy*°chrOtne P"45has been

quoted by Estabrook121 to  be 1-5. Calculations similar  to those performed

for cytochrome 33   in the  preceding section suggest, using values of 5 to

10 mm Hg for microsomal p02, that tissue pCO is 1 to 2 orders of magnitude

too low for carboxyhemoglobin at less than 15% saturation to have an effect

on the cytochrome P-450 system.

     Recent advances in our knowledge of the effects of carbon monoxide on


the cytochrome P-450 system are discussed below.  Estabrook e£ al.  found

that under conditions of rapid electron transport through the cytochrome

P-450 system, sensitivity to carbon monoxide increased.  In the presence of

high concentrations of reducing equivalents and substrate, the C0/02 ratio

necessary for 50% binding was as low as 0.2 whereas with slow electron trans-

port the system becomes almost completely refractory to carbon monoxide.

Since carbon monoxide sensitivity does vary under changing conditions,  it

is possible that in some conditions carbon monoxide sensitivity might increase

to levels where the cytochrome P-450 system is influenced by carbon monoxide

tension in tissues at low carboxyhemoglobin.

     Data are now available on the effects of elevated  carboxyhemoglobin on

the cytochrome P-450 system.  Rondia^   found that in rats exposed to 60 ppm,

there was a decreased ability of liver to metabolize 3-hydroxybenzo[a]pyrene.

                    284 285
Montgomery and Rubin   *    found prolonged sleeping time in the presence of

20% carboxyhemoglobin in rats given hexobarbital.   However,  when they compared

these effects owing to carbon monoxide poisoning to those of hypoxic hypoxia,

looking at the effects with an equivalent arterial oxyhemoglobin percent

saturation, it was discovered the effects on sleeping time were greater owing

to hypoxic hypoxia  than to carbon monoxide poisoning.  In subsequent studies,  Roth

         34.4 344c "344(1
and Rubin   '*     have found that the greater effect with hypoxic hypoxia

was due to a decrease in hepatic blood flow.  The effect of carbon monoxide is

probably due to a fall in capillary pO-, rather than carbon monoxide binding to

cytochrome P-450.  Coburn and Kane   found that 10-12% carboxyhemoglobin concentration


in anesthetized dogs resulted in inhibition of loss of hemoglobin- haptoglobin

from the plasma.  It is now known that catabolism of hemoglobin-hepatoglobin

is mediated by a cytochrome P-450 linked system but this could be due to a

blood flow effect.  Rikans333 has reported evidence for a carbon monoxide bind-

ing compound in solubilized rat hepatic microsomes that is distinct from cyto-

chrome P-450, however, the chemistry and relative affinities for carbon monoxide

and oxygen, as well as its function, have not been delineated.

     Myoglobin.  The Warburg partition coefficient for the reaction of carbon

monoxide with oxymyoglobin is 0.04.  Myoglobin is probably the intracellular

hemoprotein most likely to be involved in toxic effects of carbon monoxide,

     Coburn et al.85 have measured the ratio     MbCO/HbCO and found it to be

approximately 1 and Constant, even with  increases in blood carboxyhemoglobin,-exceed-

ing 20% saturation.  Thus, at 5 percent concentration of carboxyhemoglobin, 5

percent of the myoglobin should be bound to carbon monoxide.  It is difficult

to interpret this in terms of toxic effects of carbon monoxide on skeletal

muscle, smooth muscle or heart muscle since the function of myoglobin is not

clearly defined, nor is it known where it is located in the cell.  In the

heart, there is evidence that myoglobin buffers changes in oxygen tension

in close proximity to mitochondria during muscle contraction.-.Another

possible function of these compounds is to facilitate oxygen transport across

the cytoplasm from cell membrane to mitochondrion.   Like other intracellular

hemoproteins, carbon monoxide binding is markedly increased in the presence

of tissue hypoxia.

     There is evidence  (by Clark and Coburn75) of significant carbon monoxide

shifts out of blood, presumably into skeletal muscle, during short-term bicycle

exercise at maximal rate of oxygen uptake.




       Insufficient knowledge exists about  the biological  effects  of carbon

monoxide during intrauterine development and the newborn period.

Several studies report decreased birthweights and increased mortality

in the progeny of animals exposed to relatively high carbon monoxide concen-

trations, but few studies have reported on the more subtle effects at

lower concentrations.  Most of the evidence supporting carbon monoxide effects

on the fetus is inferred from data on maternal smoking rather than from studies

of its effects per se.  This section reviews both what is known and what is

not known about carbon monoxide effects on the developing embryo, fetus and

newborn infant; carbon monoxide exchange between the mother and the fetus

under both steady state and non-steady state conditions*and the mechanisms by

which it interferes with oxygenation of the fetus.

The Interrelations of Carboxyhemoglobin Concentrations  in the jlother and Fetus

     Maternal Carboxyhemoglobin Levels.  The carboxyhemoglobin concentration

                                                                       19 247
in the blood of normal nonsmoking pregnant  women varies from 0.5 to 1%.   '

In addition to those factors affecting carboxyhemoglobin in the nonpregnant

person f  maternal carboxyhemoglobin concentration,  [HbCOm], reflects the

endogenous carbon monoxide production by the fetus  and  its  rate of  exchange

across the placenta.^49  Fetal endogenous carbon monoxide production accounts

for about 3% of the total carboxyhemoglobin present in  the  blood of a normal

pregnant woman.

     Fetal Carboxyhemoglobin.   Under steady state conditions, the concentration

of human fetal carboxyhemoglobin, [HbCOf],  is greater than  that in maternal

blood (g ee Figure 5-9).   The wide disparity in the  reported values  of both

                                  40     ,    .60
                    0.01    0.02     0.03    0.04    0.05     0.06    0.07
FIGURE  5-9.  The relation of human maternal and fetal carboxyhemoglobin

             concentrations under steady state conditions as a function

             of both carbon monoxide partial pressure (mm Hg) an(i inspired

             air  concentrations in parts per million.  (Reprinted with

              permission  from Hill jit al.   )

human fetal carboxyhemoglobin concentrations and the ratio of fetal to maternal

carboxyhemoglobin concentrations, [HbCOf ] /[HbCOm],247 probably results from a

number of factors.  These include collecting the samples under non-steady

state conditions and using different methods to analyze for carbon monoxide.

While the fetal carboxyhemoglobin concentration varies as a function of the

concentration in the maternal blood, it also depends upon the rate of fetal

carbon monoxide production, placental carbon monoxide diffusing capacity, the

relative affinity of both fetal and maternal hemoglobin for carbon monoxide

as compared to its affinity for oxygen, and the relative affinity of blood

for these two gases.

     The relation of the fetal to maternal carboxyhemoglobin concentrations

during the steady state depends on several factors.   Assuming that the carbon

monoxide partial pressure in maternal blood, pCOm,  equals the carbon monoxide

partial pressure in fetal blood, pCOf , the Haldane  equation pCO = ([HbCO] x

        x M)  may be equated for maternal and fetal  blood.   The result when

rearranged becomes:

                        [HbCO,:]   [Hb07 ]    pOo     Mf
                                      Zf      ^
                        [HbCOm]    p02f     [HbO^]   Mm

where the ratios of oxyhemoglobin concentration to  the oxygen partial pressure

for both fetal and maternal blood equal  oxygen  affinities of  fetal and maternal

blood determined from the oxyhemoglobin  saturation  curves at  the mean oxygen

tension of blood in the placental exchange vessels;  and M£  and ^ are the relative

affinities of fetal and maternal blood. respectively for carbon monoxide  as com-

pared to oxygen.  Thus, the ratio  of the concentrations of fetal to maternal

carboxyhemoglobin, [HbCOf ]/[HbCOm] , during the  steady state depends  on both the

relative affinities of fetal and maternal hemoglobin for oxygen  and  the  ratio of

the relative affinity of fetal and maternal blood for carbon monoxide and

     Time Course of Changes in Fetal and Maternal Carboxyhemoelobin.  The

relation of carboxyhemoglobin concentrations to inspired carbon monoxide

concentrations in adult humans  has been experimentally determined in several

studies.  Recently, Longo  and Hill248  studied these relations in pregnant sheep

with catheters chronically implanted in both maternal and fetal blood vessels.

They exposed  ewes to inspired carbon monoxide concentrations of up to 300 ppm.

At 30 ppm, maternal carboxyhemoglobin  concentrations increased over a period

of 8 to 10 hr equilibrating at about   4.6 + 0.3 (SEM)%.   The fetal carboxy-

hemoglobin concentration increased more slowly, equilibrating in 36 to 48 hr

at about 7.4 + 0.5 %    (Fig. 5-10).  At 50 ppm, the time courses were similar

with maternal and fetal steady values  of 7.2+ 0.5 %   and 11. 3 + 1.1 %  re-

spectively.  At  100 ppm, maternal and  fetal steady state values were 12.2

 +1.2 %and 19.8;+ 1.4 %,  respectively.  For all three concentrations, the

half-times for carbon monoxide uptake  by maternal and fetal blood were about

2.5 and 7.5 hr respectively.

     For ethical and technical reasons, such experiments cannot be carried

out with humans.  This  same group    examined the experimental data and

theorized a mathematical   model of the interrelations of human fetal and

maternal carboxyhemoglobin concentrations.  The predicted changes in the

carboxyhemoglobin concentrations as a  function of the time of exposure to

inspired carbon  monoxide concentrations ranging from 30 to 300 ppm are

shown in Figure  5-11.

                    0   4   8   12   16  20 24 0   4   8   12 18 24
                                              0   4   8   12   16  20 24  0  4   8   12 W 24
                                                                          300 ppm for 3 hrt.




                      100 ppm
0   4    8   12   16   20 24 0

              TIME (hours)
                                                  8   12 18 24
                                                    4    8   12   U  20 24

                                                       TIME (hour*)
               FIGURE 5-10.  Time course of carbon monoxide uptake in maternal and fetal sheep exposed to
                             varying carbon monoxide concentrations.  The  experimental results for the ewe
                             (•)  and fetal lamb (o) are the mean values  (± SEM) of 9 to 11 studies at each
                             inspired carbon monoxide concentration, except in the case of 300 ppm.  Only
                             3 studies were performed at that concentration.   The theoretical predictions
                             of the changes in maternal and fetal carboxyhemoglobin concentrations for the
                             ewe and lamb are shown by the solid and interrupted lines, respectively.
                              (Reprinted with permission  from Longo and Hill *  )

4     8    12
20   24 «oO    4
    TIME (Hours)
                                                            8    12    16   20   24
FIGURE 5-11. The predicted time course of human maternal and fetal earboxy-

             hemoglobin concentrations during prolonged exposures to  30,

             50, 100, 200 and 300 ppm inspired carbon monoxide concentra-

             tions, followed by a washout period when no carbon monoxide

             is inspired.  Note the  fetal carboxyhemoglobin concentrations

             lag behind that of the  mother, but eventually reach higher

             values in most cases.   (Reprinted with permission from Hill
                  , 170v
              et al.   )

     Several points of interest are:  The equilibration for fetal carboxy-

hemoglobin was not achieved for about 30 to 36 hr  ; the half-time of the

increase in fetal carboxyhemoglobin concentration was about 7.5 hr  for all

concentrations; the time for fetal carboxyhemoglobin to equal the maternal

value varied from 12 to 15 hr ; and finally, under steady state conditions,

fetal carboxyhemoglobin concentration was about 15% greater than the maternal.

The theoretical relations in humans and the experimental data in sheep are

therefore in reasonably good agreement.

     Theoretical Prediction of Fetal._Carboxyh_emogfobin Concentration During

Intermittent Maternal Carbon Monoxide Exposure.  Mothers who smoke cigarettes

or are exposed to excessive amounts of carbon monoxide in the air are sub-

jected to fluctuating concentrations of inspired carbon monoxide.   The

previously described mathematical model^™ was used to simulate such condi-

tions and      to predict the changes in fetal and maternal carboxyhemoglobin

concentrations during exposure to various carbon monoxide concentrations for

different durations.  The time course of carboxyhemoglobin concentrations

anticipated if the mother breathed 50 ppm carbon monoxide for  a 16 hr  period

or smoked one and a half packs of cigarettes a day is  shown in Figure 5-12.

The peak fetal carboxyhemoglobin concentrations were greater than the maternal

concentrations and the mean carboxyhemoglobin concentrations were 6% and

5.4% respectively.  Similar patterns would be produced by exposure of the

mother (and fetus) to other carbon monoxide concentrations with the values

reflecting the concentrations.  While peak fetal carboxyhemoglobin concen-

trations were only about 10% greater than maternal, the mean values were

about 20% greater.  (This difference ranged from 25% greater at 5 ppm to 15%

greater at 50 ppm).   The implications for the fetus of this greater carbon

monoxide exposure are unknown.


 *""•   c
 O   5
18       24       30

FIGUKF- 5-12.  The predicted maternal and fetal carboxyhemoglobin concentrations

             when a mother breathes 50 ppm carbon monoxide for a 16-hour

             period followed by 8 hours during which no carbon monoxide

             is breathed.   This level of carbon monoxide exposure is

             equivalent  to smoking about 1 -1% packs of cigarettes per

             day, followed by an 8-hour sleep period.  (Reprinted with permission

             from Hill et al.17°)

     This mathematical model also has been used to predict fetal and maternal

carboxyhemoglobin changes following more complicated exposure patterns.  Data

on carbon monoxide concentrations obtained from the Los Angeles Air Pollution

Control District were measured at numerous sites in that city.  Figure 5-13

shows the data for a typical site in southern Los Angeles (site number 76 in

Torrance) during January 22, 23,  and 24, 1974, a Tuesday, Wednesday, and

Thursday.  The inspired carbon monoxide concentration fluctuated between 0

and 48 ppm (upper curve).  The values plotted represent hourly averages.

The lower curve shows the calculated maternal and fetal carboxyhemoglobin

concentrations, assuming a pregnant women breathed this air,  with no additional

source of carbon monoxide^ such as cigarette smoke. ™  Following the peaks of

inspired carbon monoxide, the fetal carboxyhemoglobin concentration averaged

3%, while the maternal concentration averaged 2.6%.

     These relatively low carboxyhemoglobin concentrations may seem too

small to be of much significance.   However,  several  investigators (reviewed

in other sections of this report).    have demonstrated significant reductions

in a number of physiologic functions with blood carboxyhemoglobin levels  in

the range of 4 to 5%.  For the case of a pregnant mother who  smoked 1 to  2

packs of cigarettes per day, exposure to these elevated ambient carbon monoxide

concentrations would be nearly additive.   Thus, it can be calculated that the

fetal carboxyhemoglobin concentration would be 6 to  7% in the pregnant mother

who smoked one pack of cigarettes per day and was exposed to  this level of air

pollution.         Similarly,  if  the exposed subject smoked two packs of

cigarettes per day, the fetal carboxyhemoglobin concentration may reach 10 to


     Maternal Smoking and Other Carbon Monoxide Exposure.  Perhaps the most

common source of fetal exposure to greater than normal carbon monoxide concen-

trations is maternal smoking.  Several studies have reported the carboxyhemo-

globin concentrations in the blood of mothers that smoke and their newborn

(Table 5-1).  Carboxyhemoglobin concentrations of thefetusesranged from 2 to

10%, and those of  the mothers ranged from 2 to 14%.

     The blood samples were obtained at the time of vaginal delivery or

cesarean section and probably did not accurately reflect the normal values

of carboxyhemoglobin for several reasons:  the number of cigarettes smoked

during labor may have been less than the number normally consumed; blood

                                      , \M*

samples were collected at varying time intervals following the cessation of

smoking; and many  samples were probably taken in the morning before the

carboxyhemoglobin  concentrations had built up to the values reached after

prolonged period of smoking.  Therefore, the concentrations measured in both

maternal and fetal blood may have been lower than average values for normal

smoking periods.

     The relation  between maternal smoking and low birthweight recently has

  been reviewed.  »  >   a»  5  The reports relating perinatal mortality to

-maternal smoking  habits disagree.  Several report an increased Incidence

  of  spontaneous abortion and of fetal neonatal and post-neonatal deaths

  associated with maternal smokin8.53'54'96'130'275»276»330»348,349

        314 404 454
  Others-   '    '    'reported  finding little correlation of these prob-

lems with smoking, but  this  conclusion probably was based on inadequate sample


     All recent studies using data from large population groups have concluded

that perinatal mortality is increased in the infants of mothers who smoke.

— 50


          JANUARY 22
                   JANUARY 23
                     TIME (DAYS)
FIGURE 5-13.
Measured carbon monoxide concentrations in inspired air
(upper curve) and calculated maternal and fetal carboxy-
hemoglobin concentrations (lower curves) during a 3-day
period in southern Los Angeles.  Note that fetal carboxy-
hemoglobin concentrations rise slightly higher than maternal
concentrations following each peak in carbon monoxide ex-
posure  and that the fetal carboxyhemoglobin concentrations
take longer to decline afteij^he peaks.  (Reprinted with
permission  from  Hill et  al.    )

                                        TABLE 5-1

        The Relation of the Concentrations of Fetal to Maternal Carboxyhemoglobin

                          in Mothers Who Smoke During Pregnancy


Concentration %

7.6 (SEM + 1.14)*
3.1 (+ 0.84)**

5.0 (+ 0.48)

2.4 (+ 0.30)

5.3 (+ 0.22)


3.6 (+ 0.7)

7.5 +


Concentration %

  6.2 (+ 0.75)*
  3.6 (+ 1.06)**

  6.7 (+ 0.61)

  2.0 (+ 0.31)

  5.7 (+ 0.24)


  6.3 (+1.7)




  1.2 (+ 0.2)*
  0.9 (+ 0.14)

  0.7 (+ 0.04)

  1.2 (+ 0.08)

  0.9 (+ 0.06)


  0.6 (+ 0.15)

Haddon et al.
Young and Pugh


Younoszai and Haworth^
Cole et al.
 * one or more cigarettes 1 h or less prior to delivery

** one or more cigarettes 1 to 24 h prior to delivery

 + calculated from  [HbCO  ] and the ratio of [HbCO,;] to [HbCO  ]

This increased perinatal mortality is independent of, rather than due to,

birthweight reduction5.3 ,55,96a,276,297a,302a,274a  while the mean duration

of pregnancy of smoking mothers is slightly shorter than normal, 30,250a,5Ua
the proportion of preterm births increased significantly. 53»^5^a  Goldstein

has pointed out that the lower perinatal mortality among infants weighing

less than 2500 g of women that smoke  probably reflects the increased mean

birthweight of the smokers' babies               compared with those of the

nonsmokers.  Meyer and her colleagues conclude from the Ontario, Canada data

that the independent effect of maternal smoking increases the perinatal

mortality risk 20% for light smokers (less than 1 pack per day) and 35% for

heavy smokers (1 pack or more per day) .

     Several causes have been suggested for the low birth weights and in-

creased perinatal mortality:  decreased food intake associated with smoking""

decreased placental blood flow due to the action of pharmacologic agents in

the smoke, and the effects of carbon monoxide on tissue oxygenation.  While

carbon monoxide probably adversely affects fetal growth and development,

other factors such as various chemicals in tobacco smoke and the psychologic

make-up of the mother make it difficult to assess the specific effects of

carbon monoxide per se.

     Several reports have analyzed the incidence of complications of pregnancy

and labor in smoking mothers.  There are increases in the incidence of

abruptio placenta with resulting stillbirth, !47a,274a placenta previa and

other causes of bleeding during pregnancy. 215a,302b,274a  The incidence of pre-

mature rupture of the fetal membranes is also increased^0  while that of  the

hypertensive disorders of pregnancy is decreased. oa,53,215a,348,404

     Recently, it has been observed that "breathing" movements by the fetus

are a normal component of intrauterine development.  Both the proportion of

time the fetus makes breathing movements and the character of these movements

indicate the condition of the fetus.  In women with normal pregnancies,

cigarette smoking caused an abrupt and significant decrease from a control

value of 65% to 50% in the proportion of time that the fetus made breathing
movements.           Carbon monoxide may not play an important role causing

these acute changes, however, since marked decreases in breathing were not
observed in the fetuses of women who smoked non-nicotine cigarettes.
Heron    reported a delayed onset of crying immediately after birth in the

infants of smoking mothers.  Several infants showed definite evidence of

asphyxia with irregular respiration and cyanosis.

     Long-term effects in surviving children owing to maternal smoking are

not well documented.  In a multifactorial analysis of data from over 5,000 childrei

in the British National Child Development Study, Davie et al_.     and Goldstein

found highly significant differences in reading attainment at 7 years between the

children of mothers who smoked and those who did not.  Butler and Goldstein

restudied these children at 11 years of age.  Significant differences in the

offspring of mothers who smoked 10 or more cigarettes a day   —>

included:  3,4,and 5 months of retardation in general ability,reading, and

mathematics, respectively, and a mean height 1 cm less.Tlieir data suggest a

decrement in  intelligence quotient, but the difference was not statistically
significant.  Finally, Denson £t al.     reported a syndrome of minimal

brain dysfunction in the infants of mothers who smoke.  These findings re-

quire confirmation as their implications are great.

     As noted in other portions of this report, there is a high correlation

between smoking and the development of coronary and peripheral artery disease.

Several studies indicate that carbon monoxide in tobacco smoke injures

arterial blood vessels.*•*  Exposure to low doses of carbon monoxide accelerated

athrogenesis in cholesterol-fed animals-^.46a,423 pro(jucing significant ultra-

structural changes in the aortic and coronary epithelium of rabbits     and

primates399,423 indistinguishable from early artherosclerosis.  Asmussenand

KjeldsenlSa used the human umbilical artery as a model to evaluate vascular

damage caused by tobacco smoking.  In comparison with the vessels from babies

of non-smoking mothers, the umbilical arteries from babies of smoking mothers

showed pronounced vascular intimal changes.  Scanning electromicroscopy dis-

closed swollen and irregular endothelial cells with a peculiar cobblestone

appearance and cytoplasmic protrusions or blebs on their surface.   Trans-

mission electromicroscopy showed degenerative changes including endothelial

swelling, dilation of the rough endoplasmic reticulum,  abnormal appearing

lysosomes, and extensive subendothelial edema.  In addition,  the basement

membrane was markedly thickened, a change probably indicating reparative

change.  Finally, the vessels showed focal openings of intercellular junctions

and loss of collagen fibers.   This study underscores the vulnerability of  the

fetus  to the effects of smoking by the mother.

     Animal studies showed  fetal growth retardation and increased perinatal

mortality in pregnant rats120a»^56a and rabbits351a exposed to tobacco  smoke.

Schoeneck^Sla exposed rabbits to tobacco smoke for several generations.   The

original doe weighed 3.5 kg.   One female of the first generation weighed

2.8 kg.j  another  one from the second generation weighed only  1.53  kg  and  all

attempts to breed the doe were either totally unsuccessful or resulted in

stillbirths or neonatal deaths.

     Of course, factors other than carbon monoxide in tobacco smoke may also

                                                      45 6a
result in fetal growth retardation.  Younaszai, et al.     exposed rats to

several types of smoke, including:  the smoke of tobacco leaves,

smoke from lettuce leaves plus nicotine, and smoke from lettuce leaves alone.

The body weight of rat fetuses exposed to lettuce leaf smoke decreased 9%.

Body weight of the fetuses  exposed to lettuce leaf smoke plus nicotine de-

creased about 12%, while the decrease in those animals exposed to tobacco

smoke was about 17%.  The carboxyhemoglobin concentration was maintained at

from 2 to 8% in all animals, but  the data were not given.

Biologic Effects of Carbon  Monoxide on the Developing Embryo, Fetus, and Newborn

     Experimental Studies of Mammalian Fetal Growth and Survival.  Few studies

have reported the effects of carbon monoxide on fetal growth.

Wells^3^ exposed pregnant rats to 1.5%  (15,000 ppm carbon monoxide) for from

5  to 8 min   10 times on alternate days during their 21 day pregnancy.  This

resulted in maternal unconsciousness and abortion or absorption of most fetuses.

The surviving newborns did  not grow normally.  Similar exposure to 5,900 ppm

affected only a small percentage  of animals.  This is a brief report lacking

quantitative data on the number of experimental animals and number and weight

of the fetuses.  Williams and Smith^3 exposed rats to 0.34%  (3,400 ppm)

carbon monoxide for 1 hr  daily for 3 months.  Peak carboxyhemoglobin concen-

trations in these animals varied  from 60 to 70%. The number of pregancies known to

 occur amoung the 7 exposed animals were half the number in the controls. The

 number of rats born per litter decreased and only 2 out of 13 newborns survived

 to weaning age.  No pregnancies  resulted in the 5 females exposed for 150 days-

     Aatrup,  et al.    reported quantitative data on fetal weights of 2 groups of

pregnant rabbits exposed to carbon monoxide continuously   —=>

for 30 days.   Exposure to 90 ppm resulted in maternal carboxyhemoglobin con-

centrations of 9 to 10%.  Birthweights decreased 11% from 57.7 to 51.0 g and

neonatal mortality increased to 10%, from a control value of 4.5%.  Mortality

of the young rabbits during the following 21 days increased to 25% from a

control value of 13%.  Following exposure to 180 ppm carbon monoxide, with

resulting maternal carboxyhemoglobin concentrations of 16 to 18%, birthweights

decreased 20% from 53.7 to 44.7 g, and neonatal mortality was 35% compared

with 1% for the controls.   Mortality during the following 21 days was 27%,

the same value as the controls.

     Carbon Monoxide Effects on Avian Embryogenesis. Baker and Tumasonis^^ con-

tinuously exposed fertilized chicken eggs to various carbon monoxide concen-

trations from the time they were laid for up to 18  days of incubation.  Hatch-

ability correlated  inversely with carbon  monoxide concentration.   At  425 ppm,

the apparent  "critical level", only about 75% of the eggs  hatched.  The embryos

of these eggs weighed almost the same as  those of the controls  and no congenital

anomalies were noted.  Since 0 and 100 ppm were the only lower  carbon monoxide

concentrations tested, the conclusion that 425 ppm  represents a "critical  level"

may not be justified.  In eggs exposed to 425 ppm,  the carboxyhemoglobin con-

centrations varied  from 4 to 16%.  The lowest values were  reported on the  15th

to 16th day of incubation.   At 650 ppm carbon monoxide, the percent of eggs

hatching decreased  to 46%,  and developmental anomalies of  the tibia and meta-

tarsal bones  were noted.   Subsequently, Baker  et jil.    exposed embryonated

eggs 12 and 18 days  old to 425 ppm for 24 hr.   The  carboxyhemoglobin concen-


trations averaged 7% in 12 to  14 day chick embryos, 16% in the  16-day embryos,

and 36% in the 18-day embryos.  It is not clear why the carboxyhemoglobin

increased so markedly with embryonic age.  The activities of two hepatic

mixed-function oxidase enzymes, hydroxylase and 0-demethylase increased

about 50% in the livers of the older chicks.  Chick eggs 13 days old did

not show significant enzyme increases.  The authors interpreted these results

as indicating that the increased hepatic mixed-function oxidase enzymes repre-

sent an adaptation to carbon monoxide induced tissue hypoxia in the more mature


     Exposure to Excessively High Carbon Monoxide Concentrations During the

Newborn Period.  Behrman  et al.37 reported that in 16 normal human newborns

in a nursery in downtown Chicago, the carboxyhemoglobin concentration in-

creased from about 1% to 6.98± 0.55%       and the blood oxygen capacity

decreased 13.8i 0.57 %   (control values were not given).  The authors

correlated the decreases of blood oxygen capacity with carbon monoxide con-

centrations at a Chicago air pollution station about 1.5 miles distant.

On days when the ambient carbon monoxide concentration was ^ess than 20 ppm,

blood oxygen capacity was decreased 8.4+ 1*1 %  in infants aged 24 hr  old

or younger and  11.0+0.73  %  in infants over 24 hr  old.  On days when the

atmospheric carbon monoxide was greater than 20 ppm, blood oxygen capacity

 decreased 11.4+ 1.1%    in the newborns and 13.0± 0.65 %   in babies over

24 hr  old.  The carboxyhemoglobin concentrations varied as a function

of both the concentrations of inspired carbon monoxide and the duration of

exposure.  While these observations are of interest, certain problems inherent

the study were not clarified.  A close correlation between the carbon monoxide

concentrations in the nursery and at the monitoring station 1.5 miles distant

is questionable.


The authors noted that the decrease in oxygen capacity was greater  than  could
be accounted for by carbon monoxide alone, although the error in the method
of carboxyhemoglobin determination with the spectrophotometer used  probably
invalidates such a conclusion.  No untoward clinical effects were observed
either from these carboxyhemoglobin concentrations or from the decrease  in
blood oxygen capacity.

     Apparently the only study in newborn animals is that of Smith  et al.
They exposed rats to mixtures of illuminating gas in air, such that inspired
carbon monoxide concentrations equaled 0.43%.  In 22 newborn rats 12 to 48 br
old exposed to carbon monoxide, the average survival time was about 196 min
in contrast to an average survival of about 36 min in mature animals.   McGrath
and Jaeger26' also noted that 507o of newly hatched chicks oould withstand exposure
to 1% (10,000 ppm) carbon monoxide concentration for about 32 min.   This initial
resistance to carbon monoxide decreased rapidly.  By day one, mean survival time
decreased to about 10 min, by day 4 it was 6 min, and by day 8 it was  4 min,
where it remained for all ages tested up to 21 days.   Subsequently, Jaeger and
McGrath18^ showed that decreasing the body temperature increased the time to
last gasp from a mean value of 9.8 + 0.5 min at 40 C to 20.7 + 0.1 min at 30 C.
They noted that hypothermia caused markedly reduced heart and respiratory rates
and suggested that its major benefit was a reduction in energy requiring functions.

     Possible Mechanisms by Which Carbon Monoxide Affects the Fetus and
Newborn.  Several mechanisms probably account for the effects of carbon monoxide
on developing tissue.  Undoubtedly the most important of these is the  inter-
ference with tissue oxygenation.27'* ^  As first observed by Claude Bernard
in 1857,^-> carbon monoxide decreases the capacity of blood to transport oxygen
by competing with it for hemoglobin.  When carbon monoxide binds to hemoglobin,
the oxygen affinity of the remaining hemoglobin is increased.  This shift to
the left of the oxyhemoglobin saturation curve means that the oxygen tension

of blood must decrease to lower than normal values before a given amount of
oxygen will be released from hemoglobin.  This effect may be particularly
significant for the fetus because the oxygen tension in the arterial blood
is normally relatively low, about 20-30mmHg, as compared to adult values of
about lOOmmHg.  Carbon monoxide also interferes with oxygen transport by dis-
placing oxygen from the hemoglobin in arterial blood, thus decreasing the blood
oxygen capacity.  For the pregnant woman,  these  effects on blood  oxygenation  pose
a special threat.  Not only is her oxygen consumption increased 15-25% during
          ^ 1 OQ
pregnancy,     but her blood oxygen capacity is decreased 20 to 30% or more due
to the decreased concentration of hemoglobin.  The woman   with a significant
anemia faces  even more severe compromise of her oxygen delivery.
      The  theoretical  basis  for understanding the consequences of carbon
monoxide  interaction  with oxygen in humans  is shown in Figure  5~14, in which
blood oxygen content  is  plotted  as a  function of the oxygen partial pressure.
The  oxygen affinity of fetal blood is greater than that of maternal blood,
hence its oxyhemoglobin  saturation curve  is shifted to the left.  In addition,
human fetal blood contains  more  hemoglobin  than maternal  (16.3 vs. 12 g/100 ml),
it therefore has a greater  oxygen capacity.  Under normal circumstances, ma-
ternal arterial oxygen tension is about 16.1 ml/100 ml of blood  (point A.-, in
Fig.  5-14).  The placental  exchange of oxygen extracts about 5 ml 02/100 ml
blood, producing a uterine  mixed venous oxygen tension of about  34 mmHg  (point
V^).  When the maternal blood contains carboxyhemoglobin, the oxygen capacity
is decreased and the  oxyhemoglobin curve  shifts to the left  (as  indicated by
the  curve labelled 9.4%  [HbCOm]. While  arterial  oxygen  tension  remains
essentially the same  as  under normal  conditions, the oxygen content is reduced
to 14.5 ml/100 ml (point A^) •   With  the  same placental oxygen transfer of
 5 ml/100 ml blood,  the venous  oxygen  tension would be about 27mm Hg (point V  )>
 a decrease  from normal of  about  7  mm Hg.

                Figure 5-14.   Oxyhemoglobin saturation curves  (plotted  as  blood  oxygen content vs.
                partial  pressure)  of human fetal  blood with  0% and  10%  carboxyhemoglobin and of
                maternal blood with  0 and  9.47» carboxyhemoglobin concentrations.   This figure depicts
                the  mechanism accounting for the  reduction of  umbilical artery and vein oxygen partial
                pressures and contents resulting  from elevated earboxyhemoglobin concentrations.
                (See text for details.)    (Reprinted with  permission  from  Longo246a)

     In the fetus, oxygen partial pressure in the descending aorta is normally

about 20mm Hg and oxygen content is about 12 ml/100 ml fc oint Apl) >   With an

oxygen consumption of 5 ml/100 ml, inferior vena caval oxygen tension would be

16 mm. Hg (pointv  ) •  With elev*ted fetal carboxyhemoglobin concentrations,

both arterial  (.point Ap2 ) and venous  (point  V J» oxygen tensions are reduced.

This contrasts  to the adult  in which the arterial oxygen tension remains normal.

This is because of the  lowered oxygen tension of maternal placental capillaries

with which fetal blood  equilibrates.  With normal fetal oxygen consumption,

the venous blood oxygen content is 5  ml/100 ml less  than  arterial content  pro-

ducing a venous oxygen  tension of about llmm Hg, a decrease from normal of

5  mm Hg.

     The oxygen tension of venous blood is roughly indicative of the adequacy

of tissue  oxygenation and the mean capillary partial pressure driving oxygen

into the tissues is  probably related to the oxygen partial pressure at 50%

oxyhemoglobin saturation, the  p50.  Figure 5-15 shows the changes in P50 for

maternal and  fetal blood as  a  function of the blood carboxyhemoglobin concen-

trations .

     While the effects  of carbon monoxide on venous oxygen tensions have been

considered from a theoretical  standpoint, there has been essentially no experi-

mental validation of these effects in either adults or the fetus.  Longo and

Hill248 recently examined the  changes in oxygen tension in response to various

carboxyhemoglobin concentrations in sheep in which catheters were chronically im-

planted in maternal  and fetal vessels.  About a week following their recovery

from anesthesia and  surgery  the ewes were exposed to various concentrations of

carbon monoxide.

Figure 5-15»  The partial pressure at which  the oxyhemoglobin satura-
tion is 50%, p50, for human maternal and  fetal blood as a function  of
blood carboxyhemoglobin concentration.  (Reprinted with permission
 from Longo2^73)

     Figure 5-16 shows the oxygen tensions in the descending aorta and the

inferior vena cavae below the ductus venosus as a function of carboxyhemo-

globin concentration in the fetus.  Maternal and fetal carboxyhemoglobin

levels were in quasi-steady state equilibrium.  In contrast to the adult in

which arterial oxygen tension is relatively unaffected by changes in carboxy-

hemoglobin concentrations, fetal arterial oxygen tension is particularly

sensitive to increases in maternal or fetal carboxyhemoglobin concentrations.

This is because fetal arterial oxygen tension varies with the oxygen

tension in fetal placental end-capillary blood, which in turn varies with

the  oxygen tensions  in maternal placental exchange vessels.  The oxygen

partial pressure in  the fetal descending aorta decreased from a control value

of about 20 mmHg to  15.5 mmHg at 10% fetal carboxyhemoglobin concentration

(Figure 5-16).  The  regression equation for this relation was:  p02 = 20.1 - 0.4

[HbCOf], (R = -0.96).

     Figure 5^-16  also shows the relation of oxygen tension of the inferior

vena cava, below the ductus venosus. to carboxyhemoglobin concentration of

the fetus.  At 10% carboxyhemoglobin concentration, inferior vena caval oxygen

tension decreased from a control value of about 16 to 12.5 mmHg.  The regres-

sion equation for this relation was:  p02 = 15.8 - 0.3  [HbCO-J.  (R = -0.96).

Strictly speaking, the oxygen tension of fetal venous blood is affected both

by the decreased maternal placental venous oxygen tension resulting from in-

creased maternal carboxyhemoglobin concentration and the increased fetal

 carboxyhemoglobin concentration.   Since  these  oxygen  tensions were obtained

 when maternal and fetal carboxyhemoglobin  concentrations were in a quasi-steady

 state condition,  a relation between fetal  inferior vena caval oxygen

                                  Fetal Descending Aorta
                0    2    4     6     8    10    12    14    16    18
FIGURE 5-16.
Fetal values of oxygen partial pressure as a function  of
carboxyhemoglobin concentrations during quasi-steady state
conditions.  Fetal inferior vena cava oxygen tension is a
function of both maternal and fetal carboxyhemoglobin  con-
centrations.  The oxygen partial pressure of fetal arterial
blood is chiefly a function of maternal carboxyhemoglobin
concentrations.  During steady state conditions, however,
it will also be related to the fetal carboxyhemoglobin con-
centration.  Each point represents the mean + SEM (vertical
bars) of 6-20 determinations at each concentration of  blood
          loglobin.  (Reprinted with permission from Longo
              et al.

tension and maternal carboxyhemoglobin concentration  would be expected.  If
one plots this relation,2488 the decrements in fetal oxygen tension appear
greater when plotted as a function of maternal carboxyhemoglobin than when
plotted as a function of fetal carboxyhemoglobin concentration.  This follows
because the carboxyhemoglobin concentration of the fetus exceeds that of the
mother during  equilibrium conditions.
     These oxygen partial pressures in sheep are not identical with those values
anticipated in humans, because of differences in oxygen affinities and capacities
of maternal and  fetal blood  between the species.  Differences however^are
estimated to be  no more than 2-3 mmHg.
     About 577. of the sheep fetuses in this study died when fetal carboxyhemoglobin
values were greater than 15% for 30 minutes or longer (5 of 11 died at 100 ppm,
and 3 of 3 died  at 300 pprn).2^3  These deaths presumably resulted from hypoxia
of vital tissues.  There are probably two major reasons for this.  Firstly, in
the adult, elevation of carboxyhemoglobin concentrations to 15-20% results in a
6-10 mmHg decrease in venous oxygen tension.  While this decrease is substantial,
the resultant  oxygen partial pressures probably remain well above critical values
for maintaining  oxygen delivery to the tissues.   a  In contrast, in the fetus
with normal arterial and venous oxygen tension probably close to the critical
levels, a 6-10 mmHg decrease in oxygen tension can result in tissue hypoxia or
anoxia.  Furthermore, adult subjects and animals subjected to carbon monoxide
hypoxia show increases in cardiac output,   coronary blood flow,     and presumably
of tissue blood  flow.  Apparently, such compensatory adjustments are not available
to the fetus to  any great extent.  The decreases in blood oxygen tensions measured
experimentally were similar to those predicted, assuming no increase in tissue
blood flow.  In  addition, the fetus probably cannot increase its cardiac output
significantly.as fetal cardiac output normally is about 2 to 3 times that of the
adult on a per weight basis,,      Thus, the fetus probably normally operates
near the peak  of its cardiac function curve.


     Ginsberg and Myers          studied  the  effects  of  carbon monoxide

exposure on near-term pregnant monkeys and  their  fetuses.   They acutely

exposed anesthetized animals to 0.1 to 0.3% carbon monoxide,  resulting in

maternal carboxyhemoglobin concentrations of  about 60%.  During the  1  to

3 hr studies, fetal blood oxygen  content decreased to less  than 2 ml/100

ml blood.  Fetal heart  rate decreased in proportion to the  blood oxygen

values.  These fetuses  also developed severe  acidosis (pH less  than  7.05),

hypercarbia (pCK^ = 70  mmHg or greater), hypotension and electrocardio-

graphic changes such as T-wave flattening and inversion.

     In cases of acute  carbon monoxide poisoning, the mother will develop a

high carboxyhemoglobin  concentration with a shift to the left of her oxyhemo-

globin saturation curve while the carboxyhemoglobin concentration in the

fetus remains normal or low.  Nonetheless, the fetus experiences a decrease


in pOn values, because  of the shift in the maternal dissociation curve.

                                                 O 1 O
     In a theoretical analysis, Permutt and Farhiji  calculated the effect

of elevated carboxyhemoglobin in lowering oxygen tension in the venous  blood

of an otherwise normal  adult.  Hill  et^ al.    used this approach to calculate

the effect of various carboxyhemoglobin concentrations on fetal tissue  oxygena-

tion.  The equivalent reduction in umbilical arterial oxygen tension required

to maintain normal placental oxygen exchange,  or the equivalent increase in

blood flow necessary for oxygen exchange was calculated.   The results are

plotted in Figure 5-17.   A 10% carboxyhemoglobin concentration would be

equivalent to a 27% reduction in umbilical arterial p(>2  (from 18 to 13 mm-Hg.

It would also be equivalent to a drastic reduction in blood flow.  Fetal

blood flow would have to increase 62% (from 350 to 570 ml/min) to maintain

normal oxygen exchange.   Higher levels of fetal carboxyhemoglobin require

even more dramatic compensations.

     A further carbon monorxide effect is its relation to oxygen consumption

by placental tissue.  Tanaka    used a Warburg apparatus to measure oxygen

consumption in placental slices from non-smoking and smoking mothers.  With


Figure  5-17. upper panel.  The umbilical arterial oxygen partial
pressure drop necessary to maintain normal oxygen exchange at
various carboxyhemoglobin concentrations allowing no  change in
maternal or  fetal blood flows or maternal arterial oxygen partial

Lower panel.  The increase in fetal blood flow necessary to main-
tain oxygen exchange  (with no change  in umbilical arterial oxygen
tension).  These curves indicate the  degree  of compensation neces-
sary to offset  the effects of elevations in  carboxyhemoglobin con-
centrations. (Reprinted with permission from Hill et  al.    )

tissues from normal non-smoking mothers, the oxygen consumption was about

1.9pl/mg placenta per hour.  With the tissue from smoking mothers, the rate

of oxygen consumption decreased in proportion to the concentration of carboxy-

hemoglobin in the maternal blood.  For example, it decreased about 30% to

1.3 jal/mg per hr at a maternal carboxyhemoglobin concentration of 8%.

     Hypoxia owing to carbon monoxide may further interfere with tissue

oxygenation by increasing the concentration of carboxymyoglobin, [MbCO], in

relation to carboxyhemoglobin.  As shown by Coburn and his coworkers, when

arterial oxygen tension decreases to 40 mm Hg or lower, the ratio of carboxy-

myoglobin to carboxyhemoglobin increases from a normal value of 1.0 to about

                                      84               85
2.5 in the resting dog skeletal muscle   and myocardium   of dogs.   If a

similar relation exists in the fetus, where arterial pO_ is normally 30 mm

Hg   -or less, then carboxymyoglobin concentration would be 12.5% when

carboxyhemoglobin is 5%.  The effect of this degree of carboxymyoglobin

saturation on oxygenation of the fetal myocardial cells may be significant.

     Similarities of High Altitude Hypoxia andCarbon Monoxide Hypoxia.  The

interference with tissue oxygenation caused by carbon monoxide is somewhat

similar to hypoxia at high altitudes.  Both can result  in lower oxygen

tensions  in end-capillary blood and  therefore tissue  hypoxia.   The

effect of smoking on newborn birthweight is strikingly similar to the effects

of high altitude.  A study by the Department of Health, Education,and Welfare

reported in 1954 that in the Rocky Mountain states  mean birthweights were

lower, there were a larger proportion of babies weighing less than 2500 g and

fewer weighing greater than 4000 g, and neonatal mortality was higher than in

                       O CO
the country as a whole. ~>0  A comparison of infants born in Lake County, Colorado,

elevation about 3000-3600 m (9,800-11,800 ft), with those born in Denver, about

1585 m (5,200 ft), showed that mean birthweight of the Lake County babies

                                                 f\ o n
was 290 g less than that of those born in Denver.     In Lake County, 48.3%

of newborns weighed less than 2500 g compared with 11.7% in Denver.

     A more precise analysis of the relation between altitude and birthweight

can be derived from the work of Grahn and Kratchman.1^8  These workers showed

that the proportion of low birthweights increased with increasing elevation.

The number of infants weighing less than 2500 g increased to 10.1% at 1500-1580

 (4,920-5; 180 ft) from a control value of 6.6% at sea level.  At 2,760 m  (9,055

ft), the number of such newborns further increased to 16.6%.  The decrease in

the duration of gestation, a mean difference of 0.4 week,was not enough to

explain the 190 g birthweight difference between Colorado and Illinois-Indiana

     A given carboxyhemoglobin concentration may cause a more profound effect

on tissue oxygenation at high altitudes than at sea level.  Thus, individuals

at high altitudes would be expected to be more sensitive to the effects of

carbon monoxide.  This would be particularly true of the pregnant mother, whose
oxygen requirement is greater than when not pregnant,     as well as for the

newborn infant.


     The heart requires a continuously available supply of oxygen in order

to maintain electrical and contractile integrity.  The glycolytic response

to hypoxia in the myocardium is much less than that observed in skeletal

muscle and is unable to prevent a rapid depletion of high energy phosphates.

Ectopic electrical activity, decline in contractile force and ventricular

fibrillation soon follow the induction of myocardial hypoxia.  As discussed

earlier carbon monoxide decreases the oxygen-carrying capacity of hemoglobin

and shifts the oxyhemoglobin dissociation curve to the left reducing the

oxygen tension in both the capillaries and cells (5 ee Figure 5-18).

     The concept that any alteration in intracellular oxygen tension is

rapidly corrected by an appropriate increase or decrease in coronary blood

flow is central to an understanding of the effects of carboxyhemoglobinemia

on myocardial oxygenation.   Peripheral tissues respond to increased oxygen

needs by increasing oxygen extraction and reducing venous oxygen content

and tension.   Increased myocardial oxygen extraction is almost never seen

in the normal heart and, when present, is considered evidence of myocardial


     The capacity for increased coronary blood flow in the normal heart is

demonstrated by increases of up to 300% in the coronary flow during  exercise,

in severe anemia    or in individuals exposed  to arterial hypoxemia.  •

The myocardium probably also adapts to hypoxia by increasing "functioning"

capillary density.     The  capacity of the normal heart to maintain  its

oxygenation at relatively high carboxyhemoglobin concentrations is analogous

to the situation in which a normal man climbs  to high altitudes or a patient

with chronic anemia sustains a 70% reduction in circulating hemoglobin.

      Coronary  Blood Flow,
           200i        Capacity
      MVg ,  ml/min/IOOg
      Capacity, ml/100 ml
      p50, mmHg
18      20
FIGURE 5-18.   Analogue model illustrating theoretical relationships between

               coronary blood flow and myocardial oxygen consumption (MV_ ),
               hemoglobin oxygen carrying capacity (Capacity),  and  the position

               of the oxyhemoglobin dissociation curve (p50).   Point A on the

               p50 and capacity curves represents coronary blood  flow at normal

               p50 and hemoglobin capacity; Point B illustrates theoretical

               effect of increasing carboxyhemoglobin from 0  to 20% saturation

               on coronary blood flow.  The assumptions used  in computing this

               theoretical model are given by Duvelleroy.     Adapted from

               Duvelleroy et al.

     The importance of coronary blood flow in the maintenance of myocardial

oxygen tension is emphasized in a recent theoretical model presented by

Duvelleroy et^ al.  '  Increasing myocardial oxygen consumption (by increasing

heart rate, arterial blood pressure, or contractile force), decreasing oxygen

capacity, or shifting the oxyhemoglobin dissociation curve to the left all

lead to a decrease in myocardial oxygen tension unless coronary blood flow

is increased commensurately.  Figure 5-18 shows that each alteration can

independently produce a change in coronary blood flow which restores myo-

cardial oxygen tension to normal.  Increasing carboxyhemoglobin concentra-

tions increase coronary blood flow both by decreasing the oxygen capacity

of hemoglobin and by shifting the oxyhemoglobin curve to the left, shown

in the figure as a decrease in the oxygen tension at 50% concentration, p50.

Points A and B show the effects of 0 and 20% carboxyhemoglobinemia,  re-

spectively.  Note that the relationships are curvilinear so that  progressively

greater alterations call forth disproportionately greater increases  in

coronary blood flow.   The implications of this type of response curve for

elevated baseline conditions    discussed elsewhere in this chapter.   The

cardiovascular response of the intact organism to carbon monoxide inhalation

depends on the ability of the entire coronary vascular bed to dilate and in-

crease coronary blood flow.  The random occurrence of coronary atherosclerosis

in the general population explains the variation in myocardial responses ob-

served both in humans and among different species.   It also provides a physiologic

explanation for the clinical observations described in the next section.

     The manifestations of carbon monoxide toxicity were shown to be closely

related to the carboxyhemoglobin concentration by John Haldane15' in 1895.

He did not observe serious symptoms at rest until his own hemoglobin was at

least one-third saturated with carbon monoxide.  Exertionfhowever, produced

mild dyspnea and palpitations when as little as 14% carboxyhemoglobin was

present.  Haggard158 in  1921 and Chiodi and coworkers72 in 1941 observed

increase in pulse rate and cardiac output with carboxyhemoglobin concentra-

tions between  16 and 20%.

     Ayres e£  ail.21 studied the systemic hemodynamic and respiratory re-

sponse to acute increases in carboxyhemoglobin in man, by means of measure-

ments performed during diagnostic cardiac catheterization.  Carboxyhemoglobin

concentrations between 6 and 12% were achieved by the breathing of either

5% carbon monoxide for 30-45 s or 0.1% carbon monoxide for 8-15 min.  In men

with no evidence of heart disease, cardiac output increased from 5.01 to 5.56

1/min, the minute ventilation increased from 6.86 to 8.64 1/min, and arterial

carbon dioxide pressure  pCC^, decreased from 40 to 38 mm Hg.  Systemic oxygen

extraction ratios increased from 0.27:jto 0.32:1, indicating more complete

extraction of  oxygen from perfusing arterial blood.  Mixed venous oxygen

tension decreased from 39 to 31 mm Hg as a result of the leftward shift of

the oxyhemoglobin dissociation curve and arterial pOo unexpectedly decreased

from an average of 81 to 76 mm Hg.  The decrease is probably due to enhance-

ment of the venoarterial shunt effect and is more prominent in patients who

initially have a low arterial pOo.

     Observations such as this suggest that carbon monoxide inhalation would

have a significant effect on arterial p02 in those patients with preexisting

lung disease,  in those individuals who are heavy smokers, and in patients in

coma owing to  severe carbon monoxide poisoning.  These studies were repeated

with the lower concentration of carbon monoxide (0.1% for 8-15 min).  Cardiac

output did not change significantly, although pC02 decreased, indicating  hyper-

ventilation.  Changes in arterial and mixed venous pO^ were similar  to  those

observed with the higher concentrations.

     Myocardial studies were performed before and after the administration

of either 5 or 0.1% carbon monoxide for 30-45 s or 8-15 min,respectively.

Patients were divided into two groups, those with coronary arterial disease

and those with other cardiopulmonary disorders.  In the patients with other

cardiopulmonary diseases, the lowest concentration of carbon monoxide decreased

the myocardial arteriovenous oxygen difference by an average of 6.6% and

in the patients with coronary arterial disease by 7.9%.  The higher concentra-

tion decreased the difference by 25% and 30.5%, respectively.   Coronary blood flow

increased in all but two of the studies, regardless of the dose delivered.

These changes were statistically significant.

     Neither the presence of coronary arterial disease nor the concentration

of carbon monoxide appeared to alter this reponse to increasing carboxyhemo-

globin.   Failure of the increase in coronary blood flow to compensate for the

decrease in oxygen delivery produced by carbon monoxide was suggested by a

decrease in coronary sinus oxygen tension in all but two of the 26 patients

studied.  Nine of 14 patients with coronary arterial disease had decreased

lactate extraction with increasing carboxyhemoglobin, a finding suggestive

of anaerobic metabolism.  In four of these patients,  lactate extraction ceased

and the myocardium produced lactate, indicating severe anaerobic metabolism.

The increase in their carboxyhemoglobin concentration averaged 5.05%.  These

studies were performed in patients with arteriographically demonstrated

coronary arterial disease.   The abnormal changes occurred while the patients

were, at rest.

     Adams et^ al.2 also found that low concentrations of carboxyhemoglobin

increased coronary blood flow.  These workers exposed conscious dogs to

carbon monoxide at increasing concentrations and observed a 13% increase

in coronary blood flow with a 4% increase in carboxyhemoglobin.  At 20%

carboxyhemoglobin, coronary blood flow had increased by 54%.  Coronary blood

flow measurements at 4 carboxyhemoglobin concentrations suggested a linear

relation between carboxyhemoglobin concentration and blood flow.  A threshold

value was not observed, although measurements were not made below 5% carboxy-


     Studies in experimental animals with presumably normal coronary vascular

beds are useful models for the effects of carbon monoxide on individuals with-

out coronary disease.  Since myocardial oxygen tension is maintained by in-

creasing coronary blood flow, in these studies dose-response relationships

cannot be directly used for establishing threshold levels for man.

     An interesting study of the effect of relatively low concentrations of

carbon monoxide in the cynomolgus monkey has been reported by DeBias et_ al.^97

They observed that chronically increased carboxyhemoglobin (to an average of

12.4%) produced polycythemia with an increase in hematocrit from 35 to 50%

of the volume of packed red blood cells.  All animals developed increased

amplitude of the P-wave in the electrocardiogram and some developed T-wave

inversions suggestive of myocardial ischemia.  Electrocardiographic abnormalities

were more severe in the monkeys with experimental myocardial infarction and

increased carboxyhemoglobin than in those with experimental infarction who

were breathing ambient air.

     In a later study, DeBias et al.^97 showed that increasing the carboxy-

hemoglobin concentration to an average of 8.5% in five cynomolgus monkeys

reduced the ventricular fibrillation threshold.  Fibrillation could be produced

by an average of 45 V AC delivered for 150 msec during the vulnerable period

compared to 79 V while breathing ambient air (p < 0.01).  This observation

in normal monkeys is particularly relevant to the problem of sudden death in

atheromatous man.

     The study of ultrastructural changes following exposure is another

approach to the evaluation of the myocardial toxicity of carbon monoxide.

Thomsen and Kjeldsen^OO observed myofibrillar degeneration and myelin body

formation in the mitochondria of rabbits exposed to carbon monoxide at 100 ppm

for 4 hours, an exposure sufficient to raise carboxyhemoglobin concentrations

to 8 or 9%.

     While the currently available physiologic data are insufficient by

themselves to formulate dose-response data,  they emphasize the multifarious

nature of the human response.   Sudden exertion in an individual with carboxy-

hemoglobinemia requires a substantial increase in coronary blood flow in

order to overcome the effects  of increased myocardial oxygen requirements,

decreased oxygen-carrying capacity of hemoglobin,  and rightward shift of the

oxyhemoglobin dissociation curve (Figure 5-18).   Should an appropriate in-

crease in blood flow be limited in any region of the myocardium by a rigid

vascular bed,  myocardial hypoxia may occur.   Such a multivariate formulation

emphasizes the difficulty of identifying a single threshold concentration

capable of protecting the entire population.

     Arteriosclerotic heart disease  (ASHD) is the leading cause of death in
the United States,208 with approximately 35% of all deaths directly attributable
to this disease.  It is also a major cause of morbidity.  Some of its clinical
and epidemiologic characteristics are particularly pertinent for establishing
a causal relationship between carbon monoxide exposure and ASHD.  The patho-
logic basis of the clinical disease  is a severe, diffuse narrowing of the
coronary arteries.337  Neither the frequency nor the prevalence of coronary
arterial stenosis in the population  is known.  There are, however, many people
with asymptomatic severe coronary arterial stenosis due to atherosclerosis.
They are at high risk with respect to developing clinical ASHD.
     The underlying atherosclerosis  is considered to be a chronic disease
but the clinical manifestations  are  usually acute.  Approximately 25-30%
of the individuals die suddenly, from several minutes to 24 hours  after their
first heart attack.     Certain  risk factors for clinical disease have been
identified.     They appear to act primarily in the development of the severe
underlying atherosclerotic disease and the consequent coronary stenosis.
Very little is known about the specific factors that precipitate the clinical
disease, such as myocardial infarction, angina pectoris and sudden death.
Agents that decrease the available oxygen supply to the myocardium, including
carbon monoxide, are primary  suspects as precipitating factors in heart attacks.
     Most persons who experience either angina or myocardial infarction have
severe coronary arterial stenosis.313  They are at high risk compared to the
rest of the population  for recurrent heart attacks and sudden death.^7  About
20% of the men who survive the first months after a myocardial infarction will

die within 5 years.->  Most of these deaths will be due to recurrent heart

attacks, and about 60% of the deaths will be sudden.  The specific environ-

mental and host factors that determine survival following a heart attack have

not been clearly determined.

     Environmental exposure to carbon monoxide and ASHD may be related in two

ways.  One of these is that exposure to carbon monoxide can enhance the develop-

ment of underlying atherosclerosis and subsequent coronary arterial stenosis

when associated with other risk factors such as increased cholesterol levels

and hypertension.  The other is that in the presence of severe underlying

coronary arterial stenosis due to atherosclerosis, carbon monoxide may be a

major precipitant of myocardial infarction, angina pectoris or sudden death.

Such a relation is very plausible in the light of what is known about the

epidemiology of heart disease and the possible pathophysiologic mechanisms.

     The relation between carbon monoxide and the precipitation of heart

attacks can be extended further to include new cases among individuals with

pre-existing but silent underlying atherosclerosis;      an increase in mor-

bidity}such as greater frequency and severity of chest pain,in patients with

angina pectoris:j or a reduction in survuval amoung individuals with clinical

atherosclerotic heart disease such^as decreased life expectancy following a

myocardial infarction.

The Relation Between Exposure to Carbon Monoxide,  Severity of Coronary
Atherosclerosis and Possible Precursor Vascular Lesions

     Studies of the association between the underlying coronary atherosclerosis

and exposure to carbon monoxide have been limited to laboratory investigations

with animal models (Table 5-2).17,206,399,421,423  in these studies the

experimental animals are exposed to various concentrations of carbon monoxide

either continuously or intermittently, while the controls breathe ambient air.

For some of the experiments both the experimental and the control animals are

fed a diet high in cholesterol and/or fat, and either the cholesterol content

of the artery or the extent of vascular disease is measured.  Several studies

in rabbits and primates have reported that animals exposed to relatively high

doses of carbon monoxide  (170-180 ppm) for extended periods of time have either

a higher cholesterol content in their arteries or enhanced vascular disease.

Animals in another type of study were exposed to large doses of carbon monoxide

but were not fed a high cholesterol or fat diet.  These studies in both pri-

mates and rabbits have reported finding subendothelial edema  and gaps between

the endothelial cells with increased infiltration of the cells with lipid

droplets.   (These lesions might be the early precursors of atherosclerotic


     There are no studies in humans describing the relation between exposure

to carbon monoxide and the rate of development of atherosclerotic disease.

Evidence that such a relation exists is based on the observation that

cigarette smokers, who have higher carboxyhemoglobin concentrations than

nonsmokers,  also have more advanced atherosclerosis than nonsmokers.^^

There is also evidence(however, that exposure to carbon monoxide may not be

causally related to the underlying atherosclerosis.  Heavy cigarette smokers

in countries,  such  as  Japan, where  the diets  are low in  fat  and cholesterol

do not  seem  to have a high risk of heart  attacks and probably do not have

severe  coronary atherosclerosis.

                                                       TABLE 5-2

                              The Relation  Between Atherosclerosis and Carbon Monoxide
Primates &
Primates :
Macaca Iris
CO at
Cholesterol &
170 ppm CO for
7 weeks
Atherogenic diet
plus 200-300 ppm
CO for 4 hr
5 days per week for
7 months
250 ppm CO
250 ppm every
12 hours
180 ppm CO for
2 weeks
Ambient air Increased focal degenerative and
reparative changes in intimal and
subintimal coats in CO-exposed
Cholesterol and Cholesterol content of aorta
ambient air 2 . 5 times higher than in control
Atherogenic diet Enhanced atherosclerosis in
only and com- monkey exposed to carbon
pressed air monoxide
Ambient Air Experimental group subendothelial
edema and infiltration of cells,
lipid droplets in coronary arteries
Air Exposed, local areas of partial
or total necrosis of myofibrils
and degenerative changes of
Wans tr up et al.
Astrup et al.
Webster e_t al.^23
Kjeldsen et al.

     The enhancement of atherosclerosis due to carbon monoxide exposure might

take place only in the presence of a high fat, high cholesterol diet.  Since

much of the United States population eats such a diet, exposure to carbon

monoxide could be an important determinant of the extent of coronary athero-

sclerosis.  However, because the data for such an association are preliminary

in nature, any current conclusions would be speculative.

The Relation Between Exposure to Carbon Monoxide and the Incidence of
Clinical Arteriosclerotic Heart Disease

     The relation between ambient carbon monoxide concentrations and the

incidence of clinical disease in man has been studied in two ways.  One of

these was by comparing the  spatial or temporal distribution of ambient carbon

monoxide concentrations with the occurrence of new cases (Table  5-3).  The

other was by measuring the  post-mortem carboxyhemoglobin concentrations among

ASHD sudden death subjects  and  those  of  subjects who died  suddenly from other

causes  and comparing the values with  those of normal living controls. ~  ~

     In two community studies (Los Angeles^l and Baltimore^!?) a relation

between the incidence of heart attack and ambient carbon monoxide concentrations

has not been demonstrated.  The Los Angeles study was based on hospital admissions

for myocardial  infarctions  and the Baltimore study was based on sudden and

more prolonged  deaths due to ASHD and patients admitted to the hospital because

of their  first  transmural myocardial  infarction.

     The major  problem with this type of study is the difficulty in determining

the dose  of carbon monoxide that a heart attack subject may have received owing

to exposure to  the ambient  carbon monoxide concentrations  in the community.

The onset of the acute event, myocardial infarction, or sudden death, is not

                                                         TABLE 5-3

          The Relation  Between Ambient Carbon Monoxide Levels and Incidence and Survival of Coronary Heart Disease
CO Range
     Los Angeles
Comparison of admissions and case-
fatality from heart disease in
high and low CO areas.

Above or below 8 ppm CO; 2,484
admissions in high, 596 in low
5-14 ppm weekly
  basin average
No difference in admission rates.

Increased fatality rates in high CO
areas only during times of high
pollution (8.6+ mean weekly ppm CO)
     Baltimore     Incidence of sudden death,
                   myocardial infarction, total
                   ASHD deaths compared with
                   ambient CO concentrations
                   at one station.
                                       0-4 ppm compared
                                       to 9+ ppm CO
                                       24-hour average
                    No relation between the incidence of
                    sudden death, myocardial infarction
                    or total ASHD deaths, and the ambient
                    CO concentrations.

well-defined except perhaps for instantaneous deaths.335  For many myocardial

infarction cases and sudden deaths, there is a prodromal period which may last

for several days prior to hospitalization or death.  Therefore, the ambient

carbon monoxide concentrations on the day of admission or death may have little

relation either to the onset of the myocardial infarction or to sudden death.

Can the values reported by a field-monitoring station be used to estimate the

dose or change in dose for an individual living in the community but not in the

immediate vicinity of the sampling station?  Perhaps it would be advisable to

consider the reported values as a crude estimate of changes in exposure from

day-to-day rather than as an individual dose.

     Comparing carboxyhemoglobin concentrations between subjects and controls

partially eliminates the problem of estimating specific individual doses.

Carboxyhemoglobin concentrations in subjects experiencing sudden death due

to ASHD were compared with those in subjects whose sudden death resulted from

other causes or with the concentrations in living controls.  For sudden death

the time between onset of the event and death is usually relatively brief.

Since carboxyhemoglobin concentration  is stable after death, the concentration

determined at post-mortem may be a good estimate of the concentration at the

 time  of death.

     The first such study took place in Los Angeles where carboxyhemoglobin

measurements were made at post-mortem on a 20% sample of the subjects (2,207

deaths) from the Los Angeles County Chief Medical Examiner/Coroner's cases.

Questionnaires about smoking habits were returned from the next-of-kin of

1,078 of the subjects  (Table 5-4).l03  The cause of death was determined at

the post-mortem examination.  Carboxyhemoglobin concentrations were higher

                                                 TABLE  5-4

                       The Relationship Between Carboxyhemoglobin and Cause of Death
    Los Angeles County <"!hief
      Medical  Examiner/
      Coroner  103
Baltimore Office of the
  Chief Medical Examiner217
                                20% sample,  Los Angeles County Chief
                                Medical Examiner/Coroner's case load:
                                November 1950 - June 1961, blood
                                samples 2,207 cases; smoking
Comparison of carboxyhemoglobin
concentrations in relation to cause
of death.  Age, race, sex for ASHD
deaths by smoking history detailed
pathology and length of survival;
2,366 deaths
High carboxyhemoglobin levels in
cigarette smokers
Younger smokers higher carboxyhemo-
globin than older cigarette smokers
Greater the.  amount of smoking,
higher the carboxyhemoglobin levels
Relationship between postmortem
carboxyhemoglobin and ambient carbon
monoxide among nonsmokers in most
polluted areas
Cause of death had little or no relation>
ship to the carboxyhemoglobin concent-
ion at post-mortem

Carboxyhemoglobin higher for ASHD
sudden-death subjects than in
those of sudden-death owing to
other natural causes, but no dif-
ference in comparison with
traumatic deaths
Among ASHD death subjects, carboxy-
hemoglobin higher in younger than
older age group
Among smokers, carboxyhemoglobin
higher in controls than ASHD death
subjects.  For non-smokers, carboxy-
hemoglobin greater in ASHD sudden
death than controls
No relation of ASHD death subjects'
pathology, place of activity at
onset, length of survival and
carboxyhemoglobin levels

in smokers than nonsmokers.  In the most polluted areas there was a relation

between the ambient carbon monoxide concentrations on a specific day and the

mean post-mortem carboxyhemoglobin concentrations.  The causes of death showed

little or no relation with the carboxyhemoglobin values taken at post-mortem.

     The Baltimore Study  (Table 5-5)217 measured the carboxyhemoglobin concen-

trations at post-mortem for 2,366 subjects.  These were compared by age, race,

sex and cause of death.   The post-mortem carboxyhemoglobin concentrations were

higher in those who died suddenly owing  to  ASHD  than in subjects

whose sudden death resulted from other natural  causes.  This was true for all

of the age-group comparisons.  Differences between the median concentrations

of the groups were relatively small.  No differences were noted between people

who died   mASHD and people who died suddenly from traumatic causes such as

accidents or homicides.

     Smoking histories were obtained from  a sample of the ASHD-sudden-death

subjects only and carboxyhemoglobin concentrations were found to be substantially

higher for smokers than nonsmokers.  ASHD-sudden-death subjects were compared

with living controls.  Living controls who were cigarette smokers had higher

carboxyhemoglobin concentrations than ASHD-sudden-death subjects who smoked

cigarettes prior to death.  For nonsmokers the  levels were higher for the

ASHD-sudden-death subjects than for living controls.  However, when exsmokers

were excluded there were  no differences in carboxyhemoglobin concentrations

between individuals experiencing ASHD sudden death who were life-time non-

smokers and similar living controls.  There were also no differences in post-

mortem carboxyhemoglobin  concentrations in the  ASHD-sudden-death subjects that

were related to place of  death, activity at onset, length of survival and

whether the deaths were witnessed or not.

                             TABLE  5- 5
Median and Mean Carboxyhemoglobin Levels by Age and Cause of Death
                    Baltimore Sudden Death Study

Cause of Death

Other Natural
Auto Accident
Other Accidents
Median Mean
1.5 2.0
1.2 1.6
2.0 3.2
1.0 5.4
2.4 2.9

Median Mean


Median Mean

     A detailed pathology study was done of 120 ASHD-sudden-death subjects

whose post-mortem carboxyhemoglobin concentrations were measured.  Practically

all of them had severe coronary artery stenosis.  There was no relation between

the post-mortem carboxyhemoglobin concentrations among the ASHD-sudden-death

subjects and the extent of  coronary artery stenosis, the presence of acute

pathologic lesions  such as  a  thrombosis or hemorrhage in the plaque, or recent

myocardial infarction.

     Neither of the two incidence studies, Los Angeles or Baltimore, revealed

a relation between ambient  carbon monoxide concentrations and  the number  of

new ASHD cases.  These two  studies reported that there was a strong relation

between carboxyhemoglobin concentrations and  prior smoking history, but little

or no relation to  the cause of sudden death.

The Relation Between Ambient  Carbon Monoxide  and Survival Following a

Myocardial Infarction

     There has been only one  study of the  relation between ambient carbon

monoxide concentration and  case-fatality following a myocardial  infarction.

Case-fatality  rates were compared for patients admitted with a myocardial

infarction to  35 Los Angeles  hospitals during 1958.  The carbon  monoxide

measurements were  reported  from monitoring stations operated by  the Los

Angeles County Air Pollution  Control  District.

     The hospitals were divided into  those located in the low  and those in

the high carbon monoxide pollution areas.   The low area was outside the 8 ppm

isopleths for  1955.  The majority of  hospitals were located in the high area,and

2,484 patients with myocardial infarction  were admitted in the high areas as

compared to 596 admitted in the low areas.  The areas more highly polluted

with carbon monoxide had a  greater case-fatality percentage than those  less pollutedl

during the weeks when the average carbon monoxide concentrations were in  the

highest quintile, 8.5 to 14.5 ppm weekly mean basin carbon monoxide  concentrations.

In 12 of the 13 weeks, the case-fatality percentage was greater in the higher

carbon monoxide areas (Tables 5-g and 5-7).

     A further analysis of the data suggests that the differences in case-

fatality may not be a function of a change in ambient carbon monoxide concen-

trations.  During times of high pollution, the median percentile of case-

fatalities in the high-polluted area was 30% and it was 20% in the low-polluted

area.  When the average basin pollution was low the median percentile for

case-fatalities was 26% in the high- and 27% in the low-polluted areas.   Thus,

as the mean basin carbon monoxide concentration increased, the case-fatality

percentage increased slightly in the high-polluted areas but decreased in

the low-polluted areas.   It would be very unlikely that the case-fatality

percentage would be inversely related to the carbon monoxide concentrations

in the less polluted areas.   Practically all of the high mean basin carbon

monoxide concentrations were reported in the winter while the low concentra-

tions occurred in the spring and summer.   Because of the relatively small

number of hospital admissions per week (12) in the low pollution areas,

there was a wide variation in mean case-fatality percentages (0-58%).  These

three factors suggest that the relation between ambient carbon monoxide  con-

centrations and case-fatality percentages during high pollution episodes may

be related to a seasonal factor such as an influenza epidemic, changes in

the number of hospital admissions in relation to the number of available

beds, or other undetermined  variables.

                              TABLE  5-6
Case-Fatality Percentage in High and Low Carbon Monoxide Pollution Areas
for the 13 Weeks in which the Average Basin Weekly Mean Carbon Monoxide
Concentrations were Lowest °°

23 -
High Pollution Area

Low Pollution Area

  * Estimated number of patients  with mycardial  infarction admitted  to
   hospitals per week

                               TABLE 5'.7
Case Fatality Percentage  in High and Low  Carbon Monoxide Pollution
Areas  for  the Thirteen Weeks  in Which  the Average Basin Weekly  Mean
Carbon Monoxide  Concentrations Were Highest:
High Pollution Area
Low Pollution Area
MEDIAN:                        30                        20
  *  Estimated number of patients with myocardial infarction admitted
     to hospitals per week

     Seventy percent of all ASHD deaths take place outside of the hospital.

The percentage of case-fatalities among hospital admissions is an inadequate

measure of the relation between environmental factors and the number of short-

term fatalities following a heart attack.  The percent of case-fatalities

within a hospital is a function of the incidence of heart attacks, and their

severity, and the rapidity of transfer to the hospital.  When there is fast

transportation to the hospital after heart attack, the number of case-fatalities

within the hospital might increase while the over-all percentage of case-

fatalities decreases.

     Because of the significance for public health of a possible relation

between case-fatality and ambient carbon monoxide concentrations, particularly

during periods of high pollution, replications of both the Baltimore and

Los Angeles studies should be implemented.  Such studies need to include the

number of both in-hospital and out-of-hospital case-fatalities, a description

of the criteria both for diagnosing heart disease and for the demographic

characteristics of  the subjects, and effective methods for monitoring carbon


Clinical/Experimental Studies of the Relation of Carbon Monoxide and
Morbidity Due  to Heart Disease

      Another approach to  the study of  the relation between exposure to carbon

monoxide and the natural history of ASHD is to identify high-risk subjects

and  observe  the effect of  either natural or artificial carbon monoxide exposure.

Such studies usually  involve a  technique for inducing symptoms  in the subjects,

such as  exercise  testing  (Table 5.-8).6'12'13*14*15

      The first  studies in  1971  compared subjects with angina pectoris before

and  after  smoking nicotine-free cigarettes.^  Ten male angina  pectoris  subjects

                                                     TABLE S.-8

                    Clinical  Studies of the Relation between Carbon Monoxide and Angina Pectoris
   Sample Size
Case     Control
Case    Control

   travel on
   monoxide on
   exercise in-
   duced angina

   exposure and
   onset and
   duration of

   Low  level
   onset & dura-
   t ion inter-
   mittent claud-
Evaluation of patients
with angina following
smoking non-nicotine
cigarettes, exercise
on bicycle ergometer

Comparison of breath-
ing freeway travel in
air (high carbon
monoxide) with carbon
monoxide free com-
pressed air, exercise

50 ppm carbon monoxide
2 hr for 2 mornings;
carbon monoxide free
compressed air 2
mornings; exercise

Carbon monoxide free
compressed air,
50 ppm carbon monoxide,
100 ppm carbon monoxide
 50 ppm carbon monoxide
 2 hr compared to carbon
 monoxide  free compres-
 sed air
10 subjects;
2 smoked in
2 non-smoking
10 patients with
angina pectoris,
cross-over ex-
10 patients,
angina cross
over, blind
10 patients
 10 men  cross-
7.54    0.76

8.03    1.06

5.08    0.75

2 hr   later
2.68    0.77
50 ppm (2.8)
100 ppm (4.51)
2.97    0.90
Reduction in exercise
time to development
of angina following
smoking of cigarettes.
No ECG changes

Reduction in mean
exercise time, inter-
val after freeway air,
3 patients with ST-T
Breathing freeway
air reduction in
length of time ex-
ercise until onset
of angina; no ECG

Reduct ion in t ime
to chest pain after
exercise, no differ-
ence 50-100 ppm; S-T
depression, ECG
Reduction  in  time  to
develop  claudication;
exercise on bicycle
Aronow and Rokaw
Aronow etal.
Aronow and Isbell
Anderson etal.
Aronow et al.

smoked 8 nicotine-free lettuce leaf cigarettes on two of four mornings.

Following the smoking and/or nonsmoking mornings, the subjects exercised

on a bicycle ergometer.  Their carboxyhemoglobin concentrations rose about

8.0% after smoking the cigarettes as compared to about 1% during the non-

smoking periods.

     The duration of exercise prior to the onset of angina was reduced

following cigarette smoking  (TableS -8).  Chest pain also occurred at a

lower  systolic blood pressure and heart rate than for the nonsmoking mornings.

Although prior to pain there was a wide difference in the duration of exercise,

the duration was reduced for every man who smoked.

     The next approach was to determine the effect of exposure to the high

carbon monoxide concentrations on the Los Angeles freeway. u  Ten patients

with angina pectoris rode  on the L.A. freeway for 90 minutes and then were

brought to the laboratory  for exercise testing.  Exercise testing was done

prior  to the  freeway trip, then immediately after it and finally two hours

later. Approximately three weeks later the ten subjects followed the same

testing schedule except  during the 90-minute freeway trip they breathed

carbon monoxide free compressed air.  Breathing carbon monoxide-polluted

ambient air the mean carboxyhemoglobin increased to 5% during the freeway
 trip ancr*279T~two hours later.   Ischemic  ST-T changes in the electrocardiogram

 occurred in 3 out of 10 subjects while breathing  carbon monoxide polluted

 freeway air.   There was a substantial  decrease in the duration of exercise

 time before the onset of angina (both  immediately after the trip and  2 hours

 later)  in comparison to the  exercise time prior to the freeway trip  (Table

 5-8).   Breathing carbon monoxide-free  compressed  air, there were no  changes

in exercise time before or after freeway travel.  The subjects exercised longer
after breathing the compressed air than after breathing carbon monoxide polluted
freeway air.  Systolic blood pressure and pulse were lower at the onset of
angina after breathing the freeway air.  No EGG differences were noted during
the exercise testing, before the freeway trip, after it or when breathing
compressed air.  These were not blind studies.  The investigators and subjects
both knew about the exposure to carbon monoxide.  The subjects might even
have suspected the possible occurrence of a harmful effect.
     A double-blind study was done.21  Ten subjects with angina pectoris
were exposed either to carbon monoxide at 50 ppm for two hours on two mornings,
or to compressed air for two mornings.  They then exercised on a bicycle
ergometer.  The mean carboxyhemoglobin after two hours of the carbon monoxide
exposure was 2.68% as compared to 0.77% with compressed air.   Even at relatively
low carboxyhemoglobin concentrations, the exercise time was reduced prior to
the onset of angina pectoris.  There was also a decrease in systolic blood
pressure and heart rate at the time of onset of angina.
     Anderson,  et al.6 in North Carolina did a similar double-blind study.
Ten subjects with angina pectoris were exposed to air with carbon monoxide
at either 50 or 100 ppm for 4 hours.   After exposure the subjects exercised
on a Collins treadmill.  The duration of exercise before the  onset of chest
pain was significantly shorter after exposure to carbon monoxide either at
100 ppm (mean carboxyhemoglobin 4.5%) or at 50 ppm (mean carboxyhemoglobin
2.8%).   There was however, no difference in exercise times after exposure
to either 50 and 100 ppm of carbon monoxide.   The S-T segment depression of
the electrocardiogram generally appeared earlier and was deeper after breathing
carbon monoxide.  Other measures of cardiac function such as  systolic time
intervals, left ventricular ejection time,  pre-ejection period index, and pre-
ejection peak to ejection time ratio were within normal  limits.

    Subjects with intermittent claudication  have also been studied with  this

approach.  In a double-blind study, they were exposed for 2 h either to carbon

monoxide at 50 ppm or to compressed air then they exercised.    Time until

pain, in this case intermittent claudication rather than angina pectorls, was

reduced after breathing carbon monoxide.

    Only one experimental animal study of the effects of carbon monoxide  on

the natural history of heart disease has been reported.  DeBias et al. °°

studied the effects of continuous exposure to 100 ppm carbon monoxide  (115 mg/m3)

for 24 weeks in the cynomolgus monkey.  They observed that carboxyhemoglobin

chronically raised to an average of 12.4% produced polycythemia and the

hematocrit increased from 35 to 50%.  All animals showed an increased P-wave

amplitude and T-wave inversion in their electrocardiogram suggestive of

myocardial ischemia.  Animals in which an experimental myocardlal infarction

was produced who were next  exposed to carbon monoxide had more severe electro-

cardiographic changes than  animals that breathed ambient air.

    These studies of heart  disease morbidity after carbon monoxide exposure

have  important Implications.  The first question that needs to be investigated

is whether the experimental model in man, angina pectoris subjects exposed

to low  doses of  carbon monoxide followed by exercise on  a bicycle or treadmill,

has relevance to the situation in a community.  It has been suggested  that a

carboxyhemoglobin concentration as low as 2.5%  has a deleterious health
        QTO A20
effect.-"0'^"  All cigarette smokers and about 10% of the nonsmokers  in  the

United  States frequently have carboxyhemoglobin concentrations higher  than

2.5%.  Thus  a large percentage of the United States population may be  poten-

tially  at risk.  The National Health Survey Examination  reported that  there

were  3,125,000 adults, aged 18 to 79, with definite coronary heart disease
  A  complex of symptoms  characterized  by  absence  of pain or discomfort  in  a
  limb when at  rest,  the  commencement of pain,  tension, and weakness, after
  walking  is begun,  intensification of  the condition until walking becomes
  impossible, and  the disappearance of  the symptoms after a period of rest.


and another 2,410,000 who were suspect.     Many people who have severe

coronary artery stenosis without any apparent clinical disease would also

be at high risk when the concentrations of carbon monoxide in the air

are relatively low.

    If the results of the clinical studies are applicable to this large

population at risk, then a major public health problem exists.  Taking the

current results at face value suggests only that when patients with angina

are exposed to low carbon monoxide concentrations for short periods they

cannot exercise as long on a bicycle or treadmill before developing chest

pain as those breathing compressed air.  There is no evidence from these

results that the exposure to carbon monoxide increases the frequency and

severity of chest pain or the development of other complications,  or that

it shortens life expectancy among patients with angina pectoris or other

clinical manifestations of heart disease.  We can only infer the existence

of such a relationship.

    Evidence for this association based on observations is equivocal.   People

with angina pectoris who smoke cigarettes can be expected to have  higher

carboxyhemoglobin concentrations and poorer prognosis than those who do not

smoke cigarettes. ^  The harmful effect could be due both to carboxyhemoglobin

and to    other chemicals in cigarette smoke.   A recent study has  shown, on

the other hand, that relatively few heart attacks occur while an individual

is cigarette smoking and is thus exposed to higher carbon monoxide concen-

trations. 216  There is no evidence of an increased heart attack risk,  in-

cluding myocardial infarction and sudden death^ while driving an automobile,
and there    no data showing the occurrence of  angina pectoris  pain episodes

in relation to specific activities.

    There is also no evidence suggesting a higher incidence, prevalence,

or prognosis of heart disease among industrial workers who are exposed to

high carbon monoxide concentrations.

    And finally, there is little positive or negative evidence indicating

that high ambient carbon monoxide concentrations in a community are associ-

ated either with the prevalence of angina pectoris or the natural history

of heart disease.  Furthermorej within communities, there is no evidence that

there is a relation between the frequency of angina pectoris pain episodes

and changes in the ambient carbon monoxide concentrations.  Without such

evidence, although the clinical experimental studies suggest important relation-

ships between carbon monoxide and heart disease, ambient carbon monoxide cannot

be implicated as a major causative factor of heart disease in a community.

The Relation Be.tween Cigarette Smoking and Clinical Coronary Arterial Disease

    The association between cigarette smoking and clinical coronary arterial

disease is much closer for myocardial infarction and sudden death than it

is for angina pectoris. y  At present it is not completely clear why this

is so but it may be related to specific precipitating factors rather than

to the underlying atherosclerosis.  The association of cigarette smoking

with clinical coronary arterial disease apparently depends on other risk

factors, particularly a high-fat diet and increased serum cholesterol.

This association is relatively weak in populations with low serum cholesterol

content, but is apparently stronger in younger than in older people.201

     Cigarette smoking appears to increase the risk of sudden death and

myocardial infarction among subjects with pre-existing angina pectoris.424

On the other hand, the relation between smoking after a myocardial infarction

and subsequent survival is less clear.100'425  If cigarette smoking did not

increase the mortality risk after a myocardial infarction, specifically


among those who have survived for a month or so after the initial myocardial

infarction, then the association between carbon monoxide and heart disease

would be substantially weakened.  A critical problem has been to try  to

separate the carbon monoxide effects of cigarette smoking from the effects

of other harmful substances in cigarette smoke.  After individuals free of

clinical coronary disease cease smoking their risk of heart attacks is re-

duced.   This reduction in risk takes place very soon after the cessation of

cigarette smoking.1^5

     The above observations suggest that a major effect of cigarette smoking

may be as a precipitant of heart attack rather than in the development of

the underlying atherosclerosis.   There is a small amount of data suggesting

that cigarette smokers have more extensive atherosclerosis than nonsmoking

age-related controls.  Such studies;however,  do not have data adjusted for

serum cholesterol levels or other risk factors.  The effects of cigarette

smoking on the incidence or clinical history of ASHD do not necessarily have

to be due to carbon monoxide inhalation.  Other factors in cigarette smoke,

including nicotine,  cyanide,  or trace elements may be important.

     Studies relating carbon monoxide, smoking and heart disease include

10 angina pectoris  subjects who smoked cigarettes and who had blood pressure,

heart-rate, and expired carbon monoxide measurements taken before and after

smoking high-, low-,and no-nicotine   cigarettes. ^  After they  smoked  these

cigarettes, their expired carbon monoxide concentrations increased with little

difference among the three types of cigarettes.  There was a significant in-

crease in heart rate and systolic blood pressure after smoking the high- and

low-nicotine cigarettes, but no effect after smoking nicotine-free cigarettes.

Another study compared the effects of carbon monoxide and nicotine on

cardiovascular dynamics in       8 men with angina pectoris. »    Right and

left heart catheterizations were done before and after smoking cigarettes

and then repeated after breathing carbon monoxide at 150 ppm so thau coronary

sinus carbon monoxide would be similar to that after the smoking of 3 cigarettes.

None of the subjects developed symptoms of angina pectoris either after

smoking or after exposure to carbon monoxide.  Smoking caused increases in

the aortic systolic and diastolic blood pressure and in the heart rate.

     These changes were not observed after breathing carbon monoxide.  The

left ventricular end diastolic pressure increased after both smoking and

carbon monoxide inhalation, the stroteindex was reduced with both procedures,

and   the cardiac index did not change after smoking but declined after carbon

monoxide inhalation.  Smoking and carbon monoxide exposure also reduced the

coronary sinus oxygen.  Most of the effects disappeared or were substantially

reduced within 30 minutes after the exposures.  Therefore, after smoking,

nicotine apparently increased the systolic and diastolic blood pressure

and the heart rate, while carbon monoxide caused a negative inotropic

effect, an increase in left ventricular end diastolic pressure and a decrease

in the stroke index.

     Wald e± _al.419 have attempted to determine whether cigarette smokers

with clinical coronary disease have higher carboxyhemoglobin concentrations

than smokers without disease, after adjusting for the amount of cigarette

smoking.  Volunteers (1,085) were recruited from several firms in Copenhagen,

Denmark.  They completed a questionnaire concerning their ASHD history.

All who gave a positive history of heart disease were examined and the history

validated both by examination and review of the medical records.  Smoking

histories were obtained from all of the subjects and their carboxyhemoglobin

measured.   A higher prevalence of ASHD was found with increased cigarette

smoking.   Men whose cigarette smoking was moderate to heavy and who had

higher carboxyhemoglobin concentrations had a greater prevalence of ASHD.

The relation persisted after adjusting for age, sex, duration of smoking

history,  serum cholesterol levels and cigarette consumption.  Although these

results suggest a specific carboxyhemoglobin effect, they might also just

be a measure of the inhalation of smoking products and not necessarily only

a function of carbon monoxide inhalation.

   " The  significance of carbon monoxide inhalation from smoking cigarettes

is often  dismissed when the effects of cigarette smoking on cardiovascular

disease are evaluated.   There is a good correlation with the amount of tar

and nicotine in cigarettes, but not necessarily with the amount of carbon

monoxide  produced.  A controversy exists about the effect on the increase in

carboxyhemoglobin concentrations of smoking low-as compared to high-nicotine

cigarettes.   Several investigators have suggested that the carbon monoxide

production of a cigarette be included on the package.  A safe cigarette

should be  low in tar,  nicotine,  and carbon monoxide production.  A low-carbon

monoxide-producing cigarette has been made.

     Cigarette smoking is the chief source of the high carboxyhemoglobin

concentrations in the population.   To effect major reductions in the mean

carboxyhemoglobin in the population will require both a significant reduction

in smoking and the modification of cigarettes to deliver lower doses of

carbon monoxide.

     The potentially harmful effects of low doses of carbon monoxide with

respect to cardiovascular disease should provide further impetus to efforts

 to reduce cigarette smoking.  People exposed to carbon monoxide from other

 sources,such as their occupations or concentrations in the ambient community

 air, may be at particularly high risk from cigarette smoking.   The cigarette

 smoker exposed to high carbon monoxide concentrations in the ambient air

 will also have a greater carboxyhemoglobin concentration and a concomitant

 increased risk of heart attack.

      The studies on cigarette smoking and cardiovascular disease are

summarized in Table 5-9.

                                                    TABLE  5-9

                 Cigarette Smoking,  Carbon Monoxide  and Cardiovascular Disease Studies Summarized
  Wald  et al.
1,085 volunteers, several
firms, including tobacco
  Aronow et al.
Comparison of nicotine
and carbon monoxide
effects; 10 men with
ang ina; high,low-,and
no-nicotine cigarettes
   Aronow et al.
   Cigarette smoking
   and breathing
   carbon monoxide;
   hemodynamic in
   anginal patients
8 men with angina
History of ASHD, validated by
records and exam, smoking
history and carboxy-
hemoglobin concentrations
                                Higher prevalence  of  ASHD with
                                increased tobacco  smoking, with
                                increase in carboxyhemoglobin,
                                slight effect  on carboxyhemoglobin
                                within smoking groups

                                At  older ages, carboxyhemoglobin
                                most  powerful  discriminator

(1)  Significant increase in peak systolic and  diastolic  blood
    pressure from smoking high-or low-nicotine cigarettes.

(2)  Significant increase in heart rate with high or  low  nicotine

(3)  No change in heart  rate or  systolic blood  pressure after
    smoking nicotine-free cigarette.
8 men with angina, right and
left cardiac catheterization
before, after smoking 1,2,3,
cigarettes, replicate after
150 ppm - until coronary
sinus carbon monoxide is
same as after cigarette 3
                                None of the patients  developed
                                angina pectoris.   Smoking aortic
                                systolic and diastolic  blood
                                pressure.  Blood  pressure,  no change
                                after breathing carbon  monoxide
                                Heart rate increased  with smoking,
                                not with carbon monoxide.
                                Left ventricular  end  diastolic
                                pressure increased after smoking also
                                after breathing carbon  monoxide.
                                Cardiac index unchanged after smoking>
                                decreased with carbon monoxide.
                                Stroke index decreased  after both smoking
                                and breathing carbon monoxide.
                                Increase in coronary sinus  xygen
                                after smoking cigarettes or after
                                carbon monoxide.   Most  of the hemo-
                                dynamic changes reduced after 30

                                               TABLE  5-9  continued
   Russell et al.347     22  cigarette  smokers        Comparison of  increase  in        Carboxyhemoglobin increase  after smoking

   Effects of changing                              carboxyhemoglobin after         single  strong  cigarette was 1.45%,  1.09%

   to low-tar, low-                                 smoking extra                    for small brand,  and  0.64%  for extra-mild

   nicotine cigarettes                              strong cigarettes                brand


     The experimental studies of carbon monoxide's effects on human behavior

are reviewed  under seven topical headings;  vigilance,  driving, reaction time,

time discrimination and  estimation,  coordination and tracking, sensory

processes,  and complex intellectual  behavior.   Although carbon monoxide is

probably the  most  widely studied of  all toxic  substances, our knowledge of

its effect on behavior is limited.


     Psychologists study vigilance by examining how well an individual per-

forms when detecting small changes in his environment  that take place at un-

predictable intervals and so  demand  continous  attention.2^2»253  vigilance

was first explicitly studied  during  World War  II,  when the

British government became concerned  about the  performance

of men  who spent long hours searching for submarines or aircraft.  This was a

monotonous  task and it was found that after a  while the men would miss signals

that they would not have missed  at the start of their  vigil.

     There  have been a series of reports of carbon monoxide's effects on such

a task.   Groll-Knapp £t  al.**3 exposed "20  subjects of both sexes" to carbon

monoxide at 0,  50,  100 or 150 ppm for a 2-hour period.  Whether the subjects

smoked  is not stated.  Carboxyhemoglobin concentrations were not measured

directly but  rather it was estimated that by the end of exposure their

values  would  have  reached about  3% at 50 ppm,  5.4% at  100 ppm and about 7.6%

at 150  ppm.   The subjects began  an acoustic vigilance  test one-half hour after

exposure started.   At this time  the  carboxyhemoglobin concentrations would

have reached  only  about  1.5,  2,  and  2.5%, respectively.  Pairs of short tones,

1.1 s apart;were presented in a regular sequence.  Over the 90-minute  test-

period, 200 of these pairs were transformed into signals by making  the second

tone "slightly weaker" than the first.  The subject reported their  presence

by pressing a button.  Thus, signals occurred irregularly 2.2 times per

minute.  Every subject was exposed to tich of the four conditions once with

the sequence systematically changed.

     The mean number of signals missed during the test was 26 at 0" ppm  (control),

35 at 50 ppm, 40 at  100 ppm and 44 at 150 ppm.  Apparently even the smallest

difference was  significant. (A failure to replicate these findings  is  reported

but not described  in detail in a recent symposium paper from the same  laboratory. *

Fodor and  Winneke^S carried out a second auditory signal study.  They used  a

broad band noise that las ted 0.36 s and was repeated at 2.0 s intervals. About threa

of the noises out  of every hundred were made to be slightly less intense.

These were the  signals the subjects reported by pressing a button.  Twelve

subjects  (male  and female nonsmokers, 22-35 years old) were tested  at  carbon

monoxide  concentrations of both 0 and 50 ppm, half with each order  of  presenta-

tion.  Exposure lasted 80 min before the first 45-mlnute vigilance  test.   A  five-

minute visual task followed the first vigilance test after which there was a

second 45-oinute vigilance test.  Following another five-minute visual task,

a third vigilance  test was presented.  The signals occurred randomly,  44  of

them occurring within any one 45-minute period, or about 1/min.  Carboxyhemoglobin

concentrations were not determined directly but were predicted to be 2.3  and

3.12 at the beginning and end of the first vigilance test.  The values for

the second test were 3.12 and 3.72, and 3.72 and 4.32 for the last  test.

Background carbon  monoxide was neglected.

     The  percentage of signals missed is shown in Figure 5-19.  Exposure

to carbon monoxide caused the subjects to miss signals during the first

vigilance test.   This  effect  was  not  apparent during the next two vigilance

test  periods  and the interaction  between carbon monoxide concentrations and

time  was  statistically reliable.   The data for the speed of response to the

detected  signals also  showed  a trend  toward a  maximum effect during the

first period, but this time the interaction was not significant.

     Fodor and Winneke    note that the "variation in performance after the

125th minute  of  exposure  is hardly consistent with existing theoretical and

experimental  knowledge" and go on to  speculate that "an organism exposed to

carbon monoxide  possesses the capability of compensation, which can, within

certain limits,  counterbalance a  drop in performance."  However, they recognize

that  "published  data give little  support to this ex post facto  hypothesis,

which urgently needs corroboration by replication experiments."

     Winneke     reported  another  study of the effect of carbon monoxide on

auditory  vigilance in  which the carbon monoxide data were collected along

with  data on methylene chloride,  a compound suspected of exerting at least

some  of its behavioral effects by increasing carboxyhemoglobin concentrations.

Eighteen  subjects,  nine male  and  nine female, were exposed to carbon monoxide

at 0, 50, and 100 ppm  on  different occasions for almost four h and then

tested with the  same ajudttory  vigilance task used by Fodor and Winneke.125

The results were completely negative^with differences not even in the expected

direction (see Figure  5-20).   Winneke^5 estimated that the subjects had an

average carboxyhemoglobin concentration of about 9% after the 100 ppm exposure.

At first  glance  the positive  results  with methylene chloride reinforce confidence

            70 r-
          s ?5°

           2 790

            Periods of observation
      30'   «5'    75'   30'   V5'    X'   30'   VS'
                    K'  110'

              L.J	1	j._
               130'   W  ISO'
                               •--• Control
                               .	CO
                       Minutes after beginning of exposure
                                                 1S5'  210'
Figure  5-19.  Effects  of carbon monoxide  on an acoustic vigilance
test.   Top,  number of  signals missed after exposure to carbon
monoxide at  zero and 50  ppm.  Bottom, latencies of responses to
the detected signals.  The panels show  results from three successive
45-minute vigilance tests.  Reprinted with permission from Fodor and


LJQ- 75 aO-125 130-175 l8ff-225'\

  •  Control  N=20
  D  300ppm N=12
  V  SOOppm N=14
  •  800ppm N= 6
K-125 ISO-US' 180-225}
 •  Control  N=16
 0  SOppm  N=18
 A  100ppm  N=1d
                                      ^  S



               h72 S
Figure 5-20.  Vigilance performance after exposure to methylene
chloride  (left) and carbon monoxide (right).  Reprinted from

in the reliability of the carbon monoxide data and are particularly welcome

to an area where data are rare for substances that can serve as positive

controls or reference substances.173a'222a  However, this confidence is

diminished by the inversion of the 300 ppm and 500 ppm curves and the irregular

carbon monoxide control curve.

     Horvath ^t al.173 and O'Hanlon301'302 studied carbon monoxide's effects

on a visual vigilance task.  A disk with a one-inch diameter about 3 ft (0.9 m)

from the subject was lit  for 1 s every 3 s.  The signal was a slightly brighter

pulse.  The subject pressed a button whenever he saw this brighter light.

Before starting the test, each of 10 nonsmoking, "healthy male volunteers"

21-32 years old were exposed via a mouthpiece for 1 h to the same carbon

monoxide concentration used during the test.  The subjects were then brought

to the experimental room where they were exposed further to carbon monoxide

via another mouthpiece.  Preceding the one-hour main vigilance task, there

was a short pre-test called an "alerted" test, during which 10 out of 60 light

pulses were the randomly>-interspersed, slightly-brighter pulses that were the

signal.   These were presented at a signal rate of 3.3/min.

     The 1-hour main vigilance task began after a 1-minute rest.  Twelve

hundred light pulses were shown 40 of which were the slightly brighter

signals.  These were distributed so that there were 10 among the 300 light

pulses presented during each 15-minute period.  Therefore the signal rate

was 0.67/mi*. Since each subject was exposed at 3 different times to carbon

monoxide concentrations of 0, 26, and 111 ppm, he served as his own control.

The experiment was 4 single-blind study, with the subjects being unaware of the

experimental treatment.  The carboxyhemoglobin concentration of the controls

(0  ppm) was  the  same,  0.8% after the  initial-hour exposure as at the experiment's

end.  For  the exposure to  26  ppm the carboxyhemoglobin concentration was 1.6%

after the  first  hour and 2.3% at the end;  and for the 111 ppm exposure the

concentration was  4.2% after  the first hour and 6.6% at the end.   Performance

on  the pre-test, when  the  signal rate was  3.3/min, was approximately the same

for all 3  exposures.   The  carbon monoxide  apparently had no effect.   During

the 1-hour vigilance test,  subjects  exposed to carbon monoxide at 111 ppm

made significantly fewer correct signal identifications than did  those same

subjects exposed to either 0  or  26 ppm.  The data are summarized  in Figure

5-21.  When  the  signal rate was  0.67/min performance accuracy was reduced

at  a carboxyhemoglobin concentration of about 5%.   However, during  the pre-

test, when  the signals  appeared 5 times as  frequently,  there was no such

effect.  No  measures of variability  are given.   The sudden increase in detec-

tion rate  sometimes seen near the end of the hour may be an example of the end

spurt often  reported in vigilance work when subjects have been told when to

finish their task.

    Beard and Grandstaff^2 examined the effect of carbon monoxide on a

visual vigilance task  also.   In  their double-blind experiment, the signal was

a slightly shorter flash of light than that usually programmed to occur.

The subject  was  seated in  a small audiometric booth and faced a 3 in.(7.6 cm)

electroluminescent panel 2  ft (0.6 m)  from his  eyes.   Every 2 s this panel

was lit.  On most occasions the  panel lit  up for 0.5 s but once in a while

it  remained  lit  for only 0.275 s; these  were the signals to be reported by

Pressing a button.  The subjects monitored the  light flashes for  four 30-minute

Periods during the first day's session.  The first of these periods was a control

   90 r



VIGIL (min)
     Figure 5-21.  Vigilance performance after exposure to carbon monoxide.
     Reprinted with permission from Horvath et al.l?3

period.   Carbon monoxide was then administered at concentrations of 0, 50,

175, or  250  ppm.   A second 30-minute vigilance task was begun after 10 minutes

of exposure to  carbon monoxide.   This was followed by a 10-minute rest period,

after which there was a third 30-minute vigilance test.  The carbon monoxide

was then  shut off and after  another 10-minute break, the final 30-minute

vigilance test was given.  During each of the vigilance tests, 10 signals

were randomly distributed among  a total of 722 non-signal light flashes/

making  the  signal rate  0.33/min.

    Nine nonsmoking  male and female students, 18-25 years old^were each tested

in random order eight timesV twice under each condition.  Carboxyhemoglobin

concentrations were estimated from alveolar breath samples taken both before

testing and  after the last vigilance test,  30 min after the cessation of the

carbon monoxide  exposure.  The pretest concentrations were all below 1%.

Post-test they averaged 1.8% for exposure to 50 ppm, 5.2% for 175 ppm and

7.5% for  250 ppm.   About  73% of  the signals were detected by subjects exposed

only to room air,  during  the last 3 tests.   The detection rate was about 64% for

exposure  to  both 50 and 175  ppm  and 70% for exposure to 250 ppm.   The decreases

following exposure to 50  and 175  ppm,  while small,  were statistically signifi-

cant at the  0.05  level  (one-tailed test).   The decrease following the 250 ppm

exposure  was not  significant.

    Further analysis of  the data, "  within a framework provided by signal

detection theory,  showed  that  the significant changes in performance were

related tcJ 1iot:h  an increased "degree of caution In making decisions and a

decreased sensitivity (i.e.,  beta increased and d_  decreased).  Interpretation

°f the peculiar  results found with 250 ppm would be easier if levels high

enough to produce pronounced effects had been included.

     An experiment by Lewis  et al.230a should also be noted.  These investi-

gators examined the effects of traffic pollution on vigilance by exposing

subjects to air pumped from the roadside of a busy street into the automobile

that served as the experimental "room."  Carbon monoxide levels were not

measured during the testing but sometime after the test the experimenter de-

termined that they were above 10 ppm about 40% of the time, above 30 ppm more

than 4% of the time.  The air was not analyzed for other noxious constituents.

It is not clear how long subjects were in the immediate environment and there-

fore exposed to the flow of pollution before starting on these tasks.  While

seated in the test car they performed on an auditory vigilance task in which

they were presented 0.5 s tones every 2s.  On 9 occasions during a 45-minute

session the tone was slightly shorter.  Thus the tones occurred at the rate

of about 0.2/min.

     Subjects performed on the vigilance task twice, working on a number of

other psychological tasks between runs.  The sixteen subjects used were

18-28 years old.  Order of testing was counterbalancedxwith half first being

exposed to pure air while the other half was first exposed to the polluted

air.  The subjects detected 73% of the signals while being given pure air

from a metal cylinder but only 60% while breathing polluted air from the

roadside.  This difference was reported to be reliable but, as the authors

realize, given the almost complete lack of knowledge of what the subjects

were exposed to, one can conclude nothing about the specific effects of

carbon monoxide.  For instance, Fristedt and Akesson130a found that workers

in service stations located in multistory garages reported headaches and

general fatigue at only slightly elevated lead and slightly elevated

carboxyhemoglobin levels.  They point  out  that  exposure  to  neither substance


         "of a magnitude  such as  to constitute a  primary toxicological

          hazard.  The combination of these in combination with other

          substances existing in  automobile exhausts, such as  acrolein

          and other aldehydes, alcohols,  phenols, acetone, partially

          burned hydrocarbons, oxides of  nitrogen,  sulfur  dioxide,

          and organic lead compounds,  may pose a  hazard to health,

          or a sanitation problem, manifesting itself in the form

          of discomfort.  There never has been an exhaustive investi-

          gation of the biological reaction to chronic  exposure to

          small quantities of a mixture of these  substances."

    Several investigators have looked for changes in physiological  re-

sponses that may be related to vigilance.  O'Donnell et  al   examined how

overnight exposure to low  carbon monoxide  concentrations.with carboxyhemo-

globin concentrations reaching 12.7%,  affected  sleep.  They reported small

and unreliable changes that were interpreted as a  possible  reduction in

central nervous system activation.  Their  observations agreed with the

                          450 452
findings of Xintaras et al.  '   on the evoked  response  in  the  rat and

with Colmant's   data on disturbed sleep patterns  in the rat.   Xintaras
®t al.   ,452 foun(j that effects on the visual evoked response  of  the

restrained, unanesthetized rat were similar for both carbon monoxide and

pentobarbital, the classic reference standard for depression  of reticular

activity.  They concluded that the changes induced by both substances  re-

sembled those recorded during- the normal transition from wakefulness to

sleep.  In man it has also been possible to find carbon monoxide-induced changes
in the visual evoked response.  Carboxyhemoglobin concentrations of 20 to  28%
were required,175'380 which are much higher than those associated with vigilance
changes in the studies previously cited.  It is uncertain whether recent
advances in the techniques for its measurement and analysis    will show an
increase in this response's sensitivity to carbon monoxide.
     In addition to studying behavioral vigilance Groll-Knapp et al.    measured
the slow-wave brain potentials presumed to correlate with anticipatory reactions
to an oncoming signal.  For their vigilance task subjects were presented with a
pair of brief tones.  They had to respond if the second tone was weaker than the
first.  The researchers recorded for a 4-second period starting with the first
pair of tones presented.  They reported that during a 90-minute test period that
started 30 min after exposure to a carbon monoxide concentration as low as
50 ppm, both the height reached by the anticipation wave after the first tone and
the drop after the second tone were reliably reduced.
     With vigilance as with other functions that will be discussed below,  the
interpretation of negative results with small amounts of carbon monoxide
frequently cannot be made unambiguously.  Before attacking the question of
thresholds, investigators should demonstrate the sensitivity and specificity of
their behavioral tests by employing doses high enough to produce measurable
effects.  They should also consider using previously studied drugs as reference
substances.  a»17 a  Only after sensitivity and specificity have been established,
can one have confidence in negative findings with low concentrations.
      In the earliest reported study (1937) of simulated automobile driving
Forbes et al.    exposed five subjects to enough carbon monoxide to produce
carboxyhemoglobin concentrations as high as 30%.  They reported very little effect
on a  series of reaction-time, coordination and perceptual tasks, presented within
the context of a driving skills test.  The only control observations were made
just before exposure and there was no attempt to ascertain how much performance
would have changed if room air alone had been administered.  Consequently, their
results cannot be interpreted and are only interesting historically.

     McFarland and co-workers262.264,265 have recently (1973) studied actual

driving performance,  focusing on two aspects of driving.   One of these is the

amount  of visual information required by a driver.   This  was measured by asking

the subject  to maintain a constant  speed while looking at the road as infre-

quently as possible.   The subject wore a helmet with a shield that was placed

in front of  his eyes  so that he  was prevented from seeing the road.  By de-

pressing a foot switch he could  briefly raise the shield.  He was instructed

to do this sufficiently often to keep his car within a 12-foot-wide marked

lane on a deserted superhighway  while maintaining a constant speed, of either

30 or 50 mi/h  (48  or  80 km/h)  in different trials.265  The number of steering

wheel reversals was also monitored.   Ten subjects were used in these experi-

ments with each acting as his own control.   They were either exposed only to

air or  to enough carbon monoxide at a concentration of 700 ppm (via a mouth-

piece)  to produce  a carboxyhemoglobin concentration level of 17%.  McFarland

reported that  carboxyhemoglobin  did not produce a differential effect on the

frequency of steering wheel  reversals.   His conclusion concerning visual

interruption data   is less clear.   From a significant interaction term in an

analysis of  variance,  McFarland  concluded that those subjects exposed to

carbon  monoxide while driving  at the higher speed required more roadway viewing

than those exposed only to air.

                     •jo 1
     Ray and Rockwell0-'1  examined the driving performance of three subjects

with carboxyhemoglobin concentrations of 10-20% from carbon monoxide exposure.

In this study  the  subject rode in a car yoked by a  taut wire to a second car

driven  ahead of it.   Information concerning relative-velocity.and the distance

between the two vehicles was transmitted via  the wire.  In some experiments the

subject drove while in others he was a passenger.  The authors only reported

  for an experiment in which                                 ,     .,   .
data      the subject attempted to detect slight changes in the relative

velocity of the 2 vehicles when the lead vehicle  was 200 ft (61 m) ahead of

the car in which he rode as a passenger.  The time required to respond to a

velocity change of 2.5 mi/h (4 km/h) was approximately 1.3 s for control con-

ditions, 3.3 s when the subjects had a carboxyhemoglobin concentration of

about 10%, and 3.8 s when it was about 20%.  These differences were statisti-

cally significant for a group of three subjects, each of whom had been exposed toi

three conditions.  In view of the small sample size, the authors considered

these results as only exploratory.

     Weir and Rockwell    have made extensive observations of driving behavior.

This as yet unpublished research also utilized yoking one car to another by

a thin wire.  Gas pedal positions, brake pedal applications, steering wheel

reversals, actual velocity^ and separation of the lead and following vehicles

were recorded.  The results were negative for carboxyhemoglobin concentrations

of 7% and 12%.  These were the concentrations studied with the largest numbers

of subjects and the tightest experimental designs.

     Another study of carbon monoxide's effects on driving performance used

"a standard driving simulator,"   a. device that simulated a 10-minute drive

through traffic, giving the subjects a brake pedal, an accelerator, and a

steering wheel with which to react to various realistic driving conditions

shown on a film.  Forty-four adult volunteers were used.  Half of these were

randomly allocated tc a group receiving only air and half to a group receiving

enough carbon monoxide to produce carboxyhemoglobin concentrations about 3.4%

higher than they had when tested before exposure.  Half of the experimental

group were  smokers  and  their carboxyhemoglobin concentrations just before the

second  driving simulator test averaged about 7.0% (up from about 4.4%) whereas

the nonsmokers averaged 5.6% (up  from about 1.3%).   An 80 ml dose of pure

carbon  monoxide was introduced into  the breathing system that was also used

to make carboxyhemoglobin determinations.   The subjects breathed this as a

carbon  monoxide concentration of  2%.—  The  carbon monoxide was not found to

affect  the  overall  performance on the driving simulator.

    The experimenters  divided the various  individual activities that had

been scored during  the  simulation task into 2 categories; "emergency actions,"

and "careful  driving habits."  No difference could  be detected in the way the

experimental  and  control subjects reacted on the category "emergency actions."

With the "careful driving habits," the group working under the influence of

carbon  monoxide showed  more  deterioration than the  control group.  The change

is only marginally  statistically  reliable,however,  the relevant data shown in the

-Administering carbon monoxide  at a high  concentration for a short time

 may produce greater effects on performance  than  the  customary method of

 exposing subjects to low concentrations for longer times.   While the

 carboxyhemoglobin concentration is usually  thought to  be  the most important

 physiological stimulus for determining carbon monoxide's  effects, there is
              • *

 some evidence that the rate of saturation is also important (see other section

 on  page 5-145).    Plevova and Frantik3   recently (1974)  demonstrated that

 exposing    rats to 700 ppm for 30 min produced  a greater  decrease in the

 length of a forced run on a treadmill than  did exposure to 200 ppm for 24 h,
                            *    '              f -  ^

 even though both exposures brought the average carboxyhemoglobin concentration

 at the start of the experimental task to  approximately 20%.

researchers' table*|ave a chi-square of 5.84, which with 2 degrees of freedom

yields a probability between 0.10 and 0.05.  Consequently, this finding can

only be regarded as suggestive support for their conclusion that a 3.4%

increase in carboxyhemoglobin is sufficient to prejudice safe driving.

     Runnno and Sarlanis345 also used a driving simulator to examine the effects

of approximately 7% carboxyhemoglobin concentration.  Carbon monoxide at

800 ppm was administered for 20 min prior to a 2-hour simulated drive.  By

also dosing the subjects in the same manner with air before one of the two

                        kept the subjects
experimental sessions.they4          unaware of when they received the carbon
                     '    r

monoxide.  The experimenter knew,however, which gas treatment was in effect.

The 7 subjects (6 nonsmokers, 1 smoker) tried to maintain a specified distance

between the automobile that they were "driving" and one that appeared to be

in the same lane ahead of them.  They were instructed to respond as quickly

as possible to any changes in this separation.  Both increases and decreases

in the separation were programmed  to occur at random intervals 10 times

every half-hour.  Steering wheel reversals and braking responses to a red

warning-light that appeared on the dashboard for a few seconds 8 times every

half-hour were also recorded.  An increase in the response time to changes

in the speed of the car ahead was associated with the carbon monoxide treatment.

The mean response time increased from 7.8 s to 9.6 s, a statistically significant

effect.  It was also observed that fewer steering wheel reversals were made

after carbon monoxide administration.  The lone smoker showed the opposite

effect.  The change was statistically significant only when that subject's

data were not included in the calculations.  Several other measures - reaction

time to the dashboard light,  how much space the subjects kept between cars, and
ability to maintain lateral position on the simulated road— showed no reliable
changes.   The authors pointed out one factor that may have helped them observe a
reliable decrease in the rapidity with which changes in the distance between the
two cars were detected.   The task "was designed as a vigilance task that closely
simulated a real-life situation of prolonged driving on a little traveled road under
twilight conditions.   Nearly all the subjects remarked that the driving was
                      34 5
realistic but boring."-3"
     The paucity of published research on driving and carbon monoxide is somewhat
puzzling, given the longstanding interest in the question by strong political and
economic groups.  Perhaps one impediment has been the relative difficulty in
devising both suitable and safe experimental preparations.  In light of the strong
preference on the part of some that experimenters work with the precise behavior
at issue, rather than with laboratory analogues,  '   the field is probably await-
ing the development of adequate field methods (cf., e.g., 286).
Reaction Time
     Reports on the effects of carbon monoxide on reaction time have been con-
flicting.  In the last few years several well-controlled studies were completely
negative, whereas several others equally well-controlled were positive.
                                                     449                381
     Negative studies were reported by Wright et al.,    Stewart et: al.,
                  125         445                        345
Fodor and Winneke,    Winneke,    and Rutnmo and Sarlanis.     The last study was
described above.
     In another driving simulator study also described above, Wright et al.
had subjects release the accelerator pedal and depress the brake pedal as quickly
as possible when a red light was lit by the experimenter.  No reliable difference
was found in performance on this task at carboxyhemoglobin concentrations about
3.4% higher than the subjects (both smokers and nonsmokers) had at the onset of
the experiment.

     Stewart e£ al.381 used the American Automobile Association driving

simulator in a reaction-time study.  The subject was presented one of 3

stimuli,each of which signaled a different response; turning the steering

wheel left or right  or removing the foot from the accelerator and depressing

the brake pedal.  Carboxyhemoglobin concentrations as high as about 16%

did not change the reaction time.

     During their study of vigilance, Fodor and Winneke125 carried out three

different reaction-time tests:  one in which the subject kept his finger

directly above a response button; one in which the subject had to move 25 cm

from where his finger rested between trials to the button; and one dis-

criminative reaction-time test in which 5 buttons in a circle were paired

with 5 different signals.  No significant differences were found for any

of these at carboxyhemoglobin concentrations estimated to be 2.3 and 5.3%.

The same procedures also produced negative results in a study by Winneke

in which carboxyhemoglobin concentrations were an estimated 10%.

     Ramsey325,32Jep0rted two studies in which longer reaction times with

carbon monoxide were observed.  In one study,   ^ 60 subjects were exposed via a

face mask to 300 ppm for 45 min while 20 controls breathed only air from a

tank through the mask.  The mean carboxyhemoglobin concentration of the

exposed group reached about 4.5%.  The reaction-time test is described by

the author only as "reaction-time to a visual stimulus." Subjects exposed

to carbon monoxide showed a very-small but statistically significant increase

in reaction-time.  In this experiment, the 60 experimental subjects were

of three different types; one subgroup of 20 subjects were patients with

"mild anemia;" a second 20 subjects suffered from emphysema? and the remaining

20 subjects were  in normal  health.   There was no difference in the changes

induced by carbon monoxide  in  these  subgroups.   In a second experiment,

Ramsey   exposed 20 healthy male subjects to 650 ppm for 45 min and another

20 subjects to  950 ppm  for  the same  time.   The carboxyhemoglobin concentrations

increased by 7.6% in the  first case  and  11.2% in the second.  Twenty other

subjects received only  pure air.  A  discriminative reaction-time task was

used in which the subject had  to respond  to various colored lights on different

buttons.  The increases in  reaction-time  shown by the two experimental groups

were both statistically significant.

    The interpretation of  many of the reaction time studies is difficult

because the experimental procedures  have not  been fully described.  Even

though stimulus intensity,  intertrial interval,  and the precise response

required  have  all long been known to be  important in react ion-time work-

                               OQO  QQA
they are frequently unspecified.   '

Time Discrimination and Estimation

                               f\ i

    In 1967 Beard and Wertheim  published an  account of an experiment in

which  they examined the effects of carbon  monoxide on the ability of human

subjects to discriminate between the lengths  of two tones.  In each case the

first  tone;which was 1 s longjserved as the standard.   The second tone was

presented one-half s later  and varied in duration from 0.675 s to 1.325 s.

The subject's task was to report whether the  second tone was shorter, longer

°r the same length as the first tone.  He  was given three buttons with which to

report his judgment.  The pairs were presented  in sets of 50 with approximately

7-5 s  elapsing between the  initiation of pairs.   Within each half of  this

set one-third of the comparison stimuli were identical to the standard, one-

third were longer and one-third were shorter.  These were scrambled in a

nonsystematic sequence.  It took 6 to 7 min to complete the 50 trials.  The

subject then had about 13 min during which he could either read, rest or

watch television while remaining in the small audiometric booth used as an

exposure chamber.  The next set of 50 trials was signaled by a warning light

and a tone.  This cycle of working approximately one-third of the time and

resting the other two-thirds was repeated 3 times per hour for 4 hours, so

the subject made a total of 600 judgments during each experimental session.

     Eighteen university students were each exposed 3 times to each of the

following carbon monoxide concentrations; 0, 50, 100, 175 and 250 ppm.  The

subjects were unaware of the concentrations being used but the experimenter

knew.  The experimenters presented their results in terms of "mean percent

correct responses."  Their analysis of greatest interest concerns the amount

of carbon monoxide exposure necessary to produce a marked performance decre-

ment.  Under control conditions subjects averaged about 78% correct at 0 ppm

(see Beard and Wertheim,3^ Figure 1), with a standard deviation of approximately

5 percentage points.  The authors presented data on how long an exposure to

various carbon monoxide levels was needed to produce a decrement in performance

more than two standard deviations in magnitude (Figure 5-22).   A decrease

from 78% correct to approximately 68% incorrect was produced by 50 ppm carbon

monoxide in about 90 minutes, by 100 ppm in about 50 minutes,  by 175 ppm in

about 32 minutes, and by 250 ppm in 23 minutes.  Note that the subject had

been in the booth and working for 30 minutes before the exposure began.


Carboxyhemoglobin levels were not given so they can only be estimated  at  this

time.  According to Coburn e£ al.82 these levels would have ranged from

approximately 2-1/2 to 4%.

     Beard and Wertheim3^ also plotted correct responses versus carbon

monoxide concentration in the booth.  They found a linear function with

subjects exposed to 250 ppm getting fewer than 30% correct, whereas control

subjects scored nearly 80% correct, these figures being derived from a two-

hour carbon monoxide exposure starting one hour after the subject had  been
                                        &c  _      f.
placed in the booth to start the task ana^one-ftali nour after exposure had begun.

Thus, they refer to Carboxyhemoglobin concentrations estimated to average 2%,

4%, 6% and 9% for the 50, 100, 175 and 250 ppm exposures, respectively.
     Although this experiment which showed deleterious effects from rather

short duration exposures to low carbon monoxide concentrations was reported

one      decade ago, only  three groups have attempted to replicate its  findings.

In each  case, the replication has been less than satisfactory.

     In  1970 Beard and Grandstaff33 reported that "the earlier work with

tone duration had been confirmed in 7 additional subjects."  No new data

were presented in their paper, however.  In 1971 O'Donnell e£ al.3*50 studied

the effects of overnight carbon monoxide exposure on a temporal discrimination

task patterned after that  used by Beard and Wertheim3.4  A 1-second standard

tone was delivered one-half second before a tone that varied in length between

0.675 and  1.325 s.  Neither the duration of the task nor the total number of

tone pairs was reported.   This task was given twice during a 1.5 hour battery

of tests that included 2 other tests of time-estimation, 2 tracking and moni-

toring tasks, and measures of critical flicker frequency, and mental arithmetic.

    tests were done  in  the morning  after  the subject had spent approximately

7-1/2 h  in a special chamber  exposed  to the carbon monoxide concentration

level at which he was being tested.   This was a double-blind study in which

neither  the subjects nor the  experimenters  knew which exposure was being

studied.  There were 4  subjects each  of whom produced data on 3 out of 9

nights spent in the  experimental chamber.   On the  first  4 nights the subjects

adapted  to the experimental procedures.   On the fifth night they were exposed

to carbon monoxide at either  75 ppm or 150  ppm.  On the  sixth and eighth

nights they went through all  the experimental procedures but were not exposed.

Data for these nights were not reported.  On the seventh night they were ex-

posed to the concentration alternate  to that used  on the fifth night and on

the  ninth night to zero concentration.  Because  the subjects went through

the  experimental conditions at either 75, 150,  0,  or 150, 75, 0 ppm, only the

two  higher carbon monoxide concentrations can be compared statistically.

The  carboxyhemoglobin concentrations  were 5.9% for the 75 ppm and 12.7% for

the  150 ppm exposure.  No difference  was  found between the subjects' temporal

discrimination scores for" the two  concentrations;  the means were 6.13±  1.53

at 5.9%  carboxyhemoglobin and 6.50 + 1.03 at 12.7%.

    The third experiment on  carbon monoxide's effects on auditory duration

discrimination was carried out by Stewart and co-workers.    The subjects were

exposed to carbon monoxide at different concentrations for varying lengths

°f time.  This was also a double-blind study.  Four different time estimation

methods were investigated.  One was an auditory  time-discrimination task

modeled after the one used by. Beard and Wertheim.34  Stewart e* al.380 ex-

amined performance on this task under three different conditions:  24 subjects

were tested while seated along one side of a table in an experimental room

measuring 20 x 20 x 8 ft (6 x 6 x 2.4m );5 subjects were tested one at a

time in the experimental room; and 9 subjects were tested in an audiometric

booth placed inside the larger experimental room.  Seventy-five pairs of

tones were presented in sets of 25 with 30-second pauses between the sets.

The time between the tone-pairs is not reported.  However, since the model

was Beard and Wertheim's experiment, it is assumed that similarly the pairs

were 7.5 s apart.  According to the authors, the task took approximately  15

min.  Other tasks probably took another 10 min to carry out.  Since the en-

tire task group was usually carried out once per hour during the carbon

monoxide exposure, the subjects were free to interact with one another for

at least part of every hour.  This situation was quite different from the

one in which Beard and Wertheim's subjects found themselves, alone in the

small booth working on their single task approximately one-third of the time

and resting the remaining time.

     Stewart et^ a^.    presented their results in two ways.  In one., they

combine all the pre-exposure data with the data collected after exposure  to

the control concentration, described as air containing carbon monoxide at

less than 2 ppm.  They then contrasted the score at this "base line1! concentratioa

with the mean score on the tone discrimination task at carboxyhemoglobin

concentrations as high as 20%, which were reached after exposure to various

carbon monoxide concentrations.  With this approach there was no apparent

effect.  This method of presentation does not take into consideration that

the subjects were being used as their own controls by producing both a pre-

test and a post-test for each carbon monoxide exposure.  Therefore the

experimenters could more precisely characterize each subject's reaction to

the different exposures.   Stewart  et  al.380 also presented this type of analysis,

not presenting means but presenting the results of t-tests.  The changes were

not statistically significant  for  the 2 measurement conditions in which the

subjects were not tested in  the  small booth.   The t-test was significant for

the 9 subjects who were tested in  the booth.   These subjects had a 2.9% mean

decrement in the number of correct responses.   With 8 degrees of freedom the

t-value of 2.75 was significant  at the 0.05 level.   The authors reported

that this decrement was associated with a  9.7% mean carboxyhemoglobin con-

centration.  Beard and Wertheim's3^ subjects  showed a much greater reduction

in performance at much lower carboxyhemoglobin concentrations.  However, the

results may not be comparable  because their experiment was not exactly


    Two other methods of  studying carbon  monoxide's effects on time percep-

tion have been used.  In one,  the  subject  was  requested to respond at speci-

fied regular intervals.  Beard and Grandstaff33 studied the subject's ability

to estimate the passage of either  10  or 30 s  in this way.   They reported that

exposure to as little as 50  ppm  for 64 min impaired the judgment of a 30-second

period.  Estimating a 10-second  period however,  was not modified by longer

exposures to as much as 250  ppm.   Not enough  experimental detail was reported

to evaluate the data, and  carboxyhemoglobin concentrations were not determined

                    o f\1
    O'Donnell et  al.    and Mikulka et al.279    studied the effects of low

carbon monoxide concentrations on  the ability  of men to space button presses

10 s apart.  The subjects  received no feedback concerning their accuracy.

Data are reported for 9 subjects, each exposed 3 times to 0, 50, or  125 ppm

for a 3-hour period.  The carboxyhemoglobin concentrations at the end  of  the

exposures were 1.0, 3.0 and 6.6%, respectively.  Time estimates were made for

3 minutes every half-hour.  This task was part of a 15-minute-long battery

that included some tests of tracking and ataxia.  It was a double-blind study

in that both the experimenters and the subjects were not informed about the

exposures used.  The mean time-estimates were found to be higher under the

influence of carbon monoxide.

     The experimenters carefully counterbalanced exposure to the three condi-

tions and did analyses of variance for each of the 6 time-periods.  Only  one

at 135 to 150 min after the beginning of exposure yielded a reliable difference.

There was also no apparent trend over time toward greater differences  between

the control and experimental conditions, which would be expected if the

carboxyhemoglobin concentrations were rising with carbon monoxide exposure.

This, plus the greater difference found with exposure to 50 ppm than to 125

ppm, led the investigators to conclude that they had not demonstrated  con-

sistent carbon monoxide effects on this type of performance.

     O'Donnell et al.    studied the discrimination of one-second tones

after overnight exposures and also studied discrimination of 10-second intervals.

Because they did not counterbalance zero exposure,which was always last^only

their results with exposure to 75 and 150 ppm can be compared.  These  were

completely negative.
     Stewart and co-workers   also investigated the effect of carbon monoxide

on the discrimination of 10-second time intervals.   Their general technique

is described above.  In the 10-second time-estimations, similar to those  for

30-second intervals,  the subject  held down a push-button for what he thought

was  the appropriate time.   The  mean of two such judgments was taken every hour

during  exposure.   The results for both the 10- and 30-second tests were negative.

     Stewart  and his  associates^'9,381 ftave examined carbon monoxide's effects

when the subject had  to  reproduce an interval of time by depressing a button


for  the same  length of time as  just-presented auditory or visual signal.  The

stimuli lasted either 1,  3, or  5  s,  and were presented in random order.


Three stimuli of each duration            a test.  The test lasted approximately


7 min.   In one experiment,   the  subjects were tested immediately after enter-

ing  the experimental  chamber after either 4 or 7 h exposure to carbon monoxide •,

or if the exposure was for  less than 4 h, during the last half-hour of  the

exposure.   There were no  apparent  effects at carboxyhemoglobin concentrations

as high as 25%.  In a later experiment in which carboxyhemoglobin concentrations

rose to 20%,  the results were only slightly less negative.-*'*  The earlier

study had been done with  the subjects working in one large room.  Part of the

later experiment was  carried out  under more controlled conditions.  The sub-

jects worked  either alone in a  large room or in a small audiometric booth.

This reduced  the influence  of factors extraneous to the study.   The data were

not  presented in the  published  paper.   The results of the t-tests were cited

that led the  researchers  to reject the possibility of a carbon monoxide effect.

Both experiments were conducted under double-blind conditions.

     \...t-.nou;>-. in Bea:on and  Wertheim's   original experiment the researchers

knew which exposure a subject received (it was not a double-blind study

    no  direct measurements

of the carboxyhemoglobin concentrations were made, this experiment's demonstration
of the effects of very low carbon monoxide concentrations on time discrimination
and estimation of behavior has not been refuted.  Neither has it been confirmed or
repeated by other workers.  However, since either very small or no effects on other
timing-behavior tests have been reported with carbon monoxide doses larger than
those used by Beard and Wertheim, the effect they found may have been due to aspects
of their task other than those associated with time perception.  Judging from the
frequent positive results reported by researchers studying vigilance (see above),
it is possible that Beard and Wertheim1 s use of isolated subjects who were given a
prolonged exposure to the tone-duration discrimination task was crucial to their
positive findings.  Both the research groups at Marquette University379*380'381
and at Wright-Patterson Air Field300'301 minimized boredom in their tiiaing~behavior
studies.  For instance, O'Donnell and co-workers301 reported that, "Following
90 min of exposure, the subject was allowed to walk around and stretch in order to
reduce the possibility of fatigue and boredom from being in a constant position
for 3 hr."  On the other hand, the Stanford researchers32>33,34 invariably mini-
mized external influences and ran their experiments for longer periods of time
(cf. 297), thereby turning them into vigilance experiments.  Thus, both groups may
be correct in what they are reporting, with external stimulation being the variable
that determines sensitivity to the effects of carbon monoxide.
Coordination and Tracking
     The conclusions of most of the coordination, steadiness, dexterity and track-
ing tests examined were negative.  Stewart et al.    used several tests of
manual dexterity in which, for example, subjects had to pick up small pins, place
them in little holes, and then put collars over the pins.  No effects were found,
even at carboxyhemoglobin concentrations of about 15%.  Changes in manual
dexterity were seen in a separate series of observations on two subjects when
concentrations of almost 30% were reached.  Wright et al.^49  also found
no effects on hand-steadiness at carboxyhemoglobin concentrations of 77..
Fodor and Winneke    also reported negative results on a series of coordination
tasks performed at carboxyhemoglobin concentrations estimated to be about 5%

(from  exposure  to  carbon monoxide  at  50 ppm for about 4-1/2 hours).

also reported no change  in similar task performance after doubling the exposure

level  to  100 ppm for  about the  same length of time.

                  OQ  OQ

    Bender _e_t  al .   '    reported some positive results with a coordination

task on which 42 students  were  tested.   Each subject was exposed to both 0 and

100 ppm.  The experimenter but  not the subjects was informed which exposure was

being  studied.  At carboxyhemoglobin  concentrations estimated to be about 7%

(2.5 h exposure to 100 ppm),  small but significant  decrements were observed

in how skillfully the subjects  worked on the Purdue peg board, on

which  pegs, bushings, and  rings had to be assembled.

    In 1929 Dorcus and  Weigandm reported that steadiness was  the only
characteristic tested  that  showed  any  apparent  effect after 5-hour exposures

to exhaust gas containing carbon monoxide  at  up to 400 ppm.    More recently,

in 1974, in a study of toll booth operators, Johnson et aJL.     found that a test

of eye-hand coordination correlated  significantly with the  increase in carboxy-

hemoglobin concentration produced  by exposure to automobile exhaust.

    O'Donnell e± al. 279»301 stu
had been carried out for longer than one minute.  In psychopharmacology

studies it has been found that prolonged task performances are more  likely

to be sensitive to change by chemical agents.°0,26

     Both O'Donnell and Mikulka279»301 studied a tracking task in which

subjects reacted with corrective movements to needle deflections, which

became more difficult to compensate with time.      The length of time the

subject could keep up with this increasingly difficult task was measured.

Nine nonsmoking male students were each exposed for 3 periods of 3 h duration

to 0, 50 and 125 ppm.  At the end of 3 h, the carboxyhemoglobin concentrations

were 1.0, 3.0 and 6.5%, respectively.  Although there was a significant

decrement in performance about half-way through the exposure, when the authors

examined the individual curves they concluded that the differences were not

statistically reliable.  Hanks^l using the same tracking task reported no

effects at up to 100 ppm for 4-1/2 h.  His report contains no data,  however.

     O'Donnell et_ al.301 gave their subjects the Pensacola Ataxia Battery when

they had completed the tracking task.  At carboxyhemoglobin concentrations

up to 6.6% no effects were observed on such measures of ataxia as standing

on one leg or walking a straight line both with closed eyes.

Sensory Processes

     Vision is the only sensory function that has received much attention.

The early literature abounds with case histories of the sequelae of  both

acute and chronic carbon monoxide poisoning^* but there have been surprisingly
few experimental studies.  During World War II, McFarland and co-workers

studied brightness discrimination.  They made very careful measurements with

a small  group  of well-trained subjects, thereby minimizing the variability

that  usually makes the detection of small differences difficult.  Figure  5-23

shows data  from an experiment that compared a simulated altitude of 4£94

  (15,400 ft)
meters, with various carboxyhemoglobin levels produced by graded doses of

carbon monoxide.   Also shown are the effects of administration of pure

oxygen and  a combination of 7%       and 93% carbon dioxide (carbogen).

Carbon monoxide was given via a face mask.   It is clear that even slight

increases in carboxyhemoglobin level, about 4.5%, increased the measured

visual threshold,  indicating that the subject's ability to distinguish an

increase in the intensity of a dimly lit field was diminished.  In 1970

Beard and Grandstaff-" reported that Wertheim found statistically significant

increases in the brightness thresholds of four young adults  and a decrease

in vernier  visual acuity at a carboxyhemoglobin concentration of 3% produced

by carbon monoxide exposure.

     Conversely Ramsey    failed to detect  any changes in brightness dis-

crimination in a study of 60 young adults,  20 of whom were exposed to enough

carbon monoxide to produce a carboxyhemoglobin concentration of about 8%,

20 to enough to produce a concentration of  about 12%, and 20 controls who

received only  air.   Using carbon monoxide concentrations that produced

carboxyhemoglobin percentages approximately 3.4% above those at the beginning

                          f o
of exposure, Wright et al^.    did not find any effect on several measures of

visual function;  night vision,  glare vision,  glare recovery, and depth percep-

tion.  A possible change in the speed of dark adaptation was studied in 1973

                           r)f\')  ")f\ti
by McFarland and  co-workers   '    and earlier in 1944 by Abramson and Heyman,2

with  negative  results  in both'experiments.   McFarland's study used carboxy-

hemoglobin  concentrations as  high as 17%.  However, none of these negative

                       4-5 "9-4   "15-8" 197
8-0  %COHb
                       90     120    150
                           Time (min)
Figure  5-23.  The effect of progressive increases in carboxyhemo-
globin percentage in blood on brightness discrimination thresholds.
Each point represents the mean of  10 measurements and the vertical
bars represent plus and minus one  standard deviation.  Carbon
monoxide was administered at times indicated by th*» horizontal Hnes
between arrowheads.  Reprinted with permission from Halperin et al

studies presented evidence of experimental control comparable to that shown

in the earlier work by McFarland and his colleagues.160>263»266

     Several investigators have examined the effects of low carbon monoxide

concentrations on critical flicker fusion frequency (CFF), the frequency

above which an intermittent light appears not to flicker.  In 1946 Lilienthal

and Fugitt    reported detecting the effects of carboxyhemoglobin concentrations

of 5 to 9% when subjects had also been kept at a simulated pressure correspond-

ing to an  altitude of 1,524 or 1,829 m  (5 or 6 thousand ft). No such effect was

observed at sea level,  even at carboxyhemoglobin concentrations as high as

17%.   The  actual data were not reported*rathertthe subjects'  performance

was characterized as  "depressed" or "constant1,1 which makes evaluation difficult.

In another study reported the same year^Vollmer et^ al_.     found no effect for

carboxyhemoglobin concentrations of 22% with subjects at simulated altitudes

of 4,724 m (15,500 ft).   Fodor and Winneke in 1972,125 Guest  e± al.  in 1970,156

O'Donnell  et al.  in 1971,30° Ramsey in 1973327 and Winneke in 1974445 all

reported finding no effect on CFF at carboxyhemoglobin concentrations ranging

from 10% to 12.7%.

     Besides the auditory vigilance and time perception studies discussed

above   ''     only one other study of the effects of small amounts of

carbon monoxide on auditory function has been reported.   In 1970 Guest

et al.I56  used  an auditory analogue of CFF,  an auditory "flutter fusion"

threshold.   Their subjects were required to  distinguish the point at which

an intermittently presented sound was no longer  heard as  intermittent.

They  found that a carboxyhemoglobin concentration of 10%  had  no effect on

this  threshold.

     The work by Guest et^ _al.156 is especially noteworthy in that it is one

of the few studies with carbon monoxide that included within its design

provision for gathering data on another agent with known effects, thereby

providing for some internal validation of the sensitivity of its procedures.

Complex Intellectual Behavior

     Dorcus and Weigand    used three tests of complex learned-behavior in

their 1929 study of the effects of automobile exhaust gas.  No effects were

found for carboxyhemoglobin concentrations up to 35%.  In 1963 SchulteJJJ

studied firemen carrying out a series of complex tasks, none of which is

described in detail.  He was able to produce large and regular changes.

For example, in one test subjects underlined all the plural nouns in

certain prose passages.  Between carboxyhemoglobin concentrations of 0 and

7% they averaged about 150 s to complete this task; between 10 and 15%

they averaged about 210 s.  Subjects averaged slightly less than 800 s to

complete an arithmetic test at concentrations of 8%, but took about 1,000 s

at 15%.

     Both Stewart j^t al.    and Mikulka et al.    have suggested that

Schulte's-"-* measurements of carboxyhemoglobin concentrations were probably

low since he reported a zero value under control conditions in a population

consisting mainly of smokers.  Both Guest et al.2^ and Stewart £t al.^81

have questioned the reported concentrations of 20% after exposures to carbon

monoxide at only 100 ppm.  Stewart £t al.381 suggested the possibility that

Schulte's    analytic techniques were unreliable.

     O'Donnell  es£ al.   ° also  examined  carbon monoxide's effects on ability

to do a  short series of mental arithmetic problems.   Their 4 subjects took

a mean of  89.8  s  (SE=10.56)  after  overnight  exposure to 75 ppm and 98.6S

(SE=11.52) after  exposure to 150 ppm.   The data from the control (0 ppm) measure-

ments are  not included  here  since  they  were  always taken during the last of

a series of three experimental sessions.   The difference between the scores

of the 4 subjects after the  2  different exposures (carboxyhemoglobin concen-

trations of 5.9%  and 12.7%)  is not reliable.

     Bender et^ jal.3°»39 investigated the effects of  a moderate amount: of

carbon monoxide on several complex tasks.  One of these was learning 10

meaningless syllables so that  they could be  recited  without error.   Exposure

to 100 ppm carbon monoxide for about 2.5 h (average  carboxyhemoglobin

concentration 7%)  produced a reliable decrease in accuracy.  Repeating a

series of  digits  in reverse  order  was also tested.  It too showed a reliable

decrease.  Negative results  were found  for several other tasks involving

calculation problems, analogies, shape  selection, dot counting, and  letter


     In view of the recent surge of interest among psychologists in human

perception, learning, and memory — all of which now go to make up cognitive

psychology — as  well as in  human  operant conditioning — it is surprising

that  more  work has not  been  done on these topics with carbon monoxide.  It

would be illuminating,  for instance, to see  what effects carbon monoxide has

upon  such  diverse behaviors  as simple decision making   and the complex per-

,                           126 139
tormance of aircraft pilots,   '     both of  which have been shown to be sensitive

to hypoxic hypoxia.


     It has long been known that when carboxyhemoglobin exceeds 45%, the

capacity to perform physical work is drastically reduced.  The subjects

studied by Chiodi et al.72 were unable to carry out tasks that required

low to moderate physical exertion when their carboxyhemoglobin was 40-45%.

 Attention has been directed recently to determining the influence of various

carboxyhemoglobin concentrations on maximal aerobic power (maximal oxygen

uptake*).  Most of the relatively small number of experiments carried out to

date have studied a limited population of healthy young males (see Table

5-10).  The majority of these studies have induced the requisite carboxy-

hemoglobin concentration in their subjects by first exposing them to a

relatively high concentration of carbon monoxide and then proceeding with

the maximal aerobic capacity tests while administering supplementary carbon

monoxide; alternatively it was assumed that the carboxyhemoglobin concentra-

tion remained unchanged during the test.  There have been only a few experi-

ments in which the subjects breathed carbon monoxide at the low ambient con-

centrations  encountered in urban areas.  Some studies failed to separate

the data from smoking and nonsmoking subjects.

     Within the carboxyhemoglobin concentration range of 5-35% there is a

linear relation between the decrease in aerobic capacity or power (V0?)

and the carboxyhemoglobin concentration (Figure 5-24).   This decrement
 Maximal aerobic power:  The highest oxygen uptake that an individual can

 attain during physical work breathing air at sea level (0 meters).

                                      TABLE 5-10

                  The Influence of the Presence of Carboxyhemoglobin
on Maximum Aerobic
Capacity (V09 max) *

Subjects were all male

S = smokers
; NS = nonsmokers

HbCO = Carboxyhemoglobin
8 (S,NS)
16 (S,NS)
10 (S,NS)
10 (S,NS)
7 (S)***
7 (S)***
10 (S)
10 (S)
10 (NS)
10 (NS)
9 (NS)***
9 (NS)***
4 (NS)
4 (NS)
Duration of
Max Test**
% Decrease
in V02 max
Bolus plus supplementation
Bolus plus supplementation
Bolus plus supplementation
Bolus plus supplementation
Bolus plus supplementation
Bolus plus supplementation
Bolus plus supplementation
50 ppm for duration 25 C
50 ppm for duration 35 C
50 ppm for duration 25 C
50 ppm for duration 35 C
50 ppm for duration 25 C
50 ppm for duration 35 C
50 ppm for duration 25 C
50 ppm for duration 35 C
75 ppm for duration 25 C
100 ppm for duration 25 C
Significant decreases observed at 4.8 and greater HbCO levels.

Duration of exercise time for NS to point of fatigue was consistently
in all tests when carbon monoxide was present in the ambient air.
   Middle-aged  subjects—all others  younger adults.


    40 r-
                                          % Decrease V0, m „ -  0.91 (% HbCO) + 2|
                                              ,         i max
                                          % HbCO
Figure 5-24.
          Relationship between percent carboxyhemoglobin and decrement

          in maximum aerobic power the linear regression
                             = °-91
                                                 + 2.2] obtained only
              from 5  to  36% carboxyhemoglobin.

in the maximum aerobic capacity can be predicted from the equation:

              % decrease  V02  max =0.91 [HbCO]  +2.2.

Aerobic power  was  not  significantly decreased in young nonsmokers until a

carboxyhemoglobin  concentration above 4% was attained.  In these experiments

the number  of  subjects studied was  small.  If greater numbers had been used,

some significant differences  would  probably have been noted at lower carboxy-

hemoglobin  concentrations.  There were considerable differences between the

responses of young smokers  and nonsmokers,  especially at lower carboxyhemo-

globin concentrations.  Even  though smokers had higher carboxyhemoglobin

concentrations than nonsmokers,  they often  did  not  show any decrease in

aerobic capacity.   The significance of these observations is not understood.

     Too few middle-aged  subjects were studied  to permit drawing conclusions.

                       OOQ  T9Q
In the studies reported"3   »J  * it was noted that these 41-56 year-old men

were not typical of the population.   Because of the stringent testing, subjects

were carefully screened to  eliminate those  with cardiovascular or pulmonary

disabilities.   Over half  the  subjects were  eliminated in the initial screening

process.  The  data represent  the responses  of a highly special population.

Therefore conclusions  cannot  be  extended to the general middle-aged population,

which has a risk factor related  to  the high incidence of cardiovascular disease.

Furthermore, even  though  middle-aged smokers did not  exhibit clinical signs of

cardiopulmonary disability, they had a much lower aerobic capacity (in filtered

air)  than would have been predicted on the  basis of age-related norms and so

again are different from  the .general population.  Changes in ambient temperature

(25-35 C) appear to have  only a  minimal influence on aerobic capacity at the

low carboxyhemoglobin  concentrations studied,    >328  other than that anticipated

°n the basis of an increased  thermal load.


     Recent research7 has indicated that 4.3% is a critical concentration at

which carboxyhemoglobin reduces maximum aerobic capacity.  This was also

accompanied by a reduction in total work time until maximum aerobic capacity

was reached.  Two procedures were used to raise the carboxyhemoglobin to

appropriate concentrations; a build-up method in which carboxyhemoglobin

was incrementally increased by administering ambient air containing carbon

monoxide at 75 or 100 ppm CO; and a high initial carbon monoxide exposure

followed by continued carbon monoxide inhalation to maintain the carboxy-

hemoglobin at the concentration reached in the build-up method, regardless

of the magnitude of ventilation.  The decrease in maximum aerobic capacity

was found to occur at the same carboxyhemoglobin concentration and was there-

fore independent of the procedure followed.  This observation indicated that

even low ambient carbon monoxide concentrations (23.7 ppm) would result in a

reduced maximum aerobic capacity if the individual had been previously

exposed to sufficient carbon monoxide to raise his carboxyhemoglobin concen-

tration to the critical value, 4.3%.  Clark and Coburn^ have suggested that

intracellular carbon monoxide effects may be responsible for the decrease in

aerobic capacity.

     The available data concerning the influence of various carboxyhemoglobin

concentrations on the ability of young males to perform light to moderate work

  * summarized in Table 5 -11.  The only observable effect is a slight increase

in heart-rate when the work is performed under conditions in which carboxy-

hemoglobin is increased.  Because in all but one of the studies there was

a minimal duration of effort (minutes) and relatively high percentages of

maximal capacity, it will be necessary to separate these factors in order to

                          TABLE  5-11

      The  Influence of Carboxyhemoglobin on the Capacity

                   to Perform Submaxlmal Work

                 S = Smokers; NS = Nonsmokers

                 HbCO = Carboxyhemoglobin
8 (S,NS)
8 (S,NS)
16 (S,NS)
16 (S,NS)
32 (S,NS)
24 (S,NS)
% HbCO
% Maximum
of Exer-
cise (min)V09 Uptake Reference
No change
No change
No change
No change
No change
No change
No change
No change
No change
No change
No change
These figures represent the work load at certain percentages
of the subjects'maximal aerobic capacity, i.e., submaximal work.

interpret the influence of higher carboxyhemoglobin concentrations on such


     Studies conducted with both young (22-26 yr) and older (45-55 yr) subjects,

smokers and nonsmokers, have indicated that at 35% of their maximal aerobic

capacity prolonged periods of activity (3.5 h within a 4-hour period) could

be performed with minimal changes in their physiologic responses, even though

carboxyhemoglobin concentrations were as high as 10.7 and 13.2%, for nonsmokers

                        1 3ft
and smokers respectively    (see Table 5-12).  However, heart rates were higher

during this activity of walking while breathing polluted air (ambient carbon

monoxide = 50, 75, or 100 ppm) than when walking while breathing carbon

monoxide-free air, (0 ppm).

     Previous investigations have indicated that a man could work at 35% of

his maximum aerobic capacity for a period of 8 h without evidence of physio-

logical stress,such as an increasing heart rate.  In this carbon monoxide

study, the heart rate began to increase after 2 ^indicating that the physio-

logic strain had started at an earlier time when carbon monoxide was present

in the ambient air.  Cardiac output remained constant but since there was an

increased heart rate, the heart beat volume decreased.  This alteration in

volume was enhanced when work and carbon monoxide exposures were conducted in

a warm environment (35 C).

     In summary» maximal aerobic capacity is readily affected even at fairly

low carboxyhemoglobin concentrations (5%), whereas submaximal efforts (30-

75% of maximum) can be carried out with minimal changes in efficiency, even

at relatively high carboxyhemoglobin concentrations (30%).

                                  TABLE  5-12

        Carboxyhemoglobin  Concentrations Prior  to  and at  Completion of

              4 Hours of Activity  (35% maximum  aerobic capacity)

                   at Different Concentrations of Ambient
                                                            1 O Q
               Carbon Monoxide  (2 Smokers and  2 Nonsmokers)
CO  (ppm)



Pre    Post

0.63   0.32

0.67   4.88

0.85  10.27

0.78  12.56
HbCO = Carboxyhemoglobin

       QO Uptake




Pre     Post
0? Uptake
% HbCO
liters /mil
 *Minute ventilation during walking periods averaged 17.93 liters.
  Minute ventilation during walking periods averaged 28.06 liters.

  Smokers had significantly smaller maximum aerobic capacity values, V02 max.



     Precise data on the potential scope of the problems caused by carbon

monoxide^residents and visitors at high altitude are not available.  There

are approximately 2.2 million people living at altitudes above 1,524 m  (5,000

ft)      -in the United States (see Table 5-13).  Most live in nine states,

with Colorado having the largest number.  The majority of high-altitude

residents  (95%) live at 1,524-2,134 m (5,000-7,000  ft).  The actual number of

people exposed to carbon monoxide at these altitudes may be much larger

because the tourist population in these states is high both in summer and

winter.  Since proper tuning of automobiles for high-altitude traveling is

uncommon,  the influx of visitors with cars^accompanied by an increase in the

emission of carbon monoxide and other contaminants, may be an important

factor in  increasing air pollution to an unacceptable point.

     Ambient air standards set at sea level are not applicable to high-

altitude sites.  The Environmental Protection Agency's primary standards

are expressed in milligrams per cubic meter of air.  In Denver, at 1,524 + m

 (5.,000 + ft),  each cubic meter contains about 18% less air than at sea

level.  Therefore, permissible concentrations of carbon monoxide in Denver

air would  be 18% higher  (a 10-mg/m^ maximal permissible 8-hour average is

equivalent to 8.7 ppm at sea level but 10.3 ppm at Denver's altitude)

(see Table 5-14).

     The effects of carbon monoxide and of hypoxia induced by high altitude

are similar.  Carbon monoxide produces effects that aggravate the oxygen

deficiency present at high altitudes.  When high altitude and carbon monoxide

exposures are combined,  the effects are apparently additive.  Each of

these •-a decrease in the partial pressure of oxygen in the air  and in-

creased carboxyhemoglobin- produce different physiologic  responses.   They

have different effects on the partial pressure of oxygen in the blood, on

the affinity of oxygen for hemoglobin, on the extent of oxyhemoglobin satura-

tion (carbon monoxide hypoxemia shifts the oxyhemoglobin dissociation curve

to the left, and a decrease in the alveolar oxygen partial pressure  shifts

it to the right), and on ventilation drive-

     The actual influence of a combination of increased carboxyhemoglobin

and decreased oxyhemoglobin has not been adequately documented by experi-

mental data.  The few available studies refer only to acute exposures to a

decreased oxygen partial pressure and an increased carbon monoxide partial

pressure.   The best available information on the additive nature of this

combination comes from psychophysiologic studies, but even they are inadequate.

There .   no data on the  effects of carbon monoxide on residents at high

altitudes or on their reactions when they are suddenly returned to sea level

and higher ambient carbon monoxide concentrations.

     McFarland £t al., °° in conjunction with their studies on the exposure

of young males to high altitudes,  showed that changes in visual threshold

took place at carboxyhemoglobin concentrations as low as 5% or in. a simulated

altitude of approximately 2,438 m (8,000 ft).  These observations were confirmed

by Halperin et^ al.f 16° who further observed that recovery from the detrimental

effects on visual function lagged behind the carbon monoxide elimination.

However,  there were very few   actual data given and neither the variability

among the four subjects  nor the day-to-day variations v'ere reported.  Vollmer

      / 1 Q                                               ' '
et_ a^.    studied the effects of carbon monoxide at simulated altitudes of

3,048 and 4,572 m  (10,000 and 15,000 ft)  and did not observe any additive

effects of carbon monoxide and altitude.  They suggested that carbon monoxide's

effects were masked by some compensatory mechanism.  The data reported were

not convincing.

     Lilienthal and Fugitt236 indicated that a combination of altitude

1,524 m (5,000ft) and  5-9% carboxyhemoglobin caused a decrease in flicker

fusion frequency, although either factor by itself had no effect.  They also

reported that a carboxyhemoglobin concentration of 8-10% reduced altitude


tolerance by about 1,219 m (4,000 ft). Forbes et al.   found that during light

activity at an altitude of 4,877 m (16,000 ft), carbon monoxide uptake was

increased.  This was probably the result of hyperventilation at high altitude

caused by the respiratory stimulus of  a decreased oxygen partial pressure.

Pitts and Pace    stated that if the subjects were at altitudes of 2,134 to 3,048

m (7,000-10,000 ft),   every 1% increase in carboxyhemoglobin (up to 13%) was

equivalent to a 108.2 m (355 ft) rise  in altitude.    Their observations were

based on changes in the heart-rate response to work.  These studies were

contaminated in general by such factors as poor control and the presence of

unidentified subjects who smoked.

     Two groups of investigators have  reported data comparing physiologic

responses to high altitude and carbon monoxide where the hypoxemia due to

altitude and the hypoxemia due to the  presence of carboxyhemoglobin were

approximately equivalent.  In one study,18 the mean carboxyhemoglobin con-

centration was approximately 12% (the method of carbon monoxide exposure

resulted in a carboxyhemoglobin variation of from 5 to 20%) and the altitude was

3,454 m (11,333  ft).  The second study9 compared responses at an altitude of

4,000 m (13,125  ft)    and a carboxyhemoglobin content of 20%.  In both these

studies,  the carboxyhemoglobin concentration greatly exceeded that anticipated

for typical ambient  pollution.   They both suggested,however,  that the effects

attributable to  carbon monoxide and altitude were equivalent.

    The  precise measurement of the possible additive effects of carbon

monoxide  exposures and altitudes  has not received much attention.  What

little  information is  available has been obtained by assuming simple additive

effects^'°  but these have not been verified by direct experiments.  In the

construction of  tunnels at  3,353  m (11,000 ft)  it was recommended, on theo-

retical grounds,  that  the ambient  carbon monoxide in the tunnel not exceed

25 ppm.      To elucidate this question research is needed following both

physiologic and  psychophysiologic  approaches.

                             TABLE  5-13

    Estimated U.S. Population Living at High Altitudes (9 States)

              1,524-2,134 m     2,134-2,743  m    2,743-4,572 m ^
              5,000-7,000 ft     7,000-9,000  ft    9,000-15,000 ft

      Urban     1,833,442         69,362.            4,314

      Rural*      321,100         16,770             1,150

      Total     2,154,542         86,132             5,464
       Figures based on ratio of rural to urban population.

       Total number living above 1,524 m  (5','000  ft)  = 2,246,140 or
       approximately 6.7% of the total population of these nine

       NOTE:  Total U.S. population in 1970 was 203,235,000.

                             TABLE 5-14

         Approximate Physiologically Equivalent Altitudes at

       Equilibrium with Ambient Carbon Monoxide Concentrations

Ambient                          Actual Altitude
Carbon Monoxide         ft        m    ft    m       ft      m
Concentration(ppm)'Q (sea level)  0    5,000  1,524  10,000  3,048

     Physiologically Equivalent Altitudes with Carboxyhemoglobin

  0               0 (sea level)  0    5,000  1,524  10,000  3,048

 25                  6,000     1,829   8,300  2,530  13,000  3,962

 50                 10,000     3,048  12,000  3,658  15,000  4,572

100                 12,300     3,749  15,300  4,663  18,000  5,486


     In the course of chronic or repeated carbon monoxide exposure over

periods of several weeks or months, a variety of structural and functional

changes develop.   Some of these changes serve to offset the impairment brought

about by carbon monoxide and thus can reasonably be taken to represent adaptation.

Other changes,  however, are of uncertain value to the organism, and a few are

frankly disadvantageous.  In this review, adaptive and nonadaptive changes

are considered  separately.   The reader should recognize, however, that the

separation is somewhat arbitrary and that further information may lead to re-

classification .


     The best evidence for  adaptation is that animals chronically or re-

peatedly exposed  to moderate concentrations of carbon monoxide can tolerate,with-

out apparent harm,  acute exposure to higher concentrations , which cause collapse

or death in animals not exposed previously.63,143,202,291,442  in addition^

animals chronically exposed to carbon monoxide develop tolerance for acute

exposure to simulated high  altitudes, as well as the converse.'"  This suggests

that common mechanisms are  involved.

     Whether there is symptomatic adaptation in man is much less clear; the

evidence consists mainly of anecdotal reports of industrial workers exposed

to unknown concentrations of carbon monoxide and of extensive studies con-

ducted 40 years ago by Killick^who  herself was the subject.203'20^  She

reported that in  the course of several months of 6-hour exposures at 5-8

day intervals to  carbon mono'xide at 110-450 ppm, slie became "acclimatized."

This was manifested by diminished symptoms and a smaller pulse rate response

during exposure.  She reported that after acclimatization, the equilibrium

carboxyhemoglobin concentration reached during exposure to any given inspired

carbon monoxide concentration was 30-50% lower than before.  Because "accli-

matized blood" equilibrated jin vitro showed normal relative affinities for

carbon monoxide and oxygen, and closed circuit breathing experiments appeared to

exclude the rapid metabolism of carbon monoxide,  Killick suggested that after

acclimatization,   carbon monoxide was actively transported from pulmonary

capillary blood to alveolar air.  It is unfortunate that this type of experi-

ment has not been repeated with human subjects using modern analytical teCh-

                                               AQ 99Q
niques.  Both the results of animal experiments^ '    and current evidence

that pulmonary gas exchange occurs by passive diffusion suggest that there

may have been a systematic technical error in Killick1s data.

     The mechanisms responsible for the development of symptomatic adapta-

tion are not fully understood, but it is clear that hematologic changes are

important.  Chronic or repeated exposure to carbon monoxide causes an increase

in both the hemoglobin concentration and the hematocrit (polycythemia) in a

variety of experimental animala17'48'63'76'106'143'202'229'290'291'389*442

In most studies there is a rough correspondence between the severity of the

carbon monoxide exposure and the extent of the polycythemic response.  Man

has a similar response but neither the threshold nor the time course has

been accurately quantitated.  Industrial workers exposed to high but un-

measured amounts of carbon monoxide have been found to be significantly

             1 R7
polycythemic.     Kjeldsen and Damgaard,205 in a study of eight healthy volunteers

exposed to 0.5% carbon monoxide intermittently for 8 days (mean carboxyhemo-

globin concentration was 13%), found no change in hemoglobin or hematocrit

values.   This suggests that-more severe or more prolonged exposure may be

necessary to elicit this response in man.   There is some evidence that cigarette

smokers  have higher hematocrits than nonsmokers, and in a recent survey  of  blood

donors,  hemoglobin concentration was correlated with carboxyhemoglobin concen-

tration  in both smokers and nonsmokers.

     An  increase in hemoglobin concentration increases the oxygen capacity

of the blood and improves oxygen transport to some extent.  The improvement

may be limited owing to the increase in blood viscosity that accompanies the

increased hematocrit.  °9

     A second possible hematologic adaptation involves 2,3-diphosphoglycerate

(2,3-DPG),  a phosphorylated by-product of  glycolysis that  is found in the

red blood cells of man and most other mammals.  An increase in the concentra-

tion of  2,3-DPG shifts the oxygen-hemoglobin equilibrium in the direction of

deoxygenation,   *    which lowers the effective oxygen affinity of hemoglobin

(shifts  the oxyhemoglobin dissociation curve to the right).   This shift is of

theoretical benefit during hypoxic stress  because oxygen is "unloaded" into

the tissues with a smaller drop in capillary oxygen partial pressure than

would be possible  with the normal dissociation curve.

     Red cell 2,3-DPG  concentrations are increased and the dissociation curve

is shifted to the  right in anemia and during residence at  high altitudes.199,225,306

Several  investigators  have looked for a  similar effect from carbon monoxide ex-

posure^with inconclusive results.  Dinman  et al.108 reported small increases

in 2,3-DPG in humans after 3 h at approximately 20% carboxyhemoglobin and in

rats after  exposure to variable higher concentrations of carbon monoxide.

Conversely,  Astrup16 found a small decrease in red cell 2,3-DPG in human

subjects maintained with 20% carboxyhemoglobin for 24 h.   Radford and

Kresin324 found that 2,3-DPG concentrations in rats were unaffected by 2 or

24 h exposures to carbon monoxide at 500 and 1,000 ppm.  Mulhausen et al.

studied blood from human subjects maintained for 8 days at an average carboxy-

hemoglobin concentration of 13% and found no change in the dissociation curve

from that predicted from the immediate carbon monoxide effect.  A shift

of the dissociation curve does not appear to be an important adaptation to

carbon monoxide exposure.

     Except for Killick's20^ observation that her pulse rate at any given

carboxyhemoglobin concentration was slower after acclimatization, there is

little or no information about possible adaptation of the cardiovascular

system.  Data are not available about whether tissue capillarity  increases

with prolonged carbon monoxide exposure, as it does during high altitude

residence. 03,39.)  Muscle myoglobin concentration especially in the heart

has been shown to increase at high altitude. '     Although a similar in-

crease might be expected owing to prolonged carbon monoxide exposure, no

measurements have been made.

     Montgomery and Rubin OJ in 1973 reported an example of adaptation.  In

a 1971 study    they demonstrated that exposing rats briefly (90  min) to

carbon monoxide at 250-3,000 ppm resulted in a concentration-related slowing

of the in vivo metabolism of certain drugs and a prolongation of  their pharma-

cological effects.  Hypoxia produced similar effects.  When carbon monoxide

exposure was prolonged the effect on drug metabolism became less  pronounced,

and by 24 h had almost disappeared.  Similar adaptation occurred  during ex-

posure to low inspired oxygen when hypocapnia was prevented.  The mechanism

of the adaptation is presently unknown.

Detrimental Effects of Chronic Exposure

     The general category of chronic exposure effects comprises those that

result from prolonged or repeated carbon monoxide.exposure, but which are not

caused by acute exposure to carbon monoxide concentrations in the same range.

Most  of the effects discussed are irreversible or slowly reversible.  The

effects that do not appear to be of benefit to the organism—and thus are not taken

to represent adaptation—are in this section, but whether or not an effect is

beneficial is questionable in some cases.

     Prolonged exposure to carbon monoxide concentrations higher than en-

countered in ambient air pollution in the community environment has been

                                               ft 1  7fi 90?
shown to retard growth in experimental animals.   »»     The mechanism of

growth retardation has not been extensively studied, but reduced food intake

may be involved.      The effects of carbon monoxide on fertility and on fetal

development are reviewed elsewhere in this chapter.

     A syndrome of chronic carbon monoxide intoxication has been described

by several authors,35'135'155 but is far from being established.

A wide variety of symptoms - such as: weakness, periodic loss of

consciousness with twitching,  insomnia,  personality changes, loss of libido,

and clinical and hematologic changes similar to those of pernicious anemia

have  all been attributed to chronic or repeated carbon monoxide exposure.

Dogs  exposed intermittently to carbon monoxide at  100 ppm during 11 weeks

developed a broad-based gait and other subtle neurologic abnormalities  "

but since there were no controls the results are questionable.

     Cardiac enlargement,  first reported in carbon monoxide-exposed mice

"lore  than 40 years ago,0^  has recently received renewed attention.  Theodore

et al.    reported an increase in heart weight in rats exposed at total

pressure of 5 psi to 460 mg/nr* of carbon monoxide for 71 days, followed by

575 mg/nP for 97 days.  These unusual exposure conditions resulted  in carboxy-

hemoglobin concentrations of 33-39% in dogs simultaneously exposed  with the rats,

                                                                     O 1 Q ^11

but carboxyhemoglobin was not measured in the rats.    Penney et al.    »

studied rats continuously exposed at sea level pressure to carbon monoxide

at 500 ppm (41% carboxyhemoglobin) and found that heart weight was  significantly

increased within a few days.  After 2 weeks the heart weight was 35-40%

greater than the value for control animals of the same body weight.  Both

the right and left ventricles were enlarged after 11 weeks of exposure.

This  contrasts with the predominance of right ventricular enlargement in

rats  exposed at high altitude.  Heart weight was also increased in  rats

exposed to carbon monoxide for 30 days at 200 ppm (16% carboxyhemoglobin).

In agreement with earlier reports, 118,290,384 anjjjiais that were exposed for

46 days to 100 ppm  (9% carboxyhemoglobin) did not develop significant

cardiac enlargement.

      Histologic examination of myocardial tissue from animals exposed to

90-100 ppm of carbon monoxide for long periods has revealed edema,  degenera-

tion  of muscle fibers and fibrosis.119'421  Kjeldsen et^ al.206 in 1974 de-

scribed a variety of ultrastruetural changes in the hearts of rabbits exposed

to carbon monoxide at 180 ppm (17% carboxyhemoglobin) for 2 weeks.   The func-

tional significance of these anatomic changes is not known.

      During the past several years, Astrup e£ al.   have  studied  the influence

of chronic carbon monoxide exposure on vessel walls.  In cholesterol-fed

rabbits, 8 weeks'  exposure to carbon monoxide at 170 ppm (15-20% carboxyhemo-

globin), followed by 2 weeks' exposure to carbon monoxide at 350 ppm (33%

carboxyhemoglobin), increased the cholesterol content of the aorta and

caused subendothelial edema.  Similar effects were produced in normally fed

monkeys exposed to carbon monoxide at 250 ppm (21% carboxyhemoglobin) for

2 weeks.     The significance of this observation in the pathogenesis of

human vascular disease remains to be determined.

Significance for Human Health

     Nearly all the available data on the effects of chronic CO exposure are

derived from animal experiments.  This is true both of adaptive changes and

of effects that do not seem to be of benefit.  Whereas many of these effects

are of considerable interest, it is not justifiable at present to conclude

that human beings are similarly affected.   Existing data neither establish

nor disprove a significant influence of chronic CO exposure on human health.


     This section summarizes present knowledge about the relationship of dose-

response to adverse effects in man of acute carbon monoxide exposure.

Threshold for Adverse Carbon Monoxide Effects

     Whether there is a threshold carboxyhemoglobin concentration for an

adverse effect is still unknown.  The question is of practical importance

in setting carbon monoxide air standards.  If there are adverse carbon monoxide

effects at any carboxyhemoglobin concentration (no threshold), such effects

could not be entirely prevented by legislation.  The mechanism for

adverse carbon monoxide effects is a fall in capillary oxygen partial pressure

(pO^) due to carbon monoxide binding to hemoglobin^  a pertinent question then

is whether any fall in capillary pO£, no matter how small, results  in an adverse

effect on tissues.  It is known that many tissues, in order to keep intracellular

pO2 nearly constant, can adapt to acute falls in arterial pO,, with  resultant falls

in capillary pO^.  The major  adaptation mechanism in many tissues  is probably

recruitment of capillaries to give a decrease in^  diffusion distance  between


capillary blood  and mitochondria.  If such a mechanism occurs as carboxyhemoglobia

increases» it is  unlikely that adverse carbon monoxide effects occur at  carboxy-

hemoglobin concentrations near zero, and more probable that a threshold exists at

a carboxyhemoglobin  concentration where adaptation can not compensate.

     As indicated earlier  in this chapter, the tissues most sensitive  to the

adverse effect of carbon monoxide appear to be heart, brain and exercising

skeletal muscle.  Evidence has  been obtained  that  carboxyhemoglobin concen-

trations in the  3-5% saturation range may adversely affect the ability  to

                                                             15 33  125  153
detect small unpredictable environmental changes (vigilance). ^»JJ»   •*»


There is evidence that acute increases of carboxyhemoglobin^above 4-5%

in patients with cardiovascular disease can exacerbate their symptoms »••••*»

when the carboxyhemoglobin is as low as 5%.     Maximum oxygen consumption

in exercising healthy young males has been shown to decrease when the carboxy-

hemoglobin is as low as 5%.1^^  In the studies of the effect of carbon monoxide

on vigilance and cardiovascular symptoms, there was no attempt either to

determine the effect of lower carboxyhemoglobin concentrations or to look for

a threshold.   When the aerobic metabolism of exercising skeletal muscle was

studied an apparent threshold was found.  At a carboxyhemoglobin concentration

below 5%,  a measurable effect on oxygen uptake could not be demonstrated.

Dose-Response Relationships in Man

     In recent yearSj the direction of research has been to look for adverse

effects at low carboxyhemoglobin concentrations.   Little effort has been made

to investigate dose-response relationships at carboxyhemoglobin concentrations

higher than those demonstrated to have an adverse effect.  It  is important to

define dose-response relationships in order to determine whether an increase

in carboxyhemoglobin will have the same adverse effect in a subject with a

normal carboxyhemoglobin as it does in a  subject  with a higher baseline

carboxyhemoglobin concentration, such as a smoker  or a patient  with an in-

creased rate of endogenous carbon monoxide production.  If the dose-response

curve is concave upward,  a given incremental increase in carboxyhemoglobin will

have a greater adverse effect on a subject with higher baseline carboxyhemo-

globin than on! a subject  with a normal one.   Such a subject would have a

greater risk of experiencing adverse effects from environmental carbon

monoxide.  The research mentioned above that showed the effects of sudden

small increases in carboxyhemoglobin on vigilance and on the myocardium

did not investigate the adverse effects over a wide range of carboxyhemoglobin

concentrations.  In the study of the effect of increasing carboxyhemoglobin on

aerobic metabolism in exercising muscle  (referred to above), it has been demon-

strated that over a concentration range  of 5 to 30%, there  was an almost  linear

relationship between carboxyhemoglobin and the fall in maximal oxygen uptake.

This  dose-response relationship cannot be extrapolated to carboxyhemoglobin

effects on other tissues, since under conditions of exercise  at  maximum

oxygen uptake  it is unlikely  that adaptation to carboxyhemoglobin  could  occur;

 the maximum recruitment  of  capillaries has probably already taken  place  because

 of exercise.

      It  is likely that  dose-response relationships  for  the  adverse effect of

 carbon monoxide on a given tissue are different in the presence of disease

 or in the fetus.  Mechanisms of adaptation to  carbon monoxide hypoxia may be

 altered or the tissue may be functioning under borderline hypoxia conditions

 andj theref ore} be more susceptible to the effects  of increased carboxyhemoglobin.

 Dose-response relationships may also be different when there is acute rather

 than chronic carbon monoxide exposure.

      At the present time we know that when there  are sudden increases in

 carboxyhemoglobin concentrations exceeding 20%,  there are gross adverse effects

 on the functioning of several organ systems.   Conversely,  as discussed above,

 there is evidence of adverse effects on brain, myocardium,  and skeletal muscle

 function at relatively low carboxyhemoglobin.   But virtually no data are available

 in man for other dose-response relationships.   Additional information about

 threshold carboxyhemoglobin concentrations and dose-response relationships for

various  tissues in normal and carbon monoxide-susceptible populations would be

valuable  in understanding the biologic effects of carbon monoxide  on man.


Ambient Air  Standards  for  Carbon Monoxide

    The  current  EPA standard  for carbon monoxide is 9 ppm maximum for 8 hours

average exposure,  or 35  ppm maximum for  one  hour average exposure.  Approximate

calculated carbon monoxide uptakes for varying levels of activity after exposure

to  these  concentrations  are given below.


                        Resting            Moderate Activity          Activity

 9  ppm 8  hours           1.3%  sat                 1.4% sat              1.4% sat

35  ppm 1  hour            1.3%                      2.2%                  2.9%

These  [HbCO] are  calculated with the Coburn  Forster Kane equation, using

appropriate  values for carbon  monoxide diffusing capacity, alveolar ventilation,

alveolar  pC^ and  endogenous carbon monoxide  production for resting,

moderate  or  heavy activity* The current EPA standard is mainly justified


on the basis of adverse  carbon monoxide  effects in patients with cardiac

and peripheral  vascular  disease  and effects of carbon monoxide on oxygenation

of skeletal  muscles in exercising normal human subjects.  There appears

to be  an  adequate safety factor between  the  lowest carboxyhemoglobin concentration

which  has been demonstrated to cause adverse effects and  the maximal carboxy-

hemoglobin concentration which can occur at  9 ppm carbon monoxide  for 8 hours

or 35  ppm for  one hour.   However, the existing data base on the adverse effects

of carbon monoxide exposure is not adequate  to allow a precise setting of the

carbon monoxide concentrations in ambient air due to the uncertainties discussed

 ..                                        that.

throughout this report,  and it is probable as more information becomes available,

there  can be justification for altering the  present standards.

                                   CHAPTER 6

                        EFFECTS ON BACTERIA AND PLANTS


     The limited data available indicate that relatively high carbon monoxide

concentrations can have an effect on certain airborne and soil bacteria.  But

there is no evidence that concentrations normally found in polluted atmospheres

have an effect.  Lighthart    studied carbon monoxide's effect on the survival

of airborne bacteria.  Vegetative cells of Serratia marcescens 8 UK, Sarcina

lutea, and the spores of Bacillus subtilis var. niger were held in aerosols

at 15 C and exposed to 85 ppm for up to 6 hours.  The relative humidity (RH)

was varied from 1 to 95%.  At 88% RH and above, carbon monoxide appeared to

protect the cells of
     The luminescence of some strains of aquatic bacteria has been reported

to be affected by carbon monoxide.3^  Suspensions of bacteria spotted on

luminescence sensor discs and exposed to atmospheres containing carbon

monoxide gave a detectable response with concentrations as low as 3 ppm.

     Lind and Wilson237 reported that nitrogen fixation by Azotobacter

veinlandii (a free-living nitrogen-fixing bacterium) was inhibited by carbon

monoxide concentrations from 1,000 to 2,000 ppm and totally suppressed by

concentrations from 5,000 to 6,000 ppm.  Nitrogen fixation by Nostoc muscorum

was inhibited by 1,000 ppm and it approached complete inhibition with concen-

trations of 2,500 ppm.    Bergersen and Turner^ reported that exposures to

86 ppm suppressed the nitrogen fixation of bacterial suspensions of Rhizobium

japonicum obtained from soy bean root nodules.  Higher concentrations were

required to inhibit nitrogen fixation by the intact nodules.

     Soils exposed to high carbon monoxide concentrations develop the ability

to convert it more rapidly. °*  This may be due to changes in the population

of carbon monoxide-assimilating organisms in the soil and indicates that

carbon monoxide could have an effect on the ecology of some soil organisms.

             ") 10          181
Both bacteria^1" and fungi  * are known to be able to oxidize it.


     Plants are relatively resistant to carbon monoxide.  Much higher concen-

trations are required to cause injury or growth abnormalities than for pollutants

such as sulfur dioxide, ozone and hydrogen fluoride.  For this reason data

concerning its effects on plants are extremely limited and most of the studies

have been conducted with much higher concentrations than those to which plants

are exposed in nature.  Possible damage to vegetation could be caused in three

ways:  the production of leaf injury or growth abnormalities which reduce yield,

growth or quality; the suppression of nitrogen fixation in the soil or root

nodule resulting in a deficiency of nitrogen for plant growth; and suppression

in the rate of photosynthesis over a sufficiently long time-period to cause

a significant reduction in the plant's growth rate.

     The available data indicate that carbon monoxide does not cause visible

effects on plants at concentrations found in the ambient air but at high con-

centrations it can produce various abnormalities.  Knight e_t al.    measured

a growth suppression of the etiolated epicotyl of sweet pea at 5,000 ppm.

From this study they showed that ethylene^ and not carbon monoxide , was the

major toxic component of smoke.  Zimmerman et^ _al.  °»  y conducted extensive

studies of carbon monoxide effects on plants.  Their interest was not in carbon

monoxide as an air pollutant but rather in finding a chemical that could induce

root initiation and stimulate the growth of other plant parts.  They exposed

over 100 species to artificial atmospheres containing high carbon monoxide

concentrations for periods of up to 23 days (most of the data presented are

for 10,000 ppm).  At 10,000 ppm they found a growth reduction in a number of

the species.   But the species varied widely both in their susceptibility to

carbon monoxide and in their symptom expression.   The most important responses

observed at these high concentrations were leaf epinasty and hyponasty,

leaf chlorosis; stimulation of the abscission of  leaves, flower buds,  and

fruits; hypertrophied tissue on stems and roots;  retardation of stem growth;

reduction of  leaf size;  initiation of adventitious roots from young stem or

leaf tissue;  and modification of the natural response to gravity, causing the

roots to grow upward out of the soil.  Minina e_t  ad. °" and the Heslop-
Harrisons   found that 'cucumbers and hemp exposed during critical stages of

development to carbon monoxide at 10,000 ppm induced a modification of sex

expression, even to the point of total sex reversal in genetically male hemp

plants.  McMillan et^ al.    reported that a concentration of 20,000 ppm for

24 hours caused up to 100% leaf drop in certain geographic variants of Acacia

farneniana.  They did not find this response in the two other acacia species


     It has been shown that high carbon monoxide concentrations can inhibit

nitrogen fixation of red clover plants and soy bean root nodules   '    but

the available data do not indicate that ambient concentrations would have an

adverse effect.  Lind et^ al_-    exposed red clover inoculated with Rhizobium

trifolii to carbon monoxide for periods of up to one month.  No effect was

observed at 50 ppm, there was a 20% reduction in nitrate production at 100 ppm,

andat 500 ppm nitrogen fixation was essentially halted.  Carbon monoxide com-

bines rapidly with the root nodule hemoglobin (leghemoglobin) at a rate 20

times faster than it combines with myoglobin,    and apparently inhibits oxygen

transport to the interior of the nodule.    Leghemoglobin facilitates oxygen

diffusion into the interior of the nodule.     The oxygen concentration appears

to be important for nitrogen fixation.^3

     Carbon monoxide also can limit bacterial nitrate production through

inhibiting the enzymatic process of nitrogen fixation.  Free-living nitrogen-

fixing bacteria and bacteria isolated from soy bean root nodules can be inhibited

by high concentrations^9*237 but Bergersen et _aJL.44 reported that higher con-

centrations were required to inhibit an intact soy bean root nodule than

     to inhibit a bacterial suspension prepared from nodules.  Inhibition of

nitrogen fixation by carbon monoxide has in some cases been reported to be

competitive52'245 and in others to be noncompetitive.   '

     Bidwell and Fraser   reported a carbon monoxide-induced growth suppres-

sion.  In their studies on carbon monoxide uptake by detached leaves they

measured a reversable reduction in photosynthesis in several plant species

exposed to relatively low concentrations.  For example, there was strong

inhibition of the net carbon dioxide uptake rate (a measure of the growth

rate) of grapefruit at a concentration of 1.6 ppm and complete inhibition

at 7 ppm.  These data indicate that photosynthesis might be inhibited at

concentrations commonly measured in the atmosphere.  If this is true,

carbon monoxide could be one of the major pollutants measured in ambient

air in or near large cities responsible for the suppression of plant

growth.177,274,391,398  These results reported by Bidwell and Fraser

have not been confirmed and additional studies should be conducted.

                                  CHAPTER 7

                          SUMMARY AND CONCLUSIONS

Properties and Reactions

     Carbon monoxide is a colorless, odorless gas that is chemically stable  at

ordinary temperature and pressure.  Its molecule is heteropolar, diatomic, and

diamagnetic.  It has a low electric dipole moment, short interatomic distance,

and high heat of formation from atoms.  These characteristics suggest that the

molecule is a resonance hybrid  of three valence structures.  The poisonous

nature of carbon monoxide is due to the strength of the coordination bond formed

with the iron atom, which is stronger than that of oxygen in protoheme  (a ferrous

ion complex of protoporphyrin IX that constitutes part of the hemeprotein molecules

that bind carbon monoxide).

     Chemical reactions of carbon monoxide are critical to its formation or  loss

in the atmosphere.  Its reactions with oxygen, water, nitrogen dioxide, ozone,

atomic hydrogen, and organic free radicals are probably not important.  A number

of reactions with unstable intermediates, such as atomic oxygen and hydroxyl

radicals, are important in atmospheric carbon monoxide chemistry.  It is likely

that reactions with atomic oxygen in the upper atmosphere play a role in oxidizing

carbon monoxide.  Hydroxyl radicals are generated in the atmosphere, particularly

in air heavily polluted by automobile exhausts and in the natural atmosphere.  The

major reaction generating hydroxyl radicals is believed to be solar photolysis of

aldehydes.   Carbon monoxide  is also formed during solar photolysis of aldehydes,

but this is not an important source, compared with carbon monoxide production from

internal combustion of gasoline-air mixtures in the automobile.  The generation of

hydroxyl radicals is important, because they are known to react rapidly with carbon

monoxide at ordinary temperatures.  However, such trace contaminants as hydro-

carbons, sulfur oxides, and nitrogen oxides will compete successfully  for  avail-

able hydroxyl radicals and thus reduce the significance of reaction with carbon

monoxide.  Theoretically, the rate of conversion of nitric oxide  to nitrogen

dioxide by hydroperoxyl radicals could be affected by ambient carbon monoxide con-

centrations, but carbon monoxide concentrations are affected to only a very small

extent by nitric oxide conversion.  Hydroxyl radical reaction with methane pro-

duces carbon monoxide, and this may be the most important natural source of carbon


Sources and Sinks

     There is evidence that perhaps 10 times as much carbon monoxide is formed

by natural processes as by processes related to the activities of man.  The anthropo-

genic sources are due to incomplete combustion processes.  The total anthropogenic

carbon monoxide emission is much greater than the total anthropogenic  emission of

all other criteria pollutants.  Anthropogenic carbon monoxide emission is  estimated

to have increased by about 28% in the period 1966-1970.  Global carbon monoxide

emission from combustion sources was estimated at 359 million metric tons  (tonnes)

in 1970.  Of the total anthropogenic carbon monoxide formed, the  internal-combustion

engine contributes by far the largest fraction.  There was a great increase in

carbon monoxide emission from 1940 to 1968, paralleling the increase in motor

vehicles; since 1968, the emission has decreased, owing to installation of emission

control devices.  Other sources include industrial processes, agricultural burning,

fuel combustion in stationary sources, and solid-waste disposal.  It is estimated

that 95% of the global anthropogenic carbon monoxide emission originates in the

northern hemisphere.

     Natural sources of carbon monoxide include volcanic activity, natural

gases, photochemical degradation of such organic compounds as aldehydes,

methane oxidation, and possibly solar photodissociation of carbon dioxide at

high altitudes.  Calculations of carbon monoxide formation by methane oxidation

(based on estimations  of hydroxyl radical concentration in the troposphere air)

suggest this reaction contributes about one-fourth as much carbon monoxide as

man-made sources.  The oceans are also a significant source of carbon monoxide.

Endogenous carbon monoxide production by man and animals is probably insignificant,

compared with other carbon monoxide sources, in terms of total global carbon monoxide


     Average global background concentration  of 0,1 ppm carbon monoxide in iir

                                    removal rate.  ,       ,    ,          . ,     n
reflect a balance between formation and/        Background carbon monoxide as low

as 0.025 ppm is  found in northern Pacific air and in the range of 0.04-0.08 ppm

in nonurban air  in California.  Background carbon monoxide concentrations in

unpolluted areas reflect the history of the air mass.  The residence time for

carbon monoxide  in the atmosphere has been crudely estimated at approximately

0.2 yr.

     If there were no removal of carbon monoxide the average atmospheric concen-

tration would increase at the rate of 0.06 to 0.5 ppm/yr.  The sinks include oxida-

tion by hydroxyl radicals (probably the major sink); biologic sinks, such as

microorganisms in soil, vegetation, and metabolism in animals; and removal at the

surfaces of such materials as charcoal and carbon.

Environmental Analysis and Monitoring

     There are many problems in the monitoring of any atmospheric pollutant that

are related to variation in concentration and to analytic error.   Carbon monoxide

concentration can be shown to be variable on the time and space scale of the

smallest atmospheric eddies and on all larger scales.  Furthermore, for almost

any pollutant, a monitoring device may well not record the same concentrations

as are present at a receptor a few meters away.  Carbon monoxide is taken up by

the lungs of man very slowly; after an increase in inspired carbon monoxide con-

centration, it takes nearly 24 h to reach an "equilibrium" blood carboxyhemoglobin

content.  The same is true after a decrease in inspired carbon monoxide concentra-

tion.  Because carbon monoxide concentrations in a typical urban environment are

variable in all three spatial dimensions, as well as in time, exposure of a typical

highly mobile urban dweller to carbon monoxide will vary greatly in the course of
                                              effects on           the
a day's activity, and the relationship between the/human health and concentrations

measured at monitoring stations is complex.  These complexities are partly responsi-
ble for the dearth of /    epidemiologic data on the health effects of chronic

carbon monoxide exposure.  The situation is exacerbated by the general tendency of

pollutants to vary together, owing to meteorologic factors.

     Carbon monoxide concentrations in cities exceed background concentrations

by at least an order of magnitude.  Seasonal differences are small.  Diurnal con-

centration follows diurnal traffic patterns and tends to peak during morning and

evening rush hours.

                               Concentrations decrease steeply with increasing

altitude and are also affected by persistent air circulation patterns.  Mathe-

matical models, have led to    some success in computing concentrations at loca-

tions not covered by monitoring stations.  The problem is made more difficult by

the absence of a general theory of optimal placement of monitoring devices; anthro-

pogenic emission is invariably concentrated, whereas natural emission is usually


     Carbon monoxide concentrations normally found in open air influence man

and animals on a rather long-term.  Since the turnover time or exchange rate,

or biological half-life are in hours in man, selection of brief measuring

periods (averaging times) is not necessary.

     There are a number of analytic methods for atmospheric carbon monoxide,

but only the nondispersive infrared analyzers and the electrochemical analyzer

(Ecolyzer) are in wide use.  The method consisting of catalytic reduction of

carbon monoxide to methane and gas chromatography with flame ionization detec-

tion is specific and sensitive, but it is inherently discontinuous, and most

commercial units based on this method have not been highly dependable.  Non-

dispersive infrared analyzers operate continuously, but they are significantly

less sensitive and somewhat less specific.  Specificity problems are reduced,

but apparently not totally controlled, by gas pretreatment systems.  Some of

the data generated with the nondispersive infrared analyzers have been unreli-

able.  Electrochemical methods (Ecolyzer) of monitoring carbon monoxide appear

to be highly satisfactory in regard to reliability and precision.

     Although blood carboxyhemoglobin is an almost unique biologic indicator of

exposure to carbon monoxide air pollution, its accurate measurement at low

concentrations is not easy and requires well-trained personnel.  In addition,

of course, if the plan is to sample the general public for carbon exposure,

there is a practical limitation on the willingness of people to cooperate.   For

these reasons, use of blood samples for routine monitoring of carbon monoxide

exposure of the general public is considered to be impractical.

     The efficacy of carbon monoxide monitoring networks as indexes of the

human health effects of this pollutant has not been demonstrated.  It would be

ideal to compare uptake,  as determined by blood carboxyhemoglobin measure-

ments, with measurements of carbon monoxide in air.

Effects on Man and Animals

     Carbon monoxide in the body may come from two sources:  it may be endogenous,

owing to the breakdown of hemoglobin and other heme-containing pigments; and it

may be exogenous, owing to the inhalation of carbon monoxide.  Endogenous carbon

monoxide results in blood carboxyhemoglobin saturation of approximately 0.4%  in

a normal human.  Uptake of exogenous carbon monoxide adds to this value.  The

process of uptake of exogenous carbon monoxide consists of inhalation, increase

in alveolar carbon monoxide concentration, and diffusion through the pulmonary

membrane and into the blood.  Generally, the rates of diffusion and ventilation

limit carbon monoxide uptake.  Equations developed to describe carbon monoxide

uptake in the lung include the following quantities:  diffusing capactiy of the

lung, alveolar ventilation, oxygen tension in pulmonary capillary blood, and pul-

monary capillary blood volume.  Because the carbon monoxide diffusing capacity and

pulmonary capillary blood volume vary with age and exercise, carbon monoxide up-

take at a given inspired carbon monoxide concentration also varies with age and

exercise.  Body size also influences total body hemoglobin; with a larger pool of hemo-

globin available for carbon monoxide binding, the rate of increase in carboxy-

hemogloMn will be smaller at a constant pulmonary uptake.  Obviously, the health

of the lung is important:  with decreased diffusing capacity or alveolar ventila-

tion, carbon monoxide uptake resulting from an increase in inspired-air carbon

monoxide concentration or carbon monoxide excretion resulting from a decrease in

inspired-air carbon monoxide concentration will be delayed.  Barometric pressure

is a factor, because of the effect on inspired and alveolar pO^, pulmonary capil-

lary p02, and pCO at a given carbon monoxide concentration.

     Several important questions are related to biologic effects of carbon monoxide.

For example, will a smoker with a usual carboxyhemoglobin concentration resulting

from inhalation of cigarette smoke have carboxyhemoglobin from ambient carbon

monoxide simply additive to that present from smoking?  The theory predicts that

this would be the case.  What is the influence of various parameters on the rate

at which carboxyhemoglobin changes during the daily cycle of activity?  Answers

predicted from the theory are given in Chapter 5.

     With regard to physiologic effects of carbon monoxide, the most important

chemical characteristic of carbon monoxide is that, like oxygen, it is reversibly

bound by hemoglobin and competes with oxygen for binding sites on the hemoglobin

molecule.  Because the affinity of hemoglobin for carbon monoxide is more than

200 times that for oxygen, carbon monoxide,  even at very low partial pressures,

can impair the transport of oxygen.  That is, the presence of carboxyhemoglobin

decreases the quantity of oxygen that can be carried to tissues and shifts the

oxyhemoglobin dissociation curve to the left and changes the shape of this curve,

so that capillary oxygen tensions in tissues are decreased; this is believed to

hinder oxygen transport from blood into tissues.  Although not proved, it is possible

that carbon monoxide exerts deleterious effects by combination with the intra-

cellular hemoproteins, myoglobin, cytochrome oxidase,  and cytochrome PATO-  The

evidence is best for myoglobin:  it has been demonstrated that the degree of

carboxymyoglobin saturation increases with increases in blood carboxyhemoglobin.

Recent evidence that mitochondrial  respiration is more sensitive to carbon monoxide

during tissue hypoxia and that binding of carbon monoxide to myoglobin increases

during tissue hypoxia at the same blood carboxyhemoglobin content suggests an

intracellular mechanism of carbon monoxide toxicity under this condition.

     Effects on the Fetus.  It has recently  become obvious that the fetus may be

extremely susceptible to effects of carbon monoxide carried in maternal blood.

The fetal carboxyhemoglobin content is chiefly a function of maternal carboxy-

hemoglobin and fetal endogenous carbon monoxide production, but in addition

is a function of placental carbon monoxide diffusing capacity and the factors

that affect maternal carboxyhemoglobin content.  Under steady-stage conditions,

fetal carboxyhemoglobin is about 10-15% greater than the corresponding maternal

carboxyhemoglobin concentration.  The rates of fetal uptake and elimination of

carbon monoxide are relatively low, compared with those of the mother.  After a

step change in inspired carbon monoxide concentration, the time for maternal

carboxyhemoglobin to reach half its steady-state value is about 3 h.  In con-

trast, fetal carboxyhemoglobin requires about 7.5 h to reach half its steady-

state value, and final equilibrium is not approximated for 36-48 h.  Because

of this  lag in change in fetal  carboxyhemoglobin and because the ratio of

fetal to maternal carboxyhemoglobin is greater than unity under steady-state

conditions, the mean fetal carboxyhemoglobin content is greater than that of

the mother under a wide variety of circumstances.

     Few studies have explored the effects of carbon monoxide on the growth and

development of the embryo and fetus.  Most of these used high carbon monoxide

concentrations.  The only study with a moderate carbon monoxide content showed

decreased birthweight and increased neonatal mortality in rabbits.

     As indicated above, carbon monoxide interferes with tissue oxygenation,

both by decreasing the capacity of blood to transport oxygen and by shifting

the blood oxyhemoglobin saturation curve to the left.  Blood oxygen tension

must therefore decrease to lower than normal before a given amount of oxygen

will unload from hemoglobin.  Thus, blood carboxyhemoglobin lowers tissue end-

capillary or venous p02.  This may result in tissue hypoxia if the pO- is below

a critical point and tissue blood flow does not increase appropriately.  Some


theoretical effects of blood carboxyhemoglobin on tissue oxygenation and the

effective changes in blood flow and arterial p02 values required to maintain

normal oxygen delivery are reviewed in Chapter 5.  These mechanisms may operate

either individually or together to compromise oxygen delivery to developing

cells.  If present briefly at critical periods of embryonic or fetal development

or if continued for long periods, these effects may interfere with normal


     The hypoxic effects of carbon monoxide are similar to the hypoxic effects

of high altitude; and the fetus, as well as the pregnant woman, at high altitude

may be particularly sensitive to the effects of carbon monoxide.

     The effects of carbon monoxide on fetal growth and development are of con-

siderable interest and importance, but there is a dearth of experimental data

available on them.  Because of both the short-term effects of carbon monoxide

on fetal oxygenation itself and the possible long-term sequelae (damage to the

brain and central nervous system), fundamental research in this subject is urgent.

     Maternal smoking is associated with increased blood carboxyhemoglobin in

both mother and fetus.  The decrease in mean birthweight of the infants of

smoking mothers, compared with that of infants of normal nonsmokers, may result

from relative hypoxia caused by carbon monoxide, but this is not established.

Nicotine and other chemicals in tobacco smoke may affect birthweight.

     Much of the excessive exposure of the fetus and newborn infant to carbon

monoxide results from smoking by the mother.  There is considerable evidence

that smoking during pregnancy results in increased incidences of abortion, such

bleeding problems as placenta praevia and abruptio placentae, stillbirths, and

neonatal deaths.  It is thus apparent that there is no place for cigarette-

smoking during pregnancy (it must be noted that the results of smoking during

pregnancy include not only the effects of carbon monoxide, but also the effects

of nicotine and other constituents of tobacco smoke).


     Cardiovascular Effects.  It has become clear in recent yuars that the cardio-

vascular system, particularly the heart, is susceptible to adverse effects of carbon

monoxide at low blood carboxyhemoglobin concentration.

     Considerable evidence has been obtained with experimental animals that carboxy-

hemoglobin at 6-12% saturation results in a shift from aerobic to nonaerobic metab-

olism in the myocardium and that tissue oxygen tension may be compromised.  In

addition, there apparently are ultrastructural changes in the myocardium of experi-

mental animals exposed to carbon monoxide that produces carboxyhemoglobin at 8-9%

saturation for 4 h.  There is strong evidence that patients with coronary arterial

disease are more susceptible to small increases in blood carboxyhemoglobin, in

that their physiologic responses are different from those of normal subjects.

     Arteriosclerotic heart disease is the leading cause of death and morbidity

in the United States.  Many asymptomatic people have extensive coronary athero-

sclerosis.  Experimental and clinical studies have suggested that exposure to

carbon monoxide is important in the development of atherosclerotic disease, later

heart attacks, and the natural history of heart disease.

     Persons with angina pectoris exposed to relatively low doses of carbon

monoxide for short periods—doses that raised their hemoglobin content to about

2.5%—were found to be able to exercise for a shorter peiod before the onset of

chest pain.  Similarly, exposure to the air on the Los Angeles freeway resulted

in an increase in carboxyhemoglobin concentration (mean, 5.08%) and a decrease

in exercise tolerance among angina patients, compared with those in patients who

breathed compressed air while driving on the freeway.  Patients with "intermittent

claudication"* who breathed carbon monoxide at 50 ppm for 2 h developed leg pain

sooner after the start of leg exercise and had greater duration of pain than when

they had not been exposed to carbon monoxide.
 A complex of  symptoms  characterised by  absence  of pain or discomfort  in a limb
when at rest,  the conpenefiment a£  pain,  tenpio?!,  and weakness, §f£sr walking is
begun, intensification  of  the condition until walking becomes impossible, and
the disappearance of  the symptoms  after  a  period  of rest.


     Patients with angina pectoris who continue to smoke, and therefore might

have higher carboxyhemoglobin contents than nonsmokers, have a poorer prognosis

than patients with angina pectoris who do not smoke cigarettes.

     After the cessation of cigarette-smoking by people free of clinical coronary

arterial disease, there is a reduction in their risk of heart attack.  This re-

duction in risk occurs fairly rapidly after the cessation of smoking and suggests

that cigarette-smoking (and perhaps carbon monoxide)  precipitates heart attack.

     It is not known whether results obtained in studies of effects of carbon

monoxide on patients with cardiovascular disease are pertinent to large numbers

of people with these diseases in our population.

     Exposure of selected animals, especially rabbits and primates, to carbon

monoxide and a high cholesterol diet has led to a higher incidence of athero-

sclerosis than a cholesterol diet alone.  No data are available on the increase

in atherosclerosis after carbon monoxide exposure in man.   A recent report, how-

ever, noted atherosclerotic changes in the umbilical arteries of newborns of

smoking mothers.

     Epidemiologic studies in Baltimore and Los Angeles have failed to show any

relationship between the number of heart attacks per day (including both myo-

cardial infarction and sudden death) and the daily ambient carbon monoxide con-

centrations in the community.   The case-fatality percentage was reported to be

greater in areas within the Los Angeles basin with high ambient carbon monoxide

content than areas with low content during the time when the ambient carbon

monoxide in the basin was increased.  Further analysis of  these studies suggests,

however, that other factors might have accounted for the difference between case-

fatality percentages.

     People who died suddenly from coronary arterial disease had higher postmortem

carboxyhemoglobin concentration than other sudden-death victims and living con-

trols.  The differences were related primarily to the extent of cigarette-smoking

among the different groups.  No difference in the pathologic findings between

smoking and nonsmoking people who died suddenly from arteriosclerotic heart disease

(ASHD) was noted in the Baltimore study.

     Behavioral Effects.  The behavioral effects of low concentrations of carbon

monoxide are small and variable.  The effects found most reliably in the labora-

tory are those on vigilance tasks, in which subjects are asked to report the

occurrence of occasional signals over long periods.  Four recent studies reported

a higher incidence of missed signals at very low carboxyhemoglobin—between 2%

and 5%—than under control conditions; but one study, in which the carboxyhemo-

globin was 9%, did not.  These studies all used relatively small numbers of young,

healthy subjects.  In addition, all but one relied on indirect measures of carboxy-

hemoglobin, either alveolar breath samples or estimates made from knowledge of the

duration of the exposure to carbon monoxide.  Nevertheless, taken as a group, these

studies argue that carbon monoxide does have an effect on human behavior at carboxy-

hemoglobin saturations even lower than those reached by chronic smokers.

     The finding by Beard and Wertheim in 1967, that carboxyhemoglobin estimated

at less than 5% was associated with deficits in a subject's ability to discriminate

between short tones still stands, not yet having been repeated by other workers in

a fashion that would truly challenge it.  But this finding now appears to be more

relevant to questions of vigilance than to time perception, inasmuch as some in-

vestigators have minimized the role of boredom and fatigue and others have done

the opposite—i.e., have excluded external influences and studied their subjects

for relatively long periods.  The former method has usually yielded negative

results; and the latter, positive.  Other tests of time perception have been

almost uniformly negative, so time perception itself is probably not affected

by very, low carbon monoxide content.

     Only the most tentative conclusion—that carbon monoxide has perhaps a slight

deleterious effect on driving performance—can be drawn from the few preliminary

attempts to study actual or simulated driving.  The lack of clear-cut evidence

is especially unfortunate  in view of  the  importance of  automobile  drivers  in  any

consideration of the behavioral effects of carbon monoxide.

     Low concentrations of carbon monoxide may impair brightness discrimination;

however, this finding dates to World War II and has not been adequately repeated.

     There are some hints that various verbal and arithmetic abilities and motor

coordination are lessened by carbon monoxide, but the evidence is unsatisfactory.

     Carbon monoxide may modify effects produced by other substances.  People

drive automobiles under the influence of sedatives, tranquilizers,  alcohol, anti-

histamines, and other drugs.  What would be innocuous amounts of such drugs if

taken alone may become important determinants of behavior in the presence of low

carboxyhemoglobin concentrations.   Interactions with the other constituents of

automobile exhaust may also occur.

     Carbon Monoxide Effects During  Exercise.  Data on effects of increased carboxy-

hemoglobin during exercise indicate  that aerobic capacity is compromised readily

even at fairly low carboxyhemoglobin saturation, whereas submaximal efforts

(30-75% of maximum) can be carried out with minor change in efficiency, even at

relatively high carboxyhemoglobin saturation—i.e., 30%.  The available data were

obtained for the most .part from studies on young male subjects.

     Altitude and Carbon Monoxide Effects.  Precise data on effects of carbon

monoxide in high-altitude residents and visitors are not available.  Some 2.2

million people live at altitudes above 5,000 ft in the United States.  Ambient-

air standards set at sea level are not applicable for high-altitude sites.  EPA

primary standards are expressed in milligrams per cubic meter of air, and at high

altitudes each cubic meter of space contains less air than at sea level; therefore,

allowable carbon monoxide concentrations are higher.  As noted in a previous sec-

tion, carbon monoxide and oxygen are competitive, so adverse effects of carbon

monoxide should occur at lower carbon monoxide concentrations if tissue pC>2 is

less in subjects at high altitude.  Carbon monoxide uptake during transient high

carbon monoxide concentration will be more rapid, owing to the lower alveolar pC^-

There are some data that suggest that effects of carbon monoxide are additive to

effects of hypoxia.  The most important information on carbon monoxide exposures

at altitude—the preciseness of potentially additive effects—has not received

much attention, and what little information there is has not been verified by

direct experiments.

     Chronic Carbon Monoxide Exposure.  Animals subjected to prolonged or repeated

exposure to carbon monoxide at concentrations higher than those associated with

community air pollution undergo adaptive changes, which enable them to tolerate

acute carbon monoxide exposures that cause collapse or death in animals not

previously exposed.  Polycythemia develops in the course of chronic carbon

monoxide exposure, and this probably contributes to the symptomatic adaptation.

Other compensatory mechanisms are likely, but they have not been demonstrated.

Limited evidence suggests that chronically exposed humans also develop symptomatic


     Prolonged carbon exposure also leads to detrimental effects, including growth

retardation, cardiac enlargement, and an increased rate of development of athero-

sclerosis in some experimental animals.  The extent to which similar changes occur

in humans is unknown.

     Dose-Response Relationships.  Whether there is a threshold concentration of

carboxyhemoglobin for an adverse effect is still unknown.  The question is of

practical importance in setting carbon monoxide air standards, because, if there is no

threshold and there are adverse carbon monoxide effects at any blood carboxyhemo-

globin content, it is impossible to prevent all adverse effects by legislating air

standards, but, of course, more severe effects could be prevented.

     As indicated in this document, the tissues most sensitive to adverse effects

of carbon monoxide appear to be heart, brain, and exercising skeletal muscle.

There is evidence that carboxyhemoglobin at 3-5% saturation has an adverse effect

on ability to detect small unpredictable environmental changes (vigilance).   There

is evidence that acute increases in carboxyhemoglobin to above 5% saturation in

patients with cardiovascular disease can result in exacerbation of symptoms.   Car-

boxyhemoglobin as low as 5% saturation has been shown to decrease the maximal

oxygen consumption during exercise in healthy young males.   In the studies of

effects on vigilance or cardiovascular symptoms, no attempt was made to determine

effects of lower carboxyhemoglobin saturation and to look for a threshold.  In

the studies on aerobic metabolism of exercising skeletal muscle,  an apparent

threshold was found:  measurable effects on oxygen uptake could not be demon-

strated with carboxyhemoglobin at less than 5% saturation,  but it is possible

that effects were present at lower carboxyhemoglobin saturation.

     Recent research has been directed at adverse effects of low carboxyhemoglobin

saturation, and little effort has been made to investigate dose-response relation-

ships at carboxyhemoglobin saturations greater than that demonstrated to have an

adverse effect.  Defining dose-response relationships, however, is important in

considering the question of whether increases in carboxyhemoglobin have the same adverse

effects in a subject with a normal carboxyhemoglobin as in a subject with baseline

increased carboxyhemoglobin (such as a smoker) or a patient with increased endogenous

carbon monoxide production.  The studies referred to above, showing effects of

acute small increases in carboxyhemoglobin on vigilance or the myocardium did not

explore adverse effects over a wide range of carboxyhemoglobin saturation.  It

has been demonstrated in the study of effects of increasing carboxyhemoglobin on

aerobic metabolism in exercising muscle that over about 5-30% saturation there was

an almost linear relationship between carboxyhemoglobin and the fall in maximal

oxygen uptake.

     It is likely that dose-response relationships for adverse effects of carbon

monoxide on a given tissue are different in the presence of disease.  Mechanisms

of adaptation to carbon monoxide hypoxia may be altered or the tissue may operate

under conditions of borderline hypoxia and therefore be more susceptible to effects

of increased carboxyhemoglobin.  Dose-response relationships in acute carbon

monoxide  exposure may be different  from those in chronic exposure.

     There are gross adverse effects on function in several organ systems with

acute increases in carboxyhemoglobin to over 20% saturation.  At the other end

of the spectrum, there is evidence of adverse effects on brain, myocardium, and

skeletal muscle function at relatively low carboxyhemoglobin.  But almost no data

on other dose-response relationships in man are available.  The data base is in-

adequate for determination of air quality criteria standards for carbon monoxide

with a reasonable degree of certainty.


     A great deal is known about adverse effects of e.irbon monoxide- on nomi.jJ  «md

abnormal man, and much of this knowledge supports the setting of rather stringent

air carbon monoxide standards.  The present EPA standards are designed to prevent

carboxyhemoglobin over about 2.5% saturation; this gives an adequate safety factor,

in that adverse effects of "acute" carbon monoxide exposures under experimental

conditions are demonstrated at carboxyhemoglobin above 4-6% saturation.

     This review made no attempt to survey the literature related to the possible

role of carbon monoxide in the adverse effects of cigarette-smoking.  Carboxyhemo-

globin in smokers is frequently greater than saturations that have been implicated

as having adverse effects on normal or abnormal man.  A major conclusion of this

report is that it is imperative to determine the effects of carbon monoxide in

smokers and to determine the possible role of carbon monoxide in the excess

morbidity of carbon monoxide in cigarette smokers.

Carbon Monoxide Effects on Bacteria and Plants

     Carbon monoxide reacts readily with cytochrome oxidase,  and this reaction and

the resulting effect on energy transport may be responsible for some of the observed

effects of this pollutant on plants and bacteria.   Carbon monoxide also reacts with

leghemoglobin,  affecting the oxygen transport system in the legume root nodule, and

at least some plants appear to be able to metabolize carbon monoxide.

     Carbon monoxide at 85 ppm can increase or decrease the survival rate of some

airborne bacteria,  depending on the relative humidity and the organism.  Luminescence

of some strains of  marine bacteria can be inhibited by carbon monoxide when they

are exposed to a carbon monoxide atmosphere on sensor disks.

     Carbon monoxide at about 10,000 ppm can produce various  leaf,  stem,  flower,

and root abnormalities in higher plants.   These abnormalities include retardation

of growth, epinasty and chlorosis of leaves, abscission  of  leaves  and other

plant parts, initiation of adventitious roots  from  stem  or  leaf  tissue,  modifica-

tion of the response to gravity, and modification of  sex expression.   Carbon

monoxide at about 100 ppm can inhibit nitrogen fixation  in  root  nodules, and

carbon monoxide at about 1,000 ppm  can inhibit nitrogen  fixation of  free nitrogen'

fixing bacteria.

     Inhibition of apparent photosynthesis  (a  measure of the  growth  rate) has

been measured in excised leaves  exposed to  typical  urban ambient concentrations

of carbon monoxide.  Carbon monoxide at 1-10 ppm inhibited  the apparent  photo-

synthetic rate of coleus, cabbage,  grapefruit, and  Phoenix  palm.  Inhibition of

apparent photosynthesis, if it occurs in  the field, is probably  the  only important

effect of carbon monoxide on plants at ordinary concentrations.

Closing Comments

     We would be remiss if we did not reemphasize the importance of  the  issues

associated with exposure to carbon  monoxide.   Little  research has  been done on

these problems, compared with that  in many  other important  subjects.   And the

quality of this research has sometimes been less than excellent.  Unfortunately

the public and legislatures often do not  recognize  that  the roots  of understand-
                                                         basic mechanism  of the
ing of many of the problems of clinical relevance,  such  as  the.effects of carbon

monoxide on the human organism,  lie in fundamental  research.

                             CHAPTER 8


•  We recommend that studies be supported to determine the role

   of carbon monoxide in deleterious effects of cigarette-smoking.

        The estimation of populations that are influenced by

        adverse effects of carbon monoxide is confounded by

        the lack of information about the cause of adverse

        effects of cigarette-smoking and the possible role

        of carbon monoxide.

•  We recommend that effort be directed toward greater public

   awareness of the hazards of cigarette-smoking during pregnancy.

        Although there is public awareness of the hazards of

        cigarette-smoking related to lung cancer and other

        diseases,  there has been little publicity about

        hazards to the fetus.   There is growing evidence of

        serious deleterious effects of maternal cigarette-

        smoking on the fetus.

•  We recommend expansion of the data base related to adverse

   effects of carbon monoxide  on vigilance,  on oxygenation of

   exercising skeletal muscle,  and in atherosclerotic heart disease,

   and in peripheral vascular  disease.

        Carbon monoxide standards are now set on the basis of

        available  data in these fields.   Experiments in each of

        these fields have been performed on a relatively small

        number of  human subjects.   There is a need for replica-

        tion of studies by other laboratories.   Carbon monoxide

        exposure should be varied in duration and concentration.


     Dose-response relationships should be determined.

     Studies of adverse effects on vigilance and on

     oxygen uptake during exercise should be performed

     on susceptible populations.  In studies on patients

     with atherosclerotic heart disease, it would be

     particularly useful to study:  patients with ex-

     tensive coronary arterial disease as determined by

     coronary angiography, the effects of positive exercise

     testing in people without clinical symptoms, and

     people with high risk of heart attack.

We recommend that studies be performed to determine whether

heart, brain, and exercising skeletal muscle adapt to effects

of small increases in blood carboxyhemoglobin (less than

about 5-10% saturation).

     The entire data base related to deleterious effects

     of carbon monoxide on the heart, brain, and exercising

     skeletal muscle was obtained from acute experiments.
     Yet exposure of the population to carbon monoxide can

     be chronic or intermittent.  It is necessary to de-

     termine whether subjects can adapt and therefore be-

     come less susceptible to intermittent or chronic in-

     creases in carbon monoxide in ambient air.  Susceptible

     populations should be studied.

We recommend rapid expansion of the data base relating physiologic

and ambient carbon monoxide measurements, continuation of the re-

cent approach of monitoring human exposure to carbon monoxide in

   urban communities with blood carboxyhemoglobin and alveolar

   carbon monoxide measurements, acquisition by the EPA of a

   trained team capable of measuring blood carboxyhemoglobin.

        One of the uncertainties in evaluating effects of

        environmental carbon monoxide on health is the

        relationship of environmental exposure to carbon

        monoxide uptake.  Existing methods of monitoring

        environmental carbon monoxide can be improved.

        The spacing patterns of individual measurement

        stations within a monitoring network need to be

        studied.  Comparison of air monitoring data with

        blood carboxyhemoglobin measured either directly

        or as alveolar carbon monoxide concentration,

        would allow study of the efficiency of air

        monitoring systems.

•  We recommend an increase  in the information on mechanisms

   of adverse carbon monoxide effects in man.

        It is not clear how  very small decreases in hemo-

        globin oxygen-carrying ability and computed mean

        capillary pC^ resulting from 3-5% carboxyhemo-

        globin can cause significant effects on tissue

        oxygenation.   Studies of intracellular effects

        of carbon monoxide should be performed at 5-10%


•  We recommend studies aimed at identifying susceptible popula-

   tions .

        Patients with respiratory insufficiency or anemia

        should particularly be studied.

•  We recommend research to determine the possible role of carbon

   monoxide in the increased incidence of abortion, stillbirth,

   and neonatal death associated with mothers who are heavy smokers

   and in the small-for-gestational-age infants of smoking mothers.

•  We recommend research to determine fetal susceptibility to

   carbon monoxide.

        It is known that carbon monoxide can be concentrated in

        the fetal circulation and that blood oxygen tensions in

        the fetus are very low; these factors are expected to

        increase susceptibility to the adverse effects of carbon


•  We recommend studies to determine whether increased carboxyhemo-

   globin is a factor in sudden deaths due to coronary arterial


        In previous studies, it was shown that people who died

        suddenly from coronary arterial disease had higher post-

        mortem carboxyhemoglobin saturation than other sudden-

        death victims and living controls.  The differences were

        related primarily to the amount of cigarette-smoking in

        the ASHD subjects who died suddenly.  No difference in

        the pathologic findings between smoking and nonsmoking

        people who died suddenly from ASHD was noted in the

        Baltimore study.  Studies should be performed to

        assess whether these effects are in fact relevant

        to large numbers of urban people with coronary


•  We recommend studies aimed at determining the relationship

   between carbon monoxide exposure in some industries and

   morbidity and mortality from heart disease.

        This information should give insight into the re-

        lationship between atherosclerotic heart disease

        and carbon monoxide.  A better determination of

        industrial carbon monoxide exposure is needed.

        Measurement of either carboxyhemoglobin or

        expired-air carbon monoxide among employees in

        various "high-risk" industries should be completed.

•  We recommend further research to establish the extent of

   carbon monoxide-induced decrements in vigilance.

        Much work remains before it will be possible

        to determine which aspects of performance are

        most sensitive to carbon monoxide.   For example,

        decreasing the rate of signal presentation leads

        to poorer performance,  as does increasing the rate

        of unwanted signals or increasing the length of the

        experimental session.   It would be illuminating to

        find out how carboxyhemoglobin saturation inter-

        acts with these variables.   In addition,  modern

      advances in psychophysics, particularly in signal-

      detection theory, promise to help to elucidate the

      effects of agents like carbon monoxide.

We recommend further studies of effects of increased carboxy-

hemoglobin saturation on driving performance.

      Automobile drivers probably constitute the most

      important target population when one considers the

      behavioral effects of carbon monoxide.  The task of

      driving an automobile resembles a vigilance task in

      many ways.  In the light of the findings on vigilance,

      studies on driving performance may uncover deleterious

      effects of low concentrations of carbon monoxide,

      provided that experimenters with sensitive methods

      study their subjects for a long enough period.  Two

      main targets here are probably the long-distance

      truck-driver, who performs a job that combines monotony

     with danger, and the taxi-driver, who is continuously

     exposed to some of the highest urban carbon monoxide

     concentrations.  An epidemiologic study of automobile

     accidents needs to be performed to determine the possi-

     ble role of carbon monoxide.

We recommend studies aimed at elucidating possible adverse

effects of carbon monoxide on sensory functions.

     Definitive work on the sensory effects of carbon

     monoxide has not yet been done, despite a history

     of more than 30 years of sporadic effort.  The

     important conflicts among early experimental

     findings were described by Lilienthal 25 years

     ago; they remain unresolved.  Questions concerning

     complex intellectual behavior and coordination are

     also largely unanswered, after almost 50 years of


We recommend studies of interactions between carbon monoxide

and other agents.

     Carbon monoxide may modify effects produced by other

     substances.   People drive automobiles under the in-

     fluence of sedatives,  tranquilizers,  alcohol,  anti-

     histamines,  and other drugs.  Amounts of such drugs

     that would be innocuous alone may become important

     determinants of behavior in the presence of low

     carboxyhemoglobin saturation.  Interactions may

     also occur with the other constituents of auto-

     mobile exhaust.


•  We recommend continuing research aimed at determining potentially

   additive effects of carbon monoxide and low oxygen tension.

        What little information is available has been ob-

        tained by assuming simple additive effects, but

        has not been verified by direct experiments.

        Studies involving both physiologic and psycho-

        physiologic approaches are recommended, in order

        to clarify this issue.  This information will

        allow a more rational approach to setting carbon

        monoxide air standards at high altitude.  Studies

        of effects of low carbon monoxide concentration

        on man, adapted and nonadapted to high altitude,

        are needed.

a  We recommend study of the effect of typical urban ambient

   carbon monoxide concentrations on several airborne bacteria.

•  We recommend studies on intact plants to determine the degree

   of suppression of growth or apparent photosynthesis at dif-

   ferent carbon monoxide concentrations in combination with other

   smog-induced reactions.

•  We recommend more work related to natural sources and sinks

   of carbon monoxide.

        The natural sources and sinks of atmospheric carbon

        monoxide are not well understood.   Carbon monoxide

        production occurs in soil, but little is known about

        it.   Natural sources and sinks should be measured on

        a global basis.


•  We recommend that animal research be encouraged.

        It is at this level that the behavioral, as well as

        physiologic, mechanisms of action will most likely

        be ascertained.   Emphasis on mechanisms of action

        is warranted, because such emphasis makes possible

        intelligent extrapolation to human behavior that

        may not lend itself to direct study.   The recent

        growth of sophisticated animal psychophysics has

        made possible a  concerted attack on questions in-

        volving sensory  and perceptual effects of carbon

        monoxide in animals.

                                    APPENDIX A
 Nondispersive Infrared Spectrometry (NDIR)

      In 1971  the  Environmental  Protection Agency (EPA)  designated nondispersive

 infrared spectrometry  (NDIR)  as the reference method for continuous measurement

 of  carbon monoxide. 4H This  relatively  reliable procedure that has been used

 for many years is based on  the  absorption of  infrared radiation by carbon

 monoxide.  First  infrared radiation from an emitting source passes alternately

 through a reference and a sample cell, and then  through  matched detector cells

 containing carbon monoxide.   Since  the carbon monoxide in the detector  cells

 absorbs infrared  radiation  only at  the characteristic frequencies  of  this  com-

 pound,  the detector becomes sensitive to  those frequencies.  The wall between

 the detector  cells is  a flexible diaphragm with an electrical position  trans-

 ducer attached to it.  When there is a nonabsorbing  gas  in  the  reference cell,

 and carbon monoxide-free air in the sample cell, the signals from  the detectors

 are balanced electronically.  The carbon monoxide introduced into  the sample

 cell reduces the radiation reaching the sample detector, lowering the  tempera-

 ture and pressure in the detector cell,and displacing the diaphragm.  This

 electronically detected displacement is amplified to produce an output signal.

 Such a monitoring system for carbon monoxide is shown in Figure A-l.366

     Because water is the  principal interfering substance the moisture

 control system is  particularly important.  Carbon monoxide data obtained

 using the nondispersive infrared techniqueare questionable without information

 about water  removal.  At 4.6 ym the absorption-bands  of  water and carbon

 dioxide overlap the  carbon monoxide band.  A concentration of 2.5% by volume

of water vapor can produce a response equivalent  to  6.4  ppm carbon monoxide.315


                           ANALYZER SYSTEM
                                                                                               STRIP  CHART
              SECOND STAGE
              PRESSURE GAUGE
           SECOND STAGE
                          Figure A-l    Carbon Monoxide Monitoring  System Diagram

To reduce water vapor interference, water can be removed by drying agents, by

cooling, or its effect reduced by optical filters.  A combination of these is

recommended.  Selective ion exchange resins in a "heatless air drier" system

can also be used.  In a collaborative study of the NDIR method reported by

McKee and Childers, a maximum reproducibility of + 3.5 ppm in the 0 to 50 ppm

range was found.  The minimum detectable concentration was 0.3 ppm.     The

instruments are large owing both to the long cells required for accuracy at

low concentrations and to the air cooling and drying systems for water re-

moval .

     Since the data produced by the federal, state, and local monitoring

agencies are used to make decisions which can be very costly, procedures

for validating and maintaining data quality have been developed.   Because

of the requirements of the Clean Air Act, the EPA has issued specifications

for instruments, methods,  calibration,  and data quality.      The performance

specifications for automated carbon monoxide determinations are shown in

Table A-l,     and the specifications for concentrations of interfering

substances used to.check the effects in automated analytical methods  for

carbon monoxide are summarized in Table A-2.

                                     TABLE A-l

Performance Specifications for Automated Analytical Methods for Carbon

                Range                                  0-50 ppm

                Noise                                    0.50 ppm

                Lower detectable limit                   1.0  ppm

                Interference equivalent

                     Each interfering substance        +1.0  ppm
                     Total interfering substances        1.5  ppm

                Zero drift
                     12 h                              +1.0  ppm
                     24 h                              +1.0  ppm

                Span drift, 24 h

                     20% of upper range limit          +10.0%
                     80% of upper range limit          +2.5%

                Lag time                               10 min

                Rise time                               5 min

                Fall time                               5 min


                     20% of upper range limit            0.5  ppm
                     80% of upper range limit            0.5  ppm

        Range:   Nominal minimum and maximum concentrations which a method

                is  capable of measuring.

        Noise:   The standard deviation about the mean  of  short  duration

                deviations in output  which are  not  caused by  input concen-

                tration changes.

Definitions - continued

     Lower Detectable Limit*

          The minimum pollutant concentration which produces a signal

          of twice the noise level.

     Interference Equivalent•

          Positive or negative response caused by a substance other

          than the one being measured,

     Zero Drift;

          The change in response to  zero pollutant concentration during

          continuous unadjusted operation.

     Span Drift:

          The percent change in response to an up-scale pollutant concen-

          tration during continuous  unadjusted operation.

     Lag Time;

          The time interval  between  a step  change in input  concentration

          and the first observable corresponding  change in  response.

     Rise Time:

          The time interval  between  initial response and 95 percent of

          final response.

     Fall Time!

          The time interval  between  initial response to a step decrease

          in concentration and 95 percent of final response.


          Variation about the mean of repeated measurements of the same

          pollutant concentration expressed as one standard deviation about

          the mean.


                                               TABLE A-2

                 Test Concentrations  (In Parts Per Million) of Interfering  Substances

                          for Automated Analytical Methods for Carbon Monoxide
                                                   Carbon                 Water
Measuring principles    Ammonia    Nitric oxide    dioxide    Ethylene    vapor    Methane     Ethane

Infrared photometric

  (other than reference

   method)                 -            _            750         -        20,000

Gas chromatography—

  flame —ionization

  detection                -            -             -          -        20,000      -         0.5

Electrochemical            -           0.5            -         0.2       20,000

Catalytic combustion       -            -             -                      _        _          _

Thermal detection         0.1           -            750        0.2       20,000     5.0        0.5

Infrared fluorescence      -            -            750         -        20,000      -         0.5

Mercury replacement

Ultraviolet photometric    -            -             -         0.2          -        -         0.5

Dual Isotope Fluorescence

     This instrumental method utilizes the slight difference in the infrared

spectra of isotopes.  The sample is alternately illuminated with the character-

istic infrared wavelengths of carbon monoxide-16 (CO  ) and carbon monoxide-18

(CO18).  The carbon monoxide in the sample that has the normal isotope ratio,

nearly 100% carbon monoxide-16, absorbs only the carbon monoxide-16 wavelengths.

Therefore, there is a cyclic variation in the intensity of the transmitted light

                                                               O/O 0 ^ "7 O^fl
that is dependent on the carbon monoxide content of the sample.   '   '

     Full scale ranges of 0 to 20 ppm and up to 0 to 200 with a claimed sensi-

tivity of 0.2 ppm are available in this instrument.  The response time (90%)

is 25 s, but a 1-second response time is also available.  An advantage of this

technique is that it minimizes the effects of interfering substances.

Catalytic Combustion - Thermal Detection

     Determination of carbon monoxide by this method is based on measuring

the temperature rise resulting from catalytic oxidation of the carbon monoxide

in the sample air.

     The  sample air is first pumped into a furnace  which brings it to  a pre-

set, regulated temperature and then over the catalyst bed in the furnace.   A

thermopile assembly measures the temperature difference between the  air leaving

the catalyst bed and the air entering the catalyst  bed.   The output  of the

thermopile,  which is calibrated with known concentrations of carbon  monoxide

in air, is read on a strip-chart recorder as parts  of carbon monoxide  per

million parts of air.   The sensitivity is about  1 ppm.   Most hydrocarbons

are oxidized by the same catalyst,  and will interfere unless removed.


     Carbon monoxide is measured by the means of the current produced in

aqueous solution by its electro-oxidation at a catalytically active-

electrode.  The concentration of carbon monoxide reaching the electrode

is controlled by its rate of diffusion through a membrane.  This is dependent

                                               on 01
on its concentration in the sampled atmosphere. y>     Proper selection of

both the membrane and such cell characteristics as the nature of the electrodes

and solutions  make the technique selective for various pollutants.  (£. similar

technique has been reported by Yamate et al.   )

     The generated current is linearly proportional to the carbon monoxide

concentration from 0 to 100 ppm.  A sensitivity of 1 ppm and a 10-second

response time (90%) is claimed for a currently available commercial instrument.

     Acetylene and ethylene are the chief interfering substances; one part

acetylene records as 11 parts carbon monoxide and one part ethylene as 0.25

parts carbon monoxide.  For hydrogen,  ammonia, hydrogen sulfide, nitric oxide,

nitrogen dioxide, sulfur dioxide, natural gas and gasoline vapor, interference

is less than 0.03 part carbon monoxide per one part interfering substance.

Gas Chromatography - Flame lonization

     Measured volumes of air are delivered 4 to 12 times/h: to a hydrogen

flame ionization detector that measures the total hydrocarbon content (THC).

A portion of the same air sample, injected into a hydrogen carrier  gas stream,

is passed through a column where it is stripped of water, carbon dioxide, and

hydrocarbons other than methane.  Methane is separated from carbon  monoxide by

a gas chromatographic column.  The methane, which is eluted first,  is unchanged

after passing through a catalytic reduction tube into the flame ionization de-

tector.  The carbon monoxide eluted into the catalytic reduction tube is reduced

                                                                o 1 q
to methane before passing through the flame ionization detector.JJ-y  Between

analyses the stripping column is flushed out.  Nonmethane hydrocarbon concen-

trations are determined by subtracting the methane value from the total hydro-

carbon value.  There are two possible modes of operation.  One of these is a

complete chromatographic analysis showing the continuous output from the de-

tector for each sample injection.  In the other, the system is programmed

for both automatic zero and span settings to display selected elution peaks

as bar graphs.  The peak height is then the measure of the concentration.

The first operation is referred to as the chromatographic or "spectro" mode

and the second as the barographic or "normal" mode.

     Since measuring carbon monoxide entails only a small increase in cost,

instrument complexity, and analysis time, these instruments are customarily

used to measure three pollutants; methane, total hydrocarbons and carbon


     The instrumental sensitivity for each of these three components

(methane,  total hydrocarbons and carbon monoxide) is 0.02 ppm.   The lowest

full scale range available is usually from 0 to 2 up to 0 to 5 ppm,  although

at least one manufacturer provides a 0 to 1 ppm range.  Because of the com-

plexity of these instruments, continuous maintenance by skilled technicians

is required to minimize excessive downtime.

Frontal Analysis

     Air is passed over an adsorbent until equilibrium is established between

the concentration of carbon monoxide in the air and the concentration of

carbon monoxide on the adsorbent.  The carbon monoxide is then eluted with

hydrogen,  reduced to methane on a nickel catalyst at 250 C, and determined

by flame ionization as methane.

     Concentrations of carbon monoxide as low as Q.I ppm can be measured.

This method does not give instantaneous concentrations but does give averages

over a six-minute or longer sampling period.   '

Mercury Replacement

     Mercury vapor formed by the reduction of mercuric oxide by carbon

monoxide is detected photometrically by its absorption of ultraviolet light

at 253.7 nm.  It is potentially a much more sensitive method than infrared

absorption because the oscillator strength of mercury at 253.7 nm is 2000

times greater than that of carbon monoxide at 4.6 ym.

      Hydrogen and hydrocarbons also reduce mercuric oxide to mercury and

there is some thermal decomposition of the oxide.  Operation of the detector

at constant temperature results in a regular background concentration of

mercury from thermal decomposition.  McCullough et al. recommended a tempera-

ture of 175 C to minimize hydrogen interference.36,261  ^ commercial instrument

employing these principles was made and used during the middle 1950's.

The technique has been recently used for measuring background carbon monoxide

concentrations.  Robbins  et^ al.^36 have described an instrument in which the

mercuric oxide chamber is operated at 210 C, and the amount of hydrogen inter-

ference was assessed by periodically introducing a tube of silver oxide into

the intake air stream.  At room temperature silver oxide quantitatively oxidizes

carbon monoxide but not hydrogen.  Thus the baseline hydrogen concentration can

be determined.  Additional minor improvements are discussed by Seiler and

Junge,    who gave the detection limit for carbon monoxide as 3 ppb with a 5%..

standard deviation of the calibration at 0.2 ppm.

     More recently Palanos    described a less sensitive model of this

instrument intended for use in urban monitoring.  It has a range of 0 to

20 ppm, a sensitivity of about 0.5 ppm, and a span and zero drift of less

than 2% per day.  As in other similar instruments specificity is achieved

by removal of the potentially interfering substances other than hydrogen

(which is less than 10%).

     All of these instruments assume a constant hydrogen concentration.

In unpolluted atmosphere the  hydrogen concentration is roughly 0.1 ppm.

Howeverfthe automobile is not only a source of carbon monoxidefbut also of

hydrogen.  Therefore,  if this technique is used in polluted areas, it will be

necessary to measure the hydrogen concentration frequently.

                                  APPENDIX B

     Carbon monoxide is bound chemically to a number of heme proteins in the

body, any of which could act as a measure of exposure.  But because hemoglobin

in the red blood cells binds carbon monoxide much more strongly in relation to

oxygen than do any of the other proteins, and because blood is more easily ob-

tained in a pure state than the other sources of heme proteins, blood is ob-

viously the tissue of choice for sampling carbon monoxide exposure.  Its

                                                             188 342
affinity for hemoglobin is 220 times greater than for oxygen;    '     about

40 for myoglobin and about 1 for cytochrome P-450.  For routine monitoring,

however, blood samples can be taken only under special conditions.  A more

practical method for estimating carboxyhemoglobin is to measure alveolar gas.

     The most meaningful expression for the evaluation of carbon monoxide uptake

is the percent concentration of carboxyhemoglobin [HbCO].  Alveolar carbon mon-

oxide and the fraction of total hemoglobin unavailable for oxygen transport are

both directly related to carboxyhemoglobin concentration.  The carboxyhemoglobin

concentration may be determined directly using spectrophotometric  procedures

without releasing the carbon monoxide bound with the hemoglobin.   Also,  it can

be determined by measuring total hemoglobin separately and then measuring the

amount of carbon monoxide present by liberating it as a gas.   Venous blood can

be taken by venipuncture or by pricking the earlobe or finger.   The sample should

be collected in a closed container containing an anticoagulant in  the dry form,

such as disodium ethylene diaminetetraacetic acid, EDTA (1 mg/ml of blood)

or dry sodium heparin USP (0.05 mg/ml of blood).  Use of commercial anti-

coagulant Vacutainer tubes is satisfactory.  Blood samples can be preserved

on ice (about 4 C) for several days until analyzed.  Methods which measure

gaseous carbon monoxide and hemoglobin separately require complete mixing

before aliquots are taken, which is not always easy with small blood samples".

Carboxyhemoglobin Measurement

     The amount of carbon monoxide in the blood can be determined by spectro-

photometric procedures without liberating the bound carbon monoxide from the

hemoglobin.  These techniques are designed to give the percent of carboxyhemo-

globin directly.  Alternatively, the carbon monoxide content of the blood can

be estimated by freeing all the carbon monoxide from the carboxyhemoglobin,

extracting the gas, and assaying it by one of several techniques; classical

volumetric methods, methods based on the reducing action of carbon monoxide,

infrared absorption, or combinations of these.  Many of the methods used are

quite adequate when tl\e carboxyhemoglobin percentage is greater than 20%.

The difficulty has been to develop a method that is accurate for low concen-

trations of carboxyhemoglobin, particularly in the presence of other hemo-

globin forms such as methemoglobin.

     Spectrophotometric Methods.  These methods have been popular because

they are quick and simple, but they are frequently inaccurate at low concen-

trations.  The spectrophotometric determination of carboxyhemoglobin is

dependent on the difference between the carboxyhemoglobin absorption curve

and the absorption curve  for other forms of hemoglobin that are present at

certain wavelengths of electromagnetic radiation.  Spectrophotometry can be

as simple and qualitative as observing the color of diluted blood.  More


precise and objective procedures are used to estimate the quantity of  carboxy-
hemoglobin and the degree of saturation.  A number of spectrophotometric pro-
cedures have been developed that vary in sophistication and accuracy.
Klendshoj jit. _al.    diluted blood 1:100 with dilute ammonia, added solid
hydrosulfite and measured the absorbance at both 555 and 480 nm.  The  addition
of the hydrosulfite prevents presence of any other but the two pigments,
carboxyhemoglobin and reduced hemoglobin.  Both of these have the same ab-
sorbance at 555 nm but different absorbances at 480 nm.  The ratio of  ab-
sorbance, 550/480, decreases with increasing carboxyhemoglobin.  It is evalu-
ated using a standard curve prepared by analyzing known standards.  This
method.which is simple, rapid, and sufficiently accurate to determine a
2-5% change in carboxyhemoglobin concentration, has not proven satisfactory
at low concentrations.
                 O/T C
     Small et al.  ,   using blood diluted about 1:70 in dilute ammonia, made
absorbance measurements in the Soret region (410-435 nm) at 4 wavelengths with
a 1 mm light-path.  A series of simultaneous equations were used to estimate
the percent carboxyhemoglobin, the percent methemoglobin, and by difference
the percent oxyhemoglobin.  The accuracy of this is + 0.6% at low carboxy-
hemoglobin concentrations and + 2% methemoglobin at concentrations below 20%.
This method has now been successfully adopted by other laboratories.
     Probably the  most convenient spectrophotometric procedure is automated
differential spectrophotometry carried out with a carbon monoxide-oximeter
(manufactured by the Instrumentation Laboratory Co.)  described by Malenfant
£t_ al^.      In this method measurements of the three-component system con-
taining reduced hemoglobin,  oxyhemoglobin and carboxyhemoglobin are made at
three appropriate  wavelengths; 548,568 and 578 nm.   The instrument carries out

three simultaneous absorbance measurements at the three wavelengths on an

automatically diluted, hemolyzed blood sample.  The signals are then processed

by an analog computer and displayed in digital form as total hemoglobin  and

the percent concentration of oxy- and carboxyhemoglobin.  Although this instru-

ment is commercially available and widely used, accurate measurements at low

carboxyhemoglobin concentrations (less than 5%) can be carried out only after

careful calibration.

     Measurement of carboxyhemoglobin in blood using infrared spectroscopy

of blood  (rather than of carbon monoxide extracted from blood as described

                                   o tifir*
below)'has been recently described.      This method appears to have a very

high specificity for carbon monoxide bound to hemoglobin and a precision that

is in the same range as the most precise methods in common use.

     Volumetric Methods.  A variety of methods are used to free bound carbon

monoxide  from hemoglobin before measuring the amount of released carboxyhemo-

globin.   Carbon monoxide liberation has been carried out by acidification

with various acids  (sulfuric, lactic, hydrochloric, acetic, phosphoric)

either with or without the addition of oxidizing agents (potassium ferricyanide,

potassium biiodate).  The carbon monoxide released may be determined gaso-

metrically     '     or by reaction with palladium chloride.  '

     There are three methods currently available that have sufficient accuracy

and sensitivity to detect small changes in blood carbon monoxide content on

the order of 0.02 vol % or less.  Of these, the infrared analyzer"^»^^  and

the Hopcalite carbon monoxide oxidation meter •" both require blood samples

of one ml or more.  In the infrared or Hopcalite analyzer, a maj-or problem

is the necessity for an accurate gas-phase dilution before analysis.

     The  most satisfactory current method is that of gas-solid chromatography

on molecular sieve columns.  Procedures employing thermal conductivity de-

tectors 20,110 require a gas-sample size of 1 ml or more and the detectors

must be operated at the highest sensitivity.


     Porter and Volman    showed that hydrogen f larae-ionization detection can

be used to measure carbon monoxide after in-line catalytic reduction to methane.

This technique has been applied both to the analysis of carbon monoxide both

                                 92                4
in blood and in respiratory gases   and is about 10  times more sensitive than

other chromatographic techniques.

     To calculate percent carboxyhemoglobin, total hemoglobin must be determined,

This is done by reaction to form cyanmethemoglobin.     The most satisfactory

procedure is to use Van Kampen and Zijlstra reagent    taking precautions both

to prevent the gradual loss of hydrogen cyanide from the acid reagent and to

allow sufficient time for total conversion of the carboxyheraoglobin.

     The data derived from the analysis of carbon monoxide in the blood by

various techniques are given in Table B-l.

     Measurement of Alveolar Gas.  The theory of measuring carboxyhemoglobin

by measuring alveolar gas is based on the idea that under certain conditions

the gas in the lungs will equilibrate with the blood.   Measurements of the

gas-phase can then be applied to determine carboxyhemoglobin by use of the

Haldane relationship:
                             [HbCO]  = M     [Hb02]

     To get pO_, the oxygen partial  pressure in the arterial blood can either

be measured or assumed; pCO, the carbon monoxide partial pressure, is the

measured quantity, M is the Haldane  constant (equal to 220 with very little

individual variation    at physiologic pH) ;  and oxyhemoglobin concentration

is assumed to be approximately 1 - carboxyhemoglobin concentration.

     The problem in using this method is to  approximate equilibrium conditions

when the gas composition is actually changing at all times.  Two maneuvers

                                        TABLE B-l

                    Comparison of Representative Techniques for the









Van Slyke


of Carbon Monoxide in Blood
vol. (ml)
















Sample analysis ,
time (min) CV.%b'

15 6
30 2-4

30 1.8


20C 1.8

20 1.8

30 2.0

3 1.7

aSmallest detectable difference between duplicate determinations.

 Calculated based on samples containing less than 2.0 ml CO per deciliter.

 Estimated from literature.

 Coefficient variation.

 have  been  tried  to  achieve equilibrium  conditions,  rebreathing  and  breath-

 holding.   With rebreathing, the subject breathes in and out into  a  bag.   A

 gas sample is taken after a specified number of breaths.  Since oxygen  is

 being continuously removed and carbon monoxide concentrations are a function

 of both the oxyhemoglobin present as well as the total volume of the lungs

 and bag, this is a complex system with equilibrium difficult to achieve.  For

 this reason the rebreathing method has been replaced by the breathholding


     Jones  et al.    have shown that when a subject holds his breath, alveolar

carbon monoxide concentration increases initially as carbon monoxide leaves

 the blood to equilibrate.  As the alveolar oxygen falls however, carbon monoxide

is reabsrorbed into the blood  due to the fall of oxyhemoglobin and  the Haldane

relationship.   Thus alveolar  pCO will go through a  maximum depending on the

duration of breathholding.   Jones  et^ a]^.     concluded that  20 s was the

optimum period of time for both practical  and  theoretical  reasons.  This

technique is now standard.

     The subject  expires to residual volume, inspires  maximally, holds his

breath for 20 s  and then breathes  out as far as possible.  With the  aid  of

a 3-way valve, the first 500  ml of  expirate  is discarded and the remaining

gas is collected by turning the valve to an air-tight  bag.   The gas  in the

bag is then analyzed with standard gas  analyzers.   For field  use the Ecolyzer

has proved to be a rugged and  reliable  instrument.

                                                               ft 7
     Theoretically based on the Coburn— Forster.—Kane  equation,    the slope

of the graph relating the percent concentration of carboxyhemoglobin  to alveolar

pCO in ppm should be about 0.155 at sea  level  for carboxyhemoglobin  percent

concentration values equivalent to between 0 and 50 ppm, and  progressively


lower for higher concentrations.  Various researchers have reported discrepancies
                              1 *?7
in the results.  Forbes et al.    found a ratio of 0.14,  Sjostrand364 found 0.16,

and Ringold et al.    found 0.20.  Smith  (unpublished observations) showed

that the. slope was about 0.18 and did not decrease at higher carboxyhemoglobin

percent concentrations as would be expected because equilibrium was reached

less effectively.

     This method cannot be used with persons that have chronic lung disease,

in whom the alveolar gas composition tends to be extremely variable.  Moreover,

the subject's cooperation is essential.  Despite these difficulties, skilled

personnel can achieve reliable data.


1.  Abramson, E. , and T. Heyman.  Dark adaptation and inhalation  of  carbon

         monoxide:  Acta Physiol. Scand. 7:303-305, 1944.
 2.  Adams, J. D., H. H. Erickson, and H. L. Stone.  Myocardial metabolism  during

         exposure to carbon monoxide in the conscious dog.  J. Appl. Physiol.

         34:238-242, 1973.

 3,  Allen, T. H., and W.  S. Root.  An improved palladium chloride  method for

          the determination of carbon monoxide in  blood.   J. Biol.  Chem.  '216:

          319-323, 1955.

 4.  Altshuller,  A.  P.,  and J.  J.  Bufalini.   Photochemical aspects of air

         pollution:   A review.   Environ. Sci.  Technol. 5:39-64,  1971.

 5.  Altshuller, A.  P., and J. J. Bufalini.  Photochemical aspects  of air

          pollution:  A review.  Photochem. Photobiol. 4:97-146,  1965.

5a.   Anderson,   J. G.  The absolute concentration of OH  (X2/T) in the earth's

          stratosphere.  Geophys. Res.  Lett.   3:165-168,  1976

 6.   Anderson,  E. W., R.  J.  Andelman,  J. M. Strauch, N. J. Fortuin, and  J. H.

          Knelson.  Effect of low-level carbon monoxide exposure on onset  and

          duration of angina pectoris.   A study in ten patients with ischemic

          heart disease.   Ann. Intern.  Med. 79:46-50, 1973.  •

 6a.  Andrews, J., and J.  M. McGarry.  A community study  of smoking in  pregnancy.

          J. Obstet. Gynaecol. Brit. Commonw. 79:1057-1073, 1972.

 7.   Annau,  Z.  The comparative  effects of hypoxic  and carbon monoxide hypoxia

          on behavior, pp. 105-127.  In B. Weiss  and  V.  G. Laties,  Eds.   Behav-

          ioral Toxicology.  New York:  Plenum Press,  1975.

 fi-   Anthony,  A.,  E.  -Ackcrman,  and G.  K.  Strothor.  Effects of altitude accli-
          matization  on rat myoglobin.  Changes in myoglobin content of skele-
          tal  and  cardiac muscle.   Amer.  J. Physiol. 196:512-516, 1959.
9.    Aronow, W.  S., J.  Cassidy, J. S.  Vangrow, H. March, J. C. Kern, J. R.
          Goldsmith,  M. Khemka, J. Pagano, and M. Vawtek.  Effect of cigarette
          smoking  and breathing carbon monoxide on cardiovascular hemodynamics
          in anginal  patients.   Circulation 50:340-347, 1974.                 	
 10.  Aronow, W. S.,  J.  Dendinger,  and S.  N. Rokaw.  Heart  rate  and  carbon monoxide
           level after  smoking, high-,  low-, and  non-nicotine cigarettes.  Ann.
           Intern.  Med.  74:697-702,  1971.
 11.  Aronow, W. S. , J.  R. Goldsmith, J. C. Kem, J. Cassidy, W. H. Nelson,
          L. L. Johnson, and W. Adams.  Effect of smoking cigarettes on
          cardiovascular hemodynamics.  Arch. Environ.  Health 28:330-332, 1974.
 12.   Aronow,  W.  S.,  C. N. Harris, M.  W.  Isbell, S. N. Rokaw, and B. Imparato.
           Effect  of freeway travel on angina pectoris.  Ann. Intern. Med. 77:
           669-676, 1972.
 13.   Aronow,  W.  S.,  and M. W.  Isbell.  Carbon monoxide effect on exercise-induced
           angina pectoris.  Ann.  Intern. Med. 79:392-395, 1973.
 14.  Aronow, W. S.,  and S.  N.  Rokaw.   Carboxyhemoglobin caused  by smoking non-
           nicotine cigarettes.  Effects  in angina pectoris.   Circulation  44:
           782-788, 1971.
  15.  Aronow, W.  S., E.  A.  Stemmer, and M. W. Isbell.   Effect of  carbon monoxide
           exposure  on intermittent claudication.   Circulation  49:415-417, 1974.
 15a.   Asmussen, I., and K. Kjeldsen.  Intimal ultrastructure of human umbilical
           arteries.   Observations on arteries from newborn children of smoking
           and nonsmoking mothers.  Circ.  Res. 36:579-589,  1975.
15b.   Astrup,  P.   Some physiological and pathological  effects of moderate carbon
           monoxide exposure.   Brit. Med.  J.  4:447-452, 1972.

I5c.  Astrand, I., P. Ovrum, and A.  Carlsson.  Exposure to methylene chloride.   I.
           Its concentration in alveolar air and blood during rest and exercise
           and its metabolism.  Scand. J. Work Environ. Health  1:78-94, 1975.
 16.  Astrup, P.  Intraerythrocytic 2, 3-diphosphoglycerate and carbon monoxide
           exposure.  Ann. N. Y. Acad. Sci. 174:252-254, 1970.
 17.  Astrup, P., K. Kjeldsen, and J.  Wanstrup.  Enhancing influence of carbon
           monoxide on the development of atheromatosis in cholesterol-fed rabbits.
           J. Atheroscler. Res. 7:343-354, 1967.
18.   Astrup, P., and H.  G.  Pauli,  Eds.   A comparison of prolonged exposure to
           carbon monoxide and hypoxia in man.  -Scand.  J.  Clin Lab.  Invest.
           24(Suppl. 103):1-71, 1968.
 19.  Astrup, P., D. Trolle, H. M.  Olsen, and K. Kjeldsen.  Effect of moderate
           carbon-monoxide exposure on fetal development.  Lancet 2:1220-1222,
20.   Ayres,  S.  M.,  A. Criscitiello, and S.  Giannelli,  Jr.   Determination of
           blood carbon monoxide content by gas chromatography.   J.  Appl.
           Physiol.  21:1368-1370, 1966.
 21.  Ayres,  S.  M.,  S. Giannelli, Jr.,  and  H.  Mueller.  Effects of  low concen-
           trations  of carbon monoxide.   Part  IV.  Myocardial and systemic  res-
           ponses to carboxyhemoglobin.   Ann.  N. Y.  Acad. Sci. 174:268-293,  1970.
22.   Astrup, p., H.  M. Olsen,  D. Trolle,  and K.  Kjeldsen.   Effect of moderate
           carbon-monoxide exposure on fetal development.  Lancet  2:1220-1222,
23.   Baardsen,  E. L., and R. W.  Terhune.   Detection  of OH in  the  atmosphere
           using a dye laser.   Appl. Phys.  Lett.  21:209-211, 1972.
24.   Baker,  F.  D.,  and C. F. Tumasonis.   Carbon monoxide and avian embryogenesis.
           Arch.  Environ.  Health  24:53-61, 1972.
25.   Baker,  F.  D.,  C. F.  Tumasonis, and J.  Barron.   The effect of carbon monoxide
           inhalation on  the mixed-function oxidase activity in the  chick embryo
           and the adult mouse.   Bull. Environ.  Contam.  Toxicol.  9:329-336,  1973.

 26.  Earth, C. A., A. I. Stewart, C. W. Hord, and  A. L. Lane.  Mariner   9
           ultraviolet spectrometer experiment:  Mars airglow spectroscopy
           and variations in Lyman alpha.  Icarus 17:457-468, 1972.
 27.   Bartlett, D.,  Jr.   Pathophysiology of exposure to low concentrations of car-
           bon monoxide.   Arch.  Environ. Health 16:719-727,  1968.
 28.  Baulch,  D.  L.,  D.  D. Drysdale,  and A. C. Lloyd.  Critical Evaluation of
           Rate Data for Homogeneous, Gas Phase Reactions of Interest in High-
           Temperature Systems,   No.  1.  Leeds University, England, May 1968.

29.   Bates, D. R., and A. E. Witherspoon.  The photo-chemistry of some minor
          constituents of the earth's atmosphere (CO  , CO, CH , NO).  Monthly
          Notices.  Roy. Astron. Soc.  (London) 112:101-124, 1952.
 30.   Bay,  H.  W., K.  F.  Blurton,  H.  C.  Lieb,  and  H.  G. Oswin.   Electrochemical
           measurement of carbon monoxide.   Amer.  Lab. 4(7):57-61,  1972.
31.   Bay,  H.  W., K.  F.  Blurton,  J. M.  Sedlak, and  A.  M.  Valentine.   Electrochem-
           ical technique for the measurement of  carbon monoxide.  Anal.  Chem.
           46:1837-1839,  1974.
32.   Beard, R. R.,  and N. Grandstaff.   Carbon monoxide and human functions,
           pp. 1-26.   In B.  Weiss and V. G. Laties,  Eds.  Behavioral Toxicology.
           New York:   Plenum Press,  1975.
33.   Beard, R. R.,  and N. Grandstaff.   Carbon monoxide exposure and cerebral
           function.   Ann. N. Y.  Acad.  Sci. 174:385-395, 1970.
 34.  Beard,  R.  R., and  G. A. Wertheim.  Behavioral impairment associated  with
            small doses of carbon monoxide.  Amer. J.  Public Health  57:2012-2022,
  35.  Beck, H. G.  The clinical manifestations of chronic carbon monoxide  poison-
            ing.  Ann. Clin.  Med.  5:1088-1096,  1927.

36.   Beckman,  A.  0.,  J.  D.  McCullough, and R. A. Crane.  Microdetermination of

          carbon monxide in air.   A portable instrument.  Anal Chem. 20:674-

          677, 1948.

37.  Bohrmnn,  R. E. ,  D.  E.  Fisher, and J. Pnton.  Air pollution  in  nurseries:

          Correlation with a decrease in oxygen-carrying capacity of hemoglobin.

    .-;.•••• :J. Podiatr.  78:1050-1054, 1971.

33(  Bonder,  W. , M. Gothert', and  G. Malbrny.   Effect of low carbon  monoxide

          concentrations on psychological  functions.   Staub Reinhalt.  Luft

           (Engl. Ed.) 32(4):54-60, 1972.

39.  Bender, W. , M. Gothert,  G.  Malomy, and t>. Sebbcsse.  Wirkungsbild

          neidriger Kohlenoxid-Konzentrationen beim Menschen.  Arch. Toxikol.

          27:142-158,  1971.

 40.  Benesch, R.,  and R. E.  Benesch.  The effect of organic phosphates  from the

           human erythrocyte on the allosteric properties of hemoglobin.   Biochem.

           Biophys. Res. Comm.  26:162-167, 1967.

 41.  Benesch, W. ,  M.  Migeotte,  and L. Neven.  Investigations of atmospheric

           CO  at the  Jungfraujoch.  J. Opt. Soc. Amer.  43:1119-1123, 1953.

 42.  Benson,  F. B.,  J.  J.  Henderson, and D. E. Caldwell.   Indoor-Outdoor  Air

           Pollution Relationships:  A Literature Review.   Publ. AP-112.

           Research Triangle Park, N. C.:  U. S. Environmental Protection

           Agency,  1972.  73 pp.

43. Bergersen,  F. J.   The effects of partial  pressure of oxygen upon respira-

           tion  and nitrogen  fixation by soybean root nodules.   J.  Gen. Micro-

           biol.  29:113-125,  1962.

44.  Bergersen,  F. J.,  and G. L.  Turner.  Comparative  studies of nitrogen

           fixation by soybean root nodules,  bacteroid  suspensions and cell-

           free extracts.   J. Gen. Microbiol. 53:205-220,  1968.

 45.  Bernard,  C.   Lecons  sur les Effets des Substances Toxiques et Medicamen-

           teuses.   Paris:   J.-B. Baillicre et Fils,  1857.   488 pp.

 46.  Bidwoll, R. G. S. , And D.  E. Fraser.  Carbon monoxide uptake and metabo-

           lism by  ]o;ives. Can.  J. Bot. 50:1435-1439,  1972.

46a.  Birnstingl, M.,  L. Hawkins,  and  T. McEvcn.  Experimental atherosclerosis

           during chronic  exposure to  carbon  monoxide.   Eur.  Surg. Res.   2:92-

           93, 1970.   (abstract)

 47.  Blackburn, 11., P. Canner,  W. Krol,  S. Tominaga,  and  J.  Stamler.   The

           natural  history of  myocardial  infarction  in the coronary drug pro-

           ject.  .Prognostic indicators following infarction, pp.  54-64.  In

           G.  Tibblin, A.  Keys and L.  Werko,  Eds. Preventive Cardiology.

           Proceedings of  an International Symposium held  at  Billingehus, Skovde,

           Sweden,  Aug. 21,  1971.  New York:   Halsted  Press,  1972.
 48.  Brieger, H.  Carbon monoxide polycythemia.  J. Ind. Hyg. Toxicol. 26:321-327,


48a.  Brief, R. S., R. S. Ajemian, and R. G. Confer.  Iron pentacarbonyl:  Its

           toxicity, detection, and potential for formation.  Amer. Ind. Hyg.

           Assoc. J.  28:21-30, 1967.
 49.   Brody, J. S.,*and R. F. Coburn.  Effects of elevated carboxyhemoglobin on

           gas exchange in the lung.  Ann. N. Y. Acad. Sci. 174:255-260, 1970.

 50.   Brown, F. B., and R. H. Crist.  Further studies on the oxidation of nitric

           oxide;  the rate of reaction between carbon monoxide and nitrogen

           dioxide.  J. Chem. Phys. 9:840-846, 1941.

50a.  Buncher, C. R.  Cigarette smoking and duration of pregnancy.  Amer. J.

           Obstet. Gynecol. 103:942-946, 1969.

  51.   Burns,  B.,  and  G.  H.  Gurtner.   A specific carrier for oxygen and carbon

            monoxide in the  lung and  placenta.   Drug  Metab.  Depos. 1:374-379,



 52.   Rurris,  R.  H.,  and P.  W.  Wilson.   Characteristics of the nitrogen-fixing
           enzyme; system in  Nostoc  muscorum.   Dot.  Gazette 108:254-262, 1946.
53.  Butler, N.  R., and E.  D. Alberman, Eds.  Perinatal Problems.   The Second
          Report of the 1958 British Perinatal Mortality  Survey under the
          Auspices of the National Birthday Trust Fund.   Edinburgh:   E. & S.
          Livingston, Ltd., 1969.                                             	
 54.   Butler,  N.  R. ,  and D.  G.  Bonham.   Perinatal Mortality.   The  First Report
           of  the 1958  British  Perinatal Mortality Survey, under  the Auspices
           of  the National  Birthday Trust  Fund.   Edinburgh:   E. &  S. Livingstone,
           Ltd,  1963.   308  pp.                                                	
54a.   Butler,  N.  R.,  and H.  Goldstein.   Smoking in pregnancy and  subsequent
           child  development.   Brit.  Med.  J.  4:573-575, 1973.
55^  Butler, N.  R., H.  Goldstein,  and E.  M.  Ross.  Cigarette smoking in preg-
          nancy:  Its influence  on birth  weight  and perinatal mortality.   Brit.
          Mod. J. 2:127-130, 1972.
 56.  Cahoon,  R.  L.  Simple  decision making at high altitude.  Ergonomics  15:
           157-164, 1972.
 57.   Calvert,  J. G.,  K.  L.  Demerjian,  and J.  A.  Kerr.   Computer  simulation of
           the chemistry of a  simple  analogue to the  sunlight-irradiated auto-
           exhaust polluted  atmosphere.  Environ. Lett. 4:123-140,  1973.
 58.  Calvert, J.  G., K. L.  Demerjian, and J. A.  Kerr.  The effect  of carbon
          monoxide on the chemistry of  photochemical smog systems.   Environ.
          Lett.  4:281-295,  1973.
 59.   Calvert,  J. G. ,  J.  A.  Kerr, K.  L.  Demerjian,  and  R.  D.  McQuigg.  Photo-
           lysis  of  formaldehyde  as a hydrogen  atom source in the  lower atmos-
           phere. Science  175:751-752,  1972.

 SO.  n.ilvort,  .1.  C.. ,  .in»1  j.  N.  Pitts,  Jr.   j_ The oxides of carbon:  C0_7, pp.

           27.2-223.   In  Photochemistry.   Now York:   .Tohn Wiloy & Sons, Inc., 1966.

M.  Campbell, J. A.   Growth, fertility etc. in animals during attempted acclima-

           tization to carbon monoxide.   Q.  J.  Exp.  Physiol.   24:271-281, 1934.

62.  Campbell, J. A.  Hypertrophy of the heart  in acclimatization  to  chronic

          carbon monoxide poisoning.  J. Physiol.  (Lond.)  77:8P-9P,  1933.

 63.  Campbell,  J.  A.  Tissue oxygen tension and carbon monoxide poisoning.   J.

           Physiol.   (Lond.)  68:81-96,  1929.

64.  Cnvanagh, L. A., C.  F.  Schadt,  and  E.  Robinson.   Atmospheric  hydrocarbon

           and  carbon monoxide measurements  at Point  Barrow,  Alaska.   Environ.

           Set. Tcchnol. 3:251-257,  1969.

65.  Cavender, J. II.,  D.  S.  Kircher, and A.  J.  Hoffman.  Nationwide Air

           Pollutant Emission Trends, 1940-1970.  Research Triangle Park, N. C.:

           U.  S.  Environmental Protection Agency, Office of Air and Water
           Programs, 1973.  52 pp.

 66.  Chance,  B., M. Erecinska, and  M. Wagner.   Mitochondrial responses  to  car-

           bon monoxide toxicity.   Ann.  N. Y. Acad.  Sci.  174:193-204,  1970.

 67.  Chance,  B., N. .-Oshino, T. Sugano,  and A.  Mayevsky.  "Basic principles  of

           tissue oxygen  determiniation  from mitochondrial signals,  pp.  277-292.

           In H.  I.  Bichcr and  D. F.  Bruley. Oxygen Transport to  Tissue.  Instru-

           mentation, Methods,  and  Physiology.   Advances  in Experimental Medicine

           and Biology.   Volume 37A.  New York: Plenum Press, 19.73.

 68.  Chanutin,  A.,  and R. R. Curnish.   Effect  of  organic  and inorganic  phos-

           phates on  the  oxygen equilibrium of  human erythrocytes.  Arch.

           Biochem.  Biophys. 121:96-102,  1967.

 69.  Ch.ipanis,  A.   The relevance of laboratory studies to practical situations.

           Ergonomics  10:557-577, 1967.

 70.  Chapanis, A.  The search for relevance in applied research, pp.  1-14.   In

           W. T. Singleton, J. G. Fox, and D. Whitfield, Eds.  Measurement of

           Man at Work.  An Appraisal of Physiological and Psychological Criteria

           in Man-Machine Systems.  London:  Taylor and Francis Limited, 1971.

 71.   Chevalier,  R.  B.,  R. A. Krumholz,  and  J.  C.  Ross.   Reaction of nonsmokers to

           carbon monoxide inhalation.   Cardiopulmonary  responses at rest  and

           during exercise.   J.A.M.A.  198:1061r 106.4.,  1966.

 72.   Chiodi,  H., D. B.  Dill, F.  Consolazio, and S. M. Horvath.   Respiratory

           and circulatory responses  to  acute carbon  monoxide  poisoning.   Amer.

           J.  Physiol. 134:683-693,  1941.
             • •                      l

 73.   Chovin,  P.  Carbon monoxide;  Analysis of exhaust gas investigations in

            Paris.  Environ.  Res. 1:198-216,  1967.

 74.   City of  Chicago, Department of  Environmental Control.  Engineering Services

           Division.   Chicago 1974 Emission  Inventory Summary.

74a.  City of New York, Bureau of Technical Services,  Department of Air Resources.

           Data Report Aerometric Network.   Calendar Year 1974.  37 pp.

74b.  City of  New York, Bureau of Technical Services.  Department of Air Resources.

           Estimated Emission Inventory Summary.  Technical Service Report 16.

           April  1976.

 75.   Clark, B.  J.,  and  R. F. Coburn.  Mean tnyoglobin oxygen tension  during

           exercise  at maximal oxygen uptake.  J.  Appl. Physiol.  39:135-144,


 76.   Clark, R. T.,  Jr., and A.  B. Otis.   Comparative studies  on acclimatization of

           mice to carbon monoxide and  to  low oxygen. Amer. J.  Physiol.  169:285-

            294, 1952.

77.  Clcrc, M., ami F. Barat.  Cinetique des produits  de  decomposition de  CCL
          par photolyse-cclair dans  1'ultraviolet  lointain.   J.  Chim.  Phys.
          63:1525-1529,  1966.
73.  Clerc, M., and F. Barat.  Kinetics of CO formation studies  by  Far-uv
          flash photolysis of C02.   J. Chem. Phys. 46:107-110, 1967.
79.  Coburn,  R.  F.  Enhancement  by phenobarbital and diphenylhydantoinof  carbon
          monoxide  production in normal man.  N,.Erigl* J. Med.  283:512-515,  1970.
80.  Coburn, R. F.  The  carbon monoxide stores.  Ann. N.  Y.  Acad. Sci.  174:
          11-22, 1970.
 81.  Coburn,  R.  F.; G. K.  Danielson, W. S.  Blakemore,  and R. E.  Forster II.
          Carbon monoxide  in blood:   Analytical  method and sources  of error.
          J.  Appl.  Physiol.  19:510-515, 1964.
 82.  Cobu-rn, R. F., R.  E. Forster,  and P. B. Kane.  Considerations of the physio-
           logical variables  that  determine the  blood carboxyhemoglobin concentration
           in man.  J. Clin.  Invest. 44:1899-1910, 1965.   "
 83.  Coburn, R. F., and P.  B. "Kane.  Maximal erythrocyte and hemoglobin catabolistn.
           J. Clin. Invest.  47:14.35-1446, 1968.
 84.  Coburn, R. F., and L. B. Mayers.  Myoglobin 0« tension determined  from
                    »                              ^
           measurements  of carboxymyoglobin in skeletal muscle.  Amer.  J. Physiol.
           220:66-74,  1971.
 85.   Coburn, R. F., F. Ploegmakers, P. Gondrie,  and F. Abboud.  Myocardial myoglo-
           bin oxygen tension.  Amer. J.  Physiol. 224:870-876, 1973.
 86.  Coburn,  R. F., W.  J. Williams, and R. E. Forster.   Effects of erythrocyte
           destruction on  carbon monoxide production in man.  J. Clin.  Invest.
           43:1098-1103, 1964.
 87.  Coburn, R. F., W.  J. Williams, and S. B. Kahn.  Endogenous carbon monoxide
           production in patients with hemolytic anemia.   J. Clin. Invest. 45:460-
           468, 1966.


 88.   Cohen,  S.  I.,  M.  Deane,  and  J.  R.  Goldsmith.   Carbon monoxide and survival
           from  myocardial  infarction.   Arch.  Environ.  Health 19:510-517,  1969.
 89.  Cohen, N.,  and L.  Hcicklen.  The oxidation of inorganic non-metallic com-
          pounds, pp. 1-137.  In C. H.  Bamford, and C.  F. H. Tipper, Eds.
          Comprehensive Chemical Kinetics.  Vol. 6.  Reactions of Non-Metallic
          Inorganic Compounds.  New York:  Elsevier Publishing Compnay, 1972.
90.  Cole, P. V., L. H. Hawkins, and D.  Roberts.   Smoking during pregnancy
          and its effects on  the fetus.  J. Obstet. Gynaecol. Brit. Commonw.
          79:782-787, 1972.
91.  Collins, V. G.  Isolation, cultivation and maintenance of autotrophs, pp.
          1-52.  In J.  R.  Norris and D. W. Ribbons, Eds.  Methods in Micro-
          biology.   Vol. 3B.  New York:  Academic Press, 1969.
 92.  Collison,  H.  A.,  F.  L.  Rodkey,  and J.  D.  O'Neal.   Determination of carbon
          monoxide in  blood by gas  chromatography.   Clin.  Chem.  14:162-171,  1968.
 93.  Colmant,  H. J.  Effect of low  carbon monoxide concentrations on the sleep
           and  waking pattern of the albino  rat.   Staub-Reinhalt. Luft (Engl.
           Ed.) 32(4):32-35, 1972.
 94.  Colucci,  J.  M., and  C. R. Begeman.  Carbon monoxide in Detroit, New York
           and  Los  Angeles air.  Environ. Sci.  Technol. 3:41-47,  1969.
 95.  Committee on  Sp^ce Research (COSPAR),  International Council of Scientific
           Unions.   CTRA 1965.  COSPAR International Reference Atmosphere 1965.
           Compiled by  the Members of COSPAR Working Group IV, 1965.  Amsterdam:
           North-Holland Publishing  Company,  1965.   313 pp.
 96.  Comstock, G.  W.,  and F. E. Lundin, Jr.  Parental  smoking and perinatal
           mortality.  Amer. J. Obstet. Gynecol.  98:708-718,  1967.
96a.  Comstock, G.  W.,  F.  K.  Shah, M.  B.  Meyer,  and H.  Abby.   Low birth weight
           and  neonntal mortality rate related  to maternal smoking and socio-
           economic status.   Amer. J.  Obstet. Gynecol.  111:53-59, 1971.

  97.  Conlee, C.  J.,  P.  A.  Kenline,  R.  L.  Cummins,  and  V.  J.  Konopinski.

            Motor  vehicle exhausts  at three selected sites.   Arch.  Environ.

            Health 14:429-446,  1967.

  98.  Connes, P.,  J.  Connes, L. D. Kaplan,  and W. S. Benedict.  Carbon monoxide

            in the  Venus  atmosphere.  Astrophys. J.  152:731-743, 1968.

  99.  Cooper, A.  G.   Carbon Monoxide.   A Bibliography with Abstracts.  Public

            Health  Service Publication No.  1503.  Washington,  D. C.:  U. S.

            Department of Health, Education  and Welfare, 1966.  440 pp.

 100.  Coronary Drug Project Research Group.  Factors influencing long-term prog-

            nosis after recovery from myocardial infarction--three-year findings

            of the coronary drug project.  J. Chron.  Dis. 27:267-285, 1974.

lOOa.   Crespi,  H. 1.,  D.  Huff, H. F. DaBoll,  and J. J, Katz.   Carbon Monoxide in

            the Biosphere: CO Emission by Fresh-Water Algae.   Final Report.

            (CRC-APRAC-CAPA-4-68-5)  to Coordinating Research .Council and  Air

            Pollution  Control Office,  Environmental Protection Agency.  Argonne,

            111.:   Argonne National  Laboratory, 1972.  26 pp.

 161.  Crutzcn, p.  J.  A  discussion of the chemistry of  some minor  constituents

            in the  stratosphere and troposphere.  Pure Appl.  Geophys.  106-108:

            1385-1399, 1973.

 102.  Crutzen, P.  J.  Photochemical  reactions initiated by and influencing

            ozone  in unpolluted tropospheric air.  Tellus 26:47-57,  1974.

 103.  Curphey, T.  J., L. P. L. Hood, and N. M. Perkins. Carboxyhemoglobin  in

            relation  to air  pollution and smoking.   Postmortem studies.  Arch.

            Environ.  Health  10:179-185,  1965.

 104.  Dahms, T. E., and  S. M.  Horvath.  Rapid, accurate technique  for determination

            of carbon monoxide in blood.  Clin. Chem. 20:533-537, 1974.

 105.  Dahms, T. E., S. M. Horvath, and D. J. Gray.  Technique for accurately
            producing desired carboxyhemoglobin levels during rest and exercise.
            J. Appl. Physiol. 38:366-368, 1975.
 I05a.  Davie, R.,    N. Butler, and H. Goldstein.  From Birth to Seven.  The Second
             Report  of the National Child Development Study.  (1958 Cohort)  London:
            Longman,  1972.   198 pp.
 106.  DeBias, D. A., C. M.  Banerjee,  N.  C.  Birkhead, W.  V. Harrer, and L. A. Kazal.
            Carbon monoxide inhalation effects following myocardial infarction in
            monkeys.  Arch.  Environ. Health 27:161-167, 1973.
106a.  Delwiche, C. C.   Carbonic monoxide production and utilization by higher
            plants.  Ann.  N.  Y.  Acad.  Sci.  174:116-121,  1970.
 167.  deBruin,  A.   Carboxyhemoglobin  levels  due to traffic exhaust.   Arch.
            Environ. Health  15:384-389,  1967.
107a.  Denson, R.,  J.  L. Nanson, and M.  A. McWatters.  Hyperkinesis and maternal
            smoking.  Can. Psychiatr.  Assoc.  J.  20:183-187, 1975.
 108.  Dinman, B. D., J. W. Eaton, and G. J. Brewer.  Effects of carbon monoxide
            on DPG  concentrations in the erythrocyte.  Ann. N. Y. Acad. Sci.  174:
            246-251, 1970.
 108a.  DiVincenzo,  G. D., and M. L. Hamilton.   Fate  and disposition  of    C methy-
             lene chloride in  the rat.  Toxicol. Appl.  Pharmacol.   32:385-393,  1975.
 109.   Dixon-Lewis, G., W.  E. Wilson,  and A. A. Westenberg.  Studies of hydroxyl
             radical kinetics by quantitative ERS.  J. Chem. Phys. 44:2877-2884, 1966.
 110.  Dominquez,  A. M. ,  H. E.  Christensen,  L.  R.  Goldbaum,  and V. A.  Stembridge.
            A sensitive procedure  for determining carbon monoxide in blood or
             tissue  utilizing  gas-solid chromatography.  Toxicol.  Appl. Pharmacol.
             1:135-143,  1959.
 111.  Dorcus, R. M., and G. E. Weigand.  The effect  of exhaust gas  on the per-
            formance in certain psychological tests.  J. Gen. Psychol.  2:73-96,

112.   Douglas,  C.  G.,  J.  S.  Haldane, and J. B. S. Haldane.  The laws of combination
           of haemoglobin with carbon monoxide and oxygen.  J. Physiol.  (Lond.) 44,
           275-304,  1912.
113.   Drabkin, D. L., and J. H. Austin.  Spectrophotometric  studies.   II.
           Preparations  from washed blood  cells; nitric  oxide hemoglobin
           and sulfhemoglobin.  J. Biol. Chem.  112:51-65,  1935.
114.   Drinkwater, B.  L., P. B. Raven, S. M. Horvath, J. A. Gliner, R. 0.  Ruhling,
            N.  W.  Bolduan, and S.  Taguchi.   Air pollution, exercise, and heat
            s.tress.   Arch. Environ.  Health 28:177-181, 1974.
 115.  Dubois, L., and J.  L. Monkman,   Continuous determination of carbon monoxide
           by frontal analysis.  Anal.  Chem.  44:74-76,  1972.
116.  Dubois, L. , and J.  L. Monkman.   L'emploi  de  1'analyse frontale pour
           I'echnntillonnage et le dosage  de  1'oxyde de carbone dans 1'air.
           Mikrochim. Acta  1970:313-320.
 117,   Duvelleroy, M.  A., H. Mehtnel,  and M. B. Laver.  Hemoglobin-oxygen equilib-
            rium and coronary  blood  flow:  An analog model.  J. Appl. Physiol.
            35:480-484,  1973.
 118.   Eckardt,  R. E., H. N. MacFarland, Y. C. E. Alarie,  and W.  H,  Busey.
            The biologic  effect from long-term exposure  of primates  to carbon
            monoxide.   Arch. Environ. Health 25:381-387,  1972.
 119.   Ehrich,  W.  E.,  S.  Bellet,  and F. H.  Lewey.  Cardiac changes  from  CO
            poisoning.  Amer.  J.  Med. Sci. 208:511-523,  1944.
 120.  Ekblom, B., and R. Huot.  Response  to  submaximal  and maximal exercise at
            different levels of carboxyhemoglobin.   Acta Physiol. Scand.  86:474-
            482,  1972.
I20a.  Essenberg, J. M.,  J. V. Schwind,  and A.  R.  Patras.  The effects of nico-
            tine  and cigarette smoke  on pregnant female albino rats and  their
            offsprings.   J. Lab. Clin.  Med. 25:708-717,  1940.

121.  Estabrook, R.  W. ,  M.  R.  Franklin,  and A.  G.  Hildebrandt.  Factors influenc-
           ing the inhibitory effect of carbon monoxide on cytochrome P-450-
           Ci-italyzcd mixed  function oxidation reactions.  Ann. N. Y. Acad. Sci.
           174:218-232,  1970.
122.   Faltings, K. , W.  Groth, and P. Harteck.  Photochemische Untersuchungen
           im SCIIUMANN-Ultraviolett Nr. 7.   Zur Photochemie  des  Kohlenoxyds.
           Z. Physik. Chem. Abt. B. 41:15-22,  1938.
123.   Fischer,  E. R. , and M. McCarty, Jr.  Study of  the reaction of electronically
           excited oxygen molecules with carbon monoxide.  J.  Chem.  Phys.  45:781-
           784, 1966.
124.   Flury,  F. ,  and F.  Zernik.  Kohlenoxyd, pp.  195-196.   In Schadlich Case.
           Dampfe,  Nebel,  Rauch- und Staubarten.   Berlin:   Verlag von Julius
           Springer,  1931.
125.  Fodor,  G.  G. , and G.  Winneke.  Effect  of low CO concentrations  on resis-
           tance to monotony and on psychomotor capacity.  Staub  Reinhalt. Luft
           (Engl.  Ed.) 32(4):46-54, 1972.
       Forbes, W.  H. ,  D.  B.  Dill, H. DeSilva, and F. M. Van Deventer.  The influ-
            ence of moderate carbon monoxide poisoning upon the ability to drive
            automobiles.   J. Ind. Hyg. Toxicol. 19:598-603, 1937.
 127.   Forbes, W.  H. ,  F.  Sargent, and F.  J.  W.  Roughton.   The rate of carbon
            monoxide uptake  by normal men.   Amer.  J. Physiol. 143:594-608, 1945.
 128.   Forster, R. E.   Carbon monoxide and the partial pressure of oxygen in
            tissue.  Ann. N. Y. Acad. Sci. 174:233-241, 1970.
 129.   Francisco,  D. E. ,  and J. K. G. Silvey.  The effect of carbon monoxide
            inhibition on the growth of an aquatic streptomycete.  Can. J.
            Microbiol. 17:347-351, 1971.

 130.  Frazior, T. M.,  G. II. Davis, H. Goldstein, and I. D. Goldberg.  Cigarette
            smoking and prematurity:  A prospective study.  Amor. J. Obstet.  Gynecol
            81:988-996, 1961.
130a.  Fristedt, tt.,  and B.  Akesson.  Health hazards from automobile exhausts  at
            service facitities of multistory garages.  Hyg. Revy  60:112-118,
            1971.  (in Swedish)
 131.   Gaensler, E. A., J.  B. Cadigan, Jr., M. F. Ellicott, R. H. Jones, and
            A.  Marks.  A new method for rapid precise determination of carbon
            monoxide in blood.  J. Lab.  Clin. Med. 49:945-957, 1957.
 132.  Gardner,  R. A., and  R. H. Petrucci.   The  chemisorption of  carbon  monoxide
            on metals.  J.  Amer. Chem. Soc.  82:5051-5053, 1960.
  133.  Garvin,  D.  The oxidation of carbon monoxide in the presence of ozone.  J.
             Amer.  Chem.  Soc.  76:1523-1527, 1954.
  134.   General  Electric Company.   Indoor-Outdoor Carbon Monoxide Study. • EPA-
             R4-73-020.  Philadelphia:  General  Electric Company,  1972.  /~437 pp._7
 134a.   Gennser, G.,  K. Marsf/l,  and B. Brantmark.  Maternal smoking and  fetal
             breathing  movements.  Amer.  J. Obstet.  Gynecol.  123:861-867,  1975.
 134b.   Gilbert, R.,  S. Carlson,  and B.  Bromberger-Barnea.   Comparative effects
             of low pC>2 and  carbon monoxide on myocardial blood flow and oxygen
             consumption.  Amer.  J. Physiol. (in press)
   135.  Gilbert,  G.  J.,  and G. H.  Glaser.   Neurologic manifestations of chronic
             carbon  monoxide poisoning.    N.  Engl.  J.  Med.  261:1217-1220, 1959.
  136.   Gilson,  R. D., F. E.  Guedry,  Jr.,  and A. J.  Benson.  Influence of vestibu-
             lar stimulation and display luminance  on  the  performance of a compen-
             sator:/  tracking task.  Aerosp. Med. 41:1231-1237, 1970.

 136.1.  Ginsberg, M. D., and R. E. Myers.  Fetal brain damages following maternal

            carbon monoxide intoxication:  An experimental study.  Acta Obstet.

            Gynecol. Scand.  53:309-317, 1974.

136b.   Ginsberg,  M.  D. ,  and R.  E.  Myers.   Fetal  brain injury after maternal car-

            bon monoxide intoxication:   Clinical and neuropathologic  aspects.

            Neurology   26:15-23,  1976.

 137.  Glasson, W.  A.   Effect  of  carbon monoxide  on atmospheric  photooxidation

            of nitric oxide-hydrocarbon mixtures.   Environ.  Sci.  Technol.  9:

            343-347,  1975.

 138.   Gliner, J.  A.,  P. B. Raven, S. M.  Horvath, B. L.  Drinkwater, and J. -C.

            Sutton.  Man's physiologic response to long-term work during thermal

            and pollutant stress.   J. Appl.  Physiol. 39:628-232, 1975.

 139.   Gold, R. E.,  and L.  L.  Kulak.   Effect of  hypoxia  on aircraft pilot per-

            formance.   Aerosp.  Med.  43:180-183,  1972.

 140.   Goldman, A., D. G.  Murcray, F. H.  Murcray, W. J.  Williams, J.  N. Brooks,

            and C.  M.  Bradford.  Vertical distribution of CO in the atmosphere.

            J. Geophys. Res. 78:5273-5283, 1973.

 141.  Goldsmith, J.  R. , and S. A. Landaw.   Carbon  monoxide  and  human health.

            Science 162:1352-1359, 1968.

 142.  Goldstein, II.  Cigarette smoking and low-birthweight babies.  Amer. J.

            Obstet. Gynecol. 114:570-571, 1972.    (correspondence)

 l42a.  Goldstein, H.  Factors influencing the height of seven year old children  •

            results from the National Child Development Study.  Human  Biol.  43:

            92-111, 1971.

 143.  Gorbatow, 0., and L. Noro.  On acclimatization in connection with  acute

            carbon monoxide poisonings.  Acta Physiol. Scand. 15:77-87,  1948.

 1;'*4.   Gordon,.T.   Heart  Disease  in  Adults,  United  States--1960-1962.   U.  S.
           Department  of Health,  Education,  and  Welfare,  Public  Health Service,
           National Center  for Health  Statistics.   Vital  and  Health Statistics
           Series  11,  No. 6.  Washington, D.  C.:   U.  S. Government  Printing
           Office, 1964.    43  pp.
 H5.   Gordon, T. ,  W. ft.  Kannel,  D.  Me Gee, and T. R.  Dawber.   Death  and coronary
           attacks in  men after  giving up cigarette smoking.   Lancet  2:1345-
           1348, 1974.
 1-6.   Gorse,  R.  A.,  and D.  H.  Volman.   Photochemistry of the gaseous hydrogen
           peroxide-carbon  monoxide system:  Rate constants for hydroxyl radical
           reactions with hydrogen peroxide and isobutane by competitive kinetics.
           J.  Photochem. 1:1-10, 1972.
 147.   Gothert, M., F.  Lutz, and  G.  Malorny.   Carbon monoxide partial pressure in
           tissue  of different animals.   Environ.  Res.  3:303-309,1970.
1-7.3.   Goujard, J., C.  Rumeau,  and D.  Schwartz.  Smoking during pregnancy,
           stillbirth  and abruptio placentae.  Biomedicine 23:20-22,  1975.
 "-3.  Grahn,  D.,  and J. Kratchman.  Variations in  neonatal death rate  and
           birth weight in  the United  States  and possible relations to  environ-
           mental radiation, geology,  and altitude.  Amer. J.  Hum.  Genet.  15:
           329-352, 1963.
 149.  Graven,  W.  M.,  and F.  J.  Long.  Kinetics and mechanisms  of the two opposing
           reactions  of the  equilibrium CO +  ^0 = C02 + H£.   J. Amer. Chem. Soc.
           76:2602-2607, 1954.
 150.  Green,  A. E.  S.,  T. Sawada, B. C. Edgar, and M. A. Uman.  Production of
           carbon monoxide  by charged  particle deposition.  J. Geophys. Res.
           78:5284-5292, 1973.

 151.   Gregg, D.  E.,  and 1.  C.  Fisher.   / Blood supply to the heart_7, p. 1547.

            In W. F.  Hamilton and P.  Dow, Eds.  Handbook of Physiology.  Section 2,

            Volume 2.  Circulation.  Washington, D. C.:   American Physiolgical

            Society,  1963.

152.  Greiner, N.  R.   Hydroxyl-radical kinetics by kinetic  spectroscopy.  I.

           Reactions with H2, CO, and CH, at  300°K.   J. Chem. Phys. 46:2795-

           2799, 1967.

153.   Groll-Knapp, E., H. Wagner, H.  Hauck,  and M. Haider.  Effects of low

           carbon monoxide concentrations on vigilance and computer-analyzed

           brain potentials.  Staub Reinhalt. Luft (English ed.) 32(4):64-

           68, 1972.

 154.  Groth,  W.,  W.  Pessara, and H.  J.  Rommel.  Photochemische  Untersuchungen

            im SCHUMANN-Ultraviolett  Nr.  11.   Die  photochemische Zersetzung von

            N2 und CO im Lichte  der Xenon- und Krypton-Resonanzwellenlangen.  Z.

            Physik. Chem.  (Frankfurt)  32:192-211, 1962.

155.  Grut,  A.   Chronic Carbon  Monoxide Poisoning.  A Study in  Occupational

            Medicine.  Copenhagen:  Ejnar Munksgaard,  1949.   229 pp.
       Guest, A. D. L., C. Duncan, and P. J. Lawther.  Carbon monoxide  and  pheno-

            barbitone:  A comparison of effects on auditory flutter  fusion  thresh-

            old and critical flicker fusion threshold.  Ergonomics 13:587-594,


 157.  Haddon, W.,  Jr., R. E. 1. Nesbitt, and R. Garcia.   Smoking  and pregnancy:

           Carbon  monoxide in blood during gestation  and  at term.  Obstet.  Gynecol.

           18:262-267, 1961.

158.   Haggard, H. W.  Studies in carbon monoxide asphyxia.  I.  The  behavior  of

            the htart.  Amer. J. Physiol. 56:390-403, 1921.

158n.   Hnider,  M.,  K.  Croll-Knapp, H. Holler, M. Neuberger, and H. Stidl.
            Effects of moderate CO dose on the central nervous system -—
            electrophysiological and behaviour data and clinical relevance, pp.
            217-232.  In A. J. Finkel and W. C. Duel, Eds.  Clinical Implications
          ~ of Air Pollution Research.  American Medical Association Air Pollu-
            tion Medical Research Conference, 1974.  Acton, Mass.:  Publishing
            Science Group, Inc., 1976.
 159.  Haldane, J.  The action  of carbonic  oxide on man.   J.  Physiol.   (Lond.)
            18:430-462, 1895.
 160.   Halperin, M. H., R. A. McFarland, J. I. Niven, and  F.  J. W. Roughton.
            The time course of the effects of carbon monoxide on visual thresh-
            olds.  J.  Physiol.  (Lond.)  146:583-593,  1959.
160a.   Hampson, R.  F., Jr., and D. Garvin, Eds.  Chemical  Kinetic and Photochemical
            Data for Modelling Atmospheric Chemistry.  NBS Technical Note  866.
            Washington, D. C.:  U. S. Department of Commerce, National Bureau of
            Standards, 1975.  113 pp.
160b.  Haobisch, H,  Die  Zigarette als Kohlenmonoxydquelle.   Arch. Toxic'ol.   26:
            251-261, 1970.
  161.   Hanks,  T.  G.   Human performance of a psychomotor test as a function of
             exposure  to carbon monoxide.   Ann. N.  Y.  Acad. Sci. 174:421-424,
  162.   Harteck, P., and S. Dondes.   Reaction of carbon monoxide and ozone.  J.
             Chem.  Phys. 26:1734-1737, 1957.
  163t   Heicklen, J.   Gas  phase oxidation  of perhalocarbons.   Adv.  Photochem.
             7:57-148,  1969.
 1fi4.   Heicklen, J. . K.  Westberg, and N.  Cohen.  The Conversion of NO to N02 in
            Polluted Atmospheres.  Center for Air Environment Studies Publ. 115-
            169.  University Park:  The Pennsylvania State University, 1969. 5 pp.


 165.   Hellman, L. M.,  and J. A. Pritchard.  Williams  Obstetrics.   (14th ed.)  New
           York:  Appleton-Century-Crofts,  1971.   1,242 pp.                     	
166.   Heron, H. J.  The effects of smoking  during  pregnancy:  A review with a
           preview.  N. Z. Med. J.  61:545-548, 1962.                        	
167.   Herzberg, G.  Molecular Spectra and Molecular Structure.  I.   Spectra
           of Diatomic Molecules.  (2nd ed.)  New  York:  D. Van Nostrand
   _       Company, Inc., 1950.  658 pp.                                        	
168.   Hesiop-Harrison,  J.,  and  Y.  Heslop-Harrison.   Studies on flowering-plant
           growth  and organogenesis.   II.   The modification of sex expression
           in Cannabis  sativa by carbon monoxide.   Proc.  Roy.  Soc. Edinburgh B
           66:424-434,  1957.                                               	
169.  Hesstvedt, E.  Vertical distribution of CO near the tropopause.  Nature
           225:50, 1970.
 170.  Hill, E.  P., J.  R.  Hill,  G.  G.  Power,  and L.  D.  Longo.   Carbon monoxide in
           human maternal and fetal blood.   A mathematical model.   Amer.  J.
           Physio1.  (in press)
 171.  Honig, C.  R., and  J.  Bourdeau-Martini.   Oj and  the number and arrangement of
           coronary capillaries;   effect on calculated  tissue P02>  pp.  519-524.
           In  H.  I. Bicher  and  D.  F.  Bruley,  Eds.   Oxygen Transport to  Tissue.
           Instrumentation,  Methods  and Physiology.   Advances in  Experimental
           Medicine and  Biology.   Volume 37A.   New York:   Plenum  Press,  1973.
172.  Hopfield, J. J., and R. T. Birge.  Ultra-violet absorption  and  emission
           spectra of carbon monoxide.  Phys. Rev.  29:922, 1927.   (abstract)
 173.  Horvath,  S.  M. ,  T.  E.  Dahms,  and J. F.  O'Hanlon.  Carbon monoxide and
           human vigilance.   A deleterious  effect of present  urban concentra-
           tions.   Arch.-Environ.  Health  23:343-347, 1971.

173a.   Horvath,  M.,  and E.  Frantik.  Quantitative interpretation of experimental

            toxicological data:   The use of reference substances, pp. 11-21.  in

            M.  Horvath, Ed.   Adverse Effects of Environmental Chemicals and

            Psychotropic Drugs.   Vol. 1.  Amsterdam:  Elsevier Scientific Publish-

            ing Company, 1973.

174.  Horvath,  S. M.,  P. B. Raven,  T. E.  Dahms,  and D.  J.  Gray.   Maximal

            aerobic capacity at different  levels  of  carboxyhemoglobin.   J.

            Appl. Physiol. 38:300-303, 1975.

175.  Hosko, M. J.  The  effect of  carbon  monoxide on the visual  evoked  response

            in man  and   the spontaneous electroencephalogram.  Arch.  Environ.

            Health 21:174-180, 1970.

176.  Hewlett, L., and R. J.  Shephard.  Carbon monoxide as a hazard  in  aviation,

           J. Occup. Med. 15:874-877, 1973.

177.  Hull,  H.  M.,  and F. W.  Went.   Life processes of plants as affected by air

           pollution,  pp. 122-128.   In-Proceedings of the Second National Air

           Pollution Symposium,  Pasadena,  California, 1952.  Los Angeles:  Natia

           Air Pollution Symposium, 1952.

178.  Hwang, J. C., C.  H. Chen,  and  R.  H.  Burris.   Inhibition  of  nitrogenase-

            catalyzed  reductions.   Biochim.  Biophys.  Acta 292:256-270,  1973.

1794  Ingersoll, R. B.  Soil as  a  Sink  for Atmospheric Carbon  Monoxide.  Final

           Report.  Menlo Park, Calif.:    Stanford Research Institute,  1971.

           40- pp.

180.  Ingersoll, R.  B.  The Capacity of  the Soil as a Natural Sink for Carbon

           Monoxide.   Final Report.  SRI Project LSU-1380.   Menlo Park, Calif.:

           Stanford  Research  Institute,  1972.   38 pp.

181.  Ingersoll, R.  B., R. E. Inman, and W. R. Fishor.  Soil's potential as a

           sink for atmospheric carbon monoxide.  Tellus 26:151-159,  1974.


  1S2.  Inman, R. E., and R. B. Ingersoll.  Uptake of carbon monoxide by  soil
            fungi.  J. Air Pollut. Control Assoc. 21:646-647, 1971.
  183.  Inman,  R. E., R.  B.  Ingersoll,  and E.  A.  Levy.   Soil:   A natural sink for
            carbon monoxide.   Science  172:1229-1231,  1971.
 184.  Jaegor, J. J., and J.  J. McGrath.  Effects of hypothermia on heart and
            respiratory responses of chick to carbon monoxide.  J. Appl. Physiol.
            34:564-567,  1973.
 185.   Jaffe,  L.  S.  Ambient carbon monoxide  and  its fate in  the atmosphere.   J.
            Air  Pollut. Control Assoc.  18:534-540, 1968.
 186.   Jaffe,  L.  S.  Carbon monoxide in the biosphere:   Sources, distributions,
            and  concentrations.   J. Geophys.  Res. 78:5293-5305,  1973.
 187.   Jenkins,  C.  E.  The haemoglobin  concentration of workers  connected with
            internal combustion engines.   J.  Hyg. 32:406-408,  1932.
 188,   Joels,  N., and L.  G. C.  Pugh.   The carbon monoxide  dissociation curve of
            human blood.   J.  Physiol.   (Lond.)  142:63-77,  1958.
189.   Johnson, B. L., H.  H.  Cohen, R.  Struble, J. V. Setzer, W. K. Anger, B.  D.
           Gutnik, T. McDonough, and P. Hauser.  Field evaluation of carbon
           monoxide exposed toll collectors,  pp. 306-328.   In C. Xintaras, B.
           L.  Johnson, and I. de Groot, Eds.   Behavioral Toxicology:   Early
           Detection of Occupational Hazards.  HEW Publ. No.  (NIOSH) 74-126.
           Washington, D. C.:  U. S.  Department of Health, Education,  and
           Welfare, 1974.
 190.   Johnson,  K.  L., L. H. Dworetzky,  and A. N. Heller.  Carbon monoxide and
            air  pollution from  automobile  emissions in  New York  City.   Science
            160:67-68, 1968.
 191.  Jones, R.  II., M. F. Ellicott,  J. B. Cadigan,  and E.  A. Gaensler.   The
           relationship between alveolar and blood carbon monoxide concentra-
           tions during breathholding.  Simple estimation of COHb saturation.
           J.  Lab. Clin.  Med. 51:553-564, 1958.


102.  Jungo, C. K.  Air Chemistry and Radioactivity.  Now York:  Academic  Press,

          Inc., 1963.  382 pp.

 193 <   Junge,  C.  P..   Residence time and variability of tropospheric trace

            gases.   Tellus  26:477-488,  1974.

                                                                 •* A       1 /
194.   Junge,  C. E., W. Seilor,  and  P.  Warneck.   The  atmospheric  i  CO and  i^CO

           ,budget.  J. Geophys. Res. 76:2866-2879,  1971.

195.   Kagnn,  A., B. R. Harris,  W. Winkelstein,  Jr.,  K.  G.  Johnson, H.  Kato,

           S. L. Syme, G. G.  Rhoads, M.  L. Gay,  M.  Z.  Nichaman,  H. B.  Hamilton,

           and J. Tillotson.  Epidemiologic  studies  of coronary heart  disease

           and stroke in Japanese men  living in Japan,  Hawaii  and  California:

           Demographic, physical, dietary and biochemical  characteristics.   J.

           Chron. Ms. 27:345-364,  1974.                                       	

 196.  Kahn,  A.,  R.  B.  Rutledge, D. L.  Davis, J. A. Altes, G. E. Gantner, C. A.

            Thornton, and N.  D.  Wallace.  Carboxyhemoglobin sources  in the

            metropolitan  St. Louis population.   Arch. Environ. Health 29:127-

            135,  1974.

 197.  Kaplan, L.  D., J. Connes,  and P.  Connes.   Carbon monoxide in the Martian

             atmosphere.  Astrophys. J. 157:L187-L192,  1969.   (letter)

  198.  Kaserer,  H.   Die Oxydation des  Wasserstoffes durch Mikroorganismen.

             Zentralbl.  Bakteriol. Parasitenk. Infektionskrank. Abt. II 16:

             681-696,  1906.

  199.  Kennedy,  A.  C.,  and D. J.  Valtis.  The oxygen dissociation curve  in anemia

             of various  types.  J. Clin. Invest. 33:1372-1381,  1954.

 200.  Kessler, M., H. Lang,  E. Sinagowitz,  R.  Rink,  and  J.  Hoper.  Homeostasis  of

             oxygen  supply in liver and kidney,  pp.  351-360.  In H. I.  Bicher and

             D. F. Bruley.  Oxygen Transport  to  Tissue. Instrumentation,  Methods,

             and Physiology.   Advances  in Experimental  Medicine and Biology.   Volume

             37A.  New York:   Plenum Press, 1973.


 201.  Keys, A., Ed.  Coronary Heart Disease in Seven Countries.  American Heart

           Association Monograph Number 29.  Circulation 41 Suppl. 1:1-1-1-211,


202.  Killick,  E. M.  The  acclimatization  of mice  to atmospheres  containing  low

           concentrations  of carbon monoxide.  J.  Physiol.  (Lond.)  91:279-292,


203.  Killick,  E. M.  The  acclimatization  of the human  subject  to atmospheres

           containing low  concentrations of carbon monoxide.  J.  Physiol.  (Lond.)

           87:41-55, 1936.

204. Killick, E. M.  The nature of the acclimatization occurring during

          repeated exposure of the human subject  to atmospheres containing

          low concentrations of carbon monoxide.  J. Physiol.  (Lond.) 107:

          27-44, 1948.

204a.  Kjeldsen, K.,  P.  Astrup,  and J.  Wanstrup.   Ultrastructural intimal changes

            in the rabbit aorta  after a moderate carbon monoxide exposure.

            Atherosclerosis  16:67-82,  1972.

205.  Kjeldsen, K., and  F. Damgaard.   Influence of prolonged carbon monoxide

           exposure and  high altitude on the composition of blood and urine

           in man.   Scand. J.  Clin.  Lab.  Invest.  22(Suppl.  103):20-25, 1968.

206.  Kjeldsen, K., H. K. Thomsen, and P. Astrup.  Effects of carbon monoxide

           on myocardium.  Ultrastructural changes in rabbits after moderate,

           chronic exposure.  Circ. Res. 34:339-348, 1974.

 207.  Klausen,  K., B. Rasmussen, H. Gjellerod, H. Madsen,  and  E. Petersen.

           Circulation, metabolism and ventilation during  prolonged  exposure

           to  carbon monoxide and to high altitude.  Scand. J. Clin.  Lab.

           Invest. 22(Suppl. 103):26-38,  1968.

 203.  Klebha, A. J., J. n. Maurer, and E. J. Glass.  Mortality Trends for  Lead-
            ing Causes of Death, United States--1950-1969.  U. S. Department of
            Health,  Education, and Welfare, National Center for Health Statistics.
            Vital and Health Statistics Series 20, No. 16.  Washington, D. C.:
            U. S. Government Printing Office, 1974.  75 pp.
 209.  Klcndshoj, N. C., M. Feldstein, and A. L. Sprague.  The spectrophotometric
            determination of carbon monoxide.  J. Biol. Chem. 183:297-303, 1950.
 210.  Kluyver,  A.  J., and C.  G. T. Schnellen.  On the fermentation of carbon
            monoxide by pure cultures of methane bacteria.  Arch. Biochem. 14:
            57-70, .1947.
 2ii.  Koob,  G.  F.,  Z.  Annau,  R. J. Rubin, and M. R.  Montgomery.   Effect of
            hypoxic  hypoxia and carbon monoxide on food intake,  water intake,
            and  body weight in two strains of rats.  Life Sci. 14:1511-1520,
 212.  Kortschak, H. P., and L. G. Nickell.  Photosynthetic carbon monoxide
            metabolism by sugarcane leaves.  Plant Sci. Lett. 1:213-216, 1973.
 213.  Knight, L. I.,  and W. Crocker.   Toxicity of smoke.  Contributions from the
            Hull Botanical Laboratory 171.  Bot. Gazette 55:337-371, 1913.
 214.  Krause, A.  Mcchanismus  der katalytischen Oxydation von CO mit N«0.  Bull.
            Acad. Pol on. Sci.  Scr. Sci. Chim. 9:5-6, 1961.
 215.  Krall, A. R., and N. E. Tolbert.  A comparison of the  light dependent
            metabolism of carbon monoxide by barley leaves with that of formal-
            dehyde,  formate and carbon dioxide.  Plant Physiol. 32:321-326, 1957.
215a.  Kullander, S.,  and B. Kailen.  A prospective study of smoking and preg-
            nancy.   Acta Obstet. Gynecol. Scand. 50:83-94, 1971.
215b,  Kubic, V. L.  M. W. Anders,  R. R. Engel, C. H. Barlow,  and W. S. Caughey.
            Metabolism of dihalomethanes  to carbon monoxide.  1.  In vivo studies.
            Drug. Metab. Dispos.   2:53-57, 1974.


  216.   Kullcr, L. , J. Pcrper,  and M.  Cooper.   Sudden  and  unexplained death due
             to nrteriosclerotic heart disease,  pp.  292-332.   In M.  F.  Oliver,
             Ed.  Modern Trends in Cardiology.   Vol. 3.  London:   Butterworths,
 217.  Kullor, L. H., E. P. Radford, D. Swift,  J. A. Perper,  and R.  Fisher.
           Carbon monoxide and heart  attacks.   Arch.  Environ.  Health 30:477-
           482, 1975.
 218.  Kummler, R.  II. , and T.  Baurer.  A  temporal model  of tropospheric carbon-
           hydrogen  chemistry.  J.  Geophys.  Res. 78:5306-5316,  1973.
 219.  Kummler,  R.  II.,  R.  N.  Grenda, T.  Baurer, M.  H.  Bortner, J. H. Davies, and
            J.  MacDowall.   Satellite solution of the carbon monoxide sink  anomaly.
            EOS Trans.  Amer.  Geophys.  Union 50(4):174, 1969.  (abstract)
 220.   Lambert,  J.  L.,  R.  R.  Tschorn, and  P.  A.  Hamlin.   Determination of carbon
           monoxide  in blood.   Anal. Chem. 44:1529-1530,  1972.

 221.   Lamontagne,  R. A.,  J.  W.  Swinnerton,  and V. J. Linnenbom.  Nonequilibrium
            of carbon monoxide and methane at the air-sea interface.  J. Geophys.
            Res. 76:5117-5121,  1971.
 222.   Larsen,  R.  I., and  H.  Burke.   Ambient  carbon monoxide exposures.  Paper
            69-167  Presented  at  62nd  Annual Meeting of the Air Pollution Control
            Association, New  York, 1969.   39  pp.
222a.   Laties,  V.  G.   On the use of reference substances in behavioral toxicology,
            pp. 83-88.   In M.  Horvath,  Ed.   Adverse Effects of Environmental Chem-
            icals  and Psychotropic Drugs.   Vol. 1.  Amsterdam:  Elsevier Scienti-
            fic Publishing Company, 1973.

 223.  Ratios, V. G., nnd B. Weiss.  Performance  enhancement  by  the  amphetamines:
           A new appraisal, pp.  800-808.  In  H.  Brill,  J.  0.  Cole,  P.  Deniker,
           II. Ilippius, and P. B. Bradley, Eds.   Neuropsycho-Pharmacology.   Pro-
           ceedings of the Fifth International Congress of the  Collegium Inter-
           nationale Ncuro-Psycho-Pharmacologicum.  Washington, D.  C.,  28-31
           Maxell, 1966.  International Congress  Series No. 129.  Amsterdam: .
           Excerpta Mcdica Foundation, 1967.
 224.   Leighton,  P.  A.   Other reactions of oxygen atoms,  pp.  146-150.   In Photo-
            chemistry of Air Pollution.  New York:   Academic  Press,  1961.
 225.   Lenfant, C.,  J.  Torrance,  E.  English, C.  A.  Finch, C.  Reynafarje,  J.  Ramos,
            and J. Faura.   Effect of altitude on oxygen binding by hemoglobin and
            on organic  phosphate  levels.   J. Clin.  Invest.  47:2652-2656,  1968.
 226.   Levy, H.,  II.  Normal atmosphere:   Large radical and formaldehyde concentra-
            tions predicted.  Science 173:141-143,  1971.
 227.  Levy, H., II.   Photochemistry of the  lower troposphere.   Planet. Space
            Sci. 20:919-935,  1972.
 228.  Levy, H.,  II.  Photochemistry of minor constituents in the troposphere.
            Planet.  Space Sci. 21:575-591,  1973.
 229.  Lewey, F.  11.  , and D. L. Drabkin.  Experimental chronic carbon monoxide
            poisoning of dogs.  Amer.  J. Med. Sci. 208:502-511, 1944.
 230.   Lewin, L.   Die Kohlenoxydvergiftung.  Ein Handbuch  fur Mediziner,
            Techniker und Unfallrichter.  Berlin:  Julius  Springer, 1920.
            370 pp.
230a.   Lewis, J., A. D. Baddeley, K. G. Bonham, and D. Lovett.  Traffic pollution
            and mental efficiency.  Nature 225:95-97, 1970.
 231.  Liberti, A.   The nature of particulate matter.   Pure Appl.  Chem. 24:631-
            642,  1970.

  332.  Lichty, J. A.,  R. Y. Ting, P. D.  Bruns,  and  E.  Dyar.   Studies  of babies

             born  nt high altitudes.  I.   Relation of  altitude to  birth  weight.

             A.M.A. J.  Dis. Child. 93:666-669, 1957.

  233.  Liebl, K.  II.   Dor Bodon als Senke und Quell fur das Atm. Kohlcnoxid.

             Diplomarbeit, 1971.  85 pp.

 234.   Lighthart,  B.   Survival of airborne bacteria in a high urban concentration

            of carbon monoxide.  Appl.  Microbiol.  25:86-91, 1973.

 235.  Lilionthal, J.  L., Jr.   Carbon monoxide.  Pharmacol. Rev.  2:234-254, 1950.

  236.  Lilienthal, J.  I.,  Jr., and C. H. Fugitt.  The effect of low concentra-

            tions  of carboxyhemoglobin  on the "altitude tolerance" of man.

            Amer.  J.  Physiol.  145:359-364, 1946.

 237.  Lind,  C.  J. , and P. W. Wilson.  Carbon monoxide inhibition of nitrogen

           fixation  by azotobacter.   Arch. Biochem.  1:59-72, 1942.

238.  Lind,  C. J., and P.  W. Wilson.   Mechanism of biological nitrogen  fixation.

           VIII.   Carbon monoxide as  an inhibitor for nitrogen fixation by red

           clover.  J. Amer.  Chem. Soc. 63:3511-3514, 1941.

 239.   Linderholm, II.,  T. Sjostrand,  and B.  Soderstrom.   A method  for determina-

            tion of  low carbon monoxide concentration in blood.  Acta Physiol.

            Scand. 66:1-8,  1966.

 240.   Link,  W. T.,  E.  A. McClatchie, D.  A.  Watson,  and A.  S.  Compher.  A

            fluorescent source NDIR carbon monoxide  analyzer.   Presented at

            Conference on Methods in  Air Pollution and Industrial  Hygiene

            Studies,  12th,  Los Angeles,  California,  April 6-8, 1971.

241.  Linnenbom,  J.  V., J.  W.  Swinnerton, and R. A.-Lamontagne.  The ocean as

           a source  for atmospheric carbon monoxide.  J. Geophys.  Res.  78:5333-

           5340,  1973.

 242.  Liss, P. S., and P. G. Slater.  Flux of gases across the air-sea  interface

            Nature 247:181-184, 1974.

 243.  Lissi,  E.,  R.  Simonaitis, and J. Heicklen.  The bromine atom catalyzed

            oxidation of carbon monoxide.  J.  Phys. Chem. 76:1416-1419, 1972.

 244.   Locke,  J. L.,  and L.  Herzberg.  The absorption due to carbon monoxide in

            the infrared solar sprectrum.  Can.  J. Phys.  31:504-516, 1953.

 245.  Lockshin, A.,  and R.  H. Burris.  Inhibitors of nitrogen fixation in

            extracts  from Clostridium pasteurianum.  Biochim.  Biophys. Acta

            111:1-10, 1965.

2"46.   loewus,  M.  W., artd C.  C.  Delwiche.  Carbon monoxide production, by algae.

            Plant  Physiol.  38:371-374, 1963.

246a.  Longo,  L.  D.  Carbon monoxide:  Effects on oxygenation of the fetus in

            utero.  Science   194:523-525,  1976.

 247.  Longo,  L. D.   Carbon  monoxide in the pregnant mother and fetus and its

            exchange  across  the placenta.  Ann.  N. Y.' Acad. Sci.  174:313-341, 1970.

247a.  Longo,  L.  D.  The biologic effects of carbon monoxide on the pregnant woman,

            the fetus and newborn infant.  Amer.  J. Obstet. Gynecol.  (in press)

24?b.  Longo,  L.  D., and K. Ching.  Placental diffusion  capacity for carbon

            monoxide and  oxygen in unanesthetized  sheep.   Amer. J.  Physiol.

            (in press)

 248.  Longo, L. D. ,  and E.  P.  Hill.  Carbon  monoxide uptake  and  elimination
                                                                    !-; '•?
            in fetal  and maternal sheep.   Amer J. Physiol.  (in press)

248a.  Longo,  L.  D. , E. P. Hill, and  G.  G. Power.   Theoretical analysis  of  factors

            affecting placental 02 transfer.  Amer. J. Physiol. 222:7J(X739,  1972.

 249.  Longo,  L. D.,  G. G. Power, and R.  E. Forster, II.   Respiratory fraction-

            of the placenta as determined with carbon monoxide in sheep and

            dogs.   J. Clin.  Invest. 46:812-828, 1967.

   250.  Los Angeles County Air Pollution Control District,  1974 Profile  of  Air
             Pollution Control.  Los Angeles:  County of Los Angeles, Air Pollution
             Control District, 1974.  91 pp.
 250a.  Lowe,  C.  R.   Effects  of mothers'  smoking habits  on birthweight of their
             children.   Brit. Med.  J.  2:673-676,  1959.
  251.  Luomanmaki, K.  Studies on the metabolism of carbon monoxide.  Ann.   Med.
            Exp. liiol. Fenniae 44(Suppl. 2): 1-55, 1966.
   252.  Mackworth,  J.  F.   Vigilance and Attention.   Baltimore:   Penguin Books,
             Inc.,  1970.   188 pp.
   253.  Mackworth,  J.  F.   Vigilance and Habituation.  Baltimore:  Penguin Books,
             Inc.,  1969.   237 pp.
  254.  Mackworth,  N.  H.   Researches on the Measurement  of Human Performance.
             Medical Research Council  Special Report Series No. 268.  London:
             His  Majesty's Stationery Office,  1950.   156 pp.
 255.  Madley, D. G.,  and R.  F. Strickland-Constable.  The kinetics of the
            oxidation of charcoal with nitrous oxide.  Trans.  Faraday Soc.
            49:1312-1324, 1953.
 256.  Malenfant, A. L.,  S. R.  Gambino, A.  J.  Waraksa, and E. I.  Roe.   Spectro-
           photometric determination  of hemoglobin  concentration and  per cent
           oxyhemoglobin and carboxyhemoglobin saturation.  Clin.  Chem.  14:789,
           1968.  (abstract)
256a.  Manning, F.,  E.  Wyn Pugh,  and K.  Boddy.  Effect  of cigarette smoking on
            fetal breathing movements  in normal pregnancies.  Brit. Med.  J.
            1:552-553,  1975.
256b.  Manning, F.  A.,  and C. Feyerabend.   Cigarette smoking and fetal breathing
           movements.  Brit. J.  Obstet. Gynaecol.   83:262-270, 1976.

25'ic.   Maxwell, J. C. , C. II.  Barlow, J. E. Spallholz,  and W.  S,  Caughey.   The

            utility of  infrared spcctroscopy as  a probe of  intact  tissue:   Deter-

            mination of carbon monoxide and hemeproteins in blood  and  heart muscle.

            Biochem. Biophys. Res. Commun.  61:230-236, 1974.

 257.  McClatchio, E. A.  Development of an Infrared  Fluorescent Gas Analyzer.

            EPA-R2-72-121.   Berkeley, Calif.:  Akron  Scientific Labs,  1972.  7  pp.

 258.  McClatchie, E. A., A. B. Compher, and K.  G. Williams.  A high specificity

            carbon monoxide  analyzer.  Anal. Instrum.  (Symposium)  10:67-69, 1972.

 259.    McConnell, J.  C.,  M.  B.  McElroy,  and S. C.  Wofsy.   Natural sources  of

             atmospheric CO.   Nature 233:187-188, 1971.

 260.  McCortnick, R.  A., and C.  Xintaras.  Variation of carbon monoxide concen-

            trations as related to sampling interval, traffic and meteorological

            factors.   J. Appl. Meteorol. 1:237-243, 1962.

 261.  McCullough, J. D., R. A.  Crane, and A. 0. Beckman.   Determination  of car-

            bon monoxide in  air by use of red mercuric oxide.   Ind. Eng.  Chem.

            Anal. Ed. 19:999-1002, 1947.

 262.  McFarland, R.  A,   Low level exposure to carbon monoxide  and driving

            performance.  Arch.  Environ. Health 27:355-359, 1973.

 263.  McFarland, R. A.  The effects of exposure to small quantities of carbon

            monoxide on vision.  Ann. N. Y. Acad. Sci. 174:301-312,  1970.

 264.  McFarland, R.  A., W.   H. Forbes, H. W. Stoudt, J. D.  Dougherty,  T.  J.

            Crowley,  R.  C. Moore,  and T. J. Nalwalk.  A Study of the Effects  of

            Low Levels of Carbon Monoxide upon Humans Performing Driving  Tasks.

            Final Report, CRC-APRAC Contract CAPM-9-69(2-70), June 15, 1970-

            September 15, 1972.   Boston:  Guggenheim Center  for Aerospace

            Health and Safety, Harvard School of Public Health, 1973.  88 pp.

 2f>S.  McFarl.ind, R. A., W. H. Forbos, II. W. Stoudt, J. D.  Dougherty,  A.  J.
           Morandi, and T. J. Nalwalk.  A Study of  the Effects  of  Low Levels
           of Carbon Monoxide upon Humans Performing Driving Tasks.   Final
           Report.  June 15, 1970-June 14, 1971.  Boston:  Harvard School of
           Public Health, Guggenheim Center for Aerospace  Health and  Safety,
           1971.  56 pp.
 266.  McFarl.ind, R. A., D. J. W.  Rough ton, M.  H.  Halperin, and  J.  I.  Niven.
           The  effects  of carbon  monoxide and  altitude on  visual  thresholds.
           J. Aviat. Med. 15:381-394, 1944.                                  _____
 267.  McGrath, J. J., and J. Jaeger.  Effect of iodoacetate on  the carbon monoxide
           tolerance of the chick.  Respir. Physiol. 12:71-76,  1971.
268.   McKee,  H.  C.,  and  R. E.  Childers.   Collaborative  Study  of Reference
           Method  for the Continuous  Measurement  of Carbon Monoxide in the
           Atmosphere  (Non-Dispersive  Infrared  Spectrometry).   EPA-R4-72-009.
           Houston,  Texas:   Southwest  Research  Institute, 1972.   l_ 41 pp._/
269.   McMillan,  C.,  and  J. M.  Cope.  Response to carbon monoxide by geographic
           variants  in Acacia  farnesiana.  Amer. J.  Bot.  56:600-602,  1969.
 270.  McMullen,  T.  B.   Interpreting the eight-hour national ambient air quality
           standard for carbon monoxide.  J. Air Pollut.  Control Assoc. 25:
           1009-1014,  1975.

271.   McNesby, J. R., and H. Okabe.  Vacuum ultraviolet  photochemistry.  Adv.
           Photochem.  3:157-240,  1964.
 272.  Delete

273.   Mellor, J. W.   The chemical properties of carbon monoxide, pp.  926-950.
           In A Comprehensive Treatise on Inorganic and Theoretical Chemistry.
           Vol.  V.   B, Al, Ga, In, Tl, Sc, Ce and Rare Earth Metals, C (Part  1).
           London:   Longmans,  Green and Co., 1924.


 274.  Mensor,  H. ,A.,  J.  J.  Grosso,  H.  E.  Heggstad,  and 0.  E.  Street.  Air filtra-

            tion study of "hidden" air  pollution injury to  tobacco plants.  Plant

            Physiol.  39(Suppl.)lviii,  1964.   (abstract)

274.1.  Mover,  M. B. ,  fi.  S.  Jonas,  and J.  A.  Tonascia.  Perinatal events associated

            with mnternnl smoking during pregnancy.   Amer.  J.  Epidemiol.  103:

            464-476,  1976.

 275.  Meyer,  M. B.,  and G.  W.  Comstock.   Maternal cigarette smoking and perinatal

            mortality.  Amcr.  J.  Epidemiol.  96:1-10, 1972.

 276.  Meyer, M. B. , J. A. Tonascia, and C. Buck.  The  interrelationship of

           maternal smoking and increased perinatal mortality with other risk

            factors.   Further analysis  of the Ontario perinatal mortality study,

            1960-1961.  Amer. J. Epidemiol. 100:443-452,  1974.

 277.  Migeotte, M.   The fundamental band of carbon monoxide at 4.7u in the solar

            spectrum.   Phys.  Rev.  75:1108-1109,  1949.

 278.  Migootte, M.,  and L. Neven.  Recents proges  dans  1'observation du spectre

            infrarouge  du soleil  a  la  station scientifique du  Jungfraujoch

            (Suissc).  Mem. Soc.  Roy.  Sci. Liege  12:165-178,  1952.

 279.  Mikulka, P.,  R.  O'Donnell, P. Heinig,  and  J. Theodore.   The effect of

            carbon monoxide  on human performance.   Ann.  N. Y.  Acad. Sci. 174:

            409-420,  1970.

  280.  Minina,  E. G., and L. G. Tylkina.  Physiological  study of  the effect  of

            gases upon sex differentiation in plants.  Dokl. Akad.  Nauk SSSR

            55 (2):165-168, 1947-  (in  Russian)

 281.  Miranda, J.  M., V.  J.  Konopinski,  and R.  I.  larsen.   Carbon monoxide

            control in a high highway tunnel.  Arch. Environ.  Health 15:16-25,


 232.  Mocllor, T.  Carbon monoxide, pp. 685-687.  In Inorganic Chemistry.   An
           Advanced Textbook.  New York:  John Wiley & Sons, Inc.,  1952.
283.   Montgomery,  M.  R.,  and R.  J.  Rubin.   Adaptation to the inhibitory effect
           of carbon  monoxide inhalation on drug metabolism.  J.  Appl.  Physiol.
           35:601-607,  1973.
 284.  Montgomery, M.  R. , and R. J. Rubin.  Oxygenation during inhalation of drug
           metabolism by carbon monoxide or hypoxic hypoxia.  J. Appl. Physiol.
           35:505-509, 1973.
 285.  Montgomery, M.  R., and R. J. Rubin.  The  effect of carbon monoxide inhala-
           tion on in vivo drug metabolism in the rat.  J.  Pharmacol.  Exp.  Ther.
           179:465-473, 1971.
286.   Mourant,  R.  R.,  and T.  H.  Rockwell.   Strategies of visual search by novice
           and  experienced drivers.   Human Factors  14:325-335, 1972.
287.  Mulhausen,  R. 0., P. Astrup, and K. Mellemgaard.   Oxygen affinity  and
           acid-base status  of human blood during exposure  to  hypoxia  and
           carbon monoxide.  Scand. J. Clin. Lab. Invest. 22(Suppl.  103):9-
           15, 1968.
288.   Muller,  R.  H.   A supersensitive gas detector permits accurate detection
           of toxic  or combustible gases in extremely low concentrations.   Anal.
           Chem.  26(9):39A-42A,  1954.
289.  Murray, J.  F., P. Gold, and B.  L. Johnson, Jr.  Systemic oxygen  transport
           in induced normovolemic anemia and polycythemia.  Amer. J.  Physiol.
           203:720-724, 1962.
290.  Musselman,  N.  P., W. A. Groff,  P. P. Yevich, F. T. Wilinski, M. H.  Weeks,
           and  F.  W.  Oberst.   Continuous exposure of laboratory animals to  low
           concentration of carbon monoxide.   Aerosp. Med. 30:524-529, 1959.
291.   Nasmith,  G.  G. ,  and D.  A.  L. Graham.  The haematology of carbon-monoxide
           poisoning.   J. -Physiol. (Lond.) 35:32-52, 1906.


             -- use 412.

 293 <   National Research Council.   Division of Medical Sciences.  Committee on
            Effects of Atmospheric Contaminants on Human Health and Welfare.
            Effects of Chronic Exposure to Low Levels of Carbon Monoxide on
            Human Health, Behavior,  and Performance.  Washington, D. C. :
            National Academy of Sciences,  1969.  66 pp.                  ... .  _____
 294.  Keill, U. A.  Effects of arterial hypoxemia  and hyperoxia on  oxygen  avail-
           ability  for myocardial metabolism.  Patients with  and withou-t coronary
           heart disease.  Amer. J. Cardiol.  24:166-171,  1969.
 295.   Delete  -- use 74b.
 296.   Nicholson,  S. E.   A pollution model for street-level air.  Atmos.  Environ.

            9:19-31, 1975.
  297.  Nielsen, B.  Thermoregulation during work in carbon monoxide poisoning.
            Acta Physiol. Scand. 82:98-106, 1971.
2?7a.   X-'.svander,  K. R. , and M.  Gordon.  Cigarette smoking, pp. 72-80.  In The
            Women  and Their Pregnancies.   The Collaborative Perinatal Study of
            the National Institute of  Neurological Diseases and Stroke.  Phila-
            delphia:  W.  15.  Saunders Company,  1972.
 298.  North Atlantic Treaty Organization.  Committee  on Challenges  of Modern
           Society.  Air  Pollution.  Air  Quality Criteria for Carbon  Monoxide.
           N. 10.  Brussels:  North  Atlantic Treaty Organization, 1972. /265 pp._/

  299.  Delete

 299a.  Noyes, W, A.. Jr., and ?. A. teighton.  Appendix CC-1,11, Photochemical
            reactions of carbonyls, p. 449.   In The Photochemistry  of Gases.
            New York:  Dover Publications, Inc., 1966.

  300.   O'Donnell,  R.  D., P. Chikos,  and J. Theodore.  Effect  of carbon monoxide

             exposure  on human sleep  and psychomotor performance.  J.  Appl.

             Physiol.  31:513-518,  1971.

  301.   O'Donnell, R.  D. , P. Mikulkn, P. Ileinig, and J. Theodore.  Low level

            carbon monoxide exposure and human psychomotor performance.  Toxicol,

            Appl. Pharmacol. 18:593-602, 1971.

 302.  O'Hanlon,  J.  F.  Preliminary  studies of the  effects  of  carbon monoxide

            on vigilance in man,  pp.  61-72.   In B.  Weiss  and V.  G. Laties,  Eds.

            Behavioral Toxicology.   New York:   Plenum  Press, 1975.

302a.  Ontario Department  of Health.  Perinatal Mortality Study Committee.

            Second Report  of Perinatal Mortality Study in Ten University

            Teaching Hospitals,  Ontario, Canada.  Toronto:   Ontario Department

            of Health, 1967.  274 pp.

302b.  Ontario Department  of Health.   Perinatal Mortality Study Committee.

            Supplement to  the  Second Report of the  Perinatal Mortality Study

            in Ten University  Teaching Hospitals.   Toronto, Canada:   Ontario

            Department of  Health, 1967, pp. 85-275.

 303.  Opitz,  E.   Increased vascularization of the  tissue due  to acclimatization

            to high  altitude and  its  significance for  oxygen transport.   Exp.

            Med.  Surg.  9:389-403,  1951.

 304.  Ott, W. ,  J. H.  Clark, and G.  Ozolins.   Calculating Future Carbon Monoxide

            Emissions  and  Concentrations from Urban Traffic Data.  National Air

            Pollution   Control Administration Publ. 999-AP-41.  Durham,  N.  C.:

  	    u- s- Department of  Health, Education,  and Welfare,  1967.  40 pp.

 305.  Palanos,  P.  N.   A practical design  for an ambient  carbon monoxide  mercury

            replacement analyzer.  Anal. Instrum.  (Symposium)  10:117-125,  1972.

  306.   Pfllck,  J.,  V.  Hrabcc,  L.  Mircovova",  V.  Volok,  and B.  Friedmann.   Rod cell
             carbohydrate metabolism in primary refractory anemias.   Clin.  Chim.
             Acta  18:39-45,  1967.
  307.  Park, C. D., I. Mela, R. Wharton, J. Reilly, P. Fishbein, and E. Aberdeen.
            Cardiac mitochondrial activity  in  acute and chronic cyanosis.   J.  Surg.
            Res/ 14:139-146, 1973.
 308.  Partington,  J.  R.   Carbon monoxide,  pp.  686-696.  In Textbook of Inorganic
            Chemistry for University Students.  (5th ed.)  London:   MacMil-lan &
            Co.,  Limited, 1937.
 309.  Pauling, L.   / Resonant energy of carbon monoxide_/, pp. 194-195.  In The
            Nature of the Chemical Bond and the Structure of Molecules and Crys-
            tals:   An Introduction to Modern Structural Chemistry. (3rd ed.)
            Ithaca, N. Y.:  Cornell University Press,  1960.
 310.  Penney,  D.,  M.  Benjamin,  and  E.  Dunham.   Effect of carbon monoxide on
            cardiac weight as  compared  with  altitude effects.   J.  Appl.  Physiol.
            37:80-84,  1974.
  311.  Penney, D., E. Dunham,  and M. Benjamin.   Chronic carbon monoxide exposure:
            Time course of hemoglobin, heart weight and lactate dehydrogenase
            isozyme changes.  Toxicol.  Appl. Pharmacol. 28:493-497; 1974.
  312.   Permutt, S.,  and  L.  Farhi.   Tissue hypoxia  and carbon monoxide,  pp.  18-24.
             In National  Research Council.   Division of Medical  Sciences.  Committee
             on Effects of Atmospheric  Contaminants on Human Health and  Welfare.
            Effects of Chronic Exposure to  Low Levels  of Carbon Monoxide on Human
            Health, Behavior, and Performance.  Washington, D. C.:  National
     _      Academy of Sciences, 1969.
312a.  Pernoll, M.  I., J. Metcalfe,  T.  L.  Schlenker,  J. E.  Welch,  and J. A.
            Matsumoto.  Oxygen consumption  at rest  and during exercise in preg-
            nancy.   Respir.  Physiol.  25:285-293,  1975.

  .313.  Perper, J. A. , 7,. II. Kuller, and M. Cooper.  Arteriosclerosis of coronary
            arteries in sudden, unexpected deaths.  Circulation 52(6 Suppl.  3):
            27-33, 1975.
  314.  Peterson, W.  F. , K. N. Morese, and D. F. Kaltreider.  Smoking and prema-
            turity.  A preliminary report based on study of 7740 Caucasians.
            Obstet.  Gynecol. 26:775-779, 1965.
  315.   Pierce,  J.  0., and  R.  J.  Collins.   Calibration of an infrared analyzer
             for continuous measurement  of carbon  monoxide.   Amer.  Ind. Hyg.
            Assoc.  J. 32:457-462,  1971.
  316.   Pirnay,  F.,  J.  Dujardin,  R. Deroanne, and J.  M. Petit.  Muscular exercise
             during intoxication by carbon monoxide.   J. Appl. Physiol. 31:573-575,
  317.  Pitts, G. C., and N. Pace.  The  effect  of  blood carboxyhemoglobin concen-
            tration  on hypoxia tolerance.  Amer.  J. Physiol.  148:139-151,  1947.
              s                  /'
  318.  Plevova, J.,  and E. Frantik.  The influence of various  saturation rates
            on motor performance of rats exposed  to carbon monoxide.   Activ.
            Nerv. Super. 16:101-102, 1974.
  319.  Porter, K., and D.  H. Volman.  Flame ionization detection of carbon monox-
            ide for gas  chromatographic analysis.  Anal. Chem. 34:748-749,  1962.
319a.  Power,  G.  G.,  and L. D.  longo.   Fetal circulation times and their implica-
            tions for tissue oxygenation.   Gynecol.  Invest.  6:3420355,  1975.
  320.  Pressman, J., and P. Warneck.  The stratosphere as a chemical sink for
            carbon monoxide.  J.  Atmos.  Sci. 27:155-163, 1970.
  321.  Pressman,  J.,  I.  M.  Arin,  and P.  Warneck.   Mechanisms for Removal of Carbon
            Monoxide from  the Atmosphere.   Report  GCA TR-70-6-G.   Bedford,  Mass.:
            GCA Corporation,  1970.   45 pp.

 322.  Primary  prevention  of  the atherosclerotic  diseases,  pp.  15-56.   In I.  S.
           Wright  and  D.  T.  Fredrickson,  Eds.  Cardiovascular  Diseases.   Guide-
           linos  for Prevention and  Care.   Reports  of the  Inter-Society  Commission
           for Heart Disease Resources.  Washington,  D.  C.:  U.  S.  Government
           Printing Office,  1973.
 323.  Quayle,  J. R.  The  metabolism  of one-carbon compounds  by micro-organisms.
           Adv. Microb. Physiol. 7:119-203,  1972.
 324.  Radford, E.  P.,  and E. Kresin.  Red cell diphosphoglycerate after exposure
           to  carbon monoxide.   Fed. Proc. 32:350,  1973.  (abstract)
324a.  Rahn,  H., and W. 0. Fenn.  A Graphical Analysis of the Respiratory Gas
           Exchange.   The C^-CO^ Diagram.   Washington, D.  C.:   American
           Physiological  Society,  1955.   38 pp.
 325.  Ramsey,  J. M.  Carbon  monoxide,  tissue hypoxia, and  sensory psychomotor
           response in hypoxaemic  subjects.   Clin.  Sci.  42:619-625, 1972.
 326.  Ramsey,  J.  M.  Carboxyhemoglobinemia in parking garage employees.   Arch.
           Environ. Health 15:580-583,  1967.
 327.  Ramsey,  J. M.  Effects of single exposures of carbon monoxide on sensory
           and psychomotor response. Amer.  Ind. Hyg. J. 34:212-216,  1973.
 328.  Raven, P. B., B.  L.  Drinkwater, S. M.  Horvath,  R.  0. Ruhling, J. A.  Gliner,
           J.  C. Sutton,  and N.  W. Bolduan.   Age, smoking  habits, heat stress,
           and their interactive effects with  carbon  monoxide  and peroxyacetyl-
           nitrate on  man's  aerobic  power.   Int.  J. Biometeorol.  18:222-232,  1974.
 329.  Raven, P. B., B.  L.  Drinkwater, R.  0.  Ruhling,  N.  Bolduan,  S. Taguchi,  J.
           Gliner, and S.  M.  Horvath.  Effect  of carbon  monoxide and  peroxyacetyl
           nitrate on  man's  maximal  aerobic  capacity. J.  Appl.  Physiol. 36:288-
           293, 1974.

  313.  Torpor, J. A., I. II. Kuller, and M. Cooper.  Arteriosclerosis of coronary

            arteries in sudden, unexpected deaths.  Circulation 52(6 Suppl.  3):

            27-33, 1975.

  314.  Peterson, w. F., K. N. Morese, and D. F. Kaltreider.  Smoking and prema-

            turity.  A preliminary report based on study of 7740 Caucasians.

            Obstet. Gynecol. 26:775-779, 1965.                                   _

  315.   Pierce,  J.  0.,  and  R.  J.  Collins.   Calibration of an infrared analyzer

             for continuous  measurement  of carbon  monoxide.   Amer.  Ind. Hyg.

             Assoc. J.  32:457-462,  1971.

  316.   Pirnay,  F.,  J.  Dujardin,  R. Deroanne, and  J.  M. Petit.  Muscular exercise

             during intoxication by carbon monoxide.   J. Appl. Physiol. 31:573-575,


  317.  Pitts, G. C., and N. Pace.  The  effect  of  blood carboxyhemoglobin  concen-

             tration on hypoxia tolerance.  Amer.  J. Physiol.  148:139-151,  1947.

              s                  /'
  318.  Plevova, J., and E. Frantik.  The influence of various  saturation  rates

            on motor performance of rats exposed  to carbon monoxide.   Activ.

            Nerv. Super. 16:101-102, 1974.

 319.  Porter, K., and D.  H. Volman.   Flame ionization detection of carbon monox-

            ide for gas chromatographic analysis.   Anal. Chem. 34:748-749, 1962.

319a. '  Power,  G.  G.,  and L.  D.  longo.   Fetal circulation times and their implica-

            tions  for tissue oxygenation.   Gynecol.  Invest.  6:3420355,  1975.

 320.  Pressman, J.,  and P.  Wameck.   The stratosphere as a chemical sink for

            carbon monoxide.  J.  Atmos.  Sci. 27:155-163, 1970.

 321,  Pressman,  J.,  L.  M.  Arin,  and  P.  Warneck.   Mechanisms for Removal of Carbon

            Monoxide  from  the Atmosphere.   Report  GCA TR-70-6-G.   Bedford, Mass.:

            GCA Corporation,  1970.  45  pp.

?22.  Primary  prevention  of  the  atherosclerotic  diseases,  pp.  15-56.   In I.  S.

           Wright  and  D.  T.  Fredrickson,  Eds.  Cardiovascular  Diseases.   Guide-

            linos  for Prevention  and  Care.   Reports  of the  Inter-Society  Commission

            for Heart Disease Resources.   Washington,  D.  C.:  U.  S.  Government

           Printing Office,  1973.

323.  Quayle,  J.  R.  The  metabolism  of  one-carbon compounds  by micro-organisms.

           Adv. Microb. Physiol.  7:119-203,  1972.

324.  Radford, E.  P.,  and E. Kresin.  Red cell diphosphoglycerate after exposure

            to  carbon monoxide.   Fed. Proc.  32:350,  1973.  (abstract)

324o.  Rahn,  H.,  and W. 0. Fenn.   A Graphical Analysis of the Respiratory Gas

            Exchange.   The CU-CO^ Diagram.  Washington, D.  C.:   American

            Physiological  Society, 1955.  38 pp.

 325.  Ramsey,  J.  M.  Carbon  monoxide,  tissue hypoxia, and  sensory psychomotor

            response  in hypoxaemic subjects.  Clin.  Sci.  42:619-625, 1972.

 326.  Ramsey,  J.  M.   Carboxyhemoglobinemia in parking garage employees.   Arch.

            Environ.  Health 15:580-583,  1967.

 327.  Ramsey,  J.  M.  Effects of  single  exposures of carbon monoxide on sensory

            and psychomotor response.  Amer. Ind. Hyg. J. 34:212-216,  1973.

 328.  Raven, P.  B., B. I. Drinkwater, S.  M.  Horvath,  R.  0. Ruhling, J. A.  Gliner,

            J.  C.  Sutton,  and N.  W.  Bolduan.   Age, smoking  habits,  heat stress,

            and their interactive effects  with carbon monoxide  and  peroxyacetyl-

           nitrate on  man's  aerobic  power.   Int. J.  Biometeorol.  18:222-232,  1974.

329.  Raven,  P.  B.,  B. L. Drinkwater,  R.  0. Ruhling, N.  Bolduan,  S. Taguchi,  J.

            Gliner, and S. M. Horvath.   Effect of carbon monoxide and peroxyacetyl

            nitrate on  man's  maximal  aerobic capacity.  J.  Appl. Physiol. 36:288-

            293,  1974.

  330.   Ravenholt,  R.  T.,  M.  J.  Levinski,  D.  J.  Nellist,  and M.  Takenaga.  Effects
             of  smoking  upon  reproduction.   Amor.  J.  Obstet. Gynecol.  96:267-281,
  331.   Ray,  A.  M.,  nnd  T.  H.  Rockwell.   An exploratory study of automobile driv-
             ing performance  under the  influence of low levels of carboxyhemoglo-
             bin.   Ann.  N.  Y.  Acad.  Sci.  174:396-408,  1970.
 332.   Regan, D.  Evoked Potentials in Psychology, Sensory Physiology and Clini-
            cal  Medicine.   New York:  John Wiley & Sons, Inc., 1972  328 pp.
  333.   Rikans,  L.  E., and  R.  A. van Dyke.   Evidence  for a different CO-binding
             pigment  in  solubilized  rat hepatic  microsomes.   Biochem.  Pharmacol.
             20:15-22, 1971.
  334.  Ringold,  A., J. R. Goldsmith, H. L. Helwig, R.  Finn, and  F.  Schuette.
             Estimating recent carbon monoxide exposures.  A rapid method.
             Arch. Environ. Health 5:308-318, 1962.
 335.   Rissancn, V., K.  Pyorala, and 0. P. Heinonen.  Cigarette  smoking  in rela-
            tion to coronary and aortic atherosclerosis.  Acta Path. Microbiol.
            Scand. A 80:491-500, 1972.
 336.   Robbins,  R. C., K.  M.  Borg, and E. Robinson.  Carbon monoxide in  the
            atmosphere.   J. Air Pollut. Control Assoc. 18:106-110,  1968.
 337.   Roberts,  W. C.  Coronary arteries in fatal acute myocardial  infarction.
            Circulation 45:215-230, 1972.
338.   Robinson, E.,  and C. W.  Moser.   Global gaseous pollutant emissions and
            removal mechanism,  pp.  1097-1101.   In H.  M. Englund and W.  T. Beery,
            Eds.  Proceedings of the Second International Clean Air Congress,
            Washington,  D.  C.,  1970. New York:   Academic Press, Inc.,  1971.
 339.   Robinson, E.,  and R. C.  Robbins.   Atmospheric  background concentrations
            of carbon monoxide.  Ann. N.  Y.  Acad.  Sci.  174:89-95, 1970.

  --0.  Rolinsor., F,. , ;,vt  R. C.  \:bv-:-c.   ~0'.:rc£-r.  .V urbaneo, *nd Fate- ,--f  ~      <;

            Ai-i.v^spheric Poilv.i :--rs.   SRI  I'rcjfct PR-6755.  Huntsville,  Ala..

            Stanford Research  ]>•:;_ itate,  19(>8.   i^3 pp.

 '*!.   Rodkey,  F.  L.   i-'inct ic  r.spects of  cy.uirTiC'Lhcy.iogl^Mn  formation  from c^r^oxy--

            hc.,oglobin.   Clin.  Chem.  13:2-5, 1967.

  342.  Rodkey,  F.  L. ,  J.  D. O'\'ea1, H*. A. Collison,  and  D.  E.  Uddin.   Relative

             affinitv of hemc^globin S and hemoglobin  A  for carbon monoxide and

             oxygen.  Clin. Chem. 20:83-84,  1974.

                                             /•               •*»
  343.   Rondia, D.  Abaisse.ment  de  1'activite de la benzopyrene-hydroxylase

             hepatique in vivo  apres  inhalation d'oxyde de carbone.  C. R. Acad.

             Sci.  D 271:617-619,  1970.

 343a.   Roth, R. P., R. T. Drew,  R. J. Lo,  and  J.  R.  Fouts.   Dichloromethane inhal-

             ation, carboxyhemoglobin concentrations, and drug metabolizing  enzymels

             in rabbits.  Toxicol. Appl.  Pharrcacol.   33:427-437, 1975.

  344.  Rottman, G. J., and H.  W. Moos.   The ultraviolet  (1200-1900 angstrom)

             spectrum  of  Venus.   J.  Geophys. Res.  78:8033-8048,  1973.

 344a.   Roughton,  F. J. W., and R. C. Darling.   The  effect of carbon monoxide on

             the oxyhemoglobin dissociation  curve.   Amer. J.  Physiol.   141:17-31,


344b.  Roth, R. A.,  Jr.,  snd  R.  J.  Rubin.  Role of blood flow  in carbojn monoxide-

             and hypoxic hypoxia-induced alterations  in nexobarbital metabolism

             in  rats.   Drug. Metab.  Dispos.  4:460-467, 1976.

344c.  Roth, R. A., Jr.,  and  R.  J.  Rubin.  Comparison  of the  effect of carbon  n.on-

             oxide  and  of  hypoxic hypoxia.  T.   In ± i v_o ;,i.-1 .jbol ism,  distribution

             and action of hexobarbi tal.   J. i'hannacol. Exp. Ther.  199:53-60, 1976,

 34^d.   Roth,  R.  A.,  Jr.,  and  R.  J.  Rubin.   Comparison of the effect of carbon mon-

             oxide and of  hypoxic hypoxia.   II.   Jk-xobarbital metabolism in the iso-

             lated,  perfused rat  liver.   J.  Pharmacol. Exp.  Ther.   199:61-66, 1976.
 345.   Rummo, N., and K. Sarlanis.  The  effect of carbon monoxide  on  several

            measures  of vigilance in a simulated driving task.  J.  Safety  Res.

            6:126-130, 1974.

 346.   Rush, D.   Examination of  the relationship between birthweight,  cigarette

            smoking during pregnancy   and maternal weight  gain.   J.  Obstet.

            Gynaecol. Brit. Commonw. 81:746-752, 1974.

 347.   Russell,  M.  A. H. ,  P.  V.  Cole,  C. Wilson, M.  Idle,  and C.  Feyerabend.

             Comparison of increases  in carboxyhe-r.oglobin .?fter smoking

             "extra-mild"  and  "non-mild"  cigarettes.   Lancet 2:687-690, 1973.

347a.  Russell, M. A. H.   P.  V. Cole, and E. Brown.  Absorption by non-smokers

            of carbon monoxide from room air polluted by tobacco smoke.  Lancet

            1:576-579, 1973.

 348.   Russell, C. S., R.  Taylor, and C.  E.  Law.   Smoking in pregnancy, maternal

            blood pressure, pregnancy outcome, baby weight and growth, and related

            factors.  A prospective study.   Brit.  J. Prev.  Soc. Med. 22:119-126,


 349.   Russell,  C.  S., R.  Taylor,  and  R. N.  Maddison.   Some effects of smoking

             in pregnancy.  J.  Obstet.  Gynaecol. Brit.  Co-iriionw.  73:742-746, 1966.

     .   Ssy&rs,  R.  R. ,  W.  P.  Yant,  E.  Levy,  and W.  B.  Fulton.   Effect of Rtpr.-ate-d

             Daily Exposure of Several Hours to Small  Amounts  of Automobile

             Exhaust Gas.   Public Health Bulletin No.  186.   Washington, D.  C.:

             U.  S.  Government Printing Office,  1929.   58 pp.

350a.  Schmeltz, I., D. Hoffmann, and E. I. Wynder.  The influence of  tobacco

            smoke on indoor atmospheres.  Prev. Med.  4:66-82,  1975.

 351.   Schnellen, C. G. T.   Research  on Methane  Fermentation.   Ph.D. Thesis.

            Delft,  The Netherlands:   Technische  Hoogeschool te  Delft,  1947.

            137 pp.   (in  Dutch,  summary in  English)

351a.   Schoeneck,  F. J.   Cigarette  smoking  in  pregnancy.  N.  Y.  State  J. Med.

            41:1945-1948,  1941.

 352.   Scholander,  P.  F.,  and F. J. W.  Roughton.   Micro gasoroetric  estimation  of

             the blood  gases.   II.   Carbon monoxide.   J. Biol.  Chem.  148:551-563,


 353.   Schulte, J.  H.  Effects of mild  carbon  monoxide  intoxication.   Arch.

            Environ.  Health  7:524-530,  1963.

  354.  Seiler, W.   The cycle of  atmospheric CO.  Tellus 26:116-135,  1974.

 355.   Seller, W. ,  and  C.  Junge.  Carbon monoxide  in the  atmosphere.   J.  Geophys.
            Res.  75:2217-2226,  1970.
 356t   Seller, W. ,  and  C.  Junge.  Decrease  of carbon monxide mixing ratio above
            the  polar tropopause.  Tellus 21:447-449,  1969.
 357.   Seller, W.,  and  P. Warneck.  Decrease of carbon monoxide ratio  at  the
            tropopause.  J. Geophys. Res. 77:3204-3214, 1972.
 358.   Shapiro,  S.,  and J. Unger.  Weight at Birth and Survival of  the Newborn.
            United  States, Early 1950.  Vital and  Health  Statistics Series  21,
            No.  3.   Washington, D. C.:  U.  S. Department  of Health,  Education,
            and  Welfare, 1965.  33 pp.
 359.   Shaw, J. H.   A Determination of the Abundance of Nitrous Oxide,  Carbon
           Monoxide and Methane in Ground Level Air at Several Locations Near
           Columbus, Ohio.  Scientific Report No. 1 on Contract AF19(604)2259.
           Air  Force Cambridge Research Laboratory AFCRC-TN-59-428.   Columbus:
           Ohio State  University, Research Foundation, 1959.  44 pp.
 360.   Shaw, J. H.   The abundance of atmospheric carbon monoxide above Columbus,
           Ohio.   Astrophys. J. 128:428-440, 1958.
361.   Shurtleff, D.  Some characteristics related to the incidence of cardiovascular
           disease  and death:  Framingham Study,  16-year follow-up.   Section 26.
           In W. B.  Kannel  and T.  Gordon,  Eds.  The Framingham Study.  An Epidem-
           iological Investigation of Cardiovascular Disease.   Washington, D. C.:
           U. S. Government Printing Office, December 1970.
 362.   Sidgwick, N.  V.   Carbon monoxide, pp. 545-546.  In The Chemical Elements
            and  Their Compounds.  Vol. 1.   Oxford, England:  Clarendon Press,  1950.
 363.   Sitnonaitis,  R.,  and J.  Heicklen.   Kinetics  and  mechanism of the reaction
            of 0(3P) with  carbon monoxide.   J.  Chem. Phys.   56:2004-2011, 1972.
363a.   Simpson,  W.  J.  A preliminary report on cigarette smoking and  the incidence
            of prematurity.   Amer.  J. Obstet.  Gynecol. 73:808-815, 1957.

        £ •
        Sjostrand, T.  A method  for  the  determination  of  carboxyhaemoglobin  con-

            centrations by  analysis of  the  alveolar air.   Acta  Physiol.  Scand.

            16:201-210, 1948.

  365.  Small, K. A., K. p.  Radford,  J.  M. Frazicr, F. I.  Rodkey,  and  H.  A.

            Collison.  A  rapid  method for simultaneous measurement  of carboxy-


            and mcthcmoglobin in blood.  J. Appl. Physiol.  31:154-160,  1971.

365a.  Smith,  E.,  E. McMillan,  and L. Mack.   Factors  influencing the lethal

            action of illuminating gas.   J.  Ind. Hyg.  17:18-20,  1935.

 366.  Smith,  F.,  and A. C.  Nelson,  Jr.   Guidelines  for Development of a Quality

            Assurance Program.   Reference Method for  the Continuous Measurement

            of Carbon Monoxide  in the Atmosphere.  EPA-R4-73-028a.  Research

            Triangle Park, N.C.:  Research Triangle Institute,  1973.  110 pp.

 367.  Smith,  K. A.,  J. M.  Bremner,  and  M. A.  Tabatabai.   Sorption of  gaseous

            atmospheric pollutants by soils.   Soil Sci.   116:313-139,  1973.

 368.  Smith,  I.,  Jr.,  and E. H. C.  Sie.  Response of luminescent bacteria to

            common atmospheric  pollutants.   Inst. Environ. Sci.  15:154-157, 1969.

 369.  Smith,  R.  N., and J.  Mooi.  The catalytic oxidation of carbon monoxide by

            nitrous  oxide  on carbon surfaces.   J. Phys.  Chem. 59:814-819, 1955.

369a.  Stainsby,  W.  N., and A.  B. Otis.   Blood flow,  blood oxygen tension, oxygen

            uptake,  and oxygen transport in skeletal  muscle.,  Amer. J. Physiol.

            206:858-866, 1964.

 370.  Stalker, W. W.,  and R. C. Dickerson.  Sampling  station and time require-

            ments  for urban air  pollution surveys. Part  II:   Suspended particu-

            late matter and  soiling  index.   J.  Air Pollut.  Control Assoc. 12:111-

            128, 1962.

371.  Stebbins, W.  C. ,  Ed.   Animal Psychophysics:  The Design and Conduct of

           Sensory Experireents.   New York:  Appleton-Century-Crofts, 1970.

           433 pp.

 372.  Stecher, P.  G.,  Kd.  Carbon monoxide, p. 212.  In The Merck Index of Chem-
            icals and Drups.  (7th ed.)  Rahway, N. J.:   Merck & Co., Inc., 1960.
372o.  Stechor, P.  G. ,  Ed.  Carbon monoxide, pp. 208-209.  In The Merck Index.
            An Encyclopedia of Chemicals and Drugs.   (8th ed.)  Rahway, N. J.:
            Merck & Co.,  Inc., 1968.
 373.  Stcdman, D.  H. ,  E.  D.  Morris,  Jr.,  E. E. Daby,  H.  Niki, and B. Weins.tock.
            The role of OH radicals in photochemical smog reactions.   Abstract
            WATR 026.  In Abstracts of Papers.  160th National Meeting, American
            Chemical Society,  Chicago, Illinois, September 14-18,  1970.
 374.  Stephenson,  M.  Bacterial Metabolism.   (3rd ed.)   New York:  Longmans,
            Green and Co., 1949.  398 pp.
375.  Stevens, C. M., L. Krout, D. Walling, A.  Venters,  A. Engelkemeir,  and
           L.  E. Ross.  The isotopic  composition of atmospheric carbon monoxide.
           Earth Planet. Sci. Lett.  16:147-165, 1972.
 376. Stewart,  R. D.,  E.  D. Baretta,  L. R.  Platte,  E. B.  Stewart,  H.  C.  Dodd,
            K.  K. Donohoo,  S.  A. Graff,  J.  H. Kalbfleisch,  C.  L.  Hake, B.  van
            Yserloo,  A. A.  Rimm, and  P.  E.  Newton.   "Normal"  Carboxyhemoglobin
            Levels of Blood Donors  in the  United States.   Report  No.  ENVIR MED
            MCW CRC-COHb-73-1.   Milwaukee:   The Medical College of Wisconsin,
            1973. /~252 pp._7
 377.  Stewart, R-,  D. ,  E.  D.  Baretta, L. R.  Platte,  E. B. Stewart,  J.  H.
            Kalbfleisch,  B.  van Yserloo, and A. A.  Rimm.   Carboxyhemoglobin
            concentrations  in  blood from donors in  Chicago, Milwaukee, New
            York, and Los  Angeles.   Science 182:1362-1364,  1973.
378.  Stewart, R. D.,  E. D. Baretta,  L. R.  Platte,  E. B. Stewart,  J.  H.
           Kalbfleisch, B. van Yserloo, and A.  A. Rimm.  Carboxyhemoglobin
            levels in American blood  donors.  J.A.M.A. 229:1187-1195,  1974.

378a.  Stewart, R. I)., T. N. Fisher, M. J. Hosko, J. E. Peterson, E.  D,  Baretta,

            and II. C. Dodd.  Experimental human exposure to methylene chloride.

            Arch. Environ. Health  25:342-348, 1972.

 379.  Stownrt, R. D., P. E. Newton, M. J. Hosko, and J. E. Peterson.  Effect

            of cnrhon monoxide on time perception.  Arch. Environ. Health  27:
      __    155-160, 1973.

 380.  Stewart, R. D., P. E. Newton, M. J. Hosko, J. E. Peterson, and J. W.

            Mellender.  The effect of carbon monoxide on time perception,

            manual coordination, inspection, and arithmetic, pp. 29-55.  In

            B.  Weiss and V. G.  Laties, Eds.  Behavioral Toxicology.   New York:

            Plenum Press, 1975.

 381.  Stewart, R. D., J. E. Peterson, E. D. Baretta, R. T. Bachand,  M.  J.  Hosko,

            and A. A. Herrmann.  Experimental human exposure to  carbon monoxide.

            Arch. Environ. Health 21:154-164, 1970.

 382.  Stewart, R. D. , J. E. Peterson, T. N. Fisher, M. J. Hosko, E.  D.  Baretta,

            H.  C. Dodd, and A.  A. Herrmann.  Experimental human  exposure to high

            concentrations of carbon monoxide.  Arch. Environ. Health 26:1-7, 1973.

 383.  Strickland-Constable, R. F.  Part played by surface oxides in  the oxida-

            tion of carbon.  Trans. Faraday Soc. 34:1074-1080, 1938.

 384.  Stupfel, M.,  and  G.  Bouley.  Physiological and biochemical effects  on

            rats  and mice  exposed to  small concentrations of carbon monoxide

            for long periods.  Ann. N. Y. Acad. Sci. 174:342-368, 1970.

 385.  Swlnnerton, J. W., and R. A. Lamontagne.  Carbon monoxide in the  South

            Pacific Ocean.  Tellus 26:136-142, 1974.

 386.  Swinnerton, J. W., V. J. Linnenbom, and C. H. Cheek.  Distribution  of

            methane and carbon monoxide between the atmosphere and natural

            waters.  Environ. Sci. Technol. 3:836-838, 1969.

 387.   Swinnerton, J. W. , V. J. Linnenbom, and  R. A. Lamontage.   Distribution
            of cnrbon monoxide between  the atmosphere  and  the  ocean.   Ann.
            N. Y.  Acad. Sci. 174:96-101, 1970.
 388.   Swinnorton, J. W., V. J. Linnenbom, and  R. A. Lamentagne.   The ocean:
            A natural source of carbon  monoxide.  Science  167:984-986, 1970.
 389.   Syvertson,  C. R., and J. A.  Harris.   Erythropoietin production in dogs
            exposed  to  high altitude  and  carbon monoxide.  Amer.  J.  Physiol.
            225:293-299, 1973.
 390.   Tanaka,  M.   Studios  on the etiological mechanism of fetal developmental
            disorders  caused  by maternal smoking during pregnancy.  J. Jap. Obstet.
            Gynecol. Soc.  (Nippon.  Sanka-Fujinka Gakkai Zasshi) 17:1107-1114,
            1965.   (in Japanese)
 391.   Taylor, 0.  C.  Air pollution with  relation to agronomic crops:   IV.
            Plant  growth suppressed by  exposure to  air-borne oxidants (smog).
            Agron. J. 50:556-558, 1958.
 392.   Teichner, W.  H.,  Carbon monxide and  human performance:  A methodologi-
            cal  exploration, pp.  77-103.  In B. Weiss  and  V. G. Laties,  Eds.
            Behavioral  Toxicology.  New York:   Plenum  Press, 1975.
 393.   Teichner, W.  H.   Recent studies  of simple reaction  time.   Psychol. Bull.
            51:128-149, 1954.
  394.  Teichner, W.  H., and M.  J. Krebs.  Laws of  the  simple visual reaction
            time.  Psychol. Rev.  79:344-358, 1972.
 395.   Tenney, S. M., and L. C. Ou.   Physiological  evidence  for increased tissue
            capillarity in rats acclimatized to high altitude.  Respir.  Physiol.
            8:137-150,  1970.
396.  Theodore,  J.,  R.  D.  O'Donnell,  and K.  C. Back.   Toxicological evaluation
           of carbon monoxide in humans and other mammalian species.  J. Occup.
           Med.  13:242-255,  1971.

 397.  Thomas Jefferson University.  Jefferson Medical College, Department of

            Physiology.  The Effect of Carbon Monoxide Inhalation on Induced

            Ventricular Fibrillation in the Cynomologus Monkey.  Technical

            Report.  Philadelphia:  Thomas Jefferson University, 1973.  31 pp.

 398.   Thompson,  C.  R. ,  0.  C.  Taylor, M.  D.  Thomas,  and J.  0.  Ivie.   Effects  of

            air  pollutants  on  apparent  photosynthesis  and water use  by citrus

            trees.   Environ. Sci.  Technol.  1:644-650,  1967-

 399.   Thomsen,  H.  K.   Carbon  monoxide-induced atherosclerosis in primates.   An

            electron-microscopic  study  on the coronary arteries of Macaca irus

            monkeys.   Atherosclerosis  20:233-240,  1974.

 400.   Thomsen,  H.  K., and K.  Kjeldsen.  Threshold limit for carbon monoxide-

            induced myocardial damage.   An electron microscopic study in

            rabbits.  Arch. Environ. Health 29:73-78,  1974.

 401.  Tjepkema, J. D., and C. S. Yocum.  leghemoglobin facilitated oxygen

            diffusion in soybean nodule slices.  Plant Physiol. 46(Suppl.):

            44, 1970.   (abstract)

401a.  Tokyo Metropolitan Research Institute for Environmental Protection.  Annual

            Report of the Tokyo Metropolitan Research Institute for Environmental

            Protection.  1971.  68 pp.-

 402.  Trapnell, B. M. W.  Chemisorption.  London:  Butterworths Scientific

            Publications, 1955.  265 pp.

 403.   Tunder,  R.  S.,  S. W.  Mayer,  E.  A.  Cook, and L.  Schieler.  Compilation of

            Reaction Rate Data for Non-Equilibrium Performance and Reentry Calcu-

            lation  Programs.   Aerospace Report No.  TR-1001(9210-02)-!.  El

            Segundo,  Calif.:   Aerospace Corporation,  1967.   68 pp.

 404.  Underwood, P. B., K. F. Kesler,  J. M. O'Lane, and D. A. Callagan.  Paren-

            tal smoking empirically related to pregnancy outcome.  Obstet. Gynecol.

            29:1-8, 1967.


405.  U.  S.  Bureau  of  the Census.   Social  and  Economic  Statistics  Administration.

           Department  of Commerce.   Statistical Abstract  of  the  United  States

           1974.  95th Annual  Edition.  Washington, D.  C.:   U. S.  Government

           Printing Office,  1974.   1028 pp.

 406.  U. S. Department  of  Health,  Education,  and Welfard.   Public Health Service.

            Environmental Health Service.   National Air Pollution Control Adminis-

            tration.   Air Quality Criteria for Carbon Monoxide.   NAPCA Publication

            No.  AP-62. Washington,  D. C.:   U. S.  Government Printing Office,

            1970.   /.~166 pp. ~J

407.   U. S. Department  of  Health,  Education and  Welfare.  Public Health Service.

            Vital  Statistics of the United States, 1968.   Volume 1.  Natality.

            Washington,  D.  C.:  U.  S.  Government  Printing Office,  1970.  /254 pp._/

 408.  U.  S. Environmental  Protection  Agency.  Ambient  air monitoring reference

            and  equivalent  methods.  Federal Register 40:7042-7070, 1975.

409.   U.  S. Environmental  Protection Agency.  Supplement No. 5  for Compilation

            of Air Pollutant Emission Factors.  (2nd  ed.) AP-42.   Research

            Triangle Park,  N.  C.:  U. S.  Environmental Protection  Agency, 1975.

            /~158 PP._7

410.   U.  S. Environmental  Protection Agency.  Guidelines for Air  Quality Main-

            tenance Planning and  Analysis.   Vol.  9.   Evaluating  Indirect  Sources.

            EPA 450/4-75-001.  Research Triangle  Park,  N. C.:  U.  S.  Environmental

            Protection Agency, Office of Air Quality, Planning and Standards,

            1975.   /~557 pp._7

411.  U.  S. Environmental Protection Agency.  National  primary and secondary

           ambient  air quality standards.   Federal Register 36:8186-8201, 1971.

 Alln.   U.  S.  Environmenta1 Protection Agency.  1973 National Emissions Report.

             National Emissions Data System (NEDS) of the Aerometric and Emissions

             Reporting System (AEROS).  EPA-450/2-76-007.  Research Triangle Park,

             N.  C.:   U.  S. Environmental Protection Agency, Office of Air and

            W.-istu  Management,  1976.  423 pp.

  412.  U.  S.  Environmental Protection Agency.  1972 National Emissions Report:.,

             National Emissions Data System (NEDS) of the Aerometric and Emis-

             sions  Reporting System (AEROS).  EPA-450/2-74-012.  Research Triangle

            Park,  N.  C.:   U. S.  Environmental Protection Agency, 1974.   422 pp.

4l2a.  U. S. Environmental Protection Agency.  Ambient air monitoring reference

            and equivalent methods.  Federal Register 40:7042-7070, 1975.

 413.   van  Kampen,  E. J., and W.  G.  Zijlstra.   Standardization of hemoglobin-

            ometry.   II.   The hemoglobincyanide method.   Clin. Chim. Acta 6:

            538-544,  1961.

 414.   van  Slyko, D.  D.,  A.  Hiller,  J.  R.  Weisiger,  and  W.  0.  Cruz.   Determination

            of carbon monoxide  in blood and of total and active hemoglobin by

            carbon  monoxide capacity.   Inactive hemoglobin and methemoglobin

            contents  of normal  human blood.  J. Biol.  Chem. 166:121-148, 1946.

 415.  Vaughan,  B.  E., and N. Pace.   Changes in myoglobin content of the high

            altitude  acclimatized rat.  Amer.  J. Physiol. 185:549-556,  1956.

 416.  Vogel,  J. A.,  and M. A.  Gleser.  Effect of carbon monoxide on oxygen

            transport during exercise.  J. Appl. Physiol. 32:234-239, 1972.

 417.  Vogel, J. A., M. A.  Gleser,  R. C. Wheeler, and B. K. Whitten.  Carbon  mon-

            oxide and physical work capacity.  Arch. Envir. Health  24:198-203,  1972.

 418.  Vollmer, E.  P., G. B. King,  J. E. Birren, and M. B. Fisher.  The  effects

            of carbon monoxide on three types  of performance,  at simulated  alti-

            tudes of 10,000 and 15,000 feet.   J. Exp. Psychol. 36:244-251,  1946.


 419.  Wnlcl, N., S. Howard. ?. G. Smith, and K. Kjeldsen.  Association between  ather-
            osclerotic diseases and carboxyhaemoglobin levels in tobacco smokers.
            Brit. Mod. J. 1:761-765, 1973.
 4?0.  Wallace,  N.  n.   C.  L.  Davis,  R.  B.  Rutledge,  and A.  Kahn.   Smoking and
            carboxyhomoglobin  in  St.  Louis Metropolitan population.   Theoretical
            and  empirical  considerations.  Arch.  Environ.  Health  29:136-142,  1974.
 421.  Wanstrup, J., K. Kjeldsen,  and P. Astrup.  Acceleration of spontaneous
           intitnal-subintimal changes in rabbit aorta by a prolonged moderate
           carbon monoxide exposure.  Acta Path. Microbiol. Scand. 75:353-
           362, 1969.
 422. Warneck,  P.   On the role of OH and  H02 radicals in the troposphere.
            Tellus  26:39-46,  1974.
 423.  Webster, W. S.,  T.  B. Clarkson,  and H.  B.  Lofland.   Carbon monoxide-
            aggravated  atherosclerosis  in  the squirrel monkey.  Exp.  Molec.
            Path. 13:36-50, 1970.
423a. Weast,  R.  C., Ed.   Handbook of Chemistry and Physics.   (56th ed.)
            Cleveland, Ohio:   The Chemical Rubber Company, 1975.
 424. Weinblatt, E., C. W. Frank, S. Shapiro, and R. V.  Sager.   Prognostic
            factors in  angina pectoris--a prospective study.  J.  Chron. Dis.
            21:231-245, 1968.
 425. Weinblatt, E.,  S.  Shapiro,  C.  W.  Frank, and  R.  V.  Sager.   Prognosis  of men
            after first myocardial  infarction:  Mortality and first recurrence in
            relation to selected  parameters.   Amer.  J. Public Health 58:1329-1347,
 426. Weinstock, B.  Carbon monoxide:  Residence time in  the atmosphere.
            Science 166:224-225, 1969.
 427. Weinstock, B.,  and  T. Y. Chang.  The global balance of carbon monoxide.
            Tellus 26:108-115,  1974.


 428.   Weinstoclc, B.,  and H.  Niki.  Carbon monoxide balance in nature.  Science 176:

            290-292,  1972.
 429.  Weir,  F.  W. ,  and T. H.  Rockwell.   An Investigation of the Effects of Car-

            bon  Monoxide on Hum.-ms in the Driving Task.   Final Report.  Columbus:

            The  Ohio  State University Research Foundation, 1973.  170 pp.
                  \ ;
430.  Weiss, B., and V. G. Laties.  Enhancement of human  performance  by caffeine
           and  the amphetamines.  Pharmacol. Rev. 14:1-36,  1962.

 431.   Weiss, H. R., and J. A.  Cohen.  Effects  of low levels of carbon monoxide

            on rat brain and muscle  tissue PQ_.   Environ.  Physiol.  Biochem.  4:

            31-39, 1974.
 432.  Wells, L. L.  The'prenatal effect  of carbon monoxide on  albino rats  and  the

            resulting neuropathology.  Biologist  15:80-81,  1933.

 433.   Went,  F.  W.   On the nature of Aitken condensation nuclei.  Tellus 18:

            549-556,  1966.
 434t  Went, F.  W.   Organic matter in the atmosphere and  its possible relation

            to petroleum  formation.   Proc. Nat. Acad. Sci.  U.S.A. 46:212-221,

 435.  Westberg, K., and  N. Cohen.  The Chemical  Kinetics of Photochemical  Smog

            as  Analyzed by Computer.  Aerospace Report No.  ATR-70  (8107)-!.  El

            Segundo, Calif.:  Aerospace Corporation, 1969.   29  pp.

 436.  Westberg, K., N. Cohen, and K. W.  Wilson.  Carbon  monoxide:  Its role in

            photochemical smog formation.  Science 171:1013-1015,  1971.

 437.   Westenberg, A. A.  Carbon monoxide and nitric oxide  consumption  in polluted

            air.  The carbon monoxide-hydroperoxyl reaction.  Science 177:255-256,

 438.  Westenberg,  A.  A.,  and N.  deHaas.   Steady-state intermediate concentra-

            tions and  rate constants.  Some H0« results.   J. Phys.  Chem. 76:

            1586-1593,  1972.

439.  Who Ion, W. J.  Intracellular P02 in heart and skeletal muscle.  Physiologist

            14:69-82, 1971.
440.  White, J.  J.   Carbon monoxide and its relation to aircraft.  U. S. Nav.
           Med.  Bull. 30:151-165, 1932.
441.  Wilmer, W. H.  Effects of carbon monoxid  upon the eye.  Amor. J.  Ophthalmol.
           4:73-90, 1921.
442.  Wilks, S.  S., J.  F. Tomashefski, and R. T. Clark, Jr.  Physiological effects
           of chronic exposure to carbon monoxide.  J. Appl. Physiol. 14:305-310,
443.  Williams,  I.  R., and E. Smith.  Blood picture, reproduction, and general
           condition during daily exposure to illuminating gas.  Amer. J. Physiol.
           110:611-615,  1935.
444.  Wilson, D. F.,  J.  W.  Swinnerton, and R. A. Lamontagne.   Production of car-
           bon monoxide and gaseous hydrocarbons in seawater:  Relation  to dis-
           solved organic carbon.  Science 168:1577-1579,  1970.
445.  Winnekc,  G.  Behavioral effects of methylene chloride  and  carbon monoxide
            as assessed by  sensory  and psychomotor performance, pp.  130-144.   In
           C. Xintaras,  B. L. Johnson, and I. de Groot, Eds.   Behavioral Toxicol-
           ogy:  Early Detection of Occupational Hazards.  HEW Publ. No.  (NIOSH)
           74-126.  Washington,  D. C.:  U. S. Department of  Health,  Education,
           and Welfare,  1974.
      Wittenberg, J. B., C. A. Appleby, and B. A. Wittenberg.  The kinetics of
           the reactions of leghemoglobin with oxygen and carbon monoxide.  J.
           Biol. Chem. 247:527-531, 1972.

 447.  wofsy, S. C., J. C. McConnell, and M. D. McElroy.  Atmospheric CH4, CO a
            and C02.  J. Geophys. Res. 77:4477-4493, 1972.

 448. Wohlrah,  H. ,  and G.  B.  Oguntnola.   Carbon monoxide binding studies of cyto-

            chrome  a  hemes in intact rat liver mitochondria.  Biochemistry 10:

            1.103-11.06, 1971.

 449. Wright,  G.,  P.  Randell, and R. J.  Shephard.  Carbon monoxide and driving

            skills.   Arch.  Environ. Health 27:349-354, 1973.

450.  Xintaras, C.,  B.  L.  Johnson, C.  E. Ulrich,  R.  E.  Terrill, and M. F.

            Sobecki.   Application of the  evoked response technique in air pollu-
            tion toxicology.   Toxicol.  Appl.  Pharmacol.  8:77-87, 1966.

 451.  Delete -- use 450.

452.  Xintaras, C.,  C.  E.  Ulrich, M. F.  Sobecki,  and R. E. Terrill.  Brain

            potentials studied by computer analysis.   Arch. Environ. Health

            13:223-232,  1966.

 453.  Yamate, N., and A.  Inoue.  Continuous analyzer for  carbon monoxide in

            ambient air by electrochemical technique.   Kogai to Taisaku  (J.

            Public Nuisance)  9:292-296,  1973.   (in Japanese)

  454.  Yerushalmy,  J.  Mother's cigarette smoking and survival of infant.   Amer.

            J.  Obstet.  Gynecol. 88:505-518,  1964.

454a.  Yerushalmy, J.  The relationship  of parents' cigarette smoking  to outcome

            of pregnancy - implications  as to  the problem  of inferring causation

            from observed  associations.  Amer. J. Epidemiol. 93:443-456, 1971.

 455.  Young,  I. M., and L. G.  C. Pugh.  The carbon monoxide content  of  foetal

            and maternal blood.  J. Obstet. Gynaecol. Brit.  Commonw.  70:681-

            684, 1963.

 456.  Younoszai,  M. K., and J. C. Haworth.  Placental dimensions and  relations

            in preterm, term and growth-retarded  infants.  Amer. J. Obstet.

            Gynecol. 103:265-271, 1969.

-'>56a.  Youaaszai,  M.  K. ,  J.  Peloso,  and J.  C.  Haworth.   Fegal growth retardation
            in rats  exposed  to cigarette smoke during pregnancy.   Amer.  J. Obstet.
            Gynecol.  104:1207-1213,  1969.
 457.   ZatsiorskU, M. , V. Kondrateev,  and  S.  Solnishkova.   Radiation of the flame
            of carbon monoxide and ozone and the mechanism  of this  reaction.  Zh.
            Viz. Khim.  (Leningrad) 14:1521-1527, 1940.   (in Russian)
 458.  Zimmerman, P.  W., W. Crocker,  and A.  E.  Hitchcock.  Initiation and stimu-
           lation of roots from exposure of plants to carbon monoxide.  Contrib.
           Boyce Thompson Inst. 5:1-17, 1933.
 459.  Zimmerman, P.  W., W. Crocker,  and A.  E.  Hitchcock.  The effect of carbon
           monoxide on plants.  Contrib. Boyce Thompson Inst. 5:195-211,  1933.
 460.Zorn, H.   The partial oxygen pressure  in  the brain and  liver at  subtoxic
           concentrations of  carbon monoxide.  Staub Reinhalt. Luft   (Engl. Ed.)
           32(4):24-29,  1972.

                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
                                                           3. RECIPIENT'S ACCESSION NO.

             5. REPORT DATE
               September 1977
                                                           6. PERFORMING ORGANIZATION CODE

 Subcommittee on Carbon Monoxide
                                                           8. PERFORMING ORGANIZATION REPORT NO.

 Committee on Medical and Biologic Effects of
 Environmental Pollutants
 National  Academy of Sciences
 Washington,  D.C. 20460	._
                                                           10. PROGRAM ELEMENT NO.
             11. COfJTffAtnYGRANT NO.

 Health  Effects Research Laboratory
 Office  of Research and Development
 U.S. Environmental Protection  Agency
 Research  Triangle Park. N.C. 27711


      This  document summarizes  the  carbon monoxide literature related to effects
 on man and his environment for the consideration of the Environmental Protection
 Agency in  updating the information in the Air Quality Criteria for Carbon Monoxide,
 It emphasizes recent major advances in our knowledge of carbon monoxide: chemical
 reactions  in air; biologic effects on man; problems in monitoring urban
 concentrations and relating such data to the exposure of populations; data
 concerning the identification  of susceptible populations; and evidence implicating
 carbon monoxide as a causal factor of disease.

      Not all published articles have been reviewed, but only those deemed
 to be important studies related to carbon monoxide air quality criteria.  There
 is a large literature on adverse effects of cigarette smoking and some of these
 effects may be related to carbon monoxide.
                                KEY WORDS AND DOCUMENT ANALYSIS
 Carbon monoxide
 Carbon monoxide  poisoning
 Air pollution
 chemical analysis
                            06 F, H, T

19. SECURITY CLASS (ThisReport)
20. SECURITY CLASS (Thispage}

EPA Form 2220-1 (9-73)


        Insert the EPA report number as it appears on the cover of the publication.


        Reserved for use by each report recipient.

        Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently.  Set subtitle, if used, in smaller
        type or otherwise subordinate it to.main title. When a report is prepared in  more than one volume, repeat the primary title, add volume
        number and include subtitle for the specific title.

        Each report shall carry a date indicating at least month and year.  Indicate the basis on which it was selected (e.g., date of issue, date of
        approval, date of preparation, etc.).

        Leave blank.

   7.   AUTHOR(S)
        Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.}. List author's affiliation if it differs from the performing organi-

        Insert if performing organization wishes to assign this number.

        Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.

        Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.

        Insert contract or grant number under which report was prepared.

        Include ZIP code.

        Indicate interim final, etc., and if applicable, dates covered.

        Leave blank.

        Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of. Presented at conference of,
        To be published in, Supersedes, Supplements, etc.

   16.  ABSTRACT
        Include a brief (200 words or less} factual summary of the most significant information contained in the report. If the report contains a
        significant bibliography or literature survey, mention it here.

        (a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
        concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.

        (b) IDENTIFIERS AND OPEN-ENDED TERMS  - Use identifiers for project names, code names, equipment designators, etc.  Use open-
        ended terms written in descriptor form for those subjects for which no descriptor exists.

        (c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
        jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
        endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
        the primary posting(s).

        Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited."  Cite any availability to
        the public, with address and price.

        DO NOT submit classified reports to the National Technical Information service.

        Insert the  total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.

    22.  PRICE
        Insert the  price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)