EPA-600/1-78-017
February  1978
Environmental Health Effects Research Series
                                                                                IRON
                                                         Health Effects Research Laboratory
                                                        Office of Research and Development
                                                       U.S. Environmental Protection Agency
                                               Research Triangle Park, North Carolina  27711

-------
                RESEARCH REPORTING SERIES

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.

-------
                                      EPA-600/1-78-017
                                      February 1978
                     ron
                      by

               Subcommittee on Iron
Committee on the Medical and Biologic Effects of
            Environmental Pollutants
           National Research Council
          National Academy of Sciences
                Washington, D.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
      U.S. ENVIRONMENTAL PROTECTION AGENCY
       OFFICE OF RESEARCH AND  DEVELOPMENT
       HEALTH EFFECTS RESEARCH LABORATORY
       RESEARCH TRIANGLE PARK, N.C. 27711

-------
                                 DISCLAIMER
          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 consitute endorsement or recommendation for use.
                                   NOTICE

     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 chosen for their special
competences and with regard for appropriate 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.
                                    ii

-------
                                  FOREWORD
     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 estab-
lishment 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,  environ-
mental 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 Administrator 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.  "Iron" is one of these reports.
                                      John H. Knelson, M.D.
                                            Director,
                               Health Effects Research Laboratory
                                   iii

-------
                           SUBCOMMITTEE ON IRON
                                                                                    COMMITTEE ON MEDICAL AND BIOLOGIC EFFECTS OF ENVIRONMENTAL  POLLUTANTS
CLEMENT A. FINCH, University of Washington School of Medicine, Seattle,




   WA, Chairman




ELMER B. BROWN, Washington University School  of  Medicine,  St. Louis, MO




JOHN C. BROWN, Agricultural Research Center,  Beltsville, MD




HOWARD E. BUMSTED, U. S. Steel Research Laboratory, Monroeville, PA




MERLE BUNDY, U. S. Steel Corporation, Pittsburgh, PA




JOHN D. HEM, U. S. Geological Survey, Menlo Park, CA




J. B. NEILAHDS, University of California,  Berkeley, CA




DARRELL R. VAN CAMPEN, U. S. Plant, Soil,  and Nutrition Laboratory,  Ithaca,




   NY



RAYMOND C. WANTA, Bedford, MA




MUNSEY S. WHEBY, University of Virginia School of Medicine,  Charlottesvllle,




   VA



DONALD 0. WHUTEMORE, Kansas State University, Manhattan,  KS








THOMAS H. BOTHWELL, Johannesburg Medical School, University  of  the




   Witwatersrand, Johannesburg, South Africa, Resource Person




ROBERT W. CHARLTON, Johannesburg Medical School, University  of  the




   Wltwatersrand, Johannesburg, South Africa, Resource Person
REUEL A. STALLONES, School of Public Health, University of Texas,



     Houston, Chairman



MARTIN ALEXANDER, Cornell University, Ithaca



ANDREW A. BENSON, Scripps Institution of Oceanography, University of



     California, La Jolla




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








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



EVILLE GORHAM, University of Minnesota, Minneapolis



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



IAN T. T. HIGGINS, School of Public Health, University of Michigan, Ann Arbor



JOE W. HIGHTOWER, Rice University, Houston



HENRY RAMIK, Duke University Medical Center, Durham, North Carolina



ORVILLB A. LEVANDER, Agricultural Research Center, Beltsville, Maryland



ROGER P. SMITH, Dartmouth Medical School, Hanover, New Hampshire








T. D. BOAZ, JR., Division of Medical Sciences, National Research Council,



     Washington, D.C., Executive Director
T. D. BOAZ, JR., National Research Council, Washington, DC, Staff Officer

-------
                        CONTENTS



      Introduction                                           1

 1    Iron in the Environment                                3

 2    Microorganisms and Iron                               64

 3    Iron and Plants                                       91

 4    Iron Metabolism in Humans and Other Mammals          122

 5    Iron Deficiency                                      163

 6    Acute Toxicity of Ingested Iron                      190

 7    Chronic Iron Toxicity                                196

 8    Inhalation of Iron                                   234

 9    Summary                                              244

10    Recommendations                                      253

      Appendix:     Analysis of Iron in Environmental       257
                   and Biologic Samples

      References                                           271

-------
                           INTRODUCTION









     The Subcommittee on Iron has surveyed the effects of organic




and inorganic iron that are relevant to humans and their environment.




It must be recognized that the biology and chemistry of iron are




complex and only partially understood.  Iron participates in oxida-




tion reduction processes that not only affect its geochemical mobility,




but also its entrance into biologic systems.  Hydrated ferric oxide




surfaces have adsorbent properties and may act as reaction sites and




catalysts.  In biologic systems, the iron atom is incorporated into




several protein enzymes that participate in many oxygen and electron




transport reactions.




     A first consideration is the quantity and form of iron in the




earth's crust, in the hydrosphere, and in the atmosphere, and the




degree of movement that occurs among them (Chapter 1).  Of particular




interest is the interaction between inorganic and organic forms of




the metal.  Iron is brought into the organic cycle through sophisti-




cated mechanisms that microorganisms have developed for converting




highly insoluble and unavailable forms of iron into usable ones




(Chapter 2).  Similarly, the roots of some plants have capabilities




for retrieving iron from the soil; matching this affinity to the con-




dition of local soils has led to improvement in agriculture (Chapter 3).




Vertebrates in general, despite their high iron requirements for hemo-




globin synthesis, appear to be able to achieve satisfactory iron




balance (Chapter 4).  Humans are the outstanding exception—hundreds




of millions of the world's peoples are iron-deficient because of




inadequate amounts of available iron in the diet (Chapter 5).  External

-------
factors are believed to be responsible for this borderline balance, and




deficiency may thus be considered the major iron-related environmental




health problem faced by humans.  An account is also presented of a much




smaller population that shows iron overload (Chapter 7).  The presence




of large deposits of ferritin and hemosiderin in parenchymal tissues has




been shown to result in damage to several vital organs.  Acute iron




toxicity has been reported, but only with the ingestion of large amounts




of iron salts (Chapter 6).  Pulmonary inhalation of iron compounds from




industrial exposure has not been shown to be a hazard (Chapters 1 and 8),




and 8).

-------
                               CHAPTER 1



                       IRON IN THE ENVIRONMENT






     Iron is element number 26 in the periodic table, and has an atomic



weight of 55.85.  The average isotopic composition is 5.8% of mass number



54, 91.7% of 56, 2.2% of 57, and 0.3% of 58.  It is the fourth most abun-



dant element in the earth's crust; only oxygen, silicon, and aluminum are



more common. Metallic iron occurs in a few types of rock and is a primary



constituent of the earth's core.  However, in most rocks and soils it is



combined in crystal structures either as divalent ferrous or trivalent

                                                                         o
ferric ions.  The effective ionic radius of the ferrous ion is about 0.80A.



The ferric ion has an effective radius of about 0.67A.  Fundamental aspects


                                                   SRI
of iron chemistry have been described by Nicholls.



     Iron occurs in solution in water as Fe(II) or Fe.(Ill), or as inorganic or



organic ferrous or ferric complexes.  It also can be found in small quantities



as a stable colloid or hydrosol, most commonly constituted of small dispersed



particles of ferric oxyhydroxide.  The terms "ferric oxyhydroxide" and



"hydrated ferric oxide" will be used interchangeably to mean a relatively



fresh precipitate with a poorly organized crystal structure.  Its composi-



tion is approximated by the formula Fe(OH)g, and sometimes such materials



are referred to as ferric hydroxide, which does not imply specific stoichio-



metry or crystallinity.  Upon aging, this material may achieve the structure



of goethite, a-FeOOH, hematite, ct-Fe203, or other polymorphs.


     Because of its fundamental importance to human activities, large amounts



of iron are extracted from ores each year, and converted to the metallic



form.  In this form, iron is chemically unstable under earth-surface

-------
 conditions and  is slowly oxidized and converted  to  ferrous and  ferric




 compounds.  The human contribution  to the geochemical cycle of  iron  is



                                    94
 important on a  global scale.  Bowen  estimated  that the annual amount




 of  iron mined exceeds the  amount carried to  the  ocean by natural rock




 weathering by a factor  of  8.




      The chemical behavior of ..iron  in the environment is a function  of




 the following four characteristics  of the element:




      •  Iron participates  readily in chemical oxidation and reduction




        processes.  These  processes influence the geochemical mobility




        of the  element.




      •  Iron forms organic and  inorganic complexes  that affect  its




        solubility in water and its subsequent chemical reactions.




      •  Surfaces of hydrated ferric oxide are active in absorbing



        other materials and may act as reaction  sites and catalysts.




        Small particles of the  hydrated oxide may form colloidal sus-



        pensions in water  or air and have large  surface to weight and




        surface to volume  ratios.




      •  Iron is essential  to biologic processes  and is present  in




                all living  matter.




       The material in  this chapter is set  forth in terms of  properties,



  sources, and  transport processes  that influence the environmental




  occurrence and behavior  of iron.  Both natural and anthropogenic  sources



  and processes are considered.






 NATURAL SOURCES AND THEIR  PROPERTIES




The Earth's Crust



      Rocks.  Iron is common and abundant in  igneous, metamorphic, and



 sedimentary rocks.  In most igneous and metamorphic rocks, more iron




 exists in the ferrous than in the ferric state,  whereas the reverse  is

-------
true for sedimentary rocks.   Data for iron concentrations in the earth's


                                                                     471
lithlc (lithosphere) and continental crust have been compiled by Lepp



and Parker;602 seiected values from the literature are listed in Table



1-1.



     Iron in igneous rocks is found primarily in the ferromagnesian sili-



cates, the most common of which are olivine, (Mg, Fe>2SiO/, pyroxenes,



amphiboles, and iron-containing micas.  Important iron-rich



accessory minerals are magnetite, Fe304, ilmenite, FeTi03, and pyrite,



FeS2«  Basic  (or mafic)  igneous rocks  contain more of these minerals  and



thus more iron, whereas acidic (or felsic) rocks contain less (Table



1-1).  The average concentration of iron in the lithic crust ±s close



to that of the average igneous rock, because 95% of the crust is composed



of igneous rocks or metamorphic rocks derived from them.  In the conti-



nental section of the crust, the ratio of the amounts of basic  to acidic


                     fin ?
igneous rocks is 1:20     The thinner oceanic crust consists mainly of



basic rocks.  Most of the acidic igneous rock outcrops on the continen-



tal crust are granitic; most of the basic rocks are basaltic.



     The weathering of igneous rocks oxidizes most ferrous iron to the



ferric state.  The ferric iron is then incorporated into clay minerals



or hydrolyzed to ferric oxyhydroxides.  During deposition and diagenesis



of sediments derived from the weathering products, iron is often reduced



and precipitated as sulfides.  The average content of iron in sedimentary



rocks (including sediments) given in Table 1-1 is close to that of the



average continental igneous rock.  The main difference between iron from

-------
                                      TABLE 1-1


           Average Iron Contents  of  the Earth's  Crust  and  Common  Rocks

Crust or Rock Type


Lithospherea
Continental crust
Igneous rocksc
Ultrabasic
Basic
Intermediate
Acidic
Sedimentary rocks
Average sedimentary
rock
Shale6
Sandstone-'
Limestone*?
Average metamorphic rock
Percentage by Weight
Iron (III)
Oxide
2.7
2.4

4.6
4.0
3.4
1.6

3.5

4.2
1.7

1.5
Iron (II)
Oxide
5.1
4.7

8.4
7.1
4.1
1.9

2.6

3.0
1.5

2.9
Total
Iron
5.8
5.4

9.8
8.3
5.6
2.6

4.5

5.3
2.4
0.38
3.3
aAverage of values in Poldervaart    and Ronov and Yaroshevsky.

«                                         SQQ                           £ c/
^Average of values in Pakiser and Robinson    and Ronov and Yaroshevsky.
^Average of values in Table 93 of Geochemical Tables.

^From Garrels and Mackenzie.

eFrom Lepp.471


•^From Petti John.614

      Clarke.170

"Average of values for crystalline shields in Table III of Geochemical Tables.

-------
igneous and from sedimentary rocks is the increase in the average ferric



to ferrous ratio, from 0.5 to 1.2.  The average sedimentary rock is based


                                                               295
on a calculated ratio of 81:11:8 for shale:sandstone:limestone;    these



components comprise about 97% of all sedimentary rocks.  Approximate   relative



abundances of these rocks on continents have been measured as 50% for shale,



26% for sandstone, and 20% for limestone.615  About 75% of the world's



outcrops are composed of sedimentary rocks.  Thus they are much more



abundant on land surfaces than igneous rocks.  The iron contents of



specific types of shales, sandstones, and limestones have been measured.    '



     The elemental compositions of metamorphic rocks are generally similar



to the igneous or sedimentary rocks from which they were formed.  The



average iron content of metamorphic rocks in the earth's crystalline



shields is between that of intermediate and acidic igneous rocks.



     Soils.  Iron is primarily ferric in most soils, although the



ferrous state may be predominant in some soils that are flooded and rich



in organic matter.  The principal iron-containing minerals in soils are



the ferric oxyhydroxides, which are set forth in Table 1-2.   The most



common are amorphous oxyhydroxide, goethite, and hematite.  They occur



in soils as small particles, concretions, lateritic crusts, or coatings


                             199  588
on clays and other minerals.    '     The  first minerals  formed  from the



oxidation and by hydrolysis of ferrous iron are geothite and lepidocrocite



(although less often), because of the relatively low nucleation energy



needed.  Hematite usually evolves from long-term aging of amorphous material


                                                                      442
or from dehydration of fine-grained goethite during warm, dry periods.

-------
                          TABLE 1-2





         Naturally Occurring Ferric Oxyhydroxidesa











Ferric Oxyhydroxide            Ideal Formula





Amorphous                      Indefinite: often represented

                                           as Fe(OH)3

Goethite                       ot-FeOOH



Akagane'ite                     g-FeOOH



Lepidocrocite                  y-FeOOH



Hematite                       a-Fe20-j



Maghemite                      Y-Fe0
a                                     444
 Derived from Langmuir and Whittemore.

-------
     The presence of organic compounds in soils retards the



crystallization of amorphous oxyhydroxides.      In organic-containing



soil horizons, the ferric oxyhydroxide that  forms is generally goethite;



in environments poor in organic matter, the  amorphous material alters



faster to hematite.  An increase in the organic content in a hematitic



soil could cause the dissolution of hematite through reduction and/or



complexing of iron by organic compounds.  Then the dissolved iron



could reprecipitate as poorly crystallized goethite.



     Appreciable amounts of ferrous and ferric iron in soils are in-



corporated in clay minerals as an essential  or a minor isomorphous



substitute for Mg(H) or Al(III) in the octahedral sites in their structures.



Relatively common soil clays containing iron as an essential constituent



are nontronite (a smectite), chlorites, glauconite and some illites or


                                                  149
hydromicas (clay micas), and certain vermiculites.



     Iron can also exist in some soils as unweathered ferromagnesium



silicates and in the heavy accessory minerals magnetite and ilmenite.



Soils being formed from shales containing pyrite or marcasite, FeS ,



can become quite acidic, because of the oxidation of the sulfides.  The



low pH allows some ferrous iron to remain in soil moisture before it



can oxidize to form ferric oxyhydroxides. .Small amounts of amorphous



or poorly crystalline strengite, FePO^.Zl^O  exist in many acidic



soils with high phosphate activity.  Under reducing conditions, vivianite,


                                                        390
Fe-CPO.K.SH-O, and siderite, FeCO , form in some soils.     Iron salts



are generally limited to poorly-drained, acid soils where basic ferric



sulfates may be present as a yellow mottling or yellow surface crust.

-------
                                                         390
     The iron content of most soils ranges from 0.7-4.2%.     In a



series of U. S. Geological Survey studies of 3,026 soil samples in the



United States, the total iron content ranged from 0.01-13%, with geomet-


                                                                  174
ric means for the different types of soils ranging from 0.47-4.3%.



Based on samples collected from a depth of 20 cm for 861 soils, Shacklette



et al. determined the geometric means and deviations of the iron con-



tent for the western United States to be 2.0 and 1.90%, and for the


                         685
East to be 1.5 and 2.76%.     Vinogradov reported iron contents from


                                                                      773
0.59-10.4% for 16 soils in different zones on the East European plain.



     Iron may be relatively uniformly distributed throughout soil horizons



as in the Mollisols, the main types of soils of the prairie and Great



Plains grain-growing states.  For example, the iron contents of the A,



B,, B-, and C horizons of a loam in South Dakota were 2.7, 2.8, 2.6,


                       390
and 2.7%, respectively.     Where the parent material is low in iron,



or in environments where iron has been leached from the soil, iron



contents can be small.  Iron contents averaged approximately 0.5% in



the A and B horizons of 30 uncultivated garden soils in east central


        fiftfi
Georgia.     In many moist forest soils, especially the Spodosols of the



northeastern United States and northern Michigan and Wisconsin, the



total iron concentration can reach several percent in the B horizon.



Most of it has accumulated as ferric oxyhydroxides and organic complexes,



as shown by an average of 3.6% free iron (dithionite extraction) in the


                                                220
B horizons of 10 northern Appalachian Spodosols.



     Under prolonged or intense weathering and leaching (generally in



hot, wet climates) soils known as Oxisols can form.  Iron contents
                                   10

-------
of Oxisols are high, ranging from 15-55%, and mainly in ferric




oxyhydroxides.  The higher percentages of iron in soils often referred




to as laterites are generally produced during alternating wet and dry




conditions that form more stable, less soluble, crystalline oxyhydroxides,




and dehydrate goethite to hematite.  There are no major areas, however,




where recently formed Oxisols exist in the continental United States.




     In poorly drained  soils  high in organic  matter,  ferric iron can be




reduced to the ferrous state.  Dissolved ferrous iron can be relatively




high in the soil moisture and may be controlled by the solubility of




ferrous carbonates, sulfides, or phosphates,  depending on the oxidation




potential and the activities of the anions.   Iron contents can vary widely,



                                                                     390
e.g., from 1-20%,  as measured in ash of 5 peats in the United States.



     Sediments.  Sediments are composed of more ferrous iron than soils




because most soils occur above the water table and thus are aerated,




whereas many sediments are deposited or undergo diagenesis in anaerobic,




subaqueous environments.  In a normal depositional environment, in which




the pH is generally 8+1 unit, ferric oxyhydroxides are the most impor-




tant form of precipitated iron.  During diagenesis, original amorphous




oxyhydroxides and poorly crystallized goethite generally are converted to


         iff\

hematite.     Several reviews have discussed the formation of sedimentary




iron minerals,  »'»      and concentrations of iron in average




sedimentary rocks are listed in Table 1-1.




     Ferrous compounds usually form in sediments in a reducing environ-




ment generated by the oxidation of organic matter.  Pyrite is the most




common one, although siderite can form in waters where relatively high
                                   11

-------
iron, low calcium, and very low sulfide activities exist, as in some non-




marine aqueous environments.  During diagenesis of sediments with high




organic content, available sulfate, and slow detrital addition, all iron



                                        202
compounds tend to transform into pyrite.     Thus, pyrite is common in




many clay sediments high in organic matter and equivalent sedimentary




rocks, and is often found in coals and associated strata.




     Hydrous sheet silicates, glauconite, chamosite, and greehalite also




are major constituents of some sediments.  They generally form in marine




sediments with pore waters of intermediate redox potential, low sulfide




and carbonate activity, and available silica and alumina.  In general,




sediments poorer in iron occur near coasts; iron content of sediments will




increase with ocean depth.  Berner   and James    have summarized the




expected sequences of iron minerals to be found from fresh water to pro-




gressively deeper marine waters.  Argillaceous deep-sea sediments tend to




be uniform and moderately high in iron content; the average is 5.8% for



                                               250
total iron on a carbonate and water-free basis.     Carbonaceous deep-sea



         250                                                  198
sediments    and oxide-rich sediments on the East Pacific Rise    range




up to 11 and 18% iron, respectively.  Ferromanganese oxide nodules on the




ocean floor have from 1-42% iron in them and average 19, 13, and 15% iron




for the Atlantic, Pacific, and Indian Oceans.55'198




     Iron ores.  Rock that is economical to mine for iron ranges from




20-69% total iron.  Ores may be grouped into four classes, according to




whether or not the deposits were formed by sedimentary processes, igneous




activity, hydrothermal solutions, or surface or near-surface enrichment.




The most important ores are the sedimentary iron formations and ironstones,
                                    12

-------
which are distributed throughout the world.  The largest and most abundant



deposits are around the Atlantic and Indian Oceans.^07>^28'601



     Sedimentary iron deposits are mainly marine chemical precipitates of



iron weathered and transported from land masses.  Most of these ore



deposits are very old, generally of Precambrian or early Paleozoic age.



Some were altered later by metamorphism or enriched by weathering.  On the



basis of the predominant iron minerals, the sedimentary ores can be divided


                                                                       407
into four major facies groups—oxide, silicate, carbonate,  and sulfide.



     "Iron formation"  is  the  term for a thinly  bedded  rock with layers  of



iron oxides (mainly hematite and magnetite), siderite, and iron silicates


                                                                       407
(primarily greenalite, a septachlorite) alternating with cherty layers.


                                                          471
The average total iron concentration of such rocks is 28%.     Iron forma-



tions constitute the greatest single source of iron being mined and the



largest reserves.  The biggest deposits of this type occur in the Lake



Superior region of the United States and Canada, the Labrador districts



of Canada, and areas in Brazil, Russia, and Australia.  The total iron



concentration of the ores generally ranges from 25-45%, except where the



silica has been leached and replaced by the action of surface and ground



waters.  These bodies enriched by secondary processes are known in the



Lake Superior region as "direct-shipping ores," containing soft goethite



and hematite and iron contents of 50-68%, or the "wash ores" or "semi-


          / O Q
taconite."     Where the iron formation has been metamorphosed, coarse



grains of hematite and magnetite have formed, and the silicates have



altered primarily to chlorite, minnesotaite—the iron analog of talc,
                                 13

-------
Mg3(Si4010)(OH)2—and stilpnomelane, a sheet silicate with a structure simi-



lar to talc.407  The resulting rock is taconite, the low-grade ore currently



mined in great quantities in the Lake Superior region.  Small quantities



of the amphiboles making up the cummingtonite-grunerite series are



found in metamorphosed parts of the iron formations of northern



Minnesota (primarily in the Mesabi Range), and northern Michigan.^



Most of the asbestiform fibers in the tailings dumped by a mining



company into Lake Superior and found in the water system of Duluth,


                                                   583
Minnesota, are comprised of minerals in this series.



     The term "ironstone" includes oolitic iron oxyhydroxides (goethite



and hematite) and silicates (predominantly chamosite, another septa-



chlorite).   Calcite and dolomite are the common constituents of ironstone



and its iron content ranges from 20-40%.  Thicknesses and lateral extents



of ironstone deposits are generally an order of magnitude less than those



of iron formations.  Thus, the reserves have been estimated as at least


               428
100 times less.     In certain locales, the near-surface ironstones



have been enriched by leaching of calcite.  The Clinton Formation,



extending from Alabama to New York, is one of the world's most extensive



ironstone deposits.  However, most of the high grade, secondarily enriched



ore has already been mined in the United States.  Important deposits are



still being mined in northern Europe and coastal Newfoundland.



     Other sedimentary iron ores include bog-iron deposits  (e.g., accumu-



lations of iron oxyhydroxides in swampy areas and shallow lakes in



northern Europe and North America), and blackband and clayband deposits
                                   14

-------
(e.g., the thin layers of siderite found in the coal sequences of



Appalachia and Great Britain).     Neither of  these are of much  present


                                428
or future economic significance.



        Iron deposits resulting from igneous activity or formed during



replacement by hydrothermal solutions occur primarily as magnetite and



hematite.  The magmatic segregations of magnetite in Kiurna, Sweden, and



the hydrothermally enriched iron-formation in the Quadrilatero Ferrifero



of Brazil yield high-grade ores (60-70% iron).  At present,  both deposits



are of modest economic importance to the world and contain moderate



reserves.  The contact metasomatic ores in the Cornwall area of



Pennsylvania, which contain about 40% iron, are moderately important



sources of iron now, but are potentially a large resource for the United



States.  Other igneous- or hydrothermally-related iron ores in the



United States occur in Nevada, Utah, New York, and Missouri, but are



currently of minor economic importance and of small to moderate


          428
potential.



     The fourth class of iron ores was produced by surface or near-



surface enrichment of pre-existing low-grade deposits or by the weather-



ing of iron-containing rocks into laterites.  The secondarily enriched



ores include the important "direct-shipping" and "wash ores" of the



Lake Superior region discussed above.  Iron-rich laterites generally



result from the weathering of rocks high in iron, such as serpentine,



under tropical conditions with alternating wet and dry seasons.   Such



laterites occur in Cuba and the Philippines and commonly contain 40-50%



iron, primarily as poorly crystallized goethite.  Many laterites contain
                                    15

-------
 enough nickel  to be mined  for  that  element  alone.  Although  presently



 only of minor  economic  importance,  the  unmined quantities  in the world



 are large.  The continental United  States,  however, has no major


                   428
 lateritic deposits.



 Iron in the Hydrosphere and in Solutions



     The main  features  of  the  aqueous chemistry  of iron and  the



 stability of iron compounds that     most commonly participate in



 chemical equilibria can be shown conveniently by Eh-pH diagrams. Eh



 represents the redox potential in volts, calculated for equilibrium



 conditions by  means of  the Nernst equation.  Increasing positive



 values of Eh represent  increasing intensity of oxidation.  Figure



 1-1 is one form of such a  diagram, which shows areas of dominance for



 dissolved ferric  and ferrous   species, and for  two hydroxy  com-



 plexes of each.  The shading of Figure  1-2  shows the areas of stability



 for ferric and ferrous  oxyhydroxidesi;Fe(OH)3 and Fe(OH)2, siderite,



 FeCO , pyrite, FeS., pyrrhotite, FeS, and-  metallic iron,  Fe*, and  the



 field of stability of liquid water  at 25°C  and 1 atm.  Figure 1-2 also



 illustrates the total solubility of iron as a function of  Eh and pH



 within the field of stability  of liquid water, in the system specified



 for the diagram.  Techniques for preparing  and interpreting  Eh-pH diagrams


                           294  345 347
 have been well documented.   *   *      The  system specified  for Figures



l.-l and 1-2 includes water, ferric  and  ferrous ions, sulfur, and carbon, in



 equilibrium at 25°C and  1  atm, with a total sulfur activity  of 10~  '   mol/1



 and a total activity for carbon dioxide species  of 10  '   mol/1.   These



 measurements are equivalent to  a sulfate concentration of  9.6 mg/1  and
                                    16

-------
    Eh,
  volts .4
                            Water  oxidized
           0    2    4    6     8    10   12   14
FIGURE 1-1   Areas of dominance for  dissolved ferric and ferrous
            ions and hydroxide complexes in dilute solutions  at
            25 C and 1 atm.   Drawing courtesy of J. D.  Hem.
                          17

-------
 volts
          0
                                        14
FIGURE 1-2
Fields of stability for solids and solubility of
iron between 10~3>0° and 10~9-00 mol/1 in a
system of iron, sulfur, carbon dioxide, and water
at 25°C and 1 atm. Total dissolved sulfur species
activity is fixed at 10~^-°° mol/1 (9.6 mg/1 as
sulfate ion) and t9ta-l dissolved carbon dioxide species
activity at 10      mol/1 (61 mg/1 as bicarbonate
anion).  Drawing courtesy of J. D. Hem.
                             18

-------
a bicarbonate concentration of 61 mg/1, about the worldwide average



concentration of these compounds in river water.     The stability



fields for solids shown in Figure 1-2 were extended into conditions



of low pH in which iron solubility would exceed the highest values



shown on the diagram  (10~ *   mol/1 or 56 mg/1).  For siderite to



exist below pH 6, for example, a very high dissolved iron concentration



would be required.



     Chemical thermodynamic data used in preparing the graphs were


                           777              443
selected from Wagman et al.    and Langmuir.     Their values yield


                                     —38 49
a solubility product of Fe(OH)_ of 10   *  .  The solubility product

                                          O ^             / /

for  this material can range from about 10    to about 10   , depending


                               813
on the degree of crystallinity.     The values chosen represent aged



but  not dehydrated material.  The minimum concentration of iron (10~  *



mol/1) shown in Figure 1-2 is 0.56 yg/1.  A relatively low pH or redox


                                                               -3.00
potential, attainable in oxygen-depleted water, may stabilize 10



mol/1 or more of iron in solution.  Iron sulfides generally have very



low  solubilities, but they are unstable in the presence of oxidizing



agents.  Metallic iron is unstable in water.  The species greigite ^6384)



and  mackinawite (FeS) may occur in natural systems, but they are less



stable than the forms considered in Figure 1-2.


     Silicate rocks commonly contain iron and some iron silicate



minerals  are important sources of dissolved iron.   The iron from these



rocks is released during weathering.and subsequent control over the



solubility and mobility of iron is a function of processes involving




the  secondary mineral species shown in Figure 1-2.  The silicate minerals



may not behave reverslbly under earth-surface conditions.  The thermodynamic



data for iron silicates are incomplete, but some of their effects on iron

                              258a

solubility have been examined.
                                    19

-------
     Complexes of ferric iron with fluoride, chloride, and sulfate are



strong enough to affect the chemical behavior of iron in solutions that



are enriched in these anions.  Organic complexes of iron also may



influence iron solubility.  In many systems in which organic ligands



are abundant, conditions will tend to be reducing, favoring the forma-



tion of ferrous rather than ferric complexes.  The effects of reduction



as well as of complexing must be considered in such systems.   Yet some



organic ligands are stable in the ferric species regions diagrammed in



Figure l-l.   The tartrate ion, for example, forms a strong ferric complex.



Ferrous  iron-tannic acid complexes and similar substances may be found in

                                                                     734

swamps and other natural waters that contain organic coloring matter.



Some inorganic phosphate compounds of iron have low solubilities and



may affect the behavior of the element in environments such as eutrophic



lakes, agricultural soils, or sewage.  Polynuclear complexes of iron


                                                                    703
and their relation to biologic iron transport have been reviewed.



(The simplified system portrayed in Figures 1-1 and 1-2 only considers



hydroxide complexes.)



     As a rule,the redox reactions of ferric hydroxide and ferrous and



ferric solutions are fast enough to reach equilibrium readily when the



pH approaches neutrality.  The oxidation rate of aqueous ferrous iron



in aerated water has a second-order dependence on pH,    and in acid



solutions it is very slow unless some catalytic effect is introduced.



Reactions involving sulfide species of iron generally require biologic



mediation to attain equilibrium, processes depending on an initial



oxidation or reduction of sulfur rather than iron.
                                    20

-------
     When the impositions of an external agent cause the Eh. or pH of a



solution containing equilibrated concentrations of dissolved iron to



change, the system's equilibrium will be upset (see Figure 1-2).  Where



the necessary reactants are available to restore equilibrium, the



resulting chemical processes can serve as sources of energy.   Microbiota



commonly utilize these sources to catalyze such reactions as the oxida-



tion of ferrous iron to precipitate ferric hydroxide.  Where organic



material is present, organisms also can mediate reduction of ferric iron



or sulfate.



     Electron transfers are a part of redox processes, and mechanisms



to facilitate transfers are established at active surfaces such as those



of ferric hydroxide precipitates or particles.  Such processes can



explain coprecipitation of other metals with iron.  The sorption of




various other dissolved metals by ferric  hydroxides  is Important in  controlling con-


                                                                   346
centrations of iron in soil moisture, river, and underground water.



Iron is involved in the synthesis of chlorophyll and its role as an



oxygen carrier in electron transfer is essential in the life processes



of plants and animals.  Therefore, iron is a component of almost all




organic matter.



Iron in the Atmosphere



     Iron raised from the soil by far exceeds iron residing in  the atmosphere



from all other natural sources, including meteorites and volcanic eruptions.



The Natural Environmental Research Council  of the United Kingdom began operating



several air sampling  stations in sparsely populated areas in the Lake District



at Wraymires  in April 1970.  Most of the samples were analysed  byy -ray
                                    21

-------
spectroscopy.  Indications that atmospheric iron at such sites far from


industrial activity came from the soil    were substantiated in a later


extension of the study to other places.   '     Data for about 30 trace and


more abundant elements were tabulated.  "A remarkable uniformity in the


elemental composition of the general aerosol" at seven nonurban sites chosen

                                                              608
to represent rural and more industrial exposures was reported.

                                                                          3
     Iron concentrations ranged from 67-940 ng/kg air (about 0.09-1.2/ug/m )


during January-December 1972.  At Lerwick in the Shetland Islands during

                                                             3
June 1972-May 1973, it was only 49 ng/kg air (about 0.06/ig/m ).  By taking


ratios of concentration of iron to scandium in air, and normalizing these


ratios by the ironrscandium ratio in average soil (38,000 ppm:7 ppm), an


enrichment factor was arrived at to separate elements into those that were


"soil-derived" (that is, having enrichment factors near unity), and those of


"industrial" origin, that is, enriched relative to the soil by a few to many

      608
times.


     The concentration of iron in surface air layers at nonurban sites in


the United States during 1970-1974—represented by the average of intermittent


24-h measurements throughout the year--ranged from 0.049/ig/m  at a site in

                                           3
Hawaii County, Hawaii in 1972 to 1.091/ig/m  in Park County, Indiana in 1970.


The median annual average for 80 complete site-years of record at 38 nonurban

                                         3
sites during the same 5 yr was 0.255,/ug/m .   Seventy-one percent of the annual


averages (57 out of 80) fell within the range 0.100-0.399/ig/m3.  The above


statistics are based on analysis of EPA data summaries   a by the Subcommittee


on Iron.


     The highest 3-mo average of all nonurban site averages during the same

                    3
period was 2.80/ug/m  at a site in Tom Green County, Texas during the first


quarter of 1971.  The lowest 3-mo average fell below the lowest discrimination
                                    22

-------
levels of the method of measurement (i.e., less than 0.006 to 0.101/ug/tn )

during one or more 3-mo periods at sites in Delaware, Hawaii, New Mexico,

South Dakota, and Wyoming.

     Atmospheric iron concentrations for earlier years that are based on similar

                                  193
measurements have been summarized.     The concentration of iron and its per-

centage of the total of all species in suspended particulate matter at 10

                                                         3
remote nonurban U.S. stations for 1966-1967 was 0.15/ig/m  and 0.717o, res-

          cog
pectively.     The latter figure compares with the lower end of the range of

the percentage of iron in soil.  In the absence of more direct evidence, it

can serve as a rough indicator of the order of the iron fraction (one part

in 100 or 200) in soil-derived suspended particulates.

     The gross global production rate of tropospheric particulates of all
                                                     n £-i
compositions and sizes is estimated by Hidy and Brock    to be 10.7 million

metric tons/day, of which 94% or 10 million metric tons/day is attributed

to natural sources.  Almost half of the daily production rate from natural

sources is attributed to sea spray (3 million metric tons), wind-blown dust

(20,000 to 1 million metric tons), forest fires (400,000 metric tons), and

volcanoes (10,000 metric tons).  The remaining production of tropospheric

particulates from natural sources comes from the conversion to particulates

of gaseous emissions from vegetation, the nitrogen and sulfur cycles, and

volcanoes.     Applying the ratio of iron to gross particulate from the

paragraph above to the production rate of wind-blown dust, one obtains a

rough idea of the rate of iron entry into the atmosphere from this source,
                                                             O £. I
viz., 100-10,000 metric tons/day.  Hidy and Brock's estimates    of global

tropospheric production rates of particulates await corroboration (cf=
                  Q1 A
Whelpdale and Munn   ).
                                   23

-------
MAN-MADE SOURCES OF IRON


At t;he Earth'8 Surface

     Iron Ore.  Most (95%) crude iron ore mined in the United States and in

the world (85%) in 1974 was extracted from open pits because most commercial

ore bodies lie close to the surface and have large lateral dimensions.  Sites

of underground mining in the United States have declined from at least 30 mines

in 1951 to 6 underground mines in 1974.  The present demand for iron ore of

higher iron content and greater uniformity of chemical composition and physical

structure requires that almost all crude ores mined be beneficiated before

shipment.  High-grade ores are crushed and separated according to size fractions.

Lower-grade ores are processed more extensively to increase iron concentration

as well as to produce the correct physical characteristics for iron extraction

processes.  After beneficiation, the ores with particles smaller than 0.625 cm

are agglomerated by sintering or pelleting.  The beneficiated ore, known as

usable ore, had an average of 60.6% iron in 1974 in the United States.  World

production of usable iron ore, iron ore concentrates and agglomerates, and their

total metal contents for 1974 are set forth in Table 1-3.

     Usable iron ore production in the United States has been stable since

1964.  Crude ore production, however, has slowly increased because of the rise

in the ratio of crude ore mined to usable ore produced.  As a result of mining

lower-grade ores such as taconite, the ratio rose from 1.4:1 in 1954 to 2:1 and

2.6:1 in 1964 and 1974.  Of all usable ore produced in 1974, 69% came from

Minnesota, 13% from Michigan, and the rest was divided among California, Utah,

Wyoming, Missouri, Pennsylvania, New York, Texas, and Wisconsin (Table 1-4).

The ore was produced by 35 companies operating 66 mines and 44 concentrating
       429
plants.     Figure 1-3 shows the locations of principal iron ore sources as

well as major U.S. ironmaking centers.   Approximately one-third of the usable
                                      24

-------
                               TABLE 1-3

        Worldwide Production of Iron Ore, Iron Ore Concentrates,
                   and Iron Ore Agglomerates,
                                  Gross Weight              Iron Content

North America                       141,273                    86,447

South America                       118,062                    75,772

Western Europe                      135,238                    56,387

U.S.S.R.                            224,989                   132,743

Africa                               63,560                    38,717

Asia                                111,423                    59,639

Oceania                              99,021                    62,518

TOTAL                               893,566                   511,787
aDerived from Klinger.430
^In thousands of metric tons.
                                   25

-------
                               TABLE 1-4

               Crude Iron Ore (< 5% Manganese) Mined
               in the United States By Region, 1974^
Region

Lake Superior

Southeast

Northeast

West



TOTAL
Number of Mines
Crude
Usable Ore
34
4
3
25

66
186,886
654
7,103
26,046

220,689
71,855
292
2,396
10,664
500°
85,705
Derived from Klinger.430
^In thousands of metric tons.
^Byproduct from processing of other ores.
                                   26

-------
                   •ounces or IRON out MPOSITI OF COKING COAL,
                   AND IRONMAKINQ CIMTIM OF TNI UNITED STATES
FIGURE 1-3    Sources of iron ore, deposits  of coking coal,  and
              ironmaking centers of the United States.  Reproduced
              from Klinger.429
                                  27

-------
 ore  consumed  in  the United  States each year is imported  (Table 1-5).   In 1974

 Canada  supplied  41%, Venezuela,32%, Brazil,14%, Liberia,6%, Peru,4%, and other

 countries)3%  of  the imports.  A relatively small amount  of usable iron ore was
                                     430
 exported, mainly to Canada  and Japan.

     The iron and steel industry in the United States is concentrated  in

 about twenty  metropolitan areas.  Pennsylvania, Ohio, Indiana, Illinois,

Michigan, and New York produce about three-fourths of the nation's steel.

 Iron ores produced in Minnesota and Michigan are usually shipped by railroad

 and  lake carriers to consuming centers.  However, from December to April,

 shipment of ores on the Great Lakes is impractical.  Thus, ore must be stock-

 piled,  generally in the open, at mines or shipping ports during the winter, and

 at consuming  plants during  the shipping season to maintain production  during the

 winter.  Unprotected piles  are possible sources of injections of particles high

 in iron into  the atmosphere.  In other states, most ores are transported only

 by rail.  Pipelines are now being successfully employed  to move iron ore concen-

 trates  as slurries.  If pipeline use and winter shipment of ores by lake carriers

 are  increased, steel mills  will no longer have to maintain as large inventories
                                                                       429
 of iron ore as before, which will reduce the size of unprotected piles.

Amounts of stockpiled ores  in the United States are noted in Table 1-5.

     Pollution problems related to mining iron ore are mainly associated with

 large-scale operations.  The average ratio tonnage of overburden and waste rock
                                                         429
 to crude ore mined in the United States in 1974 was 1:1.     Most of this waste,

.some of which is low-grade  iron ore, is discharged onto  large piles adjacent to

 the open mine pits.  Even larger amounts of wastes result from the beneficiation

of the crude ore into usable ore (approximately 130 million metric tons of

processing wastes were produced in 1974 in this country). Most of these wastes

are piled on  the land, although some taconite wastes have been dumped  into
                                     28

-------
                            TABLE 1-5
                    U.S. Iron Ore Statistics
                                            a
Usable Iron Ore
(< 5% manganese)

Production

Exports

Imports for consumption

Consumption
(Iron ore and agglomerates)

Stocks0

     At mines

     At consuming plants

     At U.S. docks
Average for ,
19 70-19 74fc 1974^
84,934
3,194
ion 43,099
85,705
2,360
48,797
134,069



 13,803

 51,169

  3,203
140,371



  9,555

 45,971

  3,324
aDerived from Klinger.430
^In thousands of metric tons.

CAs of December 31, 1974.
                                29

-------
Lake Superior,  Iron oxides, carbonates, or silicates in the waste piles would

either remain relatively insoluble or weather to become immobile ferric oxyhy-

droxides.  Only in sulfide-containing rock would there be serious pollution

from acidic drainage from mines or ore and waste piles.  Other environmental

problems are related mainly to the minerals associated with the iron ores, such
                                                         462
as the amphibole asbestiform fibers in taconite tailings.



     Wastes from bauxite processing.  Bauxite, the ore from which alumina and

eventually aluminum is extracted, contains from 1-25% ferric oxide.  Consequently,

the residue resulting from alumina production is mainly iron oxide.  In the Bayer

process, the waste is known as "red mud" and contains about equal amounts of

iron oxides and complex sodium-aluminum silicates.  The red mud may be fed to a

subsequent lime-soda sinter step in the processing of bauxite containing more

than 8% silica to produce brown or sinter muds, which are primarily constituted
                                     705
of iron oxides and calcium silicates.     About 5.6 million metric tons of

these residues are generated annually at 8 alumina plants located along the
                                                                 437
lower Mississippi River, the Gulf Coast, and in central Arkansas.     The muds
                                                              821
from most of the plants are impounded in large adjacent lakes.     Until several

years ago, as much as 1.7 million metric tons were discharged annually into the

Mississippi River by two alumina plants in Louisiana.  Now, one of them has

developed a method to eliminate the river disposal.  The red mud is first filtered

through sand and then stored as reserve landfill.   Red mud waste in Arkansas is
                                                                                705
occasionally used to neutralize acid waste generated at a nearby vanadium plant.

In another process, red mud is used to produce a substitute for fluorspar in
             438
making steel.

     Fertilizers.  Iron-containing fertilizer supplements are applied to soils,

to foliage in aqueous sprays, or into trunks and limbs of fruit trees.  Iron
                                    30

-------
                                                                          558
compounds have been applied experimentally in irrigation waters in Kansas.

The fertilizer substances used range from inorganic salts to organic compounds

and iron chelates, as listed in Table 1-6.  Iron salts, such as ferrous sulfate,

have proved inefficient in enriching many soils because they oxidize and then

rapidly hydrolyze to insoluble ferric oxyhydroxides.  Ferrous sulfate in solution,

however, has proved effective in sprays and is the most popular inorganic salt

for foliar application.  Fritted iron is not suitable for fertilizing alkaline
                                                             748
or calcareous soils, but it can be effective in acidic soils.     In addition to

ferrous sulfate, the iron compounds used most widely for correcting iron chlorosis

are the chelates.  Although iron chelates have been applied to soils, they are

more effective as foliar sprays.  The primary use of chelates is for correcting

iron deficiencies in citrus and other fruit trees, grain sorghum, and certain

vegetables.  Recommended rates of application for aqueous solutions of ferrous

sulfate or iron chelates range from 12-24 g of iron/citrus tree (as a solution
                                                                           558
of 0.6-1.0 g iron/1) to 0.5-1.7 kg iron/ha (as a solution of 6-8 g iron/1).


fit the Hydrosphere

     Acid mine drainage.  Acidic drainages high in dissolved iron are products of

the weathering of iron-containing sulfides in rocks exposed by mining to moist

air or oxygenated surface and ground waters.  Most acid drainage results from

the decomposition of pyrite and marcasite in the coals and associated shales of

coal mines.  During weathering, the sulfide oxidizes to sulfate, which decreases

the pH of the surface moisture or water present.  The ferrous iron released

forms salts with the sulfate (in drier weather), dissolves in rain or ground

waters, or oxidizes.  Acidic waters high in iron emerge as seeps from surface

mines or as small streams from underground mine portals.  The dissolved ferrous

iron then oxidizes and hydrolyzes to ferric oxyhydroxides,  a reaction that
                                    31

-------
                                 TABLE 1-6

                      Some Types of Iron Fertilizers
Source

Ferrous sulfate

Ferric sulfate

Ferrous oxide

Ferric oxide

Ferrous ammonium phosphate

Ferrous ammonium sulfate

Iron frits

Iron ammonium polyphosphate

Iron chelates
Iron polyflavonoids

Iron ligninsulfonates

Iron methoxyphenylpropane
Formula
FeS0
FeO

Fe203

Fe(NH4)P04.H20

(NH4)2S04-FeS04«6H20

Varies

Fe(NH4)HP20?

NaFeEDTA

NaFeHEDTA

NaFeEDDHA

NaFeDTPA
FeMPP
Approximate Percentage
	of Iron	

         19

         23

         77

         69

         29

         14

       Varies

         22

        5-14

        5-9

          6

         10

        9-10

        5-8

          5
aFrom Murphy and Walsh.558

 EDTA :  ethylenediaminetetraacetic acid
 HEDTA:  hydroxyethylenediaminetetraacetic acid
 EDDHA:  ethylene diamine di (o^hydroxyphenylacetic acid)
 DPTA :  diethylenetriaminepentaacetic acid
                                   32

-------
further acidifies the drainage.  The oxidation is ordinarily very slow at a low

pH, but it is increased by iron-oxidizing bacteria.  Dissolved ferrous iron

concentrations, which can be higher than 1,000 mg/1 in acidic waters draining

from underground coal mines, diminish downstream from the mines as waters

become neutralized by dilution or dissolved bicarbonate in non-acidic flows.

The precipitated ferric oxyhydroxldes coat the stream bed, accumulating as a

muddy orange layer that can be as deep as several centimeters near

mines.
                                                                    516
     About 17,600 km of streams have been affected by mine drainage,    primarily

in the coal-mining regions of Appalachia.  Abandoned mines contribute an estimated

60-70% of the drainage, with approximately 85% of that amount from exhausted
                  22
underground mines.    Biesecker and George found that the pH values of many

streams containing acid drainage from mines in Appalachia were below 5; 35, 83,
                                 •
and 22% of the 318 stream sites sampled had respective iron, manganese, and sul-
                                                                                  74
fate concentrations exceeding U.S. Public Health Service drinking water standards.

The streams sampled included several sites not affected by mine drainage in addi-

tion to those that were.   Future acid  mine-drainage problems will  be  greater in

the Appalachian than in the western coalfields,  because the former contain  coals

of generally higher sulfur content.

     Prevention methods such as flooding or sealing mines have been investigated

in attempts to decrease the amount of acid drainage.  Many acidic waters are
                                                    22
treated by neutralizing them with lime or limestone.    One expense of the

treatment accrues from the disposal of the voluminous amounts of sludge, con-
                                                                          821
sisting primarily of water and gelatinous, amorphous ferric oxyhydroxides.

Annual literature reviews on the treatment of coal mine drainage, as well as its

chemistry and occurrence, have been published by the Water Pollution Control
                                                                   95
Federation for over 20 years; a recent one is by Boyer and Gleason.

     Steel industry wastes.  Iron exists  in both dissolved and particulate  form

in steel industry wastewaters.  Water used to clean gases from blast and steel-

making furnaces contains very  fine, suspended particles composed  largely of

iron oxides.  Wastewaters  from continuous-casting  and hot-rolling mill operations
                                    33                         '

-------
 contain  suspended  iron  oxide  scale  that  is washed  from  the  steel.   Surface

 oxides are also removed in hot  rolling and other processes  by acid  pickling-
                                                                 657,762,821
 baths; the spent liquors have high  dissolved iron  concentrations.

 More suspended solids and slightly  less  dissolved  ferrous sulfate,  generated

 in wastes per ton  of steel produced, are created with advanced processes for

 steelmaking than with older technologies.  (The newer techniques produce more

 of certain types of pollutants because larger amounts of water are  required
                                                    762
 per day per ingot  ton than for the  older processes.)     Annual literature

 reviews of the treatment of wastes  from  the steel  industry  have been published
                                                            698
by the Water Pollution Control Federation for over 20 years.     Wastewater-

producing processes, analyses of loads of raw wastes,  and water-pollution
                                                                             98a
control practices in the carbon and alloy steel industry have been described.

 In the Atmosphere

     The  iron and  steel industry.   Man's most important metal is produced and..

 recycled  at prodigious  and increasing rates.  U.S. and world production of pig

 iron and  steel for 1974, and  averaged for the 5-yr period 1970-1974, is set
                    645a
 forth in  Table 1-7.      In 1974 the quantity of iron scrap and steel scrap

 necessary to manufacture pig  iron,  steel ingots, steel castings, and foundry

 and miscellaneous  products in the U.S. was about 10% more than the  total pig
 iron consumed.      Several descriptions of iron- and steelmaking technologies

 and their potential for emission of pollutants are available.    >    »    »      jn

 the making of iron and steel, vaporization, condensation, fracture,  and chemical

 reaction of raw materials and intermediate products result in the emission of

 iron and iron-containing particulate matter to the atmosphere, directly or in

 reduced quantity through pollution control systems.  Figure 1-4  shows the

 principal sources of particulates in iron- and steelmaking processes.
                                    34

-------
                           TABLE 1-7

Iron and Steel Production in United States and the World, 1974-
                         Average for

Pig Iron                 1970-1974-               1974-
United States               83,161               86,615

Worldwide                  466,000              514,000




Raw Steel

United States-
Carbon
Stainless
All other alloys
Total
Worldwide—
109,129
1,478
13,082
123,688
641,000
114,857
1,950
15,388
132,195
707,000
-Derived from Reno.


—In thousands of metric tons.

£
-From data of the American Iron and Steel Institute.    Includes


 ingots, continuous-cast steel, and all other cast forms.


—Ingots and castings.
                               35

-------
                                                        FLUE GAS
              (SINTER
             OPERATION)
   DUST, FINES,
    AND COAL
 SINTER
OPERATION
   (P)


DUST


H
)N 	 .





D

I

, 1

/CYCLON
                                                 SECONDARY
                                                  CLEANER
                                  PRIMARY
                                  CLEANER
       IRON ORE
                GAS
                PURIFICATION
      COAL
  COKE
OPERATION
   (P)
     LIMESTONE
                                                                           FINISHING
                                                                          OPERATIONS
                                                       SCARFING
                                                       MACHINE
FIGURE  1-4  Flow chart  of iron and  steel processes,   "(P)" denotes a
             major source of particulate emissions  if it is not  con-
             trolled.  Reproduced with revisions  from the U.S. Environ-
             mental Protection Agency.764
                                       36

-------
     Emissions of iron to the atmosphere from U.S. iron and steel mills,




which can be related to production rates for pig iron and raw steel, are




localized, as shown in Figure 1-3 above and in Tables 1-8 and 1-9.




Table 1-10 lists raw steel production by furnace type for 1974 and averaged for the




5-yr period 1970-1974.   Steel production by basic oxygen converter or electric




furnace is increasing while production by open»hearth furnace is declining.




Production data from cupola furnaces are not available, but in 1974 cupola




furnaces consumed 14,190,000 metric tons of scrap and 1,926,000 metric tons of


 .   .    227a
pig iron.




     Air pollutant emission factors are ratios of estimates of the average rate




at which a pollutant is released as a result of some activity (e.g., industrial




production), to an index of the level of that activity (e.g., production rate




or capacity).  Table 1-11, adapted from data collected for many iron- and steel-



                  764
making activities,    lists emission factors expressed as weight of particulates




per unit weight of metal produced applicable to:  iron and steel mills; ferro-




alloy production in electric smelting furnaces; gray-iron foundries; and steel




foundries.  The efficacy of various pollution control systems in reducing




emissions to the atmosphere is indicated in Table 1-11 by reductions in emission




factor for a given operation.




     That iron does enter the atmosphere from human activities will be shown




below.  A recent survey by Steiner    summarizes the relatively few measurements




available on the composition and particle-size distribution of emissions




(separately) from sinter plants, blast-furnace flues, open-hearth furnaces,




basic oxygen furnaces, electric furnaces, and foundry cupola furnaces.  Particle




sizes smaller than 5 um may represent half or more of the weight of particulates




from open-hearth and electric furnaces, and also from cupola furnaces,




especially of the hot blast type.  Typical reported    iron fractions of partic-




ulate matter emitted  in ferrous metallurgic   processes  are summarized  in




Table 1-12.

                                    37

-------
                 TABLE 1-8




U.S. Production of Pig Iron by State, 1974-
                      Production-
                                          Number of
Blast Furnaces—
3,514
6,517
15,423
6,905
-
4,237
15,843
19,681
4,621
9
19
27
9
2
11
41
50
11 (4, 4, 3)
 State




 Alabama




 Illinois




 Indiana




 Michigan




 Minnesota




 New York




 Ohio




 Pennsylvania




 California, Colorado, Utah




 Kentucky, Maryland, Texas




   West Virginia






 Total







 -Derived from Reno.





 —In thousands of metric tons.





 -Data from the  American Iron and  Steel  Institute.    Total number of blast




 furnaces as of January 1, 1975, 62 of which were out of blast.




 The total as of January 1,  1974 was 204, 42 of which were out of




 blast.





 -Does not include 2 ferroalloy blast furnaces.



—Rounded
                         9,872
                        86,615^
 18 (2, 10, 2,  4)
197-
                    38

-------
                         TABLE 1-9


        U.S. Production of Raw Steel by State,  1974-
State                                            Production—


California                                          3,895


Illinois                                           11,738


Indiana                                            20,945


Kentucky                                            2,452


Michigan                                            9,488


New York                                            4,985


Ohio                                               22,907


Pennsylvania                                       30,422


Rhode Island, Connecticut, New Jersey,


   Delaware, Maryland                               6,258


Minnesota, Missouri, Oklahoma, Texas,


   Nebraska, Iowa                                   5,217


.Virginia, West Virginia, Georgia, Florida,


   North Carolina, South Carolina, Louisiana        5,097


Arizona, Colorado, Utah, Washington, Oregon,


   Hawaii                                           4,465


Alabama, Tennessee, Mississippi, Arkansas           4,325
TotalS.                                            132,195
-Derived from data of the American Iron and Steel Institute.
                                                            23
—In thousands of metric tons.
c
"founded
                              39

-------
                                   TABLE 1-10



                U.S. Steel Production by Type of Furnace^-
Average for
1970-1974-
verter 66,528
35,162
21,999
Percentage
53.8
28.4
17.8
1974^
73,983
32,204
26,008
Percentage
56.0
24.4
19.7
Furnace





Basic oxygei



Open hearth



Electric
Total-                          123,688       100         132,195     100
a                  645a
-Derived from Reno.      Castings produced by foundries not covered in



 American Iron and Steel Institute data are excluded.
—In thousands of metric tons.



—Rounded
                                   40

-------
                                TABLE 1-11
                            Emission Factors-
Iron and Steel Mills—




Pig iron production




   Blast furnaces




      Ore charge, uncontrolled




      Agglomerates charge, uncontrolled




      Total, uncontrolled




      Settling chamber or dry cyclone




         Plus wet scrubber




         Plus venturi or electrostatic precipitator




   Sintering




      Windbox, uncontrolled




         Dry cyclone




         Dry cyclone plus electrostatic precipitator




         Dry cyclone plus wet scrubber




      Discharge, uncontrolled




         Dry cyclone




         Dry cyclone plus electrostatic precipitator




Steel Production




   Basic oxygen, uncontrolled




      Venturi scrubber




      Electrostatic precipitator




      Spray chamber
Total Particulates-
55




20




75 (65-100, range)




30




 7.5




 0.75








10




 1.0




 0.5




 0.02




11




 1.1




 0.055








25.5 (16-43, range)




 0.255




 0.255




 7.65
                                    41

-------
TABLE 1-11 (continued)
Iron and Steel Mills—




   Open hearth




      No oxygen lance, uncontrolled




         Venturi scrubber




         Electrostatic precipitator




      Oxygen lance, uncontrolled




         Venturi scrubber




         Electrostatic precipitator




   Electric arc—




      No oxygen lance, uncontrolled




         Venturi scrubber




         Electrostatic precipitator




         Baghouse




      Oxygen lance, uncontrolled




         Venturi scrubber




         Electrostatic precipitator




         Baghouse




Scarfing, uncontrolled




   Electrostatic precipitator




   Venturi scrubber




Ferroalloy Production—




Electric Smelting Furnace




   Open furnace




      50% FeSi -




      75% FeSi
 Total Particulates-









  4.15 (2.9-6.0, range)




  0.085  .




  0.085




  8.7 (4.65-11.0, range)




  0.085




  0.175




                       *»



  4.6 (3.5-5.3, range)




  0.09




      (0.14-0.37, range)




  0.045




  5.5




  0.11




      (0.165-0.44, range)




  0.055




£0.5




io.03




£0.01
100




157.5
                                     42

-------
TABLE 1-11 (continued)





Ferroalloy Production^                                 Total Particulates-



      90% FeSi                                        282.5



      Silicon metal                                   312«5



      Silicomanganese                                  97.5



   Semi-covered furnace



      Ferromanganese                                   22.5



Gray-Iron Foundries^



   Cupola furnace, uncontrolled                         8.5



      Wet cap                                           4



      Impingement scrubber                              2.5



      High-energy scrubber                              0.4



      Electrostatic precipitator                        0.3



      Baghouse                                          0.1



   Reverberatory furnace                                1



   Electric induction furnace                           0.75



Steel Foundries—



Melting



   Electric arc                                         6.5 (2-20, range)



   Open hearth                                          5.5 (1-10, range)



   Open hearth, oxygen lanced                           5   (4-5.5, range)



   Electric induction                                   0.05
a_                                                                         764
^Derived from U.S. Environmental Protection Agency data (published in 1973)



compiled by R. L. Duprey and R. Gerstle; data for iron and steel mills were



revised by W. M. Vatavuk and L. K. Felleisen.




-In kg of particles/metric  ton of product.  Emission factors are based on



 field measurements of numerous sources„




                                   43

-------
TABLE 1-11 (continued)




—In kg of particles/metric ton of metal produced, charged, or processed


 (see notes b, e_, £, j*) .


-For carbon-type electric arc furnaces; for alloy-type furnaces, multiply


 given value by 2.80.

p
—In kg of particles/metric ton of specified product.  Emission factors are


 based on few, if any, field measurements.


^Ferrosilicon with 50%, 75%, and 90% silicon, respectively.


%n kg of particles/metric ton of metal charged.  Emission factors are


 based on a limited number of field measurements.


—In kg of particles/metric ton of metal processed.  Emission factors are


 based on field measurements of numerous sources.
                                     44

-------
-Derived from Steiner.




—Expressed in percentage by weight.




-Typical



d
-Total of Fe203, FeO, and Fe components:  5-26% by weight.
                               TABLE 1-12


                                                       a.b
               Iron Fraction in Industrial Participates— —
Industrial Operation          F£o°o              FeO                 Fe
                                2. J              "™"




Pig Iron Production



   Blast furnaces



      U.S. plants                                                    36.5-50.3



      European plants                                                 5-40.0



   Sintering                  11.7-78                                5(£



Steel Production



   Basic oxygen



      N.oncombusted gas         4.0               21.4                66.7



      Combusted gas, A        90.0                1.5



                     B        80                                     56.0-57.7



                     C                           11.5-16.4           65.1-68.8



   Open hearth                61.3-96.5                              55.9-68.0



   Electric                   19-60               4-11                5-36



Ferrous Foundries



   Cupola                     d                  d                   d
                                    45

-------
      Emission factors for carbon monoxide have been estimated for uncontrolled



 blast furnaces (875 kg/metric ton), sinter discharges (22 kg/metric ton),



 basic oxygen furnaces (69.5 kg/metric ton), and electric arc furnaces (9 kg/



 metric ton), as well as for uncontrolled cupola furnaces (72.5 kg/metric


      764
 ton).     Carbon monoxide gas needs to be considered because of the potential


                                                                  219a
 for formation of the extremely toxic iron pentacarbonyl, Fe(CO),-.      Measure-



 ments of the presence or absence of such particulates in emissions from iron-



 and steelmaking seem to be lacking in the literature; c.f.  treatments of this



 compound in the petroleum industry and elsewhere.     '       Retention on



 solid surfaces of low concentrations of iron carbonyl vapors has been discussed


      .!   838a
 recently.



    Particulate matter is potentially a carrier for sorbed gases, as is



 discussed in Chapter 8.   It has been estimated that fine magnetite particles



 can carry one monolayer of sulfur dioxide at 2 ppm sulfur dioxide and 75 mono-



 layers at 66 ppm.  Thus some proposed systems of sulfur dioxide control have



 taken into account the concurrent particulate concentration.



    Other sources of iron.   Additional sources of iron and its compounds



 found in the atmosphere as a result of human activities include:  mining and



 handling iron ore; rusting and weathering of exposed iron;  weathering of iron



 pigments in surface coatings; radioactive isotopes of iron; iron-containing



 fertilizers; decay or burning of vegetation    and refuse;  and sea spray.   None



 of these sources compares  with raising of soil by wind,  or iron and steel



 manufacturing, except possibly in highly special circumstances.



    For a recent discussion of atmospheric corrosion of iron, including



 differences between marine and industrial environments,  see Spedding.      The



 artificially radioactive forms of iron are Fe-52,  -53,  -55, -59, -60, and



-61.  Their radioactivity is induced in naturally occurring iron atoms
                                    46

-------
directly, or in neighboring elements of the periodic table, such as cobalt



and manganese, which transmute to radioactive iron by decay.  Physical data



on these forms of iron are available, and a body of literature exists on



radioactive elements present in the atmosphere during and after testing of



atomic and nuclear devices.     Superphosphate fertilizers contain 2% iron,



some of which becomes airborne during or after application.  The iron



fraction in true solution in sea water has been estimated at less than


                                                             723a
2 iig/1; the total iron present may be 10 times that quantity.      Iron in



plankton may account for as much as 167o of the total iron.  Thus sea spray



associated with surf and wind introduces iron into the atmosphere.  There has



been some discussion of iron compounds as catalysts in certain chemical



reactions in the atmosphere, such as the formation of sulfuric acid mist from



sulfur dioxide and sulfurous acid, and the oxidation of nitrous oxide,    '
                                   47

-------
TRANSPORT

In Soils and Sediments

     Soils.  The weathering of rocks results in fragmented and/or chemically

altered materials that can then form a soil or be transported during erosion

as sediment or dissolved solids.  Under aerobic conditions, the amount and

distribution of iron in most soil profiles are largely dependent on the inter-

actions of the following soil-forming processes:

     1.  Release of iron from the primary minerals by chemical weathering;

     2.  Movement of iron from the surface to the subsoil by percolation

         of waters containing iron in solution or colloids;

     3.  Reaction of iron with dissolved and amorphous silica and alumina

         to form clay minerals;

     4.  Precipitation or accumulation of iron as amorphous ferric oxyhy-

         droxides in the subsoil, followed by slow crystallization to less

         soluble oxyhydroxides;

     5.  Transport of iron from the subsoil to the surface soil by plant
                                                                     46
         absorption and eventual release through plant decomposition.

     The rate of the first process generally increases with temperature and

moisture.  The weathering rate is also dependent on the parent minerals:

ferromagnesium silicates and sulfides decompose relatively rapidly, whereas

magnetite and ilmenite weather so slowly that they are found as heavy minerals

in soils and sediments.  Biochemical activity also increases weathering and

subsequent release of iron by attack of microorganisms or their metabolites.

Slightly acid waters produced by the dissolution of carbon dioxide from organic

matter decay also speed up weathering rates.  Most iron released from the

minerals is ferric, although some organic compounds form chelates with ferrous

iron.

                                     48

-------
     The most aggressive and widespread microbial agents are bacterial slimes



and acids, phenols, and certain alkaline compounds produced during the decompo-

                                                                                27

sition of organic residues or excreted by microorganisms into the soil solution.



In addition to mineral (nitric,hydrochloric, and sulfuric) acids, acidic organic



compounds are formed.  The dissolving action of the latter acid group is sometimes



stronger because they often can complex with iron.  Organic chelates found in



soil solutions are acids of low molecular weight (oxalic, citric, fumaric, and



others), fulvic and humic acids of higher molecular weight, and polyphenols.27»588



     The weathering of ferrous sulfides in rocks or sediments, e.g., some



exposed coals and adjacent shales, is aided by the bacteria Thiobacillus ferro-



oxidans.  The oxidation of ferrous iron is a source of energy to these micro-



organisms.  They are most active in an acidic environment (optimal pH range of

                                                            27

1.7-3.5), become inactive above pH 6 and die at higher pH's.    Bacterial



activity markedly increases the normally slow  rate of inorganic ferrous iron



oxidation at pH values below 4.



     The second process for distribution of iron in soils comes about more



rapidly at higher temperatures in highly organic, moist, acidic soils.  Although



microgram quantities of trivalent iron can move through acidic soils in solution



as hydroxy-complexes, perhaps  most dissolved iron  is transported as organic



Fe(III)  complexes.  Divalent  iron chelates  also  are mobilized and transported.


                                                                              588
Although it has been  claimed  that some microorganisms reduce  ferric compounds,



it appears that the organisms  isolated are  responsible  for  producing  substances


                                                             27
(by destroying organic compounds) that then reduce the  iron.     In  addition  to



the low-molecular  weight  acids,  polyphenols, and  fulvic and humic acids  mentioned



above,  it  is  possible that water-soluble hydroxamates formed  by  soil  organisms


                                778
may complex and transport  iron.      Iron  is transported through  soil  in  the  form



of colloidal  ferric oxyhydroxides and  iron-containing clays that migrate downward


                                                           27  588
under  the peptizing action of dissolved organic  compounds.   '




                                     49

-------
     The third process, iron reacting with silica and alumina to form iron-

containing clays, proceeds best at a near neutral pH.   Most of the iron

reacting is ferric.

     For the fourth process, the chemical or biologic environment of many sub-

soils must cause the iron migrating down into the subsoil to accumulate and

stabilize.  Heterotrophic bacteria of the Siderocapsaceae family, for example,

produce morphologically distinct iron precipitates.  The bacteria are widely

distributed in soils in which ferrous iron is highly unstable (with near-neutral

pH and higher oxygen pressure).  These microorganisms utilize or oxidize the

organic portion of dissolved iron chelates, releasing the iron that then will

precipitate as ferric oxyhydroxides.  Any released ferrous iron is rapidly

oxidized in such an environment.  In very moist or flooded soils with lower

oxygen pressure and neutral to slightly acidic pH, bacteria such as Gallionella

can promote the rate of ferric oxyhydroxide precipitation by increasing the
                                                                            27
oxidation rate of dissolved ferrous iron existing in ionic or chelated form.

     Iron-manganese concretions and iron intercalations are formed by micro-

organisms at specific soil horizons.  In rainy seasons, when soil capillaries

are nearly filled with water, iron-accumulating bacteria generally develop in *

thin zones where soil solutions come into contact with atmospheric oxygen.  At

moderate humidities, when capillary spaces contain both air and water, iron

bacteria develop in dispersed, separate nidi, which become centers of accumulating
                   27
iron and manganese.

     Ferric oxyhydroxides and clays accumulating in the subsoil can also increase

the deposition of iron by their absorption of iron-organic complexes.  This

accretion could continue until the surfaces were saturated and then be followed
                                   588
by microbial precipitation of iron.     The negatively charged surfaces of clays

could also adsorb positively charged ferric oxyhydroxide soils that were migrating

downward.
                                     50

-------
     Ferric oxyhydroxides deposited inorganically or biogenically in soils

are generally amorphous or very poorly crystalline.  With aging they can

slowly crystallize to more stable, less soluble oxyhydroxides.  This has the

effect of stabilizing the accumulation of iron in the subsoil.

     The rates of the five soil-forming reactions given change primarily with

climate and vegetation.  In soils (commonly in moist, forested regions) where

the first through the fourth processes operate at greater rates than the fifth

iron accumulates in the B horizon.  In northern Appalachian Spodosols, the

iron accumulated in the upper part of the B horizon is greater under coniferous
                      220
than deciduous forest.     Where chemical weathering is intense and accumulation of

iron is more important than its transport, the very high iron contents of Oxisols

can result.  These soils form in many tropical environments, especially during

dry seasons, when the ferric oxyhydroxides are dehydrated to hematite.  Oxisols

can evolve over a long period of time in a nontropical climate, which has
                                228
occurred in southeastern Africa.     In some moist, acidic soils with a high

organic content, the first and second processes may become much more important

to iron transport than the third and fourth ones, resulting in small iron con-

centrations in both upper and lower soil horizons.  An oxygen deficiency in

these soils slows the decomposition of iron-organic compounds and thus the rate

of oxyhydroxide liberation.

     When the rate of iron transport is low and the precipitation or stabiliza-

tion of iron as ferric oxyhydroxides predominates, iron distribution throughout

a soil can be relatively  uniform.   For example, in aerated, calcareous soils

in which carbonate equilibria buffer soil pH's near 8, ferric oxyhydroxides

severely limit the solubility of iron, and thus its availability to plants.

Many dry grassland soils (Mollisols) and arid soils (Aridisols) are calcareous

and cover large areas of the western half of the United States.  High phosphate


                                     51

-------
concentrations in a soil can also limit iron mobility.  At acidic pH's (where

ferric oxyhydroxides are more soluble), ferric phosphates (such as strengite)

are more stable than calcium apatites (such as hydroxyapatite) and limit iron

solubility.

     In poorly drained or flooded soils containing appreciable amounts of

organic matter, iron mobility is often much greater than in well-drained,

aerated soils.  Organic compounds are oxidized, producing an anaerobic environ-

ment in which ferric iron is reduced.  Concentrations of up to several milligrams

per liter of dissolved ferrous iron can exist in the soil solution.  The concen-

trations of other dissolved constituents that could precipitate ferrous iron,

particularly carbonate, sulfide, and phosphate, limit the amount of dissolved iron

available.  Major controls on the transport rate of dissolved iron in flooded

soils are concentration gradients for iron and the flow of water within the

soil.

     Soil pH is one of the most important regulators of iron mobility  and.by

extension, availability to plants.  The optimal soil pH for iron supply is

about 6.0-6.8 for most plants.  Within this range there is usually no iron
                     491
excess or deficiency.      Other important factors, such as the presence of

organic chelates or excessive heavy metal micronutrients such as manganese,

copper, and zinc can increase or decrease the availability of iron to plants,
             595
respectively.

     In addition to controlling the solubility of iron in most soils, ferric

oxyhydroxides are important to reactions active at soil surfaces.  Specific

surface areas of oxyhydroxides are especially large because they can coat parti-

cles of all size fractions and thus exert chemical activity much greater than
                                   410
their concentrations would suggest.     The isoelectric pH of hydrous ferric
                                     52

-------
                                              603
oxyhydroxides ranges from approximately 6.7-9,    indicating a  net  positive surface

for many soils.  This surface is shown by the association of anion adsorption
                                                               750
in soil clays with the presence of iron and aluminum compounds.     Phosphates

are fixed by specific adsorption on oxyhydroxide surfaces.  Certain trace metals

are coprecipitated in or adsorbed on the coatings.  Indeed, the hydrous oxides

of iron and manganese are thought by Jenne to be the principal control on the
                                                                               410
fixation of cobalt, nickel, copper, and zinc in soils and freshwater sediments.
                                     _o
The adsorption of molybdenum (as MoO.   ) is stronger on ferric oxyhydroxides than

on aluminum oxide and clays, and increases as soil pH decreases for positive
                      249
oxyhydroxide surfaces.

     Sediments.  The products of rock weathering can be transported by the erosional

agents/water (the most important), wind, and ice.  Iron movement in surface running

water is principally as suspended matter: colloidal ferric oxyhydroxides, ferric

oxyhydroxide coatings on silts and clays, or iron-containing clays or micas.

Concentrations of suspended iron thus depend primarily on the total amount of sus-

pended matter and range from <0.1- >1,000 mg/1.  A smaller amount of iron is

transported in the bedload of streams in heavy minerals, primarily magnetite and

ilmenite.    The average concentration of iron in suspended sediments carried by
                                                             295
the world's streams to the ocean has been estimated to be 7%.     However, when

the amount of annual iron flux is compared with the total dissolved and suspended

load of material carried by all erosional agents to the oceans, the average iron

concentration is 5.5%, about the same as' for the average continental crust. Thus,

it appears that for the world as a whole, weathering and erosion are not appre-

ciably altering the average iron concentration of the land.  Relatively high

losses of iron to suspended sediments from land subject to greater erosion are

balanced by the accumulation of iron in Oxisols of the tropics.

     The concentrations of iron in suspended sediments generally increase inversely

with particle size.  Coarser sediments deposited near the ocean shore are poorer

                                    53

-------
in iron than those in deeper sea areas, and the highest concentrations are found

in red clays far from shore.  There are, however, no present-day examples of

sedimentary iron deposits approaching in composition and extent the ancient iron
                         407
formations or ironstones.

     Because moisture and temperature control the movement of iron within aerated

sediments, transport tends to be similar to activity found in other soils in the

same climate.  Where iron mobility is low and the leaching of silica by infil-

trating meteoric waters is important, iron formations and ironstones have become

enriched in iron, sometimes to depths as great as 400 m in tropical areas.

     Most sedimentary deposits are ultimately subaqueous.  Iron moves primarily in

solution in the pore waters, either by diffusion as a response to chemical gradients

or by flow of the water.  In lake, estuary, sea basin, or lagoon-bottom sediments,

where silts and clays accumulate, often enough organic matter is decomposing

to produce reducing conditions.  Then ferric oxyhydroxides are dissolved  as  the

iron is reduced to the ferrous state.   The ferrous iron can then migrate  in  solution

and be reprecipitated, often as sulfides,  but also as carbonates and phosphates.

     Iron participates in a redox cycle at the sediment-water interface in lakes

that become stratified annually.  The cycle is an important control on dissolved

phosphate concentrations just above the interface.  In the summer, phosphate and

ferrous iron concentrations increase 'as adsorbed phosphate is released during the

reduction of ferric oxyhydroxides in the oxygen-depleted waters.  During turnover

of the water in fall, the increase in dissolved oxygen causes ferrous iron to

be oxidized.  Simultaneously, phosphate concentrations decrease by adsorption
                                          807
freshly precipitated ferric oxyhydroxides.


Transport  in the Hydrosphere

     Data on concentrations of iron present in water from various sources are
         59,480
abundant,       but they contain certain inherent biases.  Common analytic

                                     54

-------
methods are  insufficiently sensitive to detect iron concentrations much below
         104
10 yg/1;     hence.extremely low concentrations in fresh water are seldom

studied.  Concentrations of iron in colloidal particulate form are influenced
                                                                           347,419
by the method of pretreatment used in the sampling and analysis procedures.

For computing geochemical transport rates, the difference in chemical behavior

between the colloidal particulates and dissolved ions is not important, but

different forms of iron may  behave dissimilarly in plant and animal metabolism.

Hence the form in which iron occurs in water requires consideration.

     Reported concentrations of iron for the ocean range from a few tenths of
                                         59             480
a microgram per liter up to about 3 yg/1.    Livingstone    reported a mean

of 670 yg/1 for "average" river water.  Such numbers suggest a solubility

higher than that predicted in Figure 1-2.  Because ocean and rivers generally

contain substantial amounts of dissolved oxygen, the stable forms of iron should

be ferric.  Filters with pore diameters near 0.45 ym generally are used to

remove particulate matter before making analyses, and the dissolved state is

thus functionally defined as any material passable through openings of this

size.  The iron reported in rivers, lakes, and oceans is generally a colloidal
                                         419,813
suspension of ferric hydroxide particles.         In some waters, notably the

drainage from swamps or bogs, iron may be present as a stable organic complex.

Such waters commonly are yellow or brown and may contain more than 1 mg/1 iron.

In some such waters, however, the organic matter carrying the iron is also

colloidal.  Streams contaminated by acid mine drainage have a low enough pH to

retain more than 100 mg/1 of dissolved ferric and ferrous species, but chemical

reactions with sediment minerals and mixing of the water with alkaline inflows

eventually neutralize the acidity as the water moves downstream and the iron

is oxidized and precipitated as ferric oxhydroxide.

     The range of iron concentrations in groundwaters is quite wide.  In systems

where organic material is present, the oxygen content of the groundwater can be
                                     55

-------
 depleted and iron brought into solution in the ferrous form.  Concentrations



as high as 50-100 mg/1 can be held in solution in these waters at pH 6.0 and



 concentrations of 0.5-10 mg/1 are not unusual under many different kinds of


                     345a
 geologic conditions.      Groundwater with a measurable dissolved oxygen con-



 tent is likely to contain no more than a few micrograms per liter of iron if



 the pH is above 5.  Pyritic material may be attacked by oxygenated water that



 recharges the groundwater reservoir, and substantial amounts of divalent iron may



 be brought into solution as the sulfur of the pyrite oxidizes to sulfate.



 Once the oxygen is consumed, the ferrous iron can remain in solution.



      In 1962, the U.S. Public Health Service set a drinking water standard of a



 maximum of 0.3 mg/1 of iron, which the 1974 National Academy of Sciences recommen-



 dations supported.     Most public water supply systems in which iron in the raw



 water  exceeds this  limit substantially  lower  the iron  content by  treating the



 water  before it is  delivered to consumers.  However, iron  in pipes, tanks,  and



 other  equipment in  the distribution system exposed  to  water may be attacked by



 corrosion.  Thus  iron occasionally may  be present in tap water in greater



 quantities than allowable.  The 0.3 mg/1 limit was  set for primarily aesthetic



 purposes, as water  excessively high in  iron tends to stain plumbing fixtures



 and laundry and may be turbid. Compounds such as sodium phosphate are  sometimes



  added  in water  treatment to complex dissolved iron  and prevent or delay its



 precipitation.



      Mobilization, transfer, and fixation processes affecting iron in the hydro-



 sphere and controlling its removal from and return to  the lithosphere are cyclic.



 For example, iron released by weathering and biochemical processes from soil and



 rock minerals is carried to the ocean by rivers and the atmosphere.  The average


                                                                                 301
 residence time of the iron in ocean water has been estimated to be about 140 yr.



 During this period it participates in chemical and biochemical processes that



 ultimately precipitate it.   The iron next becomes a part of accumulated sediment.





                                       56

-------
The residence time of dissolved iron in the ocean is very brief compared to that
                                             236
estimated for most other metals.  Duce et^ al.    reported that the iron content

of atmospheric particulates entering the ocean from direct fallout was only

slightly enriched above the average level for mineral matter at the earth's sur-

face.  They concluded that most iron transported in this way came from natural

weathering processes.

     The importance of biota in certain redox processes influencing aqueous

chemical behavior of iron has already been noted.  Certain aspects of the topic,

however, are worthy of further consideration.  Microbes and fungi present in

soil and subsurface environments may mobilize iron and bring it into solution

by mediating chemical reactions that may serve as energy sources.  Pyrite, for

example, may be converted to ferrous and sulfate ions; usually oxygen from the

atmosphere is the oxidizing agent.  The iron is not necessarily oxidized, but

it is made available for solution through decomposition of the pyrite.  In

anoxic environments, ferric oxides or hydroxides may be reduced when certain
                                                                 590
strains of microorganisms and an organic food source are present.

     Once it has been brought into solution, aerobic species of bacteria common

in soils and surface waters may catalyze oxidation and precipitation of ferrous
                         345a
iron to ferric hydroxide.      This reaction is thermodynamically favorable in

the presence of oxygen and thus can be an energy source.  The ferric hydroxide

produced by the aid of "iron bacteria" such as Crenothrix or Leptothrix is

commonly encrusted on the cells' organic sheaths.  Colonies of these microorganisms

constitute gelatinous masses of such material and these deposits occur in many

wells, pipelines or other places in which water that contains ferrous iron is
                  589
exposed to oxygen.     The deposits are a nuisance in agricultural tile drains
                         277
in some parts of Florida.      These  iron-oxidizing  bacteria  have some
                          589
 characteristics of fungi.


                                     57

-------
      Growth  cycles  of marine  and  fresh water  algae and  related  aquatic



microorganisms  substantially  influence concentrations of  iron in  open water



bodies.  Demand for iron by algae during an algal bloom can  greatly  reduce



concentrations  in solution.   This iron is  released again  upon death  and  decay

              590

of  the plants.      Some systems may exist  in  which increased iron availability



can trigger  an  increase in algal  growth.   Possible correlations have been



advanced between "red tides"  noted along the  west coast of Florida and flooding



in  coastal streams  that was responsible for carrying organically  complexed iron

                                                422,515

into  the seawater offshore (see also Chapter  3).



Transport in the Atmosphere



    Direct transport from source to receptor.   The atmospheric dispersal  of



contaminant gases and particles of negligible  settling speed is  often treated


                                                                  127 281 697
as  simply the eddy diffusion of matter from point or line sources.   '    '    '


715 754
    '     The diffusion model in predominant use applies  the continuity equation



in  a steady (mean)  wind and permits the emergent plume to spread crosswind



vertically and horizontally while traveling downwind,  so as to yield  at  any



given distance Gaussian profiles of concentrations of  contaminant  in  those  two



directions.  Standard deviations of the two crosswind  distributions of contam-



inant that have been widely accepted were derived indirectly from a number  of



field experiments;  they increase with distance from the  source and vary with



atmospheric stability and with the duration of time over which the concentrations



are averaged.  Values for these standard  deviations ("diffusion  parameters")



are chosen to obtain estimates of contaminant  concentration,  despite  the  absence



in most places of direct measurements of  atmospheric stability,  by schemes  that



employ one or more of the meteorologic elements commonly measured  or  otherwise



known or estimated,  viz.,  wind speed,  solar irradiation, and  cloud cover,



                                                      525
together with the time of day and sometimes the season.      Another approach




                                    58

-------
to these standard deviations of contaminant distribution makes use of the



measured range of fluctuations in wind direction at a suitable height above



the earth's surface when these data are available.



    The empirical basis of the diffusion model for estimating contaminant con-



centrations at distances beyond a few tens of kilometers is weak, and other

                           901 fin*j Q-I n

methods must be considered.   '   '     Other problems of the model include its



inapplicability in near calms or during precipitation, and the all-too-common



absence of a reliable climatologic lexicon of diffusion parameters for the



height, terrain geometry and surface cover, as well as for the geographic loca-


                      788
tion being considered.     Some success has been realized in accounting for jet



and buoyancy effects on the downwind profiles of plume height versus distance,



an understanding important because of the major influence on ground-level con-



taminant concentrations that the effective height of emission exerts, especially



in the near field, i.e., within 20 or so chimney heights.   *»715  Another



important effect on plume height arises from the distortion of airflow by



buildings and other structures in the vicinity of the place of emission, in an



extreme case even resulting in recirculation to the origin.  Approaches to



estimating the effect of architecture and construction on the performance of



relatively short stacks and chimneys have been proposed.   '  ''   *     Mean-



while, allowance for or even recognition of this effect, e.g., in the fitting



of pollution models to measurements, is spotty.



    As particles increase above 10-30 urn in aerodynamic diameter, deposition on



the earth's surface and thus loss from the diffusing plume must be taken into



account.  This phenomenon is commonly treated by regarding the particulate



plume to be similar to a gaseous plume that is tilted downward at an angle



whose tangent is the ratio of the settling speed of the particle to the wind



speed,    the angle of tilt changing with particle size.
                                     59

-------
    Thus the estimation of contaminant concentrations resulting from specified

emission rates of, for example, iron-containing particles, is common practice,

but one whose results should be accepted with reserve,,  The consequences of

plausible alternatives to the model inputs advanced from meteorologic and other

perspectives might be useful to contemplate, and if even fragmentary measure-

ments of contaminant concentrations are available, patterns of correspondence

and dissimilarity should be noted for the information they provide on the dis-

persal process.

    Reentrainment of deposited dust and raising of soil by the wind.   Contam-

inants not only enter the atmosphere from stationary and mobile sources of

emission, but enter or reenter when winds are strong enough to raise the soil,

sand, or previously deposited dust.   '     Little air movement is required when

the particles are fine and dry.  The presence in the eastern U.S.  of dust from

the Great Plains in periods of drought and the sighting of Sahara  dust far out

to sea are not rare.   As a rule, as the wind speed increases, the  concentration

of airborne dust near the surface will rise— a result of increasing flux of

particles from the surface—but then  it will fall as dilatation begins to

dominate production.  With huge reservoirs of dry dust available, the high dust

concentration  and low visibilities characteristic of dust and sandstorms are

reached.  This mobility of iron-containing soil and deposited dust is neces-

sarily reflected in  the measurements  of airborne iron in remote areas discussed

above.  On  local scales, particulate  loss from ore piles may be controlled by

use  of moisture, covers, and wind breaks.  The process of reentrainment has
                                                                      /TOO
been studied and described for an  important case by Sehmel and Lloyd.
                                    60

-------
     Observed ambient concentrations.  Measurements of iron concentrations in



surface air from the National Air Surveillance Network between 1970-1974 were



summarized earlier for nonurban sites.  Similar data for 210 urban sites com-



prising 604 complete site-years of record show annual average values ranging


              3                                                           3
from 0.23^ug/m  at a site in St. Petersburg, Florida in 1972 to 12.12 ug/m  at


                                                                               2
a site in Steubenville, Ohio in 1971.  The median annual average was 1.295/ig/m



or about 5 times the median annual average for iron at nonurban sites.  (The



statistics above are based on analysis of EPA data summaries     by the Subcom-



mittee on Iron.)



     Figure 1-5 shows the cumulative frequency distribution of the annual



average iron concentrations over the 604 site-years of record at urban sites



from 1970-1974, and also corresponding data for the 80 site-years of record at



nonurban sites previously noted.  The highest 18 (3%) concentrations for urban



sites came from 14 places in Ohio, Indiana, Pennsylvania, and Kentucky (cf.



Figure 1-3).  The lowest 18 came from 13 sites all located in coastal states



including Hawaii, with the single exception of one site-year at Lincoln,


„ ,    ,   765a
Nebraska„


                                           3

     The highest 3-mo average was 16.0jug/m  at the Steubenville site in the

                                                     3

second quarter of 1971; the second highest, 14,6 ug/m , was measured at a site



in East Chicago, Indiana during the same period.  Both sites are near major



iron and steel mills.  The tendency observed in the data toward maximum con-



centrations of iron in air during the first half of the year is consistent with



the expectation that strong winds affect the contents of surface air.



     The primary 24-h and annual mean standards for total suspended particulate

                          o

matter are 260 and 75 ug/m , respectively.  Based on the EPA's composite annual



average data from 1,096 sites monitored between 1970-1974, ambient concentrations
                                    61

-------
                        210 Urban Sites
       600
c

-------
of total suspended particulates declined from 80 to 66^ig/m  during the 5 years

studied.     The percentage of air monitors reporting values exceeding the

primary standard decreased from 16 to 8% for the 24-h average, and from 50 to

23% for the annual average.

     Of 236 air-quality control regions (one or more monitors) reporting minimum

acceptable data in 1974, 99 exceeded the 24-h primary standard and 111 exceeded

the annual primary standard.  "Nonpoint sources"--defined as sources emitting

fewer than 90 metric tons/yr of total suspended particulates—also contributed

to the excessively high amounts.

     It is interesting to note that attainment and maintenance of these standards

is adversely affected by two kinds of fugitive emissions:  those generated from

industrial operations and released to the atmosphere through plant apertures

other than the primary exhaust system; and those generated by the force of the

wind or human activity on the land, including wind-raised particulates from

croplands, unpaved roads, and exposed areas at construction sites, as well as
                                                                     £ Q f\
surface material made airborne by vehicles and machines from streets,    crop-

lands, etc.  Fugitive dust is a major problem in the arid West, but it is not

confined there.  Nationwide estimates of pollutant emissions have declined

steadily in tonnage of particulates between 1970-1974.     One should expect the

"soil-derived" fraction of iron in the ambient air to increase or decrease in

quantity independently of the decline in emissions, according to cultivation

practices and the frequency of dry, windy conditions.
                                   63

-------
                                CHAPTER  2

                         MICROORGANISMS AND IRON


     Iron is thought to be a universal requirement for microbial cells whether
                                                                     572
they be prokaryotes or eukaryotes, i.e., bacterial or fungal species.     The

only possible exception is the lactic acid bacteria.  These organisms flourish

in environments, such as dairy products, which are notoriously low in absolute

and/or available iron concentrations, and they do not contain any cytochrome or
                        148
hydroperoxidase enzymes.     The absence of catalase affords a quick presumptive

test for colonies of lactic acid bacteria growing on agar surfaces.  In lactic

acid bacteria, the normal iron-containing ribotide reductase is replaced by a

functionally equivalent vitamin B   enzyme system.  Hence, in these species,

iron is not even required for synthesis of DNA.  Although it is a generally

nutritious element for microbes, certain hazardous ecologic and environmental

factors are associated with microorganisms and iron that need to be evaluated.


METABOLISM

Assimilation

     All known aerobic and facultative anaerobic organisms possess multiple

systems for acquiring this crucial nutrient.   These systems fall into

two general classes:  low affinity, or nonspecific, and high affinity, or
         575
specific.     In the first system, almost any inorganic or organic iron compound

will suffice to support growth, provided it is furnished in substantial amounts.

The biochemical mechanism of operation of this pathway is not well understood,

because no convenient way exists to probe the biochemical genetics of the system.

In contrast, the high-affinity system depends on the elaboration of specific

carrier molecules, called siderophores (formerly, siderochromes), and the cognate

surface receptors for the iron-laden form of the carrier.  Formation of sidero-

phores and their specific receptors may be induced by growth of the organism in

environments either low in absolute or available iron concentrations.

                                    64

-------
Physiologically Active Iron Compounds in Microbes

     As a rule, microorganisms contain most of the iron compounds found in the

cells of higher plants and animals.   Ferredoxins, nonheme iron-containing compounds

that serve as electron transfer agents, are abundant in all species requiring a

carrier of low-potential electrons,  such as the Clostridia, and in all organisms
                                               488
performing photosynthesis or nitrogen fixation.     It is probably physiologically

counterproductive for microbes to synthesize oxygen carriers or possibly even iron

storage compounds.  However, hemoglobin-like pigments and ferritin have been found
                   213
in certain species.     As expected, siderophores have not been detected in strict

anaerobes or lactic acid bacteria.  Table 2-1 lists iron compounds reported to

exist in microbial species.

Dissimilation

     Biologic degradation of microbial iron compounds should be similar to that

of higher species, although the subject does not seem to have been investigated

systematically.  An organism, Pseudomonas FC-1, has been isolated and shown to

be capable of growing on ferrichrome A, a fungal siderophore that was its sole
                              793
source of carbon and nitrogen.     Intracellular microbial iron compounds usually

are biologic catalysts and are present in small amounts within the organism.

No special problems with pollution or other ecologic disturbances are anticipated

in the biodegradation of these substances and no toxic substances should be pro-

duced.


DEFICIENCY

General Effects of Iron Limitation

     Among the aerobic and facultative aerobic microorganisms, the general

effect of iron deprivation will be the switch of the metabolism to a more anaer-

obic character.  Glycolysis proceeds without utilizing iron catalysts; the latter

begin to influence energy metabolism at the level of the tricarboxylic acid cycle
                                     65

-------
                          TABLE  2-1

           Iron Proteins  and Enzymes in Microorganisms
  Substance

  Oxygen-binding proteins
    Hemoglobinlike proteins
    Leghemoglobin

  Iron storage
    Ferrltin
    Phosphoproteins

  Hydroperoxidases
    Catalase
    Peroxidase

  SuperozLde dismutase

  Electron-transfer proteins
    Berne proteins
      Cytochromes
    Nonheme  proteins
      Iron-sulfur proteins
      Nitrogenase
      Glutamate synthase
      Hydrogenase
      Ribotide  reductase 82

  Iron flavoproteins
    Dlhydroorotate dehydrogenase
    Succinic  dehydrogenase
    Nitrate  reductase
    Xanthine  oxidase
    HADH° dehydrogenase
    Malate vitamin K reductase
    Adenylylsulfate reductase
    NABPIT sulfite reductase
    formate dehydrogenase

  Oxygenases
    Beme type
      Tryptophan dioxygenase
    Honheme type
      Diverse substrates

  Iron-activated enzymes
    Aconitase

    D-Altronic  acid dehydrase
    L(+)-Tartrate dehydrase

  Unclassified  iron enzymes
    Enzymes acting on amines,
    •mine acids, and other
    substrates
                                   Source
                                   Restricted distribution
                                   Rhizobia
                                   Phycomyces blakesleeanus
                                   Unknown


                                   Widely distributed in aerobes
                                   Restricted distribution

                                   Widely distributed in aexobes
                                   Widely distributed

                                   Widely distributed
                                   Kitrogen-fixing species
                                   Escherichia  coli
                                   Restricted distribution
                                   Escherichia  coli
                                   Zymobacteriun oroticiua
                                   Widely distributed
                                   Micrococcus  denitrificans
                                   Micrococcus  lactilyticus
                                   Azotobacter  vinelandii
                                   Mycobacteriug phlel
                                   Desulfovibrio vulgaris
                                   Enteric bacteria
                                   Pseudomonas  oxalaticus
                                   Widely  distributed

                                   Widely  distributed
                                   Coincident with tricarboxylic
                                   •cid  cycle
                                   Escherichia coli
                                   Pseudomonas
                                              sp.
Phosphodiesterase
                                   Widely  distributed

                      ?             Coincident  with cyclic AMP
aFrom Neilands572 uniess  otherwise  noted.
^Zinc, copper,  and manganese,  as well as  iron,  have  been
    detected in  superoxide dismutases  from different  sources.
^Nicotinamide adenine dinucleotide,  reduced.
 Nicotinamide adenine dinucleotide  phosphate,  reduced.
                                                                       317a
                           66

-------
(aconitase, succinic dehydrogenase) and the respiratory chain (heme and nonheme

iron).  Thus, a facultative anaerobe such as Escherichia coli has a substantially

higher iron requirement when offered succinate rather than glucose as an energy

source.  Microorganisms seem to be infinitely adaptable to the hardships imposed
                               537
by living in media low in iron.     Certain species of Clostridia, when grown

without iron, produce a low-molecular weight flavoprotein (flavodoxin) with the

same redox potential and biologic function as ferredoxin.  The metabolism
                                                      475
possible at low levels of iron has been characterized.

     The production of numerous commercial pharmaceutical products depends

on microbes grown at low levels of iron.  Formation of citric acid by Aspergillus

niger and rlboflavin by Eremothecium ashbyii are examples of commercial fermenta-
                                                 537
tions based on growth under iron-poor conditions.     Iron represses  the synthesis
                                10-12
of a number of bacterial toxins.       The mechanism for inhibiting synthesis of
                                                            557
the Corynebacterium diphtheriae toxin has been investigated.

Siderophores

     Chemical nature, mechanism of action, and distribution among species.  Sidero-

phores are the low-molecular weight (about 500-1,000 daltons), high-affinity

carriers found in practically all aerobic and facultative anaerobic  microorganisms
                               229,417,571-576
in which they have been sought.                 These compounds are virtually
                                                                            30
ferric-specific, have large formation constants for trivalent iron (about 10   or
                                                           573
higher), and the liganding atoms are generally all oxygens.      The complexes

are in d5 orbitals and exhibit rapid exchange kinetics.  Chemically the sidero-

phore ligands are usually classified as hydroxamates or catechols.  The former

are common to higher microorganisms, such as fungi, whereas the latter are

usually encountered in the prokaryotes.  Ferrichrome, a cyclic hexapeptide ferric

trihydroxamate, is the prototype of the hydroxamate class of siderophores.  Ferri-

chrome and ferrichrome A, and the related compounds ferrichrysin, ferricrocin,

ferrirubin, and ferrirhodin are growth-promoting iron chelates.  Hydroxamate


                                   67

-------
 siderophores are found in Neurospora, Ustilago and other basidiomycetes, and

 in species of the mold genera Aspergillus and Penicillium.  Enterobactin, also

 called enterochelin, is the prime example of the catechol class and it is pro-

 duced by the enteric bacteria.  The latter, however, may also produce a hydroxa-

 mate, such as aerobactin.

     Except for the mycobactins, siderophores tend to be exceptionally water-

 soluble.  They can usually be extracted into either benzyl alcohol or phenol-

 chloroform (1:1, vol/vol) from water or aqueous salt solution at neutral or

 acidic pH.  Diluting the organic phase with diethylether and back extraction

with water returns the siderophore to the water.  If a siderophore does not have

 a hexadentate coordination center (e.g., rhodotorulic acid), it may form poly-

nuclear complexes with iron that are practically impossible to extract into an

organic solvent.  When this happens, it is best to isolate the substance in the

 absence of iron0  Similar considerations apply to crystallization of the siderophore

 or its metal complex.  In general, the ligands and chelates may be crystallized from

 the lower alcohols.   The hydroxamates are much more stable than the catechols,

but ferric hydroxamates decompose upon excessive dry or steam heat.  Mineral

acid hydrolysis of the ferric hydroxamate will lead to disproportionation of

the hydroxylamino group and extensive degradation.  A variety of techniques are

available for removing iron: extraction with 8-hydroxyquinoline, treatment with

dilute alkali in the cold,  or continuous extraction with ether.in acidic media.

The hydroxamic acid  group is weakly acidic and has a pKa_ of around 9.   The

hydroxylamino function is easily oxidized at neutral or alkaline pH; it has a

pK.a close to 5.   The paramagnetic ferric ion causes line broadening in the nmr

spectra, but the resonances of the I aluminum or gallium complexes are sharp and
              481,482
well resolved.

     The hydroxamate siderophores give the Czaky test for bound hydroxylamines;
                                                             719
respond (inexplicably)  to the Folin-Ciocalteu phenol reagent;    are oxidized


                                    68

-------
by periodate to the acyl moiety and a cis-nitrosoalkane dimer, and by performic



acid to two carboxylic acid fragments; are reduced (if not too hindered) by



Raney nickel and hydrogen gas at 22.5 kg pressure; are hydrolytically reduced



by hydriodic acid; and are cleaved by nonreducing mineral acid to the carboxylic



acid and hydroxylamine components.  The hydroxylamine moiety will reduce alkaline



tetrazolium without heat, thus affording a sensitive spray for chromatograms.



     The catechol class of siderophores give the Arnow test and those containing



the dihydroxybenzoyl nucleus have a characteristic blue fluorescence and an



electronic absorbancy band near 310 nm in the ultraviolet region.



     The hydroxamate siderophores are well-detected by chromatography on silica


                                               375
gel, either in columns or on thin layer plates.     Enterobactin, however, is



more difficult to discern by chromatography; 57o ammonium formate (wt/vol) in 0.5%



formic acid (vol/vol) will move it slightly on cellulose thin layers or paper



and in 6% acetic acid it has an Rf of 0.29.  The iron complex is retarded



chromatographically on cellulose weak anion exchangers or on Sephadex"  LH-20.



     In both catechol and hydroxamate siderophores, the ferric ion is attached



to an asymmetric, bidentate five-membered ring.  The metal-binding center may



be chiral, provided there is optic activity in the ligand; geometric isomers


                642
are also formed.     Ferrichrome A, ferrichrysin, and ferric mycobactin have



been examined by X-ray diffraction and shown to have the  A-cis configuration,



whereas ferric enterobactin is probably  A-cis.  Isostructural Cr(III) analogs



are employed for assignment of configuration and in biologic transport experi-



ments for elucidation of the preferred optic isomer.



     The electronic absorption spectra of the ferric hydroxamate siderophores


                                                                      573
display hypsochromic and bathochromic shifts that are dependent on pH.     The



red-orange to tea-colored 3:1 complexes have a maximum absorbancy at 425-440 nm



with molar extinctions of approximately 3,000.  Upon acidifcation, tris-ferric
 Specific products and trade names are cited solely for illustrative purposes.

Mention does not imply an endorsement from the National Academy of Sciences or

the National Research Council.

                                    69

-------
acethydroxamate will decompose through the 2:1 to the 1:1 complex.  The 1:1

complex is purple, with a maximum absorbancy near 500 run and an extinction of

about 1,000.  In contrast, a trihydroxamate such as the ligand of ferrichrome,

will retain the 3:1 structure even at pH 2.

     Ferric enterobactin is a wine-colored, water-soluble, trivalent anion with

a molar absorptivity of 5,600 at 495 nm.  The ferric ion stabilizes the ligand,

although the molecule can suffer oxidation of the phenolic hydroxyls and hydro-

lysis of the ester bonds in the inner ring.

     Both enterobactin and deferriferrichrome undergo a drastic conformation
                               483
change upon metal complexation,    which enables the membrane receptors to

recognize only the metal-laden form.

     Although a very large number of molecular species transport iron in ferrous-

ferric microbial systems, predictably, the tris-catechol and tris-hydroxamate

ligands are the most efficient carriers of the ferric ion.  A microbial product

qualifies as a siderophore if:

                                                  30
     *  it has a high formation constant (about 10   or greater) for Fe(III),

        high relative avidity for Fe(III) over Fe(II), and kinetic lability;

     •  biosynthesis can be induced when iron in the microorganism's

        environment is low; and

     •  it is capable of transporting iron in microbes naturally or artificially

                                                     40
        lacking high-affinity iron transport systems.

     Substances with classic siderophoric chemical structures can be assigned

to one of seven families.  Of these families, one is constituted of catechol

and the other six are hydroxamic acids.  The following figures and captions

supply formulas and diagrams of the prototypes and a few other members of each

of the seven families.  A more detailed list of structures has been published
          571
elsewhere.


                                    70

-------
     The prototype of the catechol family is enterobactin, the cyclic trimer




of 2,3-dihydroxy-N-benzoyl-L-serine, shown in Figure 2-1.  Ferric enterobactin




has been isolated from all enteric bacteria examined for its presence.  A com-




pound of related structure, providing two catechols and one salicylic acid

                                                                             728

moiety bound to spermidine, has been isolated from Micrococcus denitrificans.




     The ferrichromes, set forth in Figure 2-2, are cyclic hexapeptides comprised




of a tripeptide of ferric S-N-acyl-S-N^hydroxy-L-ornithine and a tripeptide of




small, neutral amino acids such as glycine, serine, or alanine.  To provide the




"hairpin turn" and the cross-g structure in the cyclohexapeptide, one residue




must always be glycine (R  = H in Figure 2-2) .  The acyl group furnishing the




R in Figure 2-2 may be acetate or some small carbon piece derived therefrom,




such as trans-g-methylglutaconic acid, as in ferrichrome A, or anhydromevalonic




acid, as in ferrirubin or ferrirhodin.  In ferrichrome itself, R is methyl and



 123
R  = R  = R  = H.  Albomycin (grisein) has a close structural relation to ferri-




chrome; it too displays broad-spectrum antibiotic activity.




     The rhodotorulic acid family have in common the diketopiperazine of




N-acyl-6-N-hydroxy-L-ornithine.  Members of the rhodotorulic acid family—




rhodotorulic acid, dimerumic acid, and ferric coprogen—are diagrammed in




Figures 2-3, 2-4, and 2-5.




     The citrate-hydroxamate family (Figure 2-6) of siderophores has at least




six members, three of which have been isolated.  The remaining three theoretical




members are awaiting discovery, hence the trivial name "awaitin."




     The mycobactin family (Figure 2-7), produced by mycobacteria, has many




individual members because of the different substitutions in the R groups, as




marked in Figure 2-7.  The molecule has six potentially optically active




carbon atoms, noted as a, b, c, d, e, and f in Figure 2-7.




     The structure of the fusarinine family is shown in Figure 2-8.  Fusarinines




(also called fusigens) are obtained from species of fusaria and other fungi.
                                   71

-------
                                                               387a
FIGURE 2-1   Ferric enterobactin.  Reproduced from Isied et al.
                               72

-------
       R
©
                      R
FIGURE 2-2   Basic structure of the ferrichrome cyclohexapeptide ferric
             trihydroxamates.   In ferrichrome itself, which has a tri-
             glycine sequence. R1 = R2 = R3 = H and R = CH-.  Reproduced
             from Neilands.571
                                73

-------
                0 OH
                ii   i
          H3C-C-N
 N-C-CH:
 I  II

HO  0
                                                          571
           FIGURE 2-3   Rhodotorulic acid.  Reproduced from Neilands.
                0  OH

          CH3  C-N
HOCH2
                                                                      CH20R
                                                        571
           FIGURE 2-4  Dimerumic acid.  Reproduced from Neilands.
                                   74

-------
CH3
                                                     571
 FIGURE 2-5   Ferric coprogen.   Reproduced from Neilands.
                             75

-------
            0             R
            u             I
                          CH
             HN ^           (CH2)n
   CH2
    \
HO—C —C02H
                                             N— C— CH3
    \
                                 (CH2)n
           C
           8
FIGURE 2-6   The basic structure of the citrate-hydroxamate family.
            For the three known members, schizokinen, aerobactin
            and arthrobactin, R = H or COOH and n =  2 or 4.  Simi-
            lar structural perturbations at R and n  would afford
            three additional hypothetical members: awaitins A, B,
            and C.  Reproduced from Neilands.  •*-


                              76

-------
FIGURE 2-7   The basic structure of the mycobactin family.  Depending
             on the source organism, different substituents occur at
             R! - R^.  The ligand has six centers of potential optical
             activity (a-f).  Reproduced from Neilands.
                                77

-------
         NH2         HO  0
         I               I   II
HO-f-C-C-(CH2)3-N-C       CH2-CH2-0--H
            A                  >=<
                              H       CH3
FIGURE 2-8  The fusarinine family.  In the structure shown, n may mean
         a 1-, 2- or 3-linear arrangement, or a 3-cyclic arrangement.
         A corresponding series is acetylated on the amino group.
         Reproduced from Neilands. '
                        78

-------
     The ferrioxamines, represented in Figure 2-9, are linear or cyclic

molecules containing repeating units of l-amino-u>-N-hydroxyaminoalkane and

succinic or acetic acids.  The antibiotic ferrimycin is a further elaboration

of ferrioxamine B.

     Citrate can be considered as a primitive siderophore in that it tends
                                                                703
to form polynuclear complexes at ligand/ferric ion ratios <20:1.

     The most noteworthy feature of the siderophores is that iron regulates
                   292
their biosynthesis.     No matter what the level of available iron is, it is

likely that there can be a constitutive level of siderophore synthesis.  How-

ever, at less than about 1 >ug-atom iron/1, the substances are leaked to the

medium and, in the case of rhodotorulic acid, the siderophore of the red yeast
                                                        OQ
Rhodotorula pilimanae, production reaches almost 10 g/1.    In addition to the

siderophore itself, the appropriately specific membrane receptor proteins are

also overproduced by the culture when environmental iron concentrations are low.

Such activity may not be true for the ferrichrome receptor in enteric bacteria;

such organisms synthesize only enterobactin and not ferrichrome.527

     Biochemical genetic experiments with Salmonella typhimurium prove conclu-

sively that the biologic functions of the siderophores are to solubilize and
               619
transport iron.     Thus/mutants that are unable to biosynthesize enterobactin

can only be grown in citrate medium after addition of excess iron or an organic

agent that can decompose ferric citrate,  a complex which Salmonella cannot trans-
                                                                                575
port.  Several mechanisms exist to deliver siderophore iron in enteric bacteria.

It may be donated to the cell surface or, alternatively, the intact chelate may

penetrate the organism.  In both instances the iron is liberated by reduction,

regardless of whether the ligand is hydrolyzed.  The requirement for reduction

may obviate transport of nonnutritious metal ions such as Al(HI) •  The presence

of the ferric complex in the cytosol would provide the cell with a specific

small molecule that could act as an effector to signal the biologic compatibility

of the external environment.  Various ferrichromes and coprogen can act as
                                         375
germination factors in Neurospora crassa.

                                    79

-------
H-N'           CONH            CONH
    \          /    \          /    \       I
    (CH2)5  (CH2)2  (CH2)5  (CH2)2 (CH2)n  (CH2)2


        N-C             N-C             N-C
            ii
                     ii
      -CKC
                                              ii
                                -.0   0
        R
        I
     H-N
             CONH            CONH
\           /    \           /     \
 (CH2)5  (CH2)2   (CH2)5   (CH2)2  (CH2)n R|

   \     /          \     /         \   I
    N-C            N-C             N-C
     i   ii
  — 0^ 0
       -0
II
0
-0
                                                   II
                                                  .0
FIGURE 2-9  Cyclic and linear forms of the ferrioxamine family.
           In the ^.inear structure, various substituents occur at
           R and
       >  lines
       R1; n
may be 4 or 5.  Reproduced from Neilands.-
                           80

-------
                                              254
     Working with Ustilago sphaerogena, . Emery    proved that ferrichrome

could shuttle iron into the microorganism.  The Cr(III) complex is also incorporated

by this organism, showing that the A-cis configuration is one of the actively
                    470
transported species.

     Table 2-2 summarizes the known distribution of siderophores among species;

the list is growing rapidly.

     Phage-bacteriocin interaction.  Siderophores have been shown to protect

sensitive cells of enteric bacteria against certain phages and bacteriocins
                                             797
("protein antibiotics" or "killer proteins").     Thus ferrichrome, although

it is not made by Escherichia coli, protects that organism from phages T^, Tr,

and 80, from colicin M, and from albomycin, a structurally related antibiotic,

by a mechanism known to be one of adsorption competition for a common outer
                  492
membrane receptor.     Enterobactin protects the organism against colicin B
                                                                   796
by the same mechanism of competition for the enterobactin receptor.     In

Salmonella typhimurium, phage ES  18 and ferrichrome similarly compete for a sur-

face receptor.  These protections are quite specific because the various agents

are binding to a similar or closely related locus in the receptor complex.

     Siderophores, synthetic ferric chelates, and inorganic iron compounds
                                                                796
generally protect Escherichia coli against colicins B, V, and la.     This

process is not totally understood, except that it is not one of adsorption

competition for the outer membrane receptor.  The mechanism may amount to

repression of the surface receptor by iron, or the metal may somehow interfere

with how the colicin works.  However,  siderophores provide extremely potent pro-

tection, which is likely to be of biologic importance.

     Guinea pigs kept on a synthetic diet high in iron fare poorly.   The

gastrointestinal tract becomes distended and it represents a large fraction of

the carcass weight (G. Briggs, personal communication).  The addition of as little
                                                        424
as 5 yQ/mi enterobactin to the diet enhances weight gain    and restores the


                                    81

-------
                              TABLE 2-2
                    Known Sources of Siderophores
                                                 a
  Compound

  A.  Phenolates
      Enterobactin (enterochelin);
        Cyclo-tri-2,3-dihydroxy-N-
        benzoyl-L-serine

      2-N,6-N-di-(2,3-dihydroxy-
        benzoyl)-L-lysine

      2,3-Dihydroxy-N-benzoyl-L-
        serine
      2,3-Dihydroxybenzoylglycine
        (itoic acid)

      2,3-Dihydroxybenzoylthreonine

      N1N8-bi£- (2,3-dihydroxybenzoyl)
        spermldine

      2-Hydroxybenzoyl-N-L-threonyl-
       N4- [N^-bi^- (2,3-dihydroxy-
       benzoyl) ] spermidine

      Tri-2,3-dihydroxybenzoyl serine,
       methyl ester

  B.   Hydroxamates
      1.
Ferrichrome family
Ferrichrome
          Ferrichrome  A

          Ferrichrome  C

          Ferrichrysin


          Ferricrocin


          Ferrirubin
                                Sources
                                Aerobacter aerogenes
                                Escherichia coli
                                Salmonella typhimurium
                                Azotobacter vinelandii

                                Aerobacter aerogenes
                                Escherichia coli
                                Salmonella typhimurium
                                Bacillus subtilis
                                Klebsiella oxytoca
                                Micrococcus denitrificans
                                Micrococcus denitrificans
                                Escherichia coli
Ascomycetes
Basidiomycetes
Fungi imperfecti
  including Aspergillus,
  Penicillium, and probably
  Verticillium

Ustilago maydis and sphaerogena

Cryptococcus melibiosum

Aspergillus melleus, terreus,
  and oryzae

Aspergillus fumigatus, humicola,
  nidulans, and versicolor

Paecilomyces varioti, Penicillium
  variable, and Spicaria
 ^Compiled from Diekmann,22^ Murphy et^ al. ,559  and Neilands.572"575   The organisnl
listed are typical sources of  specific siderophores;  however,  they may  be synthe-j
sized by other species.   Thus  enterobactin (MB)  is  generally  produced  by the
enteric bacteria.

-------
                                                    TABLE 2-2 (Continued)
Compound
Sources
        Ferrlrhodin
        Albomycins
          Form 69
          Form e
          Form <$i
          Grisein
        Ferribactin

        Sake colorant A

        Verticillins

        Rhodotorulic acid family
        Rhodotorulic acid
Aspergillus nidulans and versicolor

Actinomyces griseus
Streptomyces subtropicus
Streptomyces griseus

Pseudomonas fluorescens

Aspergillus oryzae

Verticillium dahliae
Leucosporidium scottii  (allotype,
  mating type a)
Rhodosporidium toruloides  (mating
  type)
Rhodotorula glutinis
  var, dairenensis (type)
  var. glutinis graminis (type)
  rubra (type)
  rubra (type)^
Sporidiobolus
  johnsonii (type)
  ruinenii (type)
Sporobolomyces
  albo-rubescens  (type),
  hispanicus, pararoseus (type),
  and roseus
        Dimerumic acid (dimerum acid)    Fusarium dimerum
        Coprogen (compound XFe)
          Deacetyl coprogen
    3.   Citrate-hydroxamate family
        Schizokinen
        Terregens factor (arthro-
          bactin)

        Aerobactin

    4.   Mycobactins

    5.   Fusarinines (fusigens)
Neurospora crassa, Penicillium
  camemberti, chrysogenum, notatum,
  and urticae
Bacillus megaterium
Anabaena (blue-green alga) sp.
Arthrobacter pascens

Aerobacter aerogenes 62-1

Mycobacteria

Fusarium roseum; other species of
  Fusaria, Aspergillus, Gibberella
  and Penicillium
 Originally known as Rhodotorula mucilaginosa.
                                 83

-------
                                               TABLE  2-2 (Continued)
Compound
Source
    6.  Ferrioxamines
        Linear ferrioxamines:
          Ferrioxamine B
          Ferrioxamine DI

          Ferrioxamine G

          Ferrioxamine A..

          Ferrimycin A

        Cyclic ferrioxamines
          Ferrioxamine E (ferric
            nocardamine)
          Ferrioxamine D£

        Miscellaneous ferrioxamines
          Metabolite C

          Ferrioxamine Ao

C.  Other hydroxamates
          Aspergillic acids

            Aspergillie, neoasper-
            gillic, muta-aspergillic,
            hydroxyaspergillic, and
            neohydroxyaspergillic
            acids

          Mycelianamide

          Pulcherriminic acid (2,5-
            diisobutyl-3,6-dihydroxy-
            pyrazine-1,4-dioxide)
          Hadacidin
          Actinonin
Micromonospora
Nocardia
Streptomyces pilosus and other
  Streptomyces species
Nocardia
Streptomyces pilosus
Streptosporangium roseum
Chainia
Chromobacterium violaceum
Aspergillus flavus
Aspergillus sclerotiorum

Penicillium griseofulvum

Candida pulcherrima and related
  yeasts
Fabospora ashbyi, dobzhanskii,
  and lactis

Pencillium aurantio-violaceum
  F4070b and frequentans

Streptomyces sp.
                                84

-------
                                                 TABLE 2-2 (Continued)
Compound                                 Sources

          2,4-Dihydroxy-7-methoxyl-      Certain tissues of higher plants,
            l,4-bezoxazin-3-one            such as corn seedlings and
            (DIMBOA)                       seedlings of higher grasses;
                                           lettuce; leaves of tomato, cauli-
                                           flower, leek and cabbage; and
                                           extracts of carrot roots.
                                         Mycelium of Streptomyces sp.

          Viridomycin A                  Streptomyces viridaris

          Thioformin (fluopsin)          Pseudomonas fluorescens

    D.  Ferrous ion-binding compounds
        Pyrimine                         Pseudomonas GH

        Ferroverdin                      Streptomyces sp.
                                 85

-------
 gastrointestinal tract to its normal fraction of the carcass weight.  It is

 highly likely that the enterobactin alters the balance of the microbial flora

 of the gut or the metabolism of the organisms therein.  Enterobactin has been

 shown to display pacifarin activity (nutritional immunity) in mice challenged
                                                795
 with virulent strains of Salmonella typhimurium.

     The siderophore-phage-colicin interaction may determine aspects of

 how sewage systems work, but basic research on this problem has yet to be

 done.

     Transport of actinide elements.  It is estimated that 90 metric tons of

 Plutonium exist in the world, manufactured in the past quarter century by fission

 reactors.  Siderophores represent one of the few natural ligand systems that

 could coordinate plutonium effectively and initiate its insertion into the food

      132a
 chain.       Plants have been shown to concentrate plutonium from the soil.

 Animals kept on a low-iron diet have been found to take up plutonium more

 efficiently; once in the  bloodstream,  the metal is carried by transferrin.

 It may be excreted as the citrate chelate.  It is a possibility that

 this man-made element can be moved by natural ligand transport systems
 . ^   , . ,   .    132a
 intended for iron.

     Distribution in nature.  Assays with Salmonella typhimurium enb-7 have

 found substances with the biologic properties of siderophores in mold-ripened

 cheese,  such as the camembert and blue varieties (S. Ong, personal communication),

 and in soil and dung.     The presence of these substances in excrement
                                                                     357
accounts for the natural growth of the coprophylic fungus, Pilobolus.     Several

members  of the ferrichrome family of siderophores have been detected in the

Japanese fermented beverage, sake, which is prepared with the aid of Aspergillus

oryzae.   The specific siderophore schizokinen occurs in the blue-green microalgal
                693
species  Anabaena    and the capacity to synthesize hydroxamate chelates may enable
                                      86

-------
such organisms to monopolize the limited iron supply in lake waters and so


                              559
to dominate other algal forms.



     Pharmacology.  Aromatic hydroxamic acids are potent carcinogens,  but



this property has not been ascribed to the aliphatic members of the series.



If the hydroxamic acid bond were hydrolyzed, an alkyl-substituted hydroxylamine



would result.  Hydroxylamines are well known to be mutagenic and therefore



may be carcinogenic.     Little is known about the pharmacology of this



class of microbial products, although rhodotorulic acid has been touted as


                                                                          306
a possible drug for treating persons suffering from chronic iron overload.



	,   the mesylate salt of deferriferrioxamine B, is the drug of choice


                                                                           25
for treating acute episodes of accidental iron poisoning in small children.



Neither the hydroxamate ligands nor their iron complexes appear to be very



well absorbed by humans.  Hence, the substances must be injected to treat



chronic iron overload.  To control acute iron poisoning, the agents need to



be administered orally as well as injected.





TOXICITY



Iron- and Sulfur-Oxidizing Bacteria



     On the prebiotic earth, the elements were in their reduced states.  They



still retain this form in those segments of the lithosphere not contacted by


                                           94
the oxygen gas derived from photosynthesis.    Exposure of inorganic sulfur



compounds such as iron pyrite to oxygen will result in the oxidation of the



sulfur and the generation of acid.  The latter, in turn, will stabilize the



ferrous ion and
                                 87

-------
maintain it in solution as a substrate for the "iron bacteria."  These reactions


lead to acid mine water and such low pH values that few forms of life other than


the Thiobacilli survive.  The reactions bring about the precipitation of iron,


as exemplified by the gelatinous, golden-yellow ferric oxyhydroxide .


     Acid mine water is the product of a complex series of transformations of


iron pyrite and related sulfide minerals.  Some of these transformations are
                                                        336,495,839
chemical and some are catalyzed by specialized microbes .             Pyrite and
marcasite (both FeS2) are oxidized in air to a solution of ferrous sulfate and


sulfuric acid:


                 FeS2 + H20 +  3-1/2 0^          =^ FeS04 + H^


The autotrophic bacterium, Thiobacillus f errooxidans , utilizes the ferrous ion


as an energy substrate:

                    +2              j. T. f errooxidans      . o
                 ?e   +  1/4  02 -I- H+ , - '        _.._ N Fe+3  + 1/2 HO


Pyrite reduces the ferric ion, precipitating elemental sulfur (S°) :


                 2Fe+3 + FeS2                     ___ i   3Fe+2 + 2S°


The elemental sulfur may be oxidized by the ferric ion or by the sulfur -oxidizing


autotroph, Thiobacillus thiooxidans ;


                 2S° + 12Fe+3 + 8H,0                   12Fe+2 + 2S07 + 16H+
                                  z.  •                              i±
                  o                   T.  thiooxidans
                 S  + 1-1/2 02 + H20           - --- 1- 2H+ + S04


     Although the rate-limiting step in this series needs to be defined, the


process can be retarded by covering the exposed minerals  or eliminating standing


water.


Iron and Infection


     Because pathogenic microbes require iron and because available iron in host


tissues may be limiting,  organisms with a well developed  system for acquiring


this metal may have special virulence.   This subject has  been reviewed by

         799
Weinberg,    who believes that the capacity to synthesize siderophores is a prime


determinant of virulence.  This thesis is supported by his observations that:


elevated temperatures in the host (fever) inhibit synthesis of specific


                                    88

-------
                                291
siderophores (e.g., enterobactin)}  '   susceptibility to infections is enhanced

in hyperferremic animals and in hypotransferremic individuals; and devious

mechanisms are employed by the host to deny iron to invading pathogens.  Among

the infections claimed to be dependent on iron supply are gram-negative septi-

cemia and meningitis, malaria, coliform pyelonephritis, gas gangrene, listeriosis
                                                605
and systemic candidiasis.  Payne and Finkelstein    have demonstrated a role of

iron in the virulence of Neisseria gonorrhoeae.

     As noted, factors imparting nutritional immunity have been termed pacifarins,

one of which, in the case of salmonellosis in the mouse, has been characterized
                795
as enterobactin.
                                                                          132
     In surveying the role of iron in infection and immunity^ Bullen et al.

concluded that, taken as a whole, there was "ample evidence" that iron is of

crucial importance "in the mechanism of resistance to a variety of bacterial
                                       342
infections."  Yet Hegenauer and Saltman    have argued that in many instances

the observed growth-promoting effect of highly-saturated iron transferrin may

be attributed to release of metal ion bound adventitiously to nonspecific sites
                        722
on the protein.  Sussman    reviewed the relevant data and concluded that

despite the evidence from experimental animals, hyperferremia does not contribute
                                                    534
to the course of human infections.  Miles and Khimji    found no correlation

between pathogenicity and capacity of enteric bacteria to synthesize chelates.
                 836
     Yancy et al.    showed that a strain of Salmonella typhimurium rendered

defective in the biosynthesis of enterobactin displayed greatly reduced viru-

lence in the mouse.  However, the debate still continues.  Ultimately, the

concept of "optimum" iron nutrition,  as noted by Chandra,1-'3a may prove to be

correct.

Bacterial Toxins

     Although the formation of a number of bacterial exotoxins is known to be

regulated by iron, diphtheria toxin is of special interest in view of the
                                                                559
advances in knowledge of its biochemistry and molecular biology.      The toxin
                                    89

-------
acts catalytically to effect the adenosinediphosphorylribosylation of the

EF-2 protein required for polypeptide chain elongation in the translocation

process.  Toxin production is programmed by a 3-DNA phage that infects

Corynebacterium diphtherias.  In the presence of bacterial extracts, iron

binds the 3-phage of DNA to nitrocellulose filters, suggested that iron is a
                            557
corepressor of the tox gene.

     Mechanisms triggering dinoflagellate blooms, those "red tides" that cause

mass mortality of fish and other aquatic life, are as yet incompletely under-
                                                                            422
stood.  One factor triggering their activity may be the availability of iron

(cf. Chapter I).


MICROBIAL CORROSION
                                                                         389
     Microbial corrosion of iron has been reviewed thoroughly by Iverson.

Economic losses attributed to corrosion of iron and steel in the United States

has been estimated to be from $500 million-$2 billion annually.  Corrosion,

defined as "the destructive attack of a metal by chemical or electrochemical

reactions within its environment," includes rusting, tarnishing, patina forma-

tion, pitting, selective leaching, stress cracking, and intergranular corrosion.

Both fungi and bacteria are involved in corrosion processes.  The most active

bacteria are Thiobacillus, Desulfovibrio» and Desulfotomaculum, all of which

transform sulfur compounds.  By virtue of the elaboration of oxygen gas, algae

are also involved.  The chemical agents involved in biologic corrosion are

varied, and include inorganic and organic acids, hydrogen sulfide, elemental

sulfur, mercaptans, and other substances.  Microorganisms growing on the surface

or vicinity of metals may produce substances that establish electrochemical

cells leading to corrosion.  Biologic corrosion can be combated by altering

the environment in such ways as using microbial inhibitors, protective coatings,

or cathodic protection.
                                    90

-------
                                  CHAPTER 3

                               IRON AND PLANTS


     Iron is essential to the growth processes of all plants.  If a plant is

green, it usually has adequate iron.  Green plants require a continuous supply
                                                                            108,116
of iron as they grow, because iron does not move from older to newer leaves.
                                                                          114,116
Iron absorption and transport are genetically controlled by the rootstock.

For example, when the iron-efficient Hawkeye soybean top (Glycine max L., Merr.)

is grafted to the iron-inefficient PI-54619-5-1 (T203) soybean rootstock, iron

chlorosis, shown in Figure 3-1, develops in the youngest leaf because the root-

stock cannot make iron available from the soil.
                                           434
     Soils do not usually lack iron per se,    but in most calcareous soils
                                                                 592
availability of ferric ions may be insufficient for plant growth.     For each

pH unit increase above 4.0, the solubility of Fe(III) decreases by a factor of
      448,782
1,000.         When the amount of iron available to plants does not meet their

minimum needs, plants develop chlorosis, a mineral deficiency disease that will

manifest itself in the yellowing or blanching of normally green parts, such as

leaf tops, and they may die.

     Chlorosis is more common in plants grown in alkaline soils.  However, some

plants grow well on alkaline soils because they are endowed with a biochemical

mechanism that makes iron available to them from the soil, that is, they are

iron-efficient.  The plant and the soil must be compatible for maximum efficiency

and the factors involved in such a relationship are discussed in this chapter for

representative species.

     In the United States, iron deficiency is most likely to occur west of 100°
                                                          487
longitude (roughly the western half of the country).  Lock    found approximately

258 species or varieties of plants in western states exhibiting naturally occurring
                                    741
iron chlorosis.  Thorne and Peterson    estimated that 55% of the world land area
                                                                782
receives fewer than 51 cm of rain annually, and Wallace and Lunt    indicated

that 25-30% of the world's land surface is calcareous.  Hence,iron deficiency in

plants is most likely to be found in these areas and is considered a world problem.
                                  91

-------
FIGURE 3-1   Rootstocks of T203 and Hawkeye soybeans affect the
             development of iron chlorosis and absorption and
             translocation of radioiron from Quinlan soil (pH 7.5).
             Top (left to right) photograph and autoradiograph are
             of T203 top on Hawkeye rootstock; bottom photograph
             (left to right) shows Hawkeye top on T203 rootstock.
             The new leaves of T203 are green on the Hawkeye root-
             stock, whereas the new leaf of the Hawkeye soybean is
             chlorotic on T203 rootstock.  No radioiron moved from
             the old leaf to the new one (see also Figure 3-3).
             Reproduced from Brown et a
                               92

-------
IRON SUPPLY

Inorganic and Stored Iron

     The most common sources of iron for plants are the seed itself, the growth

medium, and sprays.  Germinating seeds usually contain sufficient iron to meet
                                   108             383
nutrient requirements of seedlings.     Hyde et_ al.    suggested that phyto-

ferritin in cotyledons of pea was the form of iron stored by young seedlings.

Chemical factors such as high pH and phosphate that interfere with uptake of iron

from the growth medium do not interfere with the plant's use of iron from cotyle-
     19
dons.    Most iron added as a spray remains in the plant tissue where it is

applied (see Figure 3-2), making it necessary for iron to be applied continually

to the new growth.  Iron nutrition in plants depends largely on chemical factors

in roots, e.g., release of hydrogen ions and reduction of ferric iron, that

affect absorption and translocation of iron from the growth medium.  It is not

uncommon to see two varieties of the same species growing in the same alkaline

soil, with one variety iron-deficient (chlorotic) and the other iron-sufficient

(green), as shown in Figure 3-3.  The soybean cultivars of Figure 3-3 differ in
                                        112
their ability to respond to iron stress.     The chemical reactions that release

hydrogen ions and reduce ferric iron are induced by iron stress within the plant

where the roots and soil meet and make iron available to the plant.  Plants are

classed iron-efficient if they respond to iron stress and iron-inefficient if

they do not.  Iron-inefficient plants become chlorotic and will die on many

alkaline soils unless iron is made soluble by chelation, as illustrated in Figure

3-4.


Chelated Iron

     Iron chlorosis can be corrected by adding the appropriate iron chelate to
                              153,341,373,405,469,783
the soil or nutrient solution.                         The primary role of the

chelating agent is to make the iron water-soluble and more accessible to the
                                     93

-------
FIGURE 3-2   When the chelating agent FeDTPA was sprayed on chlorotic
             leaves, it corrected the chlorosis only in spots where
             the residue accumulated (right specimen).  Note that the
             new leaf (left specimen) that developed after the plants
             were sprayed is completely chlorotic.  Reproduced from
             Holmes and Brown.
                               94

-------
FIGURE 3-3   Plant species (bottom photograph) and varieties within
             species (top photograph) differ in their susceptibility
             to iron deficiency.  Top (left to right) T203 (iron-
             inefficient) and Hawkeye (iron-efficient) soybeans grown
             on Quinlan soil (pH 7.5).  The iron-inefficient plants
             serve as indicators. Photographs courtesy of J. C. Brown.
                                  95

-------
FIGURE 3-4   Iron-inefficient T203 soybeans grown on 16 calcareous
             soils (left 4 plants) responded to 150 kg FeDTPA/ha
             added to their soils (right 4 plants). Photograph
             courtesy of J. C. Brown.
                                 96

-------
                                                               747
plant root, because roots extract iron from synthetic chelates,     and ferric

chelates must be reduced to ferrous ones before the ferrous ion can be absorbed
              154                                                            746
by the plants.     For a more detailed discussion of the subject, see Tiffin.

     Plant species differ in ability to absorb iron from iron chelates, and
                                                            124
chelating agents themselves can compete with roots for iron.     The ability of

EDDHA  [ethylene diamine di  (oj-hydroxyphenylacetic acid)] to chelate iron was

determined in nutrient solutions containing ethylenediaminetetraacetic acid

(EDTA), diethylenetriaminepentaacetic acid (DTPA), and cyclohexanediaminetetra-
                                                  124
acetic acid (CDTA), as competing chelating agents.     The apparent stability
                                                                 82
constants for FeEDTA, FeCDTA, FeDTPA, and FeEDDHA are 24.8, 29.3,   27.9, and
                283
greater than 30,    respectively.  When chelating agents and iron are in equal

molar concentration, EDDHA competes successfully with EDTA, DTPA, and CDTA for
     124
iron.     However, when the concentrations of chelating agents increase, they
                                                                       124
compete with EDDHA for iron in the following order: EDTA < DTPA < CDTA,    a

relationship diagrammed in Figure 3-5.

     Plant roots differ, as do chelating agents, in their ability to compete for
     124,125
iron.         Wheatland milo (Sorghum bicolor L., Moench) was unable to use iron

from FeEDDHA unless the iron concentration greatly exceeded the EDDHA concentra-

tion.  For example, at 2 x 10  M iron and 0.16 x 10  M EDDHA, the milo took up

120 yg iron.  At the same iron concentration (2 x 10~5M), but with the EDDHA
                                _5                                       108
concentration elevated to 1 x 10  M, the sorghum took up only 13 yg iron.

As can be seen in Figure 3-6, okra (Hibiscus esculentus L.) and wheat (Triticum

aestivum L.) developed iron-deficiency chlorosis when the molar concentration of
                                            108
DTPA exceed the molar concentration of iron.     Plant species may be able to

alter the activity of a metal ion by increasing or decreasing the concentration

of a specific chelating agent inside the plant or in the root exudate.   In this

way, the type and concentration of a chelating agent,  coupled with iron availa-

bility, determine the uptake of iron and other nutrient elements by plants.


                                      97

-------

I
Q
UJ

u.
,f
g




2.0


1.5



1.0


0.5

O O
A = CHELATING AGENT *
B =
• - 	 •- 	 r .
1==:==:^^^.
^"^^!\
V
\\\ EDTA
\\ 	 '
\\
\ \ DTPA
\^-
\
\ CDTA
A=0 .16 .5 1 2 4 6
6=4444444
C = 2 2 2 2 2 2 2
Fe
EDDHA




^-\ 	

N


M
12
4
2
x I0'5 M
x ID'5 M
xlO'5 M




•
• — -__^

-
— • — ^^

18 36
4 4
2 2
FIGURE 3-5   The capacity of EDDHA to form FeEDDHA was affected by
             the concentration of the competing chelating agents
             in solution at 4 x 10~5M iron.  No competition for
             iron existed until the molar concentration of chelating
             agents exceeded the molar concentration of iron.  Repro-
             duced from Brown et a
                                 98

-------
FIGURE 3-6   When the molar concentration of DTPA exceed the molar
             concentration of iron, DTPA competed with the roots of
             okra and wheat (top and bottom photographs, respectively)
             for iron.  The nutrient solution contained 1 x 10  M iron
             and (both genotypes, left to right) 0.16, 1, 2,_6, and
             18 x 10~-*M DTPA.  Reproduced from Brown et^ al
                                99

-------
Both the growth medium and the plant variety itself contribute to making a

continuous supply of iron available.


CONDITIONS AFFECTING IRON UPTAKE AND TRANSPORT

Soil and Environment

     The hydrogen ion concentration of a soil is a crucial factor governing

the distribution of plants in nature.  Some plants grow well only on acid soils;

others grow better on alkaline soils.  For example, bentgrass (Deschampsia

flexuosa) develops iron-deficiency symptoms in a growth medium with a pH above
  594
6,    whereas mustard (Sinapis alba L.) grows well under similar conditions.
                     782
     Wallace and Lunt    listed the following as principal causes of iron chloro-

sis: poor iron supply; excessive calcium carbonate; bicarbonate in soil or irri-

gation water; overly irrigated or high-water conditions; high phosphate; high

concentrations of heavy metals such as manganese, copper, and zinc; low or high

temperatures; high light intensities; high concentrations of nitrate nitrogen;

unbalanced cation ratios; poor aeration; certain organic additions to the soil;

viruses; and root damage by nematodes or other organisms.  All these factors are

more effective in a natural alkaline soil, and several may be operating at the

same time. • For example, the role of bicarbonate in precipitating iron depends

in part, on the interrelationship between bicarbonate, phosphate, calcium and

ferric ions in the soil.  A moist calcareous soil containing decomposing

organic matter provides a condition for maximum bicarbonate-ion accumulation that
                                                                              122
may increase phosphate availability and decrease the iron available to plants.

Bicarbonate per se does not appear to be a direct cause of iron-deficiency
          107
chlorosis.     Instances of iron chlorosis have been associated more with phosphate
                                                               107
concentration in solution than bicarbonate anion concentration.     Phosphate is
                                                          514           312
considered a ligand that competes with the plant for iron.     Greenwald    and
            596
Olsen et al.    observed that bicarbonate increased the solubility of phosphorus


                                    100

-------
in solution.  Phosphate strongly influences iron absorption and translocation
          72                                              108,553,742,782
in plants.    This phenomenon has been well characterized.
                    112
More recent research    has indicated that use of iron by plants is dependent

on the plant species or variety grown.  Plant nutrition is entering an era in

which equal emphasis is rightly being placed on the plant as well as on the

soil for improving the efficiency of crop production.

Plant Response to Iron Stress

     The term "iron stress" implies that a plant is deficient in iron.  Plants

are classed as iron-efficient if they respond to iron stress by inducing bio-

chemical reactions that make iron available in a useful form, and iron-inefficient

if they do not.  An iron-efficient plant may respond to iron stress without any

visual iron-deficiency symptoms, whereas an iron-inefficient plant develops

chlorosis.  When plants respond to iron stress, the following products or bio-

chemical reactions are more likely to occur in iron-efficient than in iron-

inefficient plants:

     •  release of hydrogen ions from the roots;

     •  release of reducing agents from the roots;

     •  reduction of ferric iron at the roots; and

     •  increases in organic acids (particularly citrate) in roots.

Response to iron stress is adaptive and is known to be determined genetically
                         54,786,805
in several plant species.
                                                           425
     Hydrogen ions released from roots.  Kirkby and Mengel,    working with

tomatoes (Lycopersicon esculentum Mill.), reported an elevated pH in nutrient

solution with nitrate, and a reduced pH with ammonium nutrition.  With nitrate

as the source of nitrogen, the pH increased to pH 7.2 for T3238fer (iron-

inefficient) but decreased to pH 4.3 for T3238FER (iron-efficient) tomatoes
                              120
when subjected to iron stress.     When supplemental iron was added to the
                                    101

-------
iron-sufficient T3238FER tomatoes, hydrogen ions were not released from their

roots and the pH increased to 7.2, the same as for T3238fer.  During an 8-h

absorption period, iron-stressed T3238FER plants absorbed and translocated

approximately 20 times more iron-59 to their tops than iron-sufficient T3238FER,
                                                               120
and 78 times more iron-59 than iron-stressed T3238fer tomatoes.     The iron-

inefficient T3238fer did not respond to iron stress under any of the experimental

conditions.

     When hydrogen ions are released from roots and the pH is lowered to make

the environment of the root more acidic, iron is made more available for plant
       594
uptake.     Although beneficial, this change is not always the ultimate solution

to iron chlorosis.  For example, iron-stressed Wheatland and "B-line" (iron-

inefficient) sorghum and Pioneer 846 and KS5 (iron-efficient) sorghum release

about the same quantity of hydrogen ions into the growth medium, but they differ
                                          118,121
in their absorption and transport of iron.         This difference is associated

                                                       121
with greater reduction of Fe(IH) to Fe(II) at the root    by the iron-efficient

than the iron-inefficient sorghum lines.  Each factor has its own effect on

how plants use iron and may differ with variety or plant species and their growth

environment.

     Reducing agents released from roots.  In addition to releasing hydrogen
                             117             114
ions, iron-efficient soybeans    and tomatoes    release "reductants" from

their roots in response to iron stress.  Reductants is a word coined to desig-

nate compounds released by roots that reduce Fe(III) to Fe(II).  Chelating agents

interfere with the activities of reducing agents, but such interference may be
                                                                        111
eliminated by increasing the concentration of the reductant in solution.

     Iron-stressed Hawkeye (iron-efficient) soybeans released more reductant

into solution than the iron-stressed T203 (iron-inefficient) variety, but
                                    102

-------
iron uptake was not increased when the T203 plants were placed in the Hawkeye
          111
solutions.     This may mean that reductants in the external solution indicate

a leaky root resulting from the release of hydrogen ions into the nutrient
                                                                            121
solution, although reductants have been found in nutrient solutions at pH 7.

More important may be the adaptive production of reductants inside the root or
                                                                       20,110
at the root surface that keeps iron in the more available ferrous form.

     Over the past 20 yr, reducing agents associated with iron transport have

been identified in microorganisms.  The excretion of metal-binding phenolic

acids by iron-stressed Bacillus subtilis might be a mechanism to correct iron
           388
deficiency.     Phenolic acids accumulated by iron-stressed Bacillus subtilis

do not seem to be involved in iron uptake, but serve to solubilize the metal in
                  613
the growth medium.     The addition of iron to growth cultures will inhibit
                                             136
phenolic acid excretion by Bacillus subtilis.     Although not identified,
                                                  20           110
the reductants released by iron-efficient soybeans   and tomato    respond

similarly to the phenolic acids released by Bacillus subtilis.

     Ferric iron reduced at root.  Reduction of Fe(III) to Fe(II) is another factor

induced by iron stress and it occurs principally in the young lateral roots of
                       20,111            110
iron-efficient soybeans       and tomato.     Sites of reduction were determined

by transferring the iron-stressed plants to nutrient solutions containing FeHEDTA
                                      110
and potassium ferricyanide, K~Fe(CN),.     A blue precipitate (referred to here

as Prussian blue) formed in the epidermal areas of the root where one form of

the ferric ions was reduced by the root, shown in Figure 3-7.  Reduction of

Fe(III) occurred in areas that were outside the root and accessible to bathophenan-
                           110
throlinedisulfonate (BPDS).     Reducing conditions were established by adding

BPDS, 10% in excess of Fe(HI), to the nutrient solutions.  As the ferric ion was

reduced, Fe(II) was trapped in solution as ferrous BPDSo.  Hence most of the iron

was not transported to the plant top.   '


                                    103

-------
FIGURE 3-7   Left photograph shows sites of reduction of Fe(III) to Fe(II)
             (dark areas are Prussian blue formations) on lateral
             roots, and on the area of elongation and maturation of
             the primary root of T3238FER (iron-efficient) tomatoes.
             No reduction can be found on the T3238fer roots.  The
             right photograph is a cross section of a lateral root of
             T3238FER.  Prussian blue crystals have formed on the per-
             iphery of the root.  The three dark spots inside the root
             reveal contamination occurring while the specimen was being
             photographed.  Reproduced from Brown and Ambler.
                                104

-------
      If  the  iron-efficient soybean roots were  given  iron  as  FeHEDTA for  20  h

 and  then rinsed  free of FeHEDTA and placed in  nutrient  solutions  containing potas-

 sium ferricyanide, Prussian blue  (indicative of  ferrous iron)  appeared throughout

 the  protoxylem of the young lateral roots, as  can be discerned in Figure 3-8,

 and  throughout the regions where  the root elongates  and matures in the primary
      20,109
 root,        (see also Figure 3-7, left).  According  to  this  test,  the metaxylem

 of the iron-efficient plants contained no ferrous ions.

       Elevation  of organic acids  (particularly citrate) in roots.  Iron-deficient
                                                                              222
  plants  usually  contain more  citric  and  malic  acids  than  normal green plants.    '
  TO c O Q £ fj. £
     '   '     Citric  acids chelate  iron  and  keep it  soluble  in an external  solu-
                                                        /: co
  tion,  and they  may  function  similarly inside  the plant.      A striking  relation-

  ship was observed to exist between  iron and citrate transported  in the  xylem
          113 123
  exudate,   '    which is plotted in Figure  3-9. When iron  increased, the  citrate

  increased;  a  decrease in iron was paralleled  by a decrease  in citrate.   This

  relationship  held if iron stress were induced in the  plant  by limiting  the iron
                                        113  123
  supply  or if  zinc,  azides, or arsenate    '     were  used  to  induce iron  stress.
         743-745
  Tiffin,        using electrophoresis  to follow the  migration of  the chelated

  metal,  identified ferric citrate in the xylem exudate of several plant  species.

  Enough  citrate  was  always present to  chelate  the metal,  and citrate in  excess  of

  that needed for iron chelation migrated as  an iron-free  fraction behind the  iron-

  citrate band.

                 169
     Clark et^ al.    showed  that  malic, acetic  and trans-aconitic acids were

ineffective in moving iron-59  electrophoretically in acetate, citrate, isocitrate,

trans-aconitate,  and malate buffers.   Citric acid moved iron to the annode whenever

present on the electrophoretogram, and competed successfully with the other acids
                       169
for iron.  Clark e_t al.    also determined that iron-efficient WF9 corn  (Zea mays

L.) absorbed and transported  more iron than iron-inefficient ys  corn, but the


                                     105

-------
FIGURE 3-8   Protoxylem of iron-stressed lateral roots of the iron-
             efficient Hawkeye soybean contained Prussian blue (dark
             areas) indicating Fe(II)  continuous in protoxylem until
             union with metaxylem of the larger root, enlarged 20
             times.  The iron-stress-response mechanism reduced Fe(III)
             to (Fell) and kept the ferrous ions available in the pro-
             toxylem.  Reproduced from Ambler et^ al.
                                 106

-------
120
£
| 10.0
X
UJ
. 8.0
H
E
« 60
u_


5 4.0


2.0



0
o>
|30.0
LU
e2QO
*o
5 n
x'
	 T203- Soybeans /'

— MA-ooyDeons x
X
X
X
X
X
X
X
X
X
X
xx
X
X

	 	
^^ ^^

„ 1 1 1
^
^^
^*>''
^**
^x^x
^^,- — -^x
1 1 	 1
                0      6   2.5        5.0                  10.0
                   xlO" M FeEDDHA in  Absorption  Nutrient
FIGURE 3-9   When the iron concentration in the stem exudate was
             increased by increasing the iron concentration in the
             nutrient solution,  the citrate concentration in the
             exudate also increased in Hawkeye (HA)  soybeans, but
             not in the iron-inefficient T203 variety.   Both culti-
             vars were subjected to iron stress before  being trans-
             ferred to the nutrient solutions containing different
             concentrations of the metal. Reproduced from Brown and
             Tiffin.123
                                107

-------
latter contained sufficient citric acid in the xylem exudate to transport  iron

in vitro.  These findings indicate that the iron-inefficlent corn roots do not

respond to iron stress and the metal is not made available for transport in  the

xylem exudate.  The translocation of iron in the plant involves more than

citrate chelation of iron in the root per se.


ROLE OF THE IRON-STRESS-RESPONSE MECHANISM

Internal Root Control

     The term "iron-stress-response mechanism" is used to denote a positive

response to iron deficiency that induces biochemical or physiologic reactions

within the plant that makes iron available for plant use.  This mechanism  operates

in green plants that grow well on calcareous alkaline soils.  Chlorotic plants

found in such land areas are iron-deficient because the iron-stress-response
                                              168
mechanism is not functioning.  Clark and Brown    found that when WF9 (iron-

efficient) and ys  (iron-inefficient) corn genotypes were grown together with

ys1:WF9 ratios of 4:0, 3:1, 2:2, 1:3, and 0:4 plants per container, WF9 in the

presence of ys-^ released more hydrogen ions, reduced more Fe(III) at the root sur-

face, and took up more iron than the ys, genotype.  All ys1 plants, regardless
                                       1      168         1
of the ratio to WF9, developed iron chlorosis.    In a similar study using Hawkeye

and T203 soybeans with T203:Hawkeye ratios of 28:0, 24:4, 16:12, 4:24, and
                      18
0*.24, Ambler and Brown   found that regardless  of ratio,  Hawkeyes took up  80%
                                         111
more iron than T203's.  In another study,     iron uptake by T203 soybeans was

not increased when they were placed in solutions in which iron-stressed Hawkeye

soybeans had released hydrogen ions and reducing agents and had absorbed and

transported 80% of the iron available.   It was  concluded that iron absorption

and transport are controlled inside the roots and iron uptake is greatest when

the response mechanism is functioning.
                                   108

-------
     Both iron-efficient and iron-inefficient roots can have several hundred

mlcrograms of iron per gram of root, but the iron-inefficient plant may die

from lack of iron in its tops.  In contrast, iron is made available to tops by
                                                    118
the iron-efficient roots in response to iron stress.     In a similar way, iron
                                                       114
may remain in the nutrient solution as a ferric chelate    or a ferric phos-
      109
phate,    and not be taken into the plant until it is made available for

absorption and transport through chemical reactions induced by iron stress.

Iron usually is used in plant tops once it is made available for transport by

the roots.


Mechanism of Iron Uptake

     Iron absorption and transport, induced in response to iron stress, involves

the release of hydrogen ions by the root, which lowers the pH at the root zone.  This

favors Fe(III) solubility and reduction of Fe(III) to Fe(II).  Iron-efficient roots

also release reductants in response to iron stress.  These agents, along with re-

duction at the root surface, reduce Fe(III) to Fe(II), which enters the root pri-

marily through the young lateral roots.  It is likely that the Fe(II) is kept

reduced in the roots by the reductant.  Ferrous irons are present throughout the

protoplasm and may or may not have entered the root by a carrier mechanism.  The

root-absorbed Fe(II) is oxidized to Fe(III) near the metaxylem, chelated by

citrate, and transported in the metaxylem to the top of the plant.


Other Chemical Reactions Affected

     The chemical reactions induced by iron deficiency may affect a plant's nitrate

reductase activity, use of iron from ferric phosphate and from FeEDDHA, and

tolerance to heavy metals.

     Nitrate reductase activity.  The products of iron stress concomitantly

increase nitrate reductase activity in roots, shown in Figure 3-10 for tomatoes.


                                   109

-------
10
 8

 6

 4i
 2
                DAY 2
                DAY 6
                 DAY 8
              O-(J   OO    O-1     O-5
                   Fe, mg/l
                                         8
                                       £3
                                             	T3238fer
                                             _ _ J3238FER
                                  O-O    O-1    O-5
                                  Fe,  mg/l
 *Statistically significant at 1%  level.
**Some iron removed from roots with EDDHA  treatment.

 FIGURE 3-10   When subjected to  different iron  concentrations  in nutrient
               solutions (NO-j-N), nitrate  reductase activity  increased  in
               T3238FER roots, and the  pH  decreased in the nutrient  solution
               in response to iron stress.  Neither the iron-sufficient
               T3238FER and T3238fer'tomato showed these responses.  The
               plants were 21 days old  when transferred to these treatments.
               Drawings courtesy  of J.  C.  Brown  and W.  E. Jones.
                                 110

-------
In both induced nitrate reductase activity and induced "reductants" activated in

response to iron-stress, a substrate is reduced: i.e., nitrate to nitrite ions
                     708                                        110
by nitrate reductase,    and Fe(III) to Fe(II) by the reductant.     The induced

nitrate reductase activity decreased when iron was made available to the

plants.
                                       153
     Use of iron from ferric phosphate.      Phosphate, the plant, and the

chelating agent, ferrous ferrozine, Fe(II)3-(2-pyridyl)-5,6-bis(4-phenylsulfonic

acid)-l,2,4,triazine, compete for the iron in nutrient solution.  Hawkeye and

T203 soybeans were used as test plants.  Within 2 days after adding 16 mg phos-

phorus as phosphate to the nutrient solution, only about 10% of the ferrous

ferrozine was colorimetrically detected in solution.  The iron was removed from

ferrous ferrozine, and it appeared as a suspension of iron phosphate.  Phosphate

was dominating the system for iron.  Four days later, all plants developed some

chlorosis in the new leaves.  The iron-efficient soybeans responded to this stress

by releasing more hydrogen into solution and reducing more ferric to ferrous ions

than the iron-inefficient variety.  During this process, ferrous ferrozine re-

appeared in the solutions containing the Hawkeye plants, but not the T203 speci-

mens.  The iron-efficient plants now dominated the competition for iron.  On

day 17 (final harvest), the green iron-efficient soybeans contained 86 yg

iron/g of tops, whereas the chlorotic iron-inefficient soybeans contained only

36 vig iron/g of tops.  In the latter, phosphate still dominated the system for

iron.
                              114
     Use of iron from FeEDDHA.     Iron-inefficient T3238fer tomato plants

developed iron chlorosis because they could not absorb iron from FeEDDHA, as in

Figure 3-11.  In contrast, the iron was available to iron-efficient T3238FER tomato

plants because they could reduce Fe(III) to Fe(II).  For plants to use Fe(III)
                                     111

-------
FIGURE 3-11
T3238fer (left) and T3238FER (right) tomatoes as they
appeared when grown in nutrient solution containing
10 yM iron supplied as FeEDDHA.  The iron-inefficient
T3238fer tomato could not use iron from FeEDDHA.
Reproduced from Brown et al.  ^
                                 112

-------
                                                                                 154
 from several ferric chelates, they first must reduce ferric to ferrous chelates,

 which generally  have  a  much  lower stability  constant than  the  ferric  compound.
                               119
     Tolerance to heavy metals.     Heavy metals added to  soils in pesticides,

 fertilizers, manures, sewage sludge, and mine wastes can cause various degrees

 (depending on genotype) of iron deficiency to develop in plants.  An  iron-efficient

 plant is more tolerant to heavy metals than an iron-inefficient one.  For example,

 copper added as  a fungal spray to citrus trees and zinc added as the  oxysulfate

 to control bacterial  spot in peach trees accumulated in the soils in  concentrations
                      119
 toxic to plant growth.     When iron is made available, it counteracts the effect
                    702
 of the heavy metals.


 GENOTYPIC DIFFERENCES AND THEIR RELATION TO IRON STRESS

 Iron Deficiency  and Excess

     A recessive gene controls the uptake of iron in iron-inefficient T203 soy-
      805              47,54                     786
 beans,    ys^/ys^ corn,      and T3238fer tomato.     These plants usually develop

 iron chlorosis on alkaline soils, and to a much lesser extent on some acid soils.

 In contrast, the most iron-efficient plants are usually green on alkaline soils,

 and may even develop symptoms of iron poisoning on some acid soils.   For example,
     594
 Olsen    found that bentgrass developed iron-deficiency symptoms in nutrient

 solutions (pH above 6), but mustard grew well under these conditions with ferric

 sulfate as the source of the metal.   When the pH of these solutions decreased

 from 6 co 4, symptoms of iron toxicity developed in the mustard, but not the bent-
                                                     594
 grass.  These same relationships were found in soils.      Iron-efficient Bragg

 soybean tops were reported to contain 45, 353 and 1,320 yg iron/g and iron-

 inefficient Forrest soybean tops had 22, 54,  and 112 pg iron/g when grown on

 alkaline Quinlan soil (pH 7.5),  acid Bladen soil (pH 4.3),  and in nutrient

 solution containing 2 mg iron/1  as FeHEDTA, respectively (unpublished observations,

J.C.  Brown and W. E. Jones).   Bragg soybeans  showed signs of iron overload


                                     113

-------
in the nutrient solution and Forrest soybeans developed symptoms of iron

deficiency on the Quinlan soil.  These relationships are illustrated by Figure

3-12.  Iron toxicity is only about 10% as prevalent as iron deficiency.  How-

ever, the increased uptake of the metal by iron-efficient plants grown on acid

soils may be injuring plant growth, although the condition has not been recog-

nized or documented.


Iron and its Relationship to Phosphorus-efficient and Phosphorus-inefficient
Plants
                    533
     Mikesell et al.    concluded that B-line sorghum was less iron-efficient

than KS5 sorghum because the B-line took up more phosphorus than the KS5 line and
                                                                          121
that phosphorus interfered with the availability of iron.  Brown and Jones

confirmed these findings and showed further that KS5 responded more to iron

stress than B-line.  They concluded that the accumulation of phosphorus and

insufficient iron-stress response both contributed to iron deficiency in B-line

sorghum.

     Other sorghum lines grown on alkaline soil developed iron chlorosis on

the phosphorus-efficient but not on the phosphorus-inefficient genotypes

(unpublished observations, J. C. Brown and W. E. Jones).  When the sorghum

lines were under phosphorus stress (deficient in phosphorus), only the phos-

phorus-efficient genotypes survived, as can be seen in Figure 3-13.  The

phosphorus-efficient genotypes took up 15 times more phosphorus-32 than the

phosphorus-inefficient plants.


Zinc Stress as an Inducer of Iron Uptake

     For some unexplained reason, zinc deficiency induces iron uptake in some
                           17,391,792
plant species or varieties.            Zinc-deficient corn plants accumulate

excessive iron in their tops; evidently the excessive iron concentration
                                   114

-------
FIGURE 3-12   Bragg (a) and Forrest (b) soybeans grown in nutrient
              solutions containing 2 mg iron/1 and grown (c and d)
              in Quinlan soil (pH 7.5).  Note the symptoms of exces-
              sive iron on Bragg (a), but not on Forrest (b) and the
              iron-deficiency symptoms on Forrest (d), but not on
              Bragg (c) .  Photographs courtesy of J. C. Brown and
              W. E. Jones.
                               115

-------
FIGURE 3-13   When SC369-3-1JB,  PI-405107 and NK212 (top left to
              right)  sorghum were grown on Quinlan soil (pH 7.5),
              SC369-3-1JB developed iron chlorosis, but NK212 did
              not.  In the bottom series of photographs, NK212
              specimens grown on Bladen soil (pH 4.3)  developed
              phosphorus-deficiency symptoms, but SC369-3-1JB did
              not.  The PI-405107 genotype was intermediate in its
              response. Photographs courtesy of J. C.  Brown, R. B.
              Clark,  and W. E. Jones.
                               116

-------
was associated with zinc stress in the plant and not with the level of iron
                     391,792                   17
in the soil solution.         Ambler and Brown,   working with Sanilac (zinc-

inefficient) and Saginaw (zinc-efficient) navy beans (Phaseolus vulgaris L.),

showed that when Sanilac plants developed zinc-deficiency symptoms, as in

.Figure 3-14, they contained twice as much iron (665 yg iron/g) and nearly  twice

as much phosphorus  (1.7%) in their tops as Saginaw.  Sanilac tops contained

18 ug zinc/g compared to 23 ug zinc/g in Saginaw tops.  Zinc-deficient cotton

(Gossypium hirsutum L.) contained 412 pg iron/g compared to 94 pg iron/g top

in the zinc-adequate plants (unpublished observations, J. C. Brown and W.  E.

Jones).


Interference of Molybdenum Stress with Iron Uptake
                                                                    360
     Molybdenum is required by flavoproteins that transfer electrons    and by
                                                       52,226,259,580
nitrate reductase, the enzyme that breaks down nitrate.
                    62
Barry and Reisenauer   determined that molybdenum affected the reductive

capacity of Marglobe tomato roots.  Iron uptake was depressed when molybdenum

concentrations were both low (less than adequate) and high (more than adequate)

in the nutrient solution.  The plants receiving adequate molybdenum were the

most effective in raising the redox potential, and those that were not supple-

mented were the least effective.  The increased redox potential corresponded

to the production of ferrocyanide, i.e., the reduction of Fe(III) to Fe(II),

and an increased uptake of iron by the plant.  Iron accumulation was maximal

at marginally adequate levels of molybdenum nutrition.


Role for Plant Breeding

     The plant breeder must select or develop iron-efficient cultivars to  be

used where the mineral deficiency is a problem in crop production.  In past

years, and with very little economic success, attempts have been made to


                                   117

-------
FIGURE 3-14
Zinc-deficiency symptoms appear on Sanilac leaves
(left) but not on Saginaw plant leaves (right), when
both are grown in a split medium of Shano soil (top)
and nutrient solution (bottom).  Reproduced from
Ambler and Brown.^
                              118

-------
correct problems of iron nutrition by changing the soil to fit the plant.

Plant breeding now offers the possibility of controlling iron chlorosis by

tailoring the plant to fit a problem soil.  In some cases, varieties, lines,

or hybrids already developed or selected will overcome the problem.  In others,

however, more iron-efficient lines will need to be developed.
           834          787
     Citrus    and grape    growers in Texas and Utah, respectively, controlled

genetically iron chlorosis by grafting iron-efficient rootstocks onto desirable
                704
scions.  Sprague    believes that genetic knowledge is adequate to provide the

necessary support for a productive cooperative effort between physiologists and

geneticists to develop iron-efficient plants.  Heslop-Harrison suggests that

"the most effective way to obtain a growth pattern efficient in a given environ-
                                                                  355
ment for a particular purpose is to breed a genotype for the job."     It is

necessary to know the nutrient requirements of crop plants before fitting them
                     4
to a particular soil.   For example, when the Hawkeye soybean was replaced by

new,iron-Inefficient soybean varieties in central and north-central Iowa (cal-
                                                                     226
careous soil), iron chlorosis developed in the new soybean varieties.     These

plants were probably not tested for iron efficiency before they were released  ...

to the field.


FORMS AND FUNCTIONS OF IRON IN PLANTS

     If iron absorption and transport depend on the induction of the iron-stress-

response mechanism, then these induced biochemical reactions should enhance the

use of the element throughout the plant.  Iron is an essential component of
                                           260,631,632
many heme and nonheme enzymes and proteins.             Nucleic acids contain
     286
iron,    and a ribosomal chromoprotein has been identified in rat liver that
                  532
contains 20% iron.     The iron concentration of this protein varied with the

physiologic state of the organism.  Storage forms of iron in plants include
                                   119

-------
             383,578,679
phytoferritin            and ferric phosphate.  Iron accumulation in different

plant parts generally follows this order: roots > old leaves > young leaves
        755
> stems.

     Some of the iron in plants is chelated.  When a cell develops iron stress,

i.e., deficiency, iron chelates with low stabilities will fail to form and the
                                              631
metal will be distributed among other ligands.     For example, in catalase,

the iron exchange is rapid, and in cytochrome a_, it is slow.  Catalase activity
                                                                         115,802
has been reduced in several plant species as a result of iron deficiency.
     631
Price    indicates that data are insufficient to decide whether or not iron

compounds in physiologic systems behave as simple chelates.  He suggests that

because chlorosis is one of the earliest symptoms of iron deficiency, the com-

pound responsible for the condition is one of the least stable among the physio-

logically essential forms of iron.  Phosphate may act as a competing ligand

for iron in such a system.

     This chapter has emphasized iron absorption and transport rather than its
                                                         221,260,358,359,563,631,63
forms and functions, which have been thoroughly reviewed.


RESEARCH NEEDS

     Plant species and varieties within species differ in their iron require-

ments and tolerance to high concentrations of mineral elements, which complicates

determining the nutrient requirements of plants so that the plant and the soil

can be made compatible.  The plant breeder will be challenged to develop plants

that are nutritionally adapted to problem soils.  The agronomist, plant physiolo-

gist, biochemist, and horticulturist will need to supply the geneticists with

techniques for identifying a desirable factor or trait for a plant.   For example,

iron-inefficient plants develop less iron-stress response than iron-efficient
                                   120

-------
plants, and this range in response is the basis of a technique for screening
                           118
plants for iron-efficiency.     A limited supply of iron and some control of

pH in the growth medium are required in this technique.   Degree of iron chlorosis

is all that is needed to give the plants an efficiency rating.  When the geneti-

cally controlled reaction is identified in the plant that responds to iron stress,

a more specific technique may be developed and used by the plant breeder to

identify iron efficiency in that particular genotype.

     Regional laboratories for testing soils and plant tissues could be estab-

lished to meet the needs of a specific area.  Soils and crops could be characterized

for what cultivars would be most suitable to the area and what supplements should

or should not be added to the soil.  The genetic control of nutrition is well

established and a genetic program is suggested to fit plants nutritionally to

problem soils, i.e., iron-deficient, saline, and manganese- and aluminum-toxic.

Recognition of the role of plants in controlling mineral absorption and transport

signals a new direction for plant nutrition.  Related programs, based on the

pooling of resources and expertise, would conserve our soils, improve the effi-

ciency of fertilizers and increase crop production.
                                   121

-------
                                  CHAPTER 4

                 IRON METABOLISM IN HUMANS AND OTHER MAMMALS


  TOTAL BODY IRON

       The essential concentration of iron in the vertebrate organism increased

  by two orders of magnitude over that of lower living forms when hemoglobin

  evolved as an oxygen transport vehicle.  Total body iron concentrations of

  various animal species, estimated on the basis of the sum of individual iron

  fractions, varies between 25-75 mg/kg body weight.   The adult human male has

  approximately 49 mg iron/kg and the adult female has approximately 38 mg/kg,

  equivalent to a total body iron content for the 80  kg male of about 4 g and

                                   89
  about 2.5 g for the 65 kg female.     Miscible iron  has been determined by
                                                       on o
  isotopic dilution studies  to be 42  mg/kg  in the  male;    this  somewhat  lower

  estimate is assumed to result  from incomplete mixing with  the pool of storage

  iron.  Variations in iron content  within individuals of one species and among

  other vertebrate species are explained by differences in circulating hemoglobin

  mass and/or by differences in  storage iron.  The general proportions of

  erythron iron, other essential tissue iron, and  storage iron in humans are

  summarized in Table 4-1.

     Essential body iron in humans is generally proportionate to lean body mass

at about 35 mg/kg, and a variable amount of  storage iron must be added to this

amount.  At birth, total body iron concentration  is increased by the elevated
                                                                        476
red cell mass and  the presence of appreciable storage iron in the liver.

During the first month of life,  stores increase still further because of the

decrease in circulating red cell mass.  Thereafter, rapid growth and greater

red cell requirements result in  a  transfer in the opposite direction so that

iron stores become virtually exhausted between the sixth and twenty-fourth
       683,701
months.         Little direct evidence exists of the size of iron stores in

later childhood, but as can be seen in Table 4-2, serum ferritin values indi-
                                                                       184,648
cate a limited iron reserve, perhaps 5 mg/kg until after the age of 15.

                                   122

-------
                       TABLE 4-1




          Body"Iron Contents of Adult Humans
Myoglobin




Cell enzymes
       Age
       6 mo
Male Female

marrow
ig erythrocytes

les
md hemosiderin
' IRON
TABLE
Mean Values for
yg/kg mg/80.kg yg/kg
2 160 2
28 2,240 26
*4 320 =3
=2 160 -2
13 1,040 5
49 4,000 38
4-2
Serum Ferritin
mg/65 kg
130
1,690
195
130
325
2,470


as a Function of Age
rborn
10
.0 yr
18 yr
•45 male
45 female
5 male
5 female
Ferritin, yg/1 .
110
5
21
22
94
25
124
89








                        123

-------
Iron reserves of some 1,000 mg in the male are created between the ages of 15

and 30; adult females maintain low iron stores of about 300 mg, but after meno-

pause they increase to the level found in males.

     Although various chemical measurements of the iron content of body tissues
                                     758
have been made in animals and humans,    such analyses are of limited utility.

The difficulty lies in the lack of attention paid to separating hemoglobin iron

from other iron, because of the very large concentrations of iron in circulating

red cells that contaminate those tissues.  Hemoglobin, ferritin-hemosiderin, and

myoglobin are the only quantitatively  important  body-iron compounds that have been

found: Therefore, any appreciable concentration of iron in the tissues is assumed

to represent one of those components.


     Body iron contents of animals vary with age, depending on the iron endow-

ment at birth, the duration of breast-feeding, the rate of growth, and repro-
                     432,758
ductive requirements.         With the exception of the pig, whose exceedingly
                                              285
rapid growth frequently outstrips iron supply,    mammals easily acquire and

maintain essential body iron, and iron deficiency is exceedingly rare.


THE NATURE OF BODY IRON

Iron Compounds of the Blood

     Red cell hemoglobin constitutes the largest fraction of body iron.  Hemo-

globin is composed of four polypeptides each weighing about 16,000 daltons, and

each possessing a heme moiety within which an iron molecule is located.  The two

different subunits of the protein together form a tetrameric molecule, (aS)^.

Each chain is coiled into an a-helix folded around the heme group in such a way

as to provide a nonpolar environment for its heme.  This folding is essential in
                                                                  611
permitting the iron to combine with oxygen without being oxidized.     In

unoxygenated hemoglobin,  the iron atoms are five-coordinated, and are forced
                                   124

-------
                                  556
out of the plane of the heme ring.     Upon oxygenation, the sixth coordinate

site is occupied by oxygen, causing the three electrons of the iron atom to

rearrange.  This alteration reduces the radius for the iron atom, enabling it

to move back into the heme plane.  These changes in structure are essential

for the normal functioning of hemoglobin as an oxygen carrier.

     Hemoglobin accounts for approximately 85% of essential iron in the human

female and 60% of the total in the male, a distinction largely related to the

difference between the size of iron stores in the two sexes.  Hemoglobin con-

centration in normal humans depends on ambient oxygen tension, hemoglobin
                                                         3
affinity for oxygen, and circulating testosterone levels.   In childhood, affinity

of hemoglobin for oxygen is somewhat decreased, perhaps related to an increased


  organic  phosphate  concentration.  The  rise  in phosphate  is thought to be respon-

  sible  for  a  slightly  lower hemoglobin  concentration 'because  it makes more oxygen
                       146
  available  to tissues.     The  sex difference  in adult  hemoglobin  concentration

  is  undoubtedly  linked  to  the effect of testosterone  on erythropoietin stimulation

  of  the marrow.     The  effect of altitude  on hemoglobin is well  established.   It

  is  proportional to decreases in arterial  oxygen saturation,  resulting from

                                   380
  changes  in ambient oxygen tension.

     The physiologic norm for an individual's hemoglobin concentration varies

considerably,  and it is extremely difficult in mild anemia to separate physiologic

variations from deviations produced by pathologic states.  It has also been compli-

cated to define mean normal values of hemoglobin concentration for normal popula-

tions.  Surveys are problematic in that they embrace a considerable number of

iron-deficient individuals, some of whom are anemic.  One study of a Swedish

population showed a hemoglobin response by 17% of the "normal" adult population
                       290
to iron administration.

     Statistical approaches have been developed for surveys to exclude individuals
                     180
with iron deficiency.      Mean hemoglobin concentrations from residents of differ-

ent parts of the world have not demonstrated differences in hemoglobin concentration

                                   125

-------
based on racial derivation, although the existence of such variances often has

been suggested.  For example, mean hemoglobin concentrations from individuals

in ten states, set forth in Table 4-3, suggest racial differences, but dissimilar

dietary customs also may have been responsible.  The values in Table 4-3 are

certainly influenced by a population with iron deficiency.  In subjects sampled

from the Pacific Northwest, about 4% of the adult males and 20% of the menstruating
                            184
females were iron-deficient,    and half of that number had demonstrable anemia.

     Normal hemoglobin values for other mammals are even less well defined, but

reports do show characteristic species variations that influence body iron
        432,758
content.         Thus, some sheep have a hemoglobin content approximately half
                                                                                  600
that of a human, apparently explained by a marked increase in oxygen dissociation.

In animals such as dogs and goats, hemoglobin concentration may fluctuate markedly

because their spleens can temporarily hold as much as one-third of the red cell

mass, although the total intravascular hemoglobin in these species is quite similar

to that of humans.

     Transferrin is the other iron compound of the blood with an important physio-
               548
logic function.     This plasma component is a glycoprotein with a molecular

weight of about 80,000; it has a prolated ellipsoid shape with an axial ratio of

1:3 and is composed of a single polypeptide chain with 2 identical carbohydrate
            6
side chains.   Although this protein is pleomorphic—with some 18 variants
                                                                              296
described in humans—these differences are not known to affect iron transport.

Transferrin is predominantly synthesized in the liver.  Its production and losses
                          296
parallel those of albumin.     The normal concentration of the protein in humans
                           180
is about 2.3 g/1 of plasma,    equivalent to an iron-binding capacity of 3.3 mg/1.

Transferrin concentration changes inversely in relation to changes in body iron
       801
stores.      In individuals with adequate body iron, the iron-binding capacity of

transferrin is only 20-45% saturated;  the remainder constitutes a latent capacity.


                                   126

-------
                                TABLE 4-3




      Median Hemoglobin in Populations of Ten States, 1968-1970a
Age
< 2
2-5
6-12
White
11.1
12.1
12.6
Black
10.7
11.3
12.0
Spanish-
Surnamed
12.3
12.5
12.9
                        Male
Female
13-16
17-44
45-59
> 59
White
14.1
14.9
14.9
14.6
Black
13.0
14.2
13.9
13.5
Spanish-
Surnamed
13.8
15.0
14.6
14.2
White
13.1
13.3
13.4
13.6
Black
12.3
12.8
12.8
12.6
Spanish-
Surnamed
13.1
12.9
13.4
13.4
Derived from data of the U.S. Department of Health, Education, and Welfare.
                                                                           761
                                   127

-------
Mean plasma iron concentration and total iron binding capacity in many animal
                                    89,758
species are quite similar to humans.        A higher plasma iron in some

animals appears to be a reflection of a much greater amount of iron absorbed—
                                                                            185
with substitution of a low-iron  diet, plasma iron  falls to about 100  ug/dl.


Essential Tissue Iron

     Three types of iron-containing compounds in the body perform essential meta-
                207
bolic functions.     The first includes all other heme iron compounds.  Of  these,
                                                                            13
myoglobin is the largest fraction, amounting to about 15 mg iron/kg muscle.

Its structure is closely related to the monomeric unit of hemoglobin: it contains

one polypeptide chain attached to a heme group with a single iron atom.  It

functions as a link in the oxygen transport chain, accepting oxygen from the

blood and storing it for utilization during muscle contraction.  Cytochromes a_,

b_, and c^ and PASO represent another set of tissue heme proteins located in  the
                                                                               582
mitochondria and other cellular membranes, and which aid in electron  transport.

Of these, cytochrome c_, a pink protein with a molecular weight of 13,000, is the

best characterized.  Catalase and peroxidase are other heme iron enzymes.  A

second category consists of metabolically active compounds with an enzymatic
                                                       318
function, but in which iron is not in the form of heme.     In mitochondria,

for example, nonheme compounds account for far more iron than do the  cytochromes.

A large portion of this iron is in a group of proteins designated as  metallo-

flavoproteins.  Metalloflavoproteins are involved in oxidative metabolism and include

reduced nicotinamlde-adenine dinucleotide, succinate, and cx-glycerophosphate

dehydrogenases.  Other enzymes of this group, such as monoamine oxidase,are not

yet purified, but are presumed to contain iron.  The final category includes

enzymes such as aconitase and microsomal lipid peroxidase, which do not contain

iron yet require it as a cofactor.  Iron in a loosely bound form is required for

hydroxylation of proline and lysine in protocollagen, steps essential in the

                                   128

-------
synthesis of collagen.  This catalog is far from complete, and it is likely that

additional compounds with important metabolic activities will be identified in

the future.


Storage Iron  ,

     Iron within the body is stored as ferritin and hemosiderin, which together

represent the second largest fraction of iron after hemoglobin.  Ferritin is a
                                                     334
specialized tissue protein designed for iron storage.     It has a molecular

weight of about 450,000 and is composed of some 24 subunits.  Within the central

cavity of ferritin, masses of iron are deposited principally as ferric hydroxide,

but they contain some phosphate as well.  Hemosiderin is considered to be,an

aggregate form of ferritin and showsup as golden brown granules when seen by

light microscopy.  More detailed examination by the electron microscope suggests

that these masses are composed of closely packed ferritin molecules.  Because a

variety of organic constituents are also included in the aggregate, it is postu-

lated that these masses actually are disintegrated, ferritin-loaded, cellular

organelles.  These iron storage compounds are widely distributed in nature
                                           197,383
among different animals, plants, and fungi.

     With a positive iron balance, the inflow of iron induces the synthesis of

apoferritin, and, when it becomes loaded with iron, the ferritin molecule is
                                      233
protected against degradation as well.     Storage ferritin is synthesized by

free polyribosomes, whereas serum ferritin may be produced by the endoplasmic
                                                           636
reticulum of the reticuloendothelial cells and hepatocytes.     The ratio between

ferritin and hemosiderin differs according to the total amount of iron stored

within the cell.  At lower concentrations of tissue iron, ferritin predominates;

at higher concentrations, most of the iron found is hemosiderin.   The

exact mechanism whereby iron loads into the ferritin molecule or is released
                                   129

-------
 is largely unknown,  although it may  involve  a Fe(III) ->• Fe(II) ->•  Fe(III) cycle.

 In mammalian species,  diverse molecular forms of ferritin have been identified;

 in humans,  different profiles are  observed,  according to  whether  the ferritin

 originates  from heart,  liver,  or other  tissues.   The  proposal  has been made

 that  these  different species  of ferritin represent various combinations  of
                     233
 dissimilar  subunits.


 IRON  BALANCE  IN HUMANS

      Iron balance represents  the relationship between iron absorbed and  iron lost

 by the individual.   The notion that  iron exchange was very small originally  came
                                      524
 from  careful  chemical balance studies.     A better understanding was achieved

 through isotopic measurements of iron losses, more accurate quantitative estimates

 of iron requirements for growth and  pregnancy, and measurements of  iron absorption

 from  food.


 Physiologic Iron Losses

      Losses in  the male.  Because of the small amounts of iron that men lose

 daily and the difficulties involved  in distinguishing true losses from the

 presence of contaminating iron not actually  excreted, it became evident that

 chemical measurements of excreted iron were  unsatisfactory.  Direct determination

 of radioactivity in urine, feces,  and sweat  after the intravenous injection

 of iron-59 permitted a distinction between excreted iron and contaminating
                                                            234
 fractions but presented considerable technical difficulties.      Such studies

 led to an estimate of about 1 mg/day of iron loss in the adult male, a figure

 that has been subsequently confirmed.  Similar figures were later obtained by
                         140,620
 counting total body iron,        but the relatively short half-life of iron-59

and the extremely limited daily loss of the isotope from the  body made precise

quantitation difficult.  Long-term studies of body iron turnover employing


                                     130

-------
iron-55, in which the specific activity of the circulating red cell mass was
                                                                  268
followed over several years, have provided the most accurate data.     A trans-

lation of these isotopic measurements into absolute amounts of iron lost,

however, involved an assumption of complete body mixing and a definition of pool

size.  Analysis of the red cell activity curve showed an initial mixing component

of about one year's duration and thereafter an exponential decrease at a rate of

about 11%/yr in the adult American male; this derivation was estimated to be
                                                                          308
equivalent to an iron loss of about 12 yg/kg/day or 1 mg for an 80 kg man.

Losses for other normal subjects in other countries studied by the same technique
                        308
were about 14 yg/kg/day.
                                                                308
     The source of these basal iron losses has been categorized.     The largest

single fraction, amounting to about 0.4 mg/day, consists of red cells entering

the gut lumen.  Additional losses of smaller magnitude, listed in Table 4-4,

are derived from the iron content of the bile and exfoliated intestinal cells.

These three sources of gastrointestinal iron are the basis for two-thirds to

three-fourths of the total body loss.  Urinary iron loss is of little consequence.

Losses from the skin could not be accurately measured by chemical techniques

because of surface contamination with iron.  Radioactive measurements permit

determination of the uptake by epithelial cells from transferrin, totaling about

2.5 yg/kg/day; this amount should represent the maximal iron available for loss

through cell exfoliation, although some individual variation may occur.  Perhaps

the greatest fluctuations occur in blood loss from the intestinal tract, which may

be increased by local irritants such as aspirin, alcohol, and other drugs.  Other

losses (skin, gastrointestinal mucosa, and urine) are affected by the level of

serum iron, and therefore decrease during states of iron deficiency and increase

in iron overload.  However, the extent of iron losses probably is not reduced to
                                  131

-------
                    TABLE 4-4




          Estimated Major Iron Lossesa











                             80 kg male
Source                  yg/kg/day      mg/day




Gastrointestinal region




   blood                   5.0           0.4




   mucosa                  1.0           0.1




   bile                    2.5           0.2




Urine                      1.0           0.1




Skin                       2.5           0.2




TOTAL                     12.0           1.0
 Adapted from Green et^ al.
                     132

-------
less than 50% in iron deficiency and not elevated more than twice normal in

iron overload.

     Special iron losses in females.  Special requirements for iron face the

adult female.  Menstrual blood losses have been shown to average about 0.6
                                     172,321
mg/day if distributed over the month.         Therefore, the mean total iron

loss in the menstruating female is about 20 yg/kg/day or 1.4 mg for the 65 kg

female.  This figure is consistent with radioactive measurements showing an
                                                     268
overall loss of red-cell specific activity of 20%/yr.     The distribution curve

for menstrual blood loss is skewed: 11% of women lose over 80 ml blood/mo,
                                                321
representing an iron loss of more than 1 mg/day.     This phenomenon is plotted

in Figure 4-1.  Thus, at least 10% of menstruating women have iron requirements

more than twice that of the male population.

     The other unique loss for the female is associated with pregnancy.  Two

sets of requirements may be drawn up for pregnancy, summarized in Table 4-5.

One represents iron needed during pregnancy, and the other, the actual iron
                                 49,195
lost with pregnancy and delivery.        When distributed over 9 months,

the requirement becomes 3.5 mg/day and the cost, 2.5 mg/day.  The increased

requirements are imposed largely during the last 6 months of pregnancy,

making the daily requirement during that period even greater.

     Requirements during infancy and childhood.  In infancy and childhood,

attention is directed to the iron required for growth.  The newborn is endowed

with sufficient excess iron by the polycythemia present at birth and the hepatic

iron stores to meet requirements for doubling body size during the first four

months.  In the premature infant, the reserve is greatly reduced—moreover,
                                  816
growth requirements are increased.     Through infancy, childhood, and adoles-

cence, requirements of growth amount to 30 mg iron/kg increase in body weight.
                                     133

-------
100-
 50-
    0
50
100           150           200   240
      MENSTRUAL BLOOD LOSS, ml
460      540
                FIGURE 4-1  Distribution of menstrual blood  loss in a population sample from
                           women in Goteborg, Sweden.  Reproduced from Hallberg et  al.321

-------
                      TABLE 4-5




       Total Iron Requirements for Pregnancy


External iron loss
Expansion of red-blood cell mass
Fetal iron
Iron in placenta and cord
Blood loss at delivery
Total requirement12
Cost of pregnancy
Average ,
mg
170
450
270
90
150
980
680
Range,
mg
150-200
200-600
200-370
30-170
90-310
580-1,340
440-1,050
Blood loss at delivery not included.




Expansion of red cell mass not included.
                          135

-------
                                                         701
 Information is limited concerning iron losses in infancy;    they are assumed

 to be at least equivalent to those of the adult on a kilogram basis.  The fre-

 quent finding of guaiac-positive stools in infants and demonstrations of increased

 losses of radioiron in the stool suggest that infants lose more iron that do
       367
 adults.     Much of this early bleeding may result from an immunologic reaction

 to cow's milk.  A composite of average iron requirements throughout life is shown
              89
 in Figure 4-2.


 Iron Intake and Absorption

     Iron absorption is the product of the amount of iron in the diet, its

 availability, the influence of various luminal factors, and the behavior of the

 intestinal mucosa.  Much of the work on absorption has been carried out in

 animals, and although this research has been useful in developing certain general

 criteria, its relevance to humans is questionable because of species differences.

     Mucosal regulation of absorption.  That the intestinal mucosa regulates
                                          316
 iron absorption has been known since 1945.     Most iron is absorbed in the duo-

 denum and upper jejunum,  although the mucosa throughout the intestine is capable

 of absorbing iron because of the favorable intraluminal conditions in that portion

 of the gut.

     Oral administration of iron salts  has permitted a more direct evaluation of

 the intestinal mucosa because highly available iron could be presented to the

mucosal cell without the modifying effects of food.  Body iron stores appear to
                                                                         91,186,616a
be the most important influence on the absorptive capacity of the mucosa.

 In subjects with marginally low iron stores whose hemoglobin or plasma iron con-
                                                       186,435
centrations were not changed, absorption was increased.         Conversely,

excess iron stores resulted in diminished absorption.

     An increased role of erythropoiesis following hemorrhage, hemolysis, or

exposure to decreased ambient oxygen tension also has been shown to increase iron

                                    136

-------
            15-
           I
             5
FIGURE 4-2   Iron requirements in humans.  The daily iron requirement
             through life is indicated by the continuous black line.
             For those over the age of 12, the line divides into the
             requirements of the female (upper line) and the male (lower
             line).   The dotted line indicates available iron in the
             normal diet.  The shaded area during the first year of life
             indicates a period of negative iron balance when the infant
             utilizes iron stores.  The black arrows indicate the two
             critical periods when intake and loss are of similar magni-
             tude.  Reproduced from Bothwell and Finch.
                               137

-------
           91,177,340,530,710
absorption.                    A return to more normal conditions will tend to
                                            91,530,803
decrease erythropoiesis and iron absorption.            Whereas it has been

proved in animals that hemolytic anemia increases iron absorption, data for

humans are less convincing.  Only with disorders in hemoglobin synthesis, i.e.,

thalassemia and sideroblastic anemia, does iron absorption appreciably increase.
                                                               531          671
Anemia per se has been described as increasing iron absorption.     Schiffer

found that markedly less iron was absorbed in patients with aregenerative anemia

when their hemoglobin was restored by transfusion.  Still another condition

affecting absorption is inflammation, during which iron transport across the

mucosa is reduced, accompanied by a. similar decrease in iron release by hepa-

tocytes and reticuloendothelial cells.

     Although these factors and possibly others are known to modify iron absorp-

tion, the means by which it is accomplished is not known.  Considerable effort

has gone into the search for a humoral factor in the blood, yet convincing
                               752
results have not been obtained.     Iron absorption may be regulated by iron
                                                            178,179,804
within the intestinal cells received during their formation.             How-

ever, others have observed that the iron content of mucosal cells in man is not
                                                       16,42a
appreciably changed during iron deficiency or overload.        Even if this

suggestion is correct, it remains unexplained as to how the iron content of

these cells is modified, inasmuch as plasma iron does not diminish with depleted

iron stores, yet absorption increases.

     The nature of the absorptive process is as yet unclear, although interesting
                                           377
observations have been made.  Hubers et^ .aJU    found that iron-deficient rats

have twice as much iron in their brush border as those of iron-adequate rats,

suggesting that the brush border might be one point of control.  Within the cell,

several iron-binding substances have been found in the particle-free fraction of
                                 138

-------
                         363,606,622,831
mucosal cell homogenates.                  A transferrin-like protein has been

demonstrated that is found in greatest concentration in the upper part of the

small intestine of rats.  It is increased by iron deficiency and by pretreatment

of iron with phenobarbital.  The administration of phenobarbital also results

in iron absorption, but it is not demonstrated in the intestinal mucosa of mice
                                                         378
with sex-linked anemia who have impaired iron absorption.     This protein is

not precipitated by antiferritin, antitransferrin, or antilactoferrin antisera.
                                                622
A similar protein has been described by Pollack.     However, proof adequate

to establish any given protein as an important transport mechanism for iron has

not yet been provided.  Notable amounts of iron are held as ferritin within

the mucosal cell in inverse relationship to the amount absorbed.  This binding,

however, is not thought to regulate iron absorption, but merely to hold iron

not designed to enter the body, since the metal is lost when the mucosal cell

is sloughed.

     Lmnlnal factors.  The amount of dietary iron absorbed depends not only on

the absorptive behavior of the intestinal mucosa, but on the effect of various

luminal secretions and gastrointestinal motility.  It has been demonstrated

both in humans and experimental animals that absorption is usually reduced when
                                                183,512,561
gastric acid secretion is reduced or eliminated.             Similarly; the

administration of hydrochloric acid to individuals with achlorhydria increases
                    183,403             811
absorption in humans        and animals.     Because inorganic iron compounds

are more readily solubilized in acidic than in neutral or alkaline media, it

is not surprising that the pH of the stomach is important to absorption.

Gastric juices may contain other substances facilitating or inhibiting iron

absorption, but on the basis of present data, it is impossible to identify any-

thing other than a nonspecific absorption of iron to mucoproteins secreted
                                   139

-------
                   550
within the stomach.     Pancreatic secretions have also been thought to modify

iron absorption, but the most recent animal and clinical studies do not support

this idea.  Thus it is inappropriate to assume any effect other than alkaliniza-
                                  41,225,415,560
tion of the pH in the upper bowel.

     Alkalinization precipitates iron and makes it unavailable for absorption.

At the same time, the digestion of food in the upper duodenum by pancreatic

enzymes might well be expected to elaborate chelates such as cysteine and histi-

dine, which might improve availability of iron for absorption.  Studies of the

physical form of iron in the stomach and duodenum indicate that iron is bound

predominantly to macromolecules in the stomach, but micromolecular iron is found
                400
in the duodenum.     Ascorbic acid secreted into the stomach of some animals and

into the duodenum from the bile has also been proposed as an aid to iron absorp-
                                                        175
tion, but the acid's quantitative importance is unknown.     The role of the

digestive process and of possible chelates secreted into the lumen is unclear.

     Iron absorption also is affected by the motility of the gastrointestinal

tract.  Retarded gastrointestinal motility is accompanied by increased absorption,

whereas operative procedures on the stomach that shorten emptying time are asso-
                                 141,511
ciated with decreased absorption.         Such observations indicate that the

"hopper" function of the stomach is essential for normal iron absorption.


IRON IN THE DIET

     Humans ingest less iron than other animal species.  Average dietary iron

intake in developed countries is about 6 mg/1,000 cal; in developing countries,
                                                              49,761
the content is more variable, averaging about 10 mg/1,000 cal.        The higher

values in developing countries may be a function of extrinsic iron that is con-

taminating food during its procurement and preparation.  The iron content of

various foods is listed in Table 4-6.
                                   140

-------
                                      TABLE 4-6
                               Iron Content of Foods
                                                    a
 I.  Low Iron Content (<0.7 mg iron/100 kcal)
                                                mg iron/
     Food                                       100 kcal

     Apples, raw                                  0.6
     Avocado, raw                                 0.4
     Beer                                        trace
     Bologna, frankfurters                        0.6
     Bread, white, unenriched                     0.3
     Butter                                        0
     Cheese, cheddar                              0.3
     Cheese, cottage, creamed                     0.3
     Chocolate, semisweet                         0.5
     Codfish, broiled                             0.6
     Corn, fresh, cooked                          0.7
     Egg whites, chicken                          0.2
     Honey                                        0.2
     Ice cream                                   trace
     Lamb, loin chop, broiled                     0.4
     Margarine                                     0
     Milk                                        trace
     Oil, salad or cooking                         0
     Orange juice                                 0.4
     Peaches, canned in syrup                     0.4
     Peanut butter                                0.3
     Peanuts, roasted                             0.4
     Pears, raw                                   0.5
     Potato chips                                 0.3
     Rice, brown, cooked                          0.4
     Rice, white, unenriched, cooked              0.2
     Salad dressing, mayonnaise                   0.1
     Salad dressing, mayonnaise-like              0.1
     Sugar, white, granulated                    O.I
     Sweet potatoes, baked                        0.6
     Wheat flour, white, unenriched               0.2
                mg iron/100 g
                edible portion
                    of food

                     0.3
                     0.6
                    trace
                     1.8
                     0.7
                      0
                     1.0
                     0.3
                     2.6
                     1.0
                     0.6
                     0,
                     0,
                     0.1
                     1.3
                      0
                    trace
                      0
                     0.2
                     0.3
                     2.0
                     2.2
                     0.3
                     1.8
                     0.
                     0.
                     0.5
                       ,2
                       ,1
                     0.9
                     0.8
II.  Medium Iron Content (0.7-1.9 mg iron/100 kcal)
     Almonds, dried
     Apricots, raw
     Beef, ground, cooked
     Beef, round, broiled
     Beef, T-bone steak, fried
     Bread, white, enriched
     Bread, whole wheat
     Carrots, raw
0.8
1.0
1.1
1.9
0.6
0.9
0.9
1.7
aPrepared by Elaine Monsen.
4.7
0.5
3.2
3.7
2.7
2.5
2.3
0.7
                                        141

-------
      TABLE 4-6
      (Continued)

                                                 rag iron/
      Food                                       100 kcal

      Cereals, prepared breakfast
        enriched at 2-7% USRDA per oz.            0.4-1.0
      Chicken, dark, cooked                        1.0
      Chicken, white, cooked                       0.8
      Cranberries,  raw                             1.1
      Eggs, chicken, whole                         1.4
      Eggs, chicken, yolk                          1.6
      Gingerbread,  made with enriched flour        0.7
      Liverwurst                                   1.8
      Molasses, light (first extraction)            1.7
      Melon, cantaloupe                            1.3
      Oatmeal, cooked                              1.1
      Onions, mature, cooked                       1.4
      Oranges, raw                                 0.8
      Peaches, raw                                 1.3
      Pork, medium-fat, roasted                    0.8
      Potatoes, white, baked                       0.8
      Prunes, dehydrated, uncooked                 1.3
      Raisins, uncooked                            1.2
      Rice, white,  enriched, cooked                0.8
      Sardines, canned in oil, drained              1.4
      Soup, canned, vegetable beef, ready
        to serve                                   0.9
      Squash, winter, baked                        1.3
      Sugar, brown                                 0.9
      Tuna, canned in oil, drained                 1.0
      Turnips, cooked                              1.7
      Walnuts, black                               1.0
      Wheat flour,  white, enriched                 0.8
      Wheat flour,  whole grain                     1.0

III.  High Iron Content (>2.0 mg iron/100 kcal)

      Apricots, dried                              2.1
      Artichokes, cooked                           4.2
      Asparagus, cooked                            3.0
      Baby cereals, enriched                     >13.5
      Beans, green, cooked                         2.4
      Beans, kidney, cooked                        2.0
      Broccoli, cooked                             3.1
      Caviar, granular                             4.5
      Cereals, prepared breakfast
        enriched at 25% U.S. RDA per oz.            3.8
      Clams, hard shell                            9.4
      Coffee, instant, dried                       4.3
      Fish flour, from fish fillets                2.0
      Fish flour, from whale fish                 12.2
      Giblets, chicken, fried                      2.6
mg iron/100 g
edible portion
    of food
   1.4-4.0
     1.7
     1.3
     0.5
     2.3
     5.5
     2.3
     5.4
     4.3
     0.4
     0.6
     0.4
     0.4
     0.5
     2.9
     0.7
     4.4
     3.5
     0.9
     2.9

     0.3
     0.8
     3.4
     1.9
     0.4
     6.0
     2.9
     3.3
     5.5
     1.1
     0.6
   >50.0
     0.6
     2.4
     0.8
    11.8

    15.0
     7.5
     5.6
     8.0
    41.0
     6.5
                                      142

-------
TABLE 4-6
(Continued)
Food

Heart, beef, braised
Kidneys, beef, braised
Lettuce, iceberg
Lettuce, romaine
Liver, calves, fried
Molasses, blackstrap (third extraction)
Mung bean sprouted seeds, cooked
Oysters, fried
Peas, frozen, cooked
Soybeans, mature, cooked
Spinach, fresh, cooked
Squash, summer, cooked
Tomatoes
Turnip greens, cooked
Wheat bran, commercially milled
Wheat germ, commercially milled
mg iron/
100 kcal

  3.1
  5.2
  3.8
  7.8
  5.4
  7.6   .
  3.2
  3.4
  2.8
  2.1
  9.6
  2.9
  2.3
   .3
   .0
5,
7.
  2.6
mg iron/100 g
edible portion
    of food

     5.9
    13.1
     0.5
     1.4
    14.2
    16.1
     0.9
     8.1
     1.9
     2.7
     2.2
     0.4
     0.5
     1.0
    14.9
     9.4
                                  143

-------
     Employing previously discussed Estimates of the amount of iron required

to maintain balance, adult males in developed countries should be absorbing

about 6% of dietary iron, and adult females should require about 12%.  In

developing countries, 4% and 8% absorption, respectively, would seem adequate.

These requirements do not appear difficult to achieve when compared to absorp-

tion data obtained by the use of iron salts in fasting subjects.  However,

they are large when the limited availability of dietary iron is considered.

In the past, the total iron content of the diet has been emphasized, and there-

fore on the iron contents of individual foods.  However, it is not practical to

modify the iron content of a diet greatly by changing its composition except

through fortification, and food iron content per se: does not correlate well with

the amount of iron absorbed.
 Availability of Dietary Iron

      Until recently, direct attempts to measure dietary iron availability have

 had limited success.  Chemical balance studies have been technically difficult

 and imprecise.  To understand the availability of food iron in humans, it was

 necessary to develop a methodology capable of overcoming certain problems.
                                                                                186
 One complicating factor was the variation in iron absorption among individuals,

 which included differences in iron balance affecting the mucosal setting for

 iron absorption and meal-to-meal fluctuations.  The latter presumably was

 related to physiologic differences in secretions and in motility of the gastro-

 intestinal tract.  Differences caused by variations in iron stores were controlled

 by carrying out comparisons in the same individual, employing two isotopes of

 iron.  Sporadic variations were dealt with by studying about a dozen individuals

 of the same age and sex, or by administering repeated doses of isotope so as to
                                               103
 determine absorption from the number of meals.

      Employing  an adequate  experimental  design,     the  absorption of  iron from

 single  foods was evaluated.  When  compared to other  forms  of  iron,  hemoglobin
                                    144

-------
was found to be particularly well absorbed by humans, and more important,

was found to be uninfluenced by chelates that could block absorption of nonheme
     176
iron.     Heme is taken intact into the mucosal cell, where it is catabolized
                                               808
by heme oxygenase with the release of its iron.

     Many foodstuffs have been biosynthetically labeled and their absorption

measured, as charted in Figure 4-3.  Vegetal iron was poorly absorbed in normal
                                                                452
subjects (1-10%), whereas meat iron was better absorbed (5-20%).     When these

studies were extended to evaluate mixtures of tagged foods, it became apparent

that individual foods interacted, affecting the availability of all the different

forms of nonheme iron present.  Indeed, virtually all nonheme iron in a single

meal had the same availability.  When added to a complex meal, a tracer dose of

radioiron salt was absorbed in the same level as nonheme iron in individual
              187
food articles.     Similarly, a tracer dose of radioactive heme was found to be
                                                 320,454
absorbed to the same degree as heme iron in food.         These two extrinsic

tags provided the first accurate means of determining iron absorption from the

normal diet.

     The two-pool, intrinsic tag method (heme and nonheme) has yielded estimates

of food iron absorption that closely agree with estimates based on normal body

iron losses.  For example, meals composed of aliquots of all food consumed in a

typical 6-wk diet and doubly tagged with radioactive heme and nonheme iron were
                             77
administered to 32 young men.    The total daily intake of iron in these men was

17.4 mg, of which only 1 mg was in the form of heme iron.  Total absorption

averaged 1.25 mg/day.  Absorption of nonheme iron averaged 5.3% or represented

0.88 mg/day, whereas absorption of heme iron averaged 37% or 0.37 mg of iron.
                                                ,           454
Similar data were obtained by Layrisse and Martinez-Torres,    who fed their sub-

jects a meal of meat, black baansv maize, and rice containing a total of 4.5 mg

iron.  In their normal subjects, absorption from 1.5 mg of heme iron was 27%

(0.34 mg),  compared to an absorption of 3 mg of nonheme iron of 6% (0.12 mg) .

                                  145

-------

Dose of
food Fe
N° coses
20-
15-
-^ 10-

-------
Total absorption from the meal was 0.46 rag iron,,  These studies underline the



important contribution made by the heme iron to the diet.



     If individual components affecting the availability of nonheme iron in a meal



could be identified, absorption from any type of diet should be  predictable.  A



number of inhibitors and enhancers of food iron absorption have been identified.



Phytic acid is often considered a potent inhibitor of iron absorption; however, the



effects of phytate on iron utilization are difficult to interpret because of several



intervening factors.  For example, the number of iron atoms sequestered by the



phytate molecule have an effect.  Iron from certain iron-phytate complexes is less



available for hemoglobin formation than that from more soluble iron salts.   '   a>



    '      However, Morris and Ellis     recently reported that iron from monoferric



phytate is readily available to rats.  The research indicates  that much of the iron



in wheat is monoferric phytate.  Moreover, wheat iron is readily available to



rats     and humans.     Evidently different iron-phytate complexes have differing



availabilities, which must be taken into account when considering the effect of



phytate on iron availability.



     Another complication is that inorganic phosphates also depress iron utiliza-


     343
tion.     Thus, in experiments in which the total phosphorus content of the diet



is not constant, it is not possible to separate the effects of phytate from those



resulting from increased phosphate content of the diet.   a Phytate in vegetal food



sources (such as legumes) is often complexed to protein or other components of the



plant and may be less reactive than soluble phytate salts.  The addition of soluble



salts of phytic acid to diets has reduced iron availability, but naturally-occurring

                                                               f Q oa

phytates may not act the same way.  For instance, Sharpe e£ al.     found that adding



sodium phytate to milk reduced the iron absorption of human subjects.  However, when



the phytate was supplied by oats, no correlation was observed  between the phytate



content of the oats and iron availability.  Similarly,  Welch and Van Campen   a



found that the phytate concentration of soybeans was not correlated with the avail-
                                      147

-------
 ability  of  the  soybean  iron to rats.  Thus  it  is not at all certain  if naturally-



 occurring phytates have the same  influence  on  iron utilization as soluble phytate



 salts.



     Finally, recent reports of the effect  of  fiber on iron utilization have  fur-



 ther complicated analysis.  Several investigators have reported that substitution



 of whole-wheat  bread for white bread or the addition of increasing quantities of


 ,   „,     _  _,   ,. _  .,, ,       .     .....  _.    75,2803,4093,6883,8153
 wheat bran  to the diet  will depress iron utilization,,   '     '



 This phenomenon had been attributed to the phytate content of the whole mesl or


                                                                           643 a
 bran products.  Yet phytin-free fiber'can reduce iron utilization in humans     and



 rats.       Since wheat  bran contains:both fiber and phytate, it is hard to discern



 which is the critical factor in reduc'lhg iron utilization.  At present it is



 difficult,  if not impossible,  to assess the individual effects of fiber, phytate,



 and phosphate on iron utilization.



     Undoubtedly, a considerable number of substances may interfere with iron



 absorption.  Calcium and phosphate salts and ethylenediaminetetraacetic acid (EDTA),



 which are added to American diets as food preservatives, have been shown to reduce


                  192 543
 iron avsilability.   '     At  present, two food components have been shown to



 enhance absorption.  The availability of nonheme iron increases when the meal con-



 tains animal tissues.   '     For example, the substitution of 100 g beef for an



 equivalent amount of egg albumin in a test meal increased absorption more than


        191
 5 times.     Thus meat  is important not only as a source of heme iron, but as an



 enhancer of absorption  of nonheme iron.  Muscle from animals and fish and liver



have this property, whereas milk and cheese do note  Ascorbic acid is another



potent enhancer of iron absorption because it can reduce iron and form a chelate



with ferric iron at low pH, effects that maintain solubility at the higher pH of



 the duodenum.  Recent studies  have shown an enhancing effect on nonheme iron



 absorption by relatively small amounts of ascorbic acid that were either contained



 in or added to food during its preparation.     For example, 60 mg
                                      148

-------
 ascorbic acid added to a meal of rice more than tripled absorption of iron,




 and  150 g of papaya containing 66 mg ascorbic acid increased iron absorption


                                                    456
 more than fivefold  when taken with a meal of maize.     Such studies of




 dietary iron availability employing the extrinsic tag have clarified information




 on iron absorption and loss and pointed out that availability may be even more




 important than content in determing the amount of iron absorbed from food.



      Some limitations in the extrinsic tag technique as applied to nonheme




 iron absorption have become evident.  These shortcomings do not invalidate




 its use as described, but need to be considered in special situations.




 Certain iron salts used in fortification, such as pyro- and orthophosphates




 are incompletely miscible with the nonheme dietary iron pool.190 Certain forms




 of food iron—such as ferritin and hemosiderin with their masses of ferric




 hydroxide are also incompletely miscible—and it is presumed that soil iron


                      457
would be even more so.




 Balance and Stores



     From the information available on iron requirements and available iron in




 the diet, humans appear to be uniquely restricted in external iron exchange.




 Basal exchange in the adult male of 12 yg/kg/day may diminish to as little as



 6 yg/kg/day with iron deficiency, and the adult menstruating female, with an




 average requirement of about 20 yg/kg/day may decrease her daily requirement to




 about 15 yg/kg.  An iron-deficient American male should be able to absorb a




 maximum of about 50 yg of iron/kg/day from his average diet, whereas the iron-




 deficient menstruating female might be expected to absorb about 35 yg/kg/day.



 These availability estimates for either sex may be halved, if the diet does




 not  contain enough meat or ascorbic acid.  It is apparent that the male has a




 considerable margin of safety, whereas a major portion of the female population




 must be at risk.  A prevalence of a 4% iron deficiency in the adult male popula-



 tion  (most of which is related to pathologic bleeding), compared to a 20% iron

                                                                     184

 deficiency in the adult female population, reflects this distinction.     The




 inability of most pregnant women to meet iron requirements is acknowledged by




                                    149

-------
 the routine therapeutic administration of iron salts.  A high prevalence of

 iron deficiency is also found in infancy, in which requirements of growth out-
                              195
 strip available dietary iron.     There is no question that iron balance in

 menstruating and pregnant women and in infants is perilous.

     One of the best indicators of an individual's iron balance is the extent

of  ferritin and hemosiderin stores.  They are normally found equally divided
                                                                      749
among the hepatocytes, reticuloendothelial cells, and striated muscle.

These stores have been evaluated in various ways.  The most direct but least

practical method is phlebotomy, which mobilizes iron stores; approximately 1 g of

storage iron has been measured in the adult male and one-third that much in the
       42a,597,634
female.             Another approximation of storage iron is provided by
                                            707
examining a marrow aspirate for hemosiderin.     In healthy individuals, a

negative correlation has been shown between the hemosiderin of the marrow and
                                              800
the total iron binding capacity of the plasma.     However, the effect of other

factors, particularly protein malnutrition, on the iron binding capacity, vitiate

the usefulness of this measurement in evaluating stores.  One of the more in-

formative procedures has been the assay of a postmortem liver specimen for

nonheme iron.  The results of some 4,000 determinations in 18 different coun-
                                            160
tries have been reported by Charlton et al.,    and mean values for their male

subjects are summarized in Table 4-7.  Assuming that the liver iron of such

individuals represented about one-third of storage iron, estimates of the size

of body iron stores could be made as well as comparisons among different popu-

lations .
                                                                       401,535
     Immunologic methods have been developed to measure serum ferritin.

The values obtained in several of experimental situations showed a close rela-

tionship to iron stores.  That is,  the geometric mean of about 90 in males as

compared to 30 in females was proportionate to the difference between iron

stores in the two sexes.  A relationship has been shown between serum ferritin
                   188
and iron absorption    and between serum ferritin and iron stores as determined
              785
by phlebotomy.     Only in hepatic disease and inflammation does the ferritin
                                               478
level appear to deviate from this relationship.

-------
                                 TABLE 4-7
                   Hepatic Storage Iron Concentrations
                                                      a
                           Males
                                    Females
Countries
Great Britian
Sweden
Czechoslovakia
United States
South Africa
Caucasian
Bantu
No. of
Subjects
182
422
187
232

97
318
Mean, mg/kg
158
159
212
188

258
818
No. of
Subjects
128
300
187
121

42
133
Mean, mg/kg
156
129
182
144

161
270
Rhodesia




   Bantu




Nigeria
142
681
76
228
Bantu
India
New Guinea
Venezuela
Brazil
Mexico
87
203
110
204
74
151
189 96
112 67
106
203
187
198
187
79




 Adapted from Charlton et al.
                                     151

-------
INTERNAL IRON METABOLISM

     Internal iron exchange in mammals is dominated by iron requirements for

hemoglobin synthesis.  The iron content of the erythron is about 1 g/kg tissue,

some 2 orders of magnitude greater than the essential iron content of other

body tissues.  In the design of internal iron exchange, therefore, some means

had to be created whereby the disparate needs of individual tissue could be met

and excess iron could be stored, accomplished through the function of a plasma
                     548
protein, transferrin.     The phylogenic appearance of this plasma protein co-

incided with the appearance of hemoglobin in red cells.  When, as a genetic

oddity, transferrin is absent in a human, the unique finding of iron deficiency

anemia despite iron overload of other body tissues is observed.  This phenomenon
                                                                       304
is clear evidence of the protein's essential function in iron exchange.

     The first steps in binding appear to be the interaction of transferrin with

iron, and the release of three protons; the second step involves the inclusion

of bicarbonate in the iron transferrin complex and isf associated with the

development of a salmon-pink color.  The reverse reaction occurs with iron

release.  Iron binding above pH 7.2 is maximal, with a binding constant in plasma

                   24  -1 9
of approximately 10   M  ;  as the pH is reduced to < 6.5, iron starts to disso-
                                            721
ciate, and it is nearly completed at pH 4.5.     In vivo transferrin shuttles iron

back and forth between body tissues without being used up itself.  The behavior

of the two iron-binding sites has been the subject of some controversy: it is

undecided if they participate independently and equally in the binding and

release of iron, or if differences in affinity exist between them and certain binding
        8
tissues.   Recent evidence favors the homogeneous behavior of transferrin iron,
                                                                   625
with specific tissue receptors decisively determining the exchange.     The ability
                                                               7,730,828
of transferrin to bind a variety of trace metals is noteworthy.           However,
                                     152

-------
the patterns in which these metals are released are quite different from that

of iron and they are not assimilated by immature erythrocytes in any appreciable

amount.

     Current methods for measuring plasma iron largely exclude hemoglobin and
                                                             386a
are therefore assumed to represent isolated transferrin iron.      However, they

will measure part of parenterally injected preparations such as iron dextran.

The normal level of human plasma iron is about 100 yg/dl and saturation of trans-
                              89
ferrin with iron is about 33%.    Whereas transferrin concentration is relatively

stable from hour to hour and day to day, serum iron in the normal individual may

vary over 24 h between 50-200 yg/dl.  Diurnal variations are normal, with morn-

ing values at least 50% higher than evening values.  The higher levels of plasma

iron commonly found in animals are related to a much greater iron absorption,
                                                     185
and fasting will reduce the level to about 100 yg/dl.     Decreases in plasma

iron are associated with exhaustion of iron stores, inflammation, increased
                                                            89
erythropoiesis, and reduction in transferrin concentrations.    Elevations in

plasma iron are caused by increased absorption, hepatic damage, increased blood

destruction (in particular, ineffective erythropoiesis), and the parenteral

administration of iron.   Figure 4-4 sets forth relationships between plasma iron

and transferrin concentration observed in various diseases.

     The erythron contains 90% of essential body iron in most mammalian systems

and is therefore the primary user of transferrin iron.  Iron is taken up by the

immature red cell, incorporated into its hemoglobin, and remains within the

cell for its life-span.   Uptake initially involves the adherence of the trans-

ferrin iron complex to membrane receptors.  Then the complex is internalized

in microtubles and iron is removed, followed by the return of transferrin to
                       348,413
the surrounding plasma.          Within the erythroid cell, more than 80% of the
                                      153

-------
                              TRANSFERRIN
Unbound
\ 3



1

1

Hemolyfic anemia
(ineffective erythropoiesis) °
Idiopathic J
hemochromatosis
Reticuloendothelial \[
iron overload
Hypoproteinemia [ 4

Bound
Normal

g Iron deficiency

ra Infection

^^^^^ Hemolytic anemia

iililllll

>0^^^^^^^

56^^^^^^

^

                              Saturation (%)
FIGURE 4-«4   Relationships between transferrin and plasma iron in
             disease.  Reproduced from Bothwell and Finch.
                                                           89
                                154

-------
                                                                             586
iron is delivered to the mitochondria, where it is converted into hemoglobin.

Excess iron is deposited in the cytoplasm in the form of ferritin aggregates.

Although a coordination exists between heme and globin synthesis that keeps

their production balanced, the ratio of production of these moieties has little

effect on the uptake of iron by the developing red cell.  If hemoglobin synthesis

is reduced, excess iron will accumulate in ferritin stores with aberrant globin

synthesis or in mitochondria with abnormal heme synthesis.

     Usually, transferrin iron above 50 yg/dl provides the erythroid marrow with

the amount of iron it needs for hemoglobin synthesis.  However, the adequacy of

iron supply is influenced by changes in requirement by the erythroid marrow as well

as by fluctuations in plasma iron concentration.  As production increases, the

minimal level of plasma iron required to support erythropoiesis increases propor-
          404
tionately.     The adequacy of iron supply versus erythron needs may be monitored

by red cell protoporphyrin, which increases as the ratio of plasma iron supply to
                     441
erythron needs falls.

     Quantitative aspects of iron exchange between transferrin and the erythron
                                               270
have been studied in detail with iron isotopes.     The basal turnover of this

fraction depends on the life-span of the red cell and varies from about 3%/day

in mice to 0.8% in humans.  However, with the wastage iron of erythropoiesis,

iron turnover within the erythron is greater than that calculated from red cell

life-span.  Each day in the adult human, some 25 mg of iron is used by the
         270
erythron.     The anatomic location of the erythroid marrow may be visualized by
                                               770
the positron camera after injection of iron-52,    and the uptake and release of

iron may be measured by surface probes placed over marrow-rich areas such as the
       271
sacrum.

     The other component of the erythron iron circuit is the reticuloendothelial

cell, which processes the iron of senescent or defective red cells and returns it
                                   155

-------
to plasma transferrin.  The wastage iron of erythropoiesis is released to the

reticuloendothelial cell at the time of red cell maturation.  Within the reticulo-

endothelial cell, red cell hemoglobin is catabolized.  Its iron is either directed

to the cell membrane to be taken up by transferrin, an exchange in which cerulo-
                     637
plasmin participates,    or it is deposited within the cell as ferritin.  In
                                                                       266
dogs, about one-half of catabolized red cell iron goes by each pathway;    in
                                          267
humans, only one-third is held in storage.     The partition between these two

pathways is inconstant, and the variation is thought to be largely responsible

for the diurnal variation in plasma iron.  During inflammation, the greater

deposition of ferritin iron within the reticuloendothelial cell results in lowered
                              266,353
iron concentrations in plasma.

     An alternate pathway for transferrin iron uptake involves the liver and
                          270
other parenchymal tissues.     The hepatocyte accepts and stores iron when uptake

by the erythron decreases, as illustrated by in vivo profiles obtained in patients
                     270
with aplastic anemia.     Studies in the rat show some localization in muscles,

skin, and subcutaneous tissues, but the largest amount resides in the intestinal
                                                                               162
mucosa, reflecting the ability of that species to excrete iron through the gut.

The only other well-delineated pathway in humans is the exchange of plasma iron
                                        549
with the extravascular transferrin pool.     The total format of internal iron

exchange is shown in Figure 4-5.   The diagram does not illustrate the uptake by

hepatocytes of the hemoglobin iron released through intravascular hemolysis of
          351
red cells.

     Isotopic studies of plasma iron kinetics have provided a better appreciation
                                                                  150
of these various pathways and their functional interrelationships.     More

detailed data on kinetics is available for humans than for other mammals.

Iron turnover through the plasma is considerably greater than was expected for

the needs of the circulating red cell mass, an excess largely explainable by a
                                    156

-------
                              INTERNAL IRON EXCHANGE
                                  Circulating
                                 Erythrocytes
                       Gut
Extrovascular
  exchange
FIGURE 4-5   Internal iron exchange.   Reproduced from Hillman  and
              Finch.
                     365
                                  157

-------
                                 624
reflux of radioiron from tissues.     Part of this reflux represents the wastage

iron of erythropoiesis, amounting to about 24% of plasma iron turnover in humans
                                 189
with a half-time of about 7 days.     The second portion of this reflux, 8% of

plasma iron turnover with a half-time of about 7 h, relates to extravascular

exchange of transferrin iron.  In humans, absorbed iron comprises only about 3%

of plasma iron turnover and has little effect on internal exchange.  By contrast,
                                                                                 185
as much as 50% of plasma iron turnover in the rat may be derived from absorption.

Detailed ferrokinetic analyses of humans have shown that the nonerythron turnover

fraction, including extravascular flux and parenchymal uptake, is proportionate
                             189
to the amount of plasma iron.     Thus, by an empirical calculation of this frac-

tion and its subtraction from the plasma iron turnover, the amount of erythron

iron turnover can be ascertained accurately.  The erythron in a variety of dis-
                                                                                  270
orders has shown a large range of iron uptake: from 0-4 mg/kg (normal, 0.4 mg/kg).

Parenchymal uptake is more limited and reaches maximal values of about 0.6 mg/kg,

and only when plasma iron is high and transferrin is saturated.  It is not possible

to measure iron stores by short-term kinetic measurements, because the early

mixing of radioiron with stores is limited.  As erythron iron cycles through the

reticuloendothelial cell, it will mix with ferritin, but approximately a year
                                            308
is required for stores to reach equilibrium.     Hemosiderin stores may not

exchange iron.  Kinetic measurements revealed the difference between reticulo-
                                         270
endothelial and parenchymal iron storage.     Reticuloendothelial iron stores

are the products of red cell catabolism, whereas parenchymal iron stores are

created when elevated plasma iron leads to an increased hepatocyte uptake.  Hepa-

tocyte loading may be further augmented in hemolytic states when the hemoglobin

and ferritin are transported to them.
                                     158

-------
IRON BALANCE IN ANIMALS

     Information on iron balance in mammals is more fragmentary than data for

humans, but marked differences are known to exist that must be taken into

account when interspecies inferences are drawn.  For example, the rat has an

active excretory mechanism  whereby plasma iron is taken up by the gut mucosa
                                                         162,178
and lost from the body when mucosal cells are exfoliated.         Under basal

conditions, 10-15% of the plasma iron turnover follows this pathway, and the

amount may be increased when an animal is given iron supplements.  A rat's

dietary iron intake is greater than 100 times that of a human's when expressed
                    185
on a per kilo basis.     Absorption of nonheme iron is more efficient than in

humans, whereas heme iron is less well absorbed.  As much as 50% of the iron

entering a growing rat's plasma may be derived from absorption, and the plasma

iron is elevated to 2-3 times that of humans as a consequence.  Growth require-

ments are much greater, and erythropoiesis in the newborn rat is 2-4 times

greater than that of the fully grown animal when expressed per kilogram body

weight because of the requirements imposed by expansion of blood volume.  These

differences make comparison between rats and humans difficult.  The use of the

rat to evaluate iron availability of foods is equally suspect.  From the stand-

point of iron balance, rats are far better off than humans.  Not only is iron

deficiency virtually nonexistent, but it has been impossible to produce a pattern

of tissue damage similar to that seen in humans with iron overload.

     These differences observed in the rat apply to a greater or lesser degree
                        758
to other animal species.     Differences are to be expected in the adequacy

of iron balance at different ages, the rate of iron exchange, and the ability

to absorb heme versus nonheme iron.  Some animals, such as the pig, are born

with very low iron stores and others, such as the rabbit, with very adequate

reserves.  Animals generally seem to have little problem in maintaining iron
                                   159

-------
balance unless removed from their natural environment and placed on an arti-

ficial diet.

     Although information concerning iron metabolism of most farm animals is

limited, enough concern has arisen about their nutritional needs to define dietary

iron requirements. Swine appear to have the greatest problem in maintaining iron
                                                                         758
balance, because of their very rapid growth and low iron stores at birth.

Estimates of food iron required for baby pigs range from 60-125 mg iron/kg dry
                     521
diet.  Matrone e_t al.    reported that shoats fed a milk substitute needed
                                                                               98
125 mg/kg of food iron to maintain normal growth and hemoglobin.  Braude et al.

stated their estimates in another way: baby pigs must retain 21 mg of iron/kg

live weight increase to maintain normal iron status.  All these estimates of

iron requirements are complicated by the variable utilization of dietary iron.
              521
Matrone et al.    found that 30% of the iron in swine diets was utilized for
                                           756
hemoglobin synthesis, whereas Ullrey et al.    found that 82, 61, and 50% of

the food iron was utilized when the diets contained 25, 35, and 125 mg iron/kg
                                  98
diet, respectively.  Braude et^ al.   reported from 30-60% retention of orally

administered iron, according to the method of dosing.  As a result of these and

other studies, the National Academy  of  Sciences has  recommended  that  diets  for
                                                       569
growing pigs contain at least 80 mg iron/kg dry matter.     Requirements for

adult swine are not known; however,  because parasitism is not a problem,

natural feed ingredients evidently provide adequate iron intake.

     The iron requirements of poultry for growth have been studied.  Hill and
       364
Matrone    suggested on the basis of blood data that a diet containing 50 mg
                                                                       216
iron/kg would meet the chick's requirement.  Subsequently, Davis et al.

reported that 4-wk-old chicks required between 65 and 105 mg iron/kg diet to
                                   i
maintain normal weight gains and hemoglobin levels.  They later observed that

chicks fed a soybean protein or a casein-gelatin based diet required 75-80 mg
                                   160

-------
                217
iron/kg of diet.     Thus, it appears that diets for growing chickens should

contain a minimum of 50 mg iron/kg and that 80-100 mg of the element are re-

quired to provide an adequate safety margin.  Reliable estimates for mature

chickens are not available; however, the demand for iron by laying hens is

very high.  An average hen's egg contains about 1 mg of iron—thus laying hens

require some 200 mg iron/yr for egg production.  Although hemoglobin levels

will decline slightly with the onset of laying, no striking reduction of body iron
                                        638
stores has been measured in laying hens.     Because iron deficiency is not con-

sidered a practical problem in poultry, it should be safe to assume that diets com-

posed of natural feedstuffs supply enough iron for growing and laying chickens.

     Veal calves fed unsupplemented whole-milk diets rapidly developed anemia.
              520
Matrone et al.    fed calves a diet of whole milk that provided 30 or 60 mg

iron/day depending on whether or not iron supplements were added. Both diets were

adequate in maintaining normal growth and hemoglobin levels from birth to

40 wk of age.  In this case, 30 mg iron/day was sufficient for calves in the

age group studied; however, the dose is considerably smaller than the value of
                                      79                      99
100 mg/day suggested by Blaxter et al.    Bremner and Dalgarno   reported that

for calves on a fat-supplemented skim milk diet, 40 yg iron/g diet prevented

all but a very mild anemia when fed from 17 days to 11 wk of age.  The hemo-

globin levels of these calves were increasing at the end of the experiment, and

the authors speculated that they would have been normal had the dietary regime

been continued to the calves' normal slaughter weights.  They found that similar

results were obtained when iron was supplied as ferrous sulfate, ferric citrate,

or ferric EDTA; however, iron phytate was not utilized as well as the more solu-

ble iron compounds.  Requirements for mature cattle are not known; however, it

has been estimated that milking cows require 50-60 mg iron/day and that pregnant
                       432
cows need 60-80 mg/day.     These levels should be met by virtually any diet

normally consumed by these animals.

                                    161

-------
                                                                   450
     Data on iron requirements of sheep are limited.  Lawlor et al.    experi-

mented with sheep fed purified diets containing levels of iron ranging from

10-280 ppm.  They found that 10 ppm was clearly inadequate for growing lambs,

25 ppm did not support normal growth, and 40 ppm appeared to be adequate.  They

concluded that the minimal requirement for iron in growing lambs was in excess

of 25 ppm but no more than 40 ppm.  No experimentally derived estimates of iron

requirements for mature sheep are available; however, mature sheep need about
                     432
10-15 mg of iron/day.     This requirement is met when the animal's feed averages

10-15 ppm iron.  As forages and feed grains uniformly contain considerably more

than 15 ppm iron, healthy sheep fed normal diets should have no trouble in

meeting their iron requirements.

     There are several extensive compilations of feed analyses that include data
                536,567-569           757
on iron content.             Underwood    summarized the iron content of common

animal feeds: grasses, 100-250 mg/kg; legumes, 200-300 mg/kg; cereal grains,

30-60 mg/kg; and oilseed meals, 100-200 mg/kg.  Except for milk and milk by-

products, feed byproducts of animal origin contain high levels of iron.  The

following figures are representative of the iron content of some of these

materials: bloodmeal, 3,108 mg/kg; fishmeal, 381 (range,210-800); meatmeal,

439 (range, 180-920); dried skim milk, 52 (range,5-104) mg/kg dry material.

Additionally, many inorganic materials fed to animals as sources of calcium and

phosphorus—i.e., ground limestone, rock phosphate, ground oyster shells, and
                                                                                 757
dicalcium phosphate—contain high levels of iron, frequently more than 500 mg/kg.

The high levels of iron in common animal feeds explain why iron deficiency is not

generally a practical problem in farm animals, except in young animals consuming

only milk or milk-product diets.  Specific information, however, on the availa-

bility of these different iron salts is lacking.
                                   162

-------
                                CHAPTER 5

                             IRON DEFICIENCY


PREVALENCE

     Worldwide and in the United States, iron deficiency is the most

common nutritional problem and cause for anemia.  However, estimates of

prevalence are difficult to determine with precision because of problems

in defining what constitutes iron deficiency.

     Three stages of iron deficiency can be defined.  The mildest form,

depletion of iron stores, sometimes termed pre-latent iron deficiency, is

characterized by decreased or absent storage iron.  No other abnormalities

are identifiable.  To detect pre-latent iron deficiency, iron stores must

be measured in one of three ways: by phlebotomy to the point of iron-deficient
               339,785
erythropoiesis,        which yields a quantitative assessment of iron stores

(the amount of iron in the hemoglobin removed); by a semiquantitative esti-

mate derived from the amount of stainable hemosiderin in bone marrow
                      641
aspirates or biopsies;    or by measuring an increase in intestinal iron
                                                            344,830
absorption, which is correlated with decreasing iron stores.         Because

none of these methods has been practical for large population studies, con-

siderable interest has arisen in the recently introduced serum ferritin

measurements.  These sensitive imraunoradiometric assays appear to reflect
            691,718,760,761,785
iron stores.                     Although few reports have been published to

date on the efficacy of serum ferritin concentrations to detect iron deficiency
                            134,184,829
in large population studies,            the approach seems a promising one.

     The second stage, latent iron deficiency, ensues after iron stores are

depleted.  It is characterized by reduced serum iron concentration, elevated

transferrin concentration, and a resultant drop in transferrin iron saturation
                 35
to less than 15%.    Concomitantly, hemoglobin synthesis is impaired by iron


                                    163

-------
lack,causing erythrocyte protoporphyrin concentration to rise to more than
         441
70 yg/dl.     However, in this stage of iron deficiency, circulating hemo-

globin concentrations remain within the normal range, although treatment with

                                            290
iron results in slight hemoglobin increases.

     Overt iron deficiency anemia, the third stage, is triggered when the

restricted hemoglobin synthesis contributes to a measurable decrease in the

concentration of circulating hemoglobin or in the volume of packed erythro-

cytes (hematocrit).  Tests of large populations have been able to discern

only overt or latent iron deficiency.  Studies that add serum ferritin

measurements to transferrin saturations and erythrocyte protoporphyrin

determinations may detect the earliest stage of iron deficiency, which will

provide increased precision in defining iron deficiency in population sur-
     134,184,401,402,478
veys.

     Another difficulty in estimating the prevalence of iron deficiency in

population studies is the uncertainty about what constitutes an exact range

of normal hematologic values.  Anemia is difficult to define with precision

for different age groups.  Most "normal" values are arbitrary and represent

the best judgments of committees of experts; indeed, sharp demarcation of
                                       49
continuous variables may be impossible.    Another problem stems from

sampling biases:  populations with varying  susceptibilities to  iron

deficiency must be  included  to obtain  a valid distribution rather than

projecting population data from easily sampled groups.  Older studies often

suffered from the loose attribution of iron deficiency as the cause for most

of the anemias detected.  Multifactorial causes of anemia and the difficulty

of separating serum iron changes associated with infection or chronic

disease from those of iron deficiency contribute to the problems of inter-
                                                                     49
preting prevalence data.  These complexities are discussed by Beaton,   in

a thoughtful review of the epidemiology of iron deficiency.


                                  164

-------
     When values of transferrin iron saturation of less than 15% are used
                                  35
as a criterion of iron deficiency,   the prevalence of iron deficiency in
                                                                830
selected populations of various countries is shown in Table 5-1.     These

results—from relatively small numbers of subjects in each group (48-220)—

emphasize a population particularly at risk (pregnant women), but they also

show the incidence of iron deficiency in men and nonpregnant women.  The

study also demonstrates the discrepancy between iron deficiency as defined

by low transferrin saturation, and anemia as defined by hemoglobin levels

below arbitrary standards.  In addition, other or coincidental causes of

anemia, such as those based on deficiencies of vitamin 8,2 or folic acid,

are identified.  By the criterion of low transferrin saturation, 40-99% of

the pregnant women sampled from the 6 countries studied were iron-deficient.

Nonpregnant women showed 11.4-42.5% iron deficiency, and men showed a.range

of 0-8.8% iron deficiency.

     Other large population studies have confirmed the prevalence of iron
                                          829
deficiency in many countries of the world.     In the United States, infor-

mation about the prevalence of iron deficiency has been derived largely
                                 195
from patients seen by physicians,    but two large populations have been
         760,761
surveyed.         The first was designed to obtain information about nutri-
                                                                    761
tional deficiencies in a low-income population of the United States.

The second study, the "HANES" preliminary report, provides a more balanced
                                            760
appraisal of nutrition in the United States.     Again, discrepancies are

apparent between different criteria of anemia (such as hemoglobin and hema-

tocrit values) and raise questions concerning the standardization of methods,

the criteria of iron deficiency, and differences among populations studied.

Iron deficiency, estimated from a serum transferrin saturation of less than
                                                       760
15%, was observed in about 10% of children aged 1-5 yr.     Projecting this

percentage to the total population of children in the United States, the


                                  165

-------
                                TABLE 5-1
                          Prevalence of Anemia a
and of Iron, Vitamin Bi->, and Folate Deficiencies*

Country
Israel


Poland
India
(Delhi)

India
(Vellore)


Mexico


Venezuela



Subjects
Women
pregnant
nonpregnant
Men
Women, pregnant
Women
pregnant
nonpregnant
Women
pregnant
nonpregnant
Men
Women
pregnant
nonpregnant
Men
Women
pregnant
nonpregnant
Men

Propor-
tion
of cases
with
anemia, 7.
47.0
29.0
13.6
21.8
80.0
64.3
56.0
35.0
6.0
26.6
11.7
0.9
37.0
14.9
1.9




Proportion (2) of cases with the fol-
lowing deficienies
Serum
iron,
< 50 uR/dl
36.0
15.0
12.1
31.4
10.0
9.5
99.0
51.0
7.1
30.5
22.1
6.2
56.8
14.0
7.5
Trans -
ferrin
satur-
ation,
< 157.
46.3
11.4
8.8
40.0
51.7
25.8
99.0
42.5
5.1
61.2
28.1
3.6
59.7
18.9
0
Serum
vitamin
B12>
< 80 pg/ml
2.0
0
0
0.5
49.0
26.7
0
3.0
0
7.1
0
1.1
23.0
1.0
2.0
Serum
folatej
< 3 ng/ml
6.3
5.1
1.6
1.4


9.0
0
2.0
6.5
6.0
3.5
15.1
9.5
18.8
 The following hemoglobin concentrations per deciliter of blood are
 considered to indicate the presence of anemia:

         Children, 6 mo-6 yr         < 11 g/dl
         Children, 6-14 yr           < 12 g/dl
         Adult males                 < 13 g/di
         Adult females, nonpregnant  < 12 g/dl
         Adult females, pregnant     < 11 g/dl
^Derived from the World Health Organization.
830
                                    166

-------
authors estimated that a total of 383,000 children met this criterion of

iron deficiency.  Adolescent girls and women during their reproductive

years were far more likely to be iron-*deficient than men, consistent with

physiologic iron losses of menses and pregnancy (see Chapter 4).   Consider-

able additional information is included from separate analyses of subgroups

of different races and economic status.  However,  some information of

interest to epidemiologists is hidden by analyses  that define low trans-

ferrin saturation differently (
-------
CAUSES OF IRON DEFICIENCY

     Most cases of iron deficiency in humans are caused by insufficient

dietary iron, iron losses, or malabsorption of iron.  Although each of

these causes has important health implications, the problem of insufficient

iron assimilation is probably the problem most amenable to solution.

Insufficient Dietary Iron

     People fail to assimilate enough iron from the food that they eat

because of absence of iron in the food itself; insufficient quantities of

iron-containing foods; ingestion of cleaner food to which no extraneous

iron has been added during processing; interference with absorption because

of natural or artificial inhibitors of dietary iron absorption; and reduced

intake of substances that promote iron absorption.  Often a combination of

these factors will play a role in reducing the assimilation of iron from

food (see Chapter 4).

     As can be seen from Table 4-6, many foods are poor in iron.  Prominent in

this list are milk and milk products, many fruits and vegetables, sugar,

and salad and cooking oils.  There is little wonder that infants and small

children, whose major caloric intake is from cow's milk, are at risk of

iron deficiency.  Intake of mother's milk instead of cow's milk improves

the absorption of iron.  Selecting foods low in iron may

be a function of habit, economic necessity, or of religious and cultural

reasons, but the result is the same: an iron intake insufficient to balance

losses and eventual development of iron deficiency.

     The quantity of food consumed by most people is based on their caloric
             49
requirements.    Accordingly, men and boys eat more food than women and

girls.  Although iron content and availability vary considerably in indi-

vidual foods, the mixtures of foods people eat provide a fairly constant
                                                             49
iron concentration in relationship to caloric intake.  Beaton   has
                                 168

-------
tabulated this iron:kilocalorie ratio from calculated and measured patterns

of dietary iron consumption in many countries.   In the United States, the

iron concentration in milligrams per kilocalorie ranged from    4.9-6.3 in
         49                              760
one study   and was about 6.0 in another.      Iron intake calculated in re-

lation to energy requirements for men expending 3,000 kcal yields about

18 mg dietary iron from the average American diet.  Women whose energy

expenditure is approximately 2,000 kcal would be expected to consume about

12 mg dietary iron.  This latter value is  close to that actually measured in
                            50
a group of 97 Canadian women   and exceeded the mean value (9.9 mg) recorded
                            544
for 13 young American women.     Caloric intakes of children are related to
                                                   760
age, but vary within the range of 1,500-2,000 kcal,    which would provide

9-12 mg iron from diets similar to those consumed by adult members of a

family.  Thus, lower caloric intakes in women and children place them at

risk of insufficient dietary ingestion of  iron.

     The possibility that iron intake is somewhat dependent on the addition

of extraneous iron during food preparation was emphasized by the major dis-

crepancies found in calculated and chemically analyzed iron concentrations

of diets sampled primarily from institutional dietary kitchens in a variety
                     49
of foreign countries.    Contamination of  acid foods from iron cooking pots
                              545
has been demonstrated by Moore    and the  addition of iron to Kaffir beer
                                         83
consumed by the Bantu is well documented.     The converse, reduced iron con-

tamination of foods from the use of stainless steel and glass cooking vessels,

seems to be more widespread in developed countries where modern food processing

prevents marked contamination.        Several studies have showed close corre-
                                                                    50,544,832
lation when  iron content of diets from actual chemical measurements
                                                         794
is  compared to calculated iron content from food tables.
                                  169

-------
     Availability of iron in foods depends considerably on absorption

from single food sources as well as the interaction of foods on the ab-

sorption potential of the iron compounds they contain.  The graph of foods

labeled with radioactive iron shown in Figure 4-3 indicates a wide range
                                        451
of variation in absorption of food iron.     As a rule, iron in cereals

and vegetables is poorly absorbed, whereas iron from animal sources,

especially heme-containing compounds, is absorbed much better.  Furthermore,

absorption of vegetable food iron is almost doubled when it is ingested
                                  458,517,519a
with heme from meat, liver or fish             or when ascorbic acid is
      77,139,435,667
added.                Iron absorption reflects that of the least well

absorbed vegetable when two vegetables are mixed, and is the same as the

vegetable iron when up to 60 mg of a ferric or ferrous iron salt is mixed
                 78,187,454,455
with a vegetable.                Addition of a chelating agent (desferrioxa-
                                                  187,435
mine) diminishes the absorption of vegetable iron.         Heme iron compounds

from meat, liver and fish are not affected by desferrioxamine or ascorbic
     187,454
acid.

     These results have been interpreted as evidence of two major iron

pools, i.e., one a heme and one nonheme pool with the various iron compounds

in each pool having the same percentage of absorption.  Each pool is affected

by substances present in food that inhibit or enhance the metal's absorption.

Furthermore, iron compounds used for food fortification are absorbed to the
                                         187,455
same extent as foods in the nonheme pool.

     The distinction of dietary iron according to iron pools has permitted

measurements of iron absorption from complex dietary mixtures in which the

interactions of iron from many sources were once difficult to discern.  A

tracer dose of iron salt such as iron-59 chloride to label the nonheme pool
                                  170

-------
mixed with iron-55-labeled hemoglobin provides separate measurements of

the nonheme and heme pools of a dietary mixture and an accurate estimate

of percentage and total absorption from the two forms of iron.  This

extrinsic method of labeling also makes it possible to measure absorption

from iron compounds used for food fortification as well as from the naturally

occurring iron of foods.  Diets with an adequate total amount of iron but

containing little heme, which enhances absorption of the rest of the iron

intake, may not be adequate to meet nutritional requirements.  Thus, forti-

fication of foods with iron will be dependent not only on the amount of iron

added, but also on the overall availability of iron in the diet.
     Natural dietary inhibitors that decrease iron absorption include phy-
                  343
tates, phosphates,    oxalates, and carbonates, which form insoluble pre-
          760                              231      76,455,518
cipitates.     Substances in tea (tannins),    corn,           milk and
    76,139,546
eggs           also reduce absorption.  Artificial substances that may

interfere with iron absorption include ethylenediaminetetraacetic acid
                                                       192,278
(EDTA), which is added to many foods as a preservative.

     The antacids given with various types of therapeutic iron salts reduce
                          247,339,667
absorption of the mineral.             No information is available about

the effects of antacids on food iron absorption per se, although decreased

absorption of the nonheme component would be likely.  Some types of clay
        538
and dirt    consumed accidentally or intentionally by subjects with pica

will reduce iron absorption.  Pancreatic secretions (probably because of
                                                       56
the bicarbonate component) also reduce iron absorption.

Iron Losses

     Iron balance in the human subject can be upset by iron losses of two

sorts: common physiologic losses that exceed iron intake and pathologic

iron losses.  Iron deficiency may be a consequence of either.
                                  171

-------
     The obligatory physiologic losses in the urine, feces, sweat, and

desquamation of cells from epithelial surfaces are discussed in Chapter 4,

as is normal menstrual blood loss.  Unrecognized menorrhagia is one of the

most frequent causes of iron deficiency in women during their reproductive

years.  Using analyses of subtle changes in hemoglobin concentrations, mean

corpuscular hemoglobin concentrations, and serum iron levels, Scandinavian

investigators concluded that the upper range of normal menstrual losses was
               321
about 60-80 ml.     Above this level, bleeding was labeled abnormal or
                                                                        659
menorrhagic.  Factors increasing menstrual flow include genetic factors,
                    658                                    838
multiple childbirth,    and the use of intrauterine devices    for contra-
                                                              171,584
ception.  Oral contraceptives tend to reduce menstrual losses.

     As mentioned, pregnancy and delivery contribute to physiologic iron

losses.  Lactation depletes the maternal iron stores by approximately
               195
0.5-1.0 mg/day,    which may be partially counterbalanced by the absence of

menses during lactation.

     When pathologic causes of iron loss are examined, most of them

involve blood loss.  In sheer numbers, the menorrhagia suffered by 10-25%

of women is the most common form of abnormal blood loss throughout the world.

Other obvious forms of blood loss are nosebleeds, wound bleeding, hematuria,

and bleeding hemorrhoids. Less  overt  are the many sources of gastrointestinal

bleeding (shown in Table 5-2) that cannot be detected solely by clinical means.

In men and postmenopausal women, occult blood loss into the gastrointestinal

tract is the most common cause of iron deficiency.  The frequency of lesions
                                                         69
responsible for this bleeding was estimated by Beveridge.    He found that

hemorrhoids were responsible in 10% of the cases, salicylate ingestion in

8%, peptic ulceration and hiatus hernia in 7% each, diverticulosis in 4%,

neoplasms in 2%, and undefined causes in 16%.
                                   172

-------
                              TABLE  5-2
            Sources of Blood Loss in the Alimentary Tracta
Esophagus




     Varices




     Hiatus hernia









Stomach




     Varices




     Ulcer




     Carcinoma




     Gastritis




     Le iomyoma




Small bowel




     Ulcer




     Aberrant pancreas




     Meckel's diverticulum




     Telangiectasia




     Polyp




     Carcinoma




     Regional enteritis




     Helminthiasis




     Vascular occlusion




     Intussusception




     Volvulus




     Le iomyoma
Biliary tract




      Trauma




      Cholelithiasis




      Neoplasm




      Aberrant pancreas




      Ruptured ancurysm




      Intrahepatic bleeding




Colon




      Ulcerative colitis




      Amebiasis




      Carcinoma




      Telangiectasia




      Diverticulum




Rectum




      Hemorrhoids




      Ulceration




      Carcinoma
  aFrom Fairbanks et al.
                        264
                                 173

-------
     Infestation with hookworm is a common cause of blood loss and iron
                                 459
deficiency in tropical countries.     It is estimated that a worm load

sufficient to produce 1,000 eggs/g of feces will result in the loss of

2.1 ml blood (0.8 mg iron)/day when Necator americanus is the parasite
                                                                           652
and 4.5 ml blood (1.8 mg iron)/day when Ancylostoma duodenale is the agent.

Further studies set an upper limit on the degree of parasitism -~ 2,000 eggs/g

feces in women, and 4,000-5,000 eggs/g in men--that would not undermine the hemo-
                                      460
globin concentrations of a population.     Parasitism of greater degrees pro-

duced an anemia directly correlated with egg counts because blood and iron

losses exceeded the compensatory iron absorption mechanisms.

     Other less common forms of abnormal iron loss include blood donation

and disorders that produce intravascular hemolysis followed by iron loss as

hemoglobin or hemosiderin in the urine.  Blood donation results in the loss

of about 200 mg iron per unit removed (assuming a 450-ml unit of blood with an

average hemoglobin concentration of 14 g/dl).   Repeated blood donations at

intervals of about two months can be tolerated by many male donors whose body '
                         597
iron stores are depleted,    but who avoid anemia by increasing their absorp-
                                               323
tion through a generous intake of dietary iron.     Women donors, at risk

because of menstrual losses as well as the lesser intake of dietary iron,
                                                                            282
frequently develop iron deficiency unless they receive iron supplementation.

     Iron can be lost through hemosiderinuria resulting from intravascular

hemolysis.  In intravascular hemolysis, depletion of haptoglobin leads to

filtration of hemoglobin dimers through the glomerulus and iron deposition

in tubular cells.  These cells, laden with hemosiderin, are later excreted

into the urine.  Although uncommon in the general population, this process

may induce profound iron deficiency in patients with paroxysmal nocturnal
                                 174

-------
hemoglobinuria, prosthetic heart valves, atrial myxoma, or malaria.  Urinary
                                                                     262
iron losses from intravascular hemolysis may be as high as 34 mg/day.

     An even rarer cause of iron deficiency is pulmonary siderosis, brought

on by hemorrhage into pulmonary alveoli.  Iron, deposited as hemosiderin,

remains in the alveoli and is not resorbed into the body to any major degree.

Thus, large deposits of iron may be found in the lung although the individual

is iron-deficient.

Malabsorption

     Compared to blood loss, impaired gastrointestinal absorption of iron is

a rare cause of iron-deficiency anemia.  Two major types of impaired iron

absorption are recognized.  One results from gastrointestinal surgery and

the other from various disorders commonly lumped together under the term

"malabsorption syndrome."
                                                                           2
     Iron-deficiency anemia occurs in most patients after total gastrectomy

and in 10-50% of patients who have a partial gastrectomy, especially those
                                                               36,324,368
who undergo the Billroth II surgery that bypasses the duodenum.

Typically, the anemia develops progressively during a 10-year period after

the operation and, predictably, it is more frequent and severe in menstru-
            36
ating women.    Studies with iron radioisotopes indicate that food iron
                       324,368,616
absorption is impaired;            the subjects' absorption of medicinal

iron was normal but did not increase appropriately in response to the
       38,324,700
anemia.

     In malabsorptive syndromes such as celiac disease caused by gluten entero-

pathy, epithelial cells are damaged and villi in the proximal small intestine
                                                                               31,32
atrophy.  Consequently, iron and various other nutrients are not well absorbed.

Clinically detectable improvement following treatment with a gluten-free diet
                                                                     31,138
results in improved iron absorption .                in most patients.
                                  175

-------
DELETERIOUS EFFECTS OF IRON DEFICIENCY

     Several tissue changes and metabolic abnormalities have been observed

in association with iron deficiency in human beings and in experimental

animals.  In some cases the causal relationship of iron deficiency is clear,

but in many instances the association is merely inferential.  In the follow-

ing discussion, an effort will be made to distinguish between the two.

Specific Tissue Changes

     Iron deficiency, despite its wide prevalence and its associated morbidity,

rarely causes death.  Consequently, information from autopsies is scarce
                                    723,823
about     pathologic tissue changes.         More information is available

for tissues that can be studied readily through biopsy during life, such as

peripheral blood lymphocytes, bone marrow, the oral cavity, and the gastro-
                 394
intestinal tract.     These tissues contain rapidly proliferating cells that

are more likely to show a lack of iron earlier than cells with a slow turn-
     207
over.

     Bone marrow.  The most obvious change in the bone marrow of an iron-

deficient patient is the absence of sideroblasts and of stainable iron

granules in reticuloendothelial cells as storage iron deposits become
          288,331,641
mobilized.             The erythroid precursors in a hyperplastic marrow
         203,822
increase,        but hypoplasia may also be observed, accompanied by in-
                                              67,126
creased numbers of reticulum and plasma cells.        Poor correlation

often exists between the degree of severity of the anemia and the degree of
                      67,784
erythroid hyperplasia.        Normoblasts with ragged edges and reduced
                                                      822
cytoplasm are characteristic of iron-deficient marrow.     However, most of

these changes are too general or variable to be pathognomonic.

     Gastrointestinal tract.  Lesions connected with iron deficiency have

been described in all portions of the gastrointestinal tract.  The predomi-

nant lesion is mucosal cell atrophy, which sometimes includes a component


                                 176

-------
                                                                      395,396
of mononuclear cell infiltration.  In cells of the oral mucosa, Jacobs

found abnormally thin epithelia, increased mitoses in the prickle cell layer,

decreased melanin, greater numbers of inflammatory cells in the subepithe-

lia, and keratinization.  Similar changes have been described in biopsies
                                            37
of tongues with atrophied filiform papillae.     Both lesions are reversed
                                                                 37
with iron therapy, although the tongue changes require 7-14 days.

     The clinical manifestations of these lesions in the oral cavity include

angular cheilitis, a sore red tongue, pallor, and often soreness of the

buccal mucosa.  These changes are not specific, however, because similar

abnormalities are exhibited in disorders such as deficiencies of folic acid

and vitamin B  .

     Additional lesions in the hypopharynx associated with iron deficiency

are mucosal atrophy and web formation in the post-cricoid area of the

pharynx and upper esophagus.  This syndrome, characterized by difficulty in

swallowing, is termed sideropenic dysphagia.  When the clinical features of

gastric achlorhydria and gastric atrophy are involved, especially in middle-

aged women, this constellation is termed the Paterson-Kelly or Plummer-Vinson
         418,542,604,723,774
syndrome.                     In some patients (5-15%), the disorder has

                        5,163,382,392,397,835
progressed to carcinoma.                       However, many investigators

believe  that the causal connection of iron deficiency with the various mani-
                                                               164,165,252,
festations of the Plummer-Vinson syndrome is far from explicit.
392,397,398,835
                 Especially troublesome  to clearcut explanation is the geo-

graphic  distribution of the syndrome, which predominates in Northern Europe

and  the  United States.  Almost no reports of it exist from Africa, South

America,  or the Far East, areas where iron deficiency is far more common.

Moreover, similar lesions have been  described  in the absence of any current
                                     252
or past  evidence of iron deficiency.     Whether factors other than iron
                                  177

-------
 deficiency predispose to the post-cricoid webs  and gastric atrophy is a

 moot question,  particularly because the incidence of Plummer-Vinson syndrome

 has decreased in recent years.

      Gastritis, gastric atrophy,  and diminished acid and intrinsic factor
                                                                           33,
 secretions are  additional abnormalities related to iron deficiency anemia.
 214,397,399,467,823
                      Histamine-fast achlorhydria has been found  in about 50%
                                         399,467,713
 of  patients with iron-deficiency  anemia,             and an even  higher inci-
                                      33,214,412,466
 dence of  gastritis  has  been reported.                 Whether  the gastritis

 and gastric atrophy are a cause or  a consequence of the iron  deficiency is
           33,214,466
 debatable;            either explanation is possible for individual cases.

 Iron therapy has improved the appearance  of the  gastric
                                                               33,214,399,467,713
 lesions and returned acid secretion to  normal in some patients,
                         126,468
 especially  young people,        chronic cases,  and patients whose  gastric

 atrophy is  not  severe.   However,  recovery is uncommon in patients  older than
    399,467,713
 30.              Antibodies  circulate to gastric  parietal cells in  about a

 third of  patients with  iron-deficiency  anemia and  histamine-fast achlorhy-
      203a,513
 dria.           Again, the causal  relationship of antibodies to iron  deficiency

 is  obscure.


     Atrophy of small intestinal mucosal cells in children with iron-
                                     314,562
deficiency anemia has been described.         Treatment  with iron restored
                                                39
these mucosal lesions to normal.   However, Baker   criticized the adequacy
                                        562
of the biopsy specimens in these studies    because he could detect no marked

morphologic abnormalities in 20 patients with severe iron-deficiency anemia.
                                        423
The observations of Kimber and Weintraub    may be of greater importance.

They found intestinal malabsorption of radioiron by severely iron-deficient

children and puppies, a condition reversible after iron repletion.  They

attributed the phenomenon of malabsorption to decreased iron-containing

intestinal enzymes.

                                  178

-------
     Integument.  The progressive changes in fingernails from brittleness,

thinning, flattening to eventual concavity (koilonychia or spoon nails) are
                                                                       372
associated with iron deficiency and respond completely to iron therapy.

Decreased cystine content of the nails has been reported in individuals with
            406
koilonychia.     It is unknown whether iron deficiency interferes with cystine

metabolism or if another defect alters its intake or metabolism.  Koilonychia

is rare in the United States and is no longer a common disorder in other
                                                                        332
countries.  Hair loss has been associated with iron deficiency in women,

but such a symptom is not very specific.

     Neurologic changes.  Various neurologic symptoms have been attributed

to iron deficiency, including numbness, paresthesia, and pain, yet all with-

out objective clinical findings.  However, papilledema and increased intra-

cranial pressure not due to other causes and which disappear with iron
                            145,431
therapy are well documented.

     The high concentrations of nonheme iron found in certain parts of the

brain (substantia nigra, globus pallidus, and the red nucleus) are of the
                                 325
same order as those in the liver.     Some of this nonheme iron is ferritin;

the rest is not clearly characterized by either type or function.  In the

development of iron deficiency, nonheme iron concentrations in the brain do not
                                            325
drop, as do the nonheme hepatic iron stores.     Yet a brief period of

severe iron deficiency in the young rat results in a deficit of brain iron

that persists in the adult animal, despite a subsequent adequate intake of
     212
iron.

     Nasal mucosal atrophy (ozena).  Ozena, a disorder that occurs with iron
                                         57,58
deficiency in eastern European countries,      is practically unknown in the

United States and the rest of the world.  The association of atrophic rhinitis

with iron deficiency may be fortuitous, as others have reported failure of
                                      554.
iron therapy to correct the condition.

                                  179

-------
Metabolic Defects

     No exact measurement of organ or tissue dysfunction exists that

clearly divides the effects of iron deficiency from those of anemia.  The

question of whether or not patients without anemia but with definite deple-

tion of stores and possible depletion of some key tissue iron component
                                                              68
have symptoms related to their iron depletion is unresolvable.    Clinicians

have observed two discordant patterns.  Some patients without overt anemia

but with depleted iron stores and borderline lowering of serum iron concen-

tration and transferrin saturation exhibit symptoms of profound fatigue and

apathy that are alleviated after iron therapy.  Yet other people with

moderately severe anemia and clearcut iron deficiency may be vigorous,

asymptomatic, and completely unaware of their anemia.  Faced with these

inconsistencies and uncertain about which key iron-containing biochemical

systems to study, investigators have been relatively slow to probe the

metabolic defects of iron deficiency with the same enthusiasm accorded to

the more easily and successfully measured aspects of the anemia.


Biochemical  Compounds and Reactions Altered by Iron Deficiency

     Heme proteins.  Heme proteins, which function in the process of oxi-

dative metabolism,  include hemoglobin, myoglobin, the cytochromes, catalase,

and peroxidase.  Because iron is central to the function of these proteins,

its insufficiency would be expected to affect all cells profoundly.  In the

absence of adequate iron, hemoglobin and myoglobin synthesis is decreased,

and the function of the heme proteins is changed quantitatively rather than

qualitatively.  Much evidence confirms the decrease in hemoglobin in iron-

deficient subjects; little is known about the effect of iron deficiency on

human myoglobin.  Myoglobin concentrations in various muscles of experimental

animals respond differently to iron deficiency.  In rat skeletal muscle,
                                  180

-------
                                           210
myoglobin concentration markedly decreases,    but the concentration of myo-

globin in the animal's cardiac muscle is unchanged, and absolute amounts of
                                                    210
myoglobin appear to increase during iron deficiency.     This phenomenon may

be related to the workload of the organ, because myoglobin is similarly spared

in the diaphragm.

     Cytochromes function in the electron transport system.  Cytochromes

a. (cytochrome oxidase), b_, and £ are located in the cristae of mitochondria

and are responsible for transforming cellular energy into adenosine tri-

phosphate.  Cytochrome P450 is located in microsomal membranes of the liver

and aids in the oxidative degradation of drugs and various metabolites.

Cytochrome b^ is found in many membranes; it functions in a variety of

metabolic activities unique to its cell of residence.  Levels of cytochrome

activity in iron deficiency vary in different organs.  Biochemical and

histochemical measurements have shown that cytochrome oxidase is reduced
                                                               204,212a,393
in the buccal and intestinal mucosa of iron-deficient patients.

In iron-deficient rats, the enzyme is markedly decreased in intestinal

mucosa and skeletal muscle, virtually unaffected in brain and heart
                                                           207
muscle, and only slightly below normal in kidney and liver.     The micro-
                                                                            209
somal cytochromes P450 and ^ are hardly affected by severe iron deficiency.

     Catalase and peroxidase are involved in the reduction of endogenously

produced hydrogen peroxide.  These enzymes are widely distributed in many

cells, and they are usually contained in small organelles such as peroxi-

somes of liver cells or granules of leukocytes.  Congenital absence or

marked deficiency of catalase is remarkably asymptomatic in many people
                                                                           1
with acatalasia, although others are afflicted with necrotic mouth lesions.
                                                                   42,509,661
Human red blood cell catalase is lowered in iron deficiency anemia;

measurements in other tissues have been made in experimental animals.  Rat
                                                 65
liver catalase remains normal in iron deficiency.
                                  181

-------
     Metalloflavoproteins.  Metalloflavoproteins include succinate dehydro-

genase, a-glycerophosphate dehydrogenase, and NADH-dehydrogenase of mito-

chondria, and monoamine oxidase, xanthine oxidase,   and aldehyde dehydro-

genase in the cell sap.  These enzymes all contain nonheme iron and are

present in many tissues of the body.  Knowledge of their activities in iron

deficiency is virtually nil in human studies and meager in experimental
        30,66,725
animals.

     Enzymes with iron as a cofactor.  This class of enzymes includes

aconitase (citrate or isocitrate hydrolase).  They are deficient in whole
                                                724
blood and leukocytes of iron-deficient patients,    and are normal or
                                                       64
decreased in various tissues of the iron-deficient rat.    Hydroxylation

of proline and lysine in collagen synthesis is dependent upon a poorly
                                        635
identified iron-requiring enzyme system.      The ferrous iron chelating

agent a, a -dipyridyl interferes with collagen synthesis when given
                           812
parenterally to normal rats    and iron-deficient rats show a similar deteriora-
     167
tion.     Impaired collagen formation has been proposed as a possible cause
                      724
for poor wound healing    that accompanies iron deficiency.  However, it
                                                                            510
is more likely that the defect is a nonspecific consequence of the disorder.

     Other enzyme systems.  Enzymatic activities in which iron is neither a
                                                                           51
component nor a cofactor may be diminished in iron-deficient rats.  Becking

showed a 50% reduction of the activity of glucose-6-phosphate reductase,

phosphogluconate dehydrogenase, and malate dehydrogenase in iron-deficient

rats compared to normal control animals.  Iron deficiency led to reduced

disaccharidase activity in the brush border of dog intestinal epithelial
      369                                                                354
cells.     DNA synthesis  in marrow is inhibited in human iron deficiency,
                                                        650
and a similar impairment has been observed in HeLa cells    treated with

the chelating agent desferrioxamine, perhaps because of interference with
                          370
ribonucleotide reductase.
                                  182

-------
Mitochondrial Damage

      Injury  to mitochondria in several tissues of iron-deficient experimental

animals and  humans has been demonstrated by electron microscopy and other
                    207-209,302,409,647
analytic  techniques.                     The mitochondria are uniformly

enlarged  and radiolucent.  In human bone marrow they also appear to be in-

creased.            Similar findings in the heart muscle of iron-deficient

rats  have suggested that greater numbers of enlarged mitochondria may account

for the cardiac hypertrophy in excess of that caused by the increased work-

load  imposed by the anemia.  Mitochondria from the small intestine of iron-
                                      647
deficient rats are abnormally fragile.     Swelling, vacuplation, and breakdown

of cristae in mitochondria from lymphocytes of patients with iron-deficiency
      409                                                       209
anemia    and in the hepatocytes of severely iron-deficient rats    attest to

the diffuse  abnormalities produced in these important subcellular organelles.

However,  translation of these morphologic abnormalities into specific func-

tional defects has not been accomplished.

Functional Abnormalities

      Although many symptoms of iron deficiency remain unexplained,  measur-

able  functional abnormalities associated with iron deficiency have been

documented in humans and experimental animals.  Most of these abnormalities

involve complex interactions between a number of organs and biochemical

mechanisms.  Iron deficiency impinges upon growth, physical and mental per-

formance, reproduction, and susceptibility to infection.  Descriptions of

these and other conditions are supplied below.

     Growth.  The effect of iron deficiency on growth rate appears to have

been  studied much more carefully in laboratory animals than in humans,

perhaps because of the difficulty in separating nutritional iron deficiency
                                                                206,211,523
from other dietary insufficiencies in children.   Studies in rats

indicate that depressed growth is a late consequence of iron deficiency


                                183

-------
that appears long after the onset of severe anemia.  Two studies found no
                                                                     48,135
association between hemoglobin concentrations and weight in children,

although addition of iron to the diet of male infants produced an increase
                                                          412a
in weight gain not seen in female infants.  Judisch e_t^ ad.     observed two

patterns of infant growth.  The first was a rapid weight gain that slowed

when iron deficiency developed; the second was a normal weight gain that

accelerated when iron was supplied.  They concluded that iron administration

was associated with increased body weight in infancy, but they were uncertain

if such a weight gain were desirable.

     Physical performance.  Although lassitude and tendency to be fatigued
                                                                       68
easily are common in patients with iron deficiency even without anemia,

documentation of decreased physical performance from iron deficiency alone

is rare.  It has been difficult to define for humans effective endpoints
                                                       539
of tests of endurance or maximal response to workloads.
                  215
     In one study,    reduced heart rate following exercise was observttd after
                                                                    194
correction of iron-deficiency anemia with iron therapy.  In another,

pulmonary efficiency was depressed during fairly vigorous exercise in

patients with iron-deficiency anemia.  Respiratory function was improved by
                                                  775                   293
iron therapy but not by placebo.  Viteri and Torun    and Gardner et al.

showed that the capacity for physical work was impaired even with small

(1-2 g/dl) decreases of hemoglobin concentration within the normal range.

However, Hallberg and his colleagues produced iron deficiency by phlebotomy

in otherwise normal student volunteers and were unable to detect any appre-

ciable differences in the performance of various standardized physical tasks

between the phlebotomized subjects and controls (personal communication,

L. Hallberg).
                                  184

-------
     In rats running a treadmill, forced and voluntary running time declined

with iron-deficiency anemia 25-30% below that of normal controls.  Perform-

ance improved 3 days after iron therapy and was restored to normal 7 days
                244                   272
after  treatment.     In another study,    the treadmill running of iron-

deficient rats was markedly impaired as compared to normals.  The effects

of anemia were eliminated  by equalizing the hemoglobin concentrations in

both groups of animals.  Attempts to identify a biochemical defect in

skeletal muscle disclosed that the rate of mitochondrial oxidative phos-

phorylation with a-glycerophosphate as substrate was uniquely responsive

to iron replacement and paralleled recovery of treadmill performance in

iron-deficient rats treated with iron.  Reduced concentrations of cytochromes

and myoglobin and low rates of oxidative phosphorylation with other sub-

strates were observed in these iron-deficient rats, but none of the functions

improved rapidly after iron administration.

Spontaneous basal activity diminished in rats in the presence of iron-
                  297
deficiency anemia.     This reduction was independent of the severity of

anemia, and it was rapidly reversed after treatment with iron.   In school
                                         289
children studied by Gandra and Bradfield,    daily energy expenditure was

not dependent upon the degree of anemia.  A similar observation was made in
                                                              39
studies of iron-deficient and normal children in an orphanage.

     Mental performance.   Efforts to show quantitative reduction in tests

of scholastic achievement or other mental performance in relationship to
                                                          623
iron deficiency have been notoriously difficult to assess.      Significantly

lower test scores were reported in a group of iron-deficient adolescent boys
                                                        798
and girls compared with matched normal control subjects.     Another study,
                              376
reported only in summary form,    compared 2 groups of black children aged

3-5 from low-income areas of Philadelphia.   One group was anemic (9-10 g
                                 185

-------
hemoglobin/dl blood), and the other was injected with enough iron to raise

the hemoglobin concentration to 12 g/dl.  Comparisons of the two groups by

a battery of tests indicated that iron-deficient children exhibited strikingly

decreased attentiveness, more aimless manipulation, narrower attention span,

and less complex and purposeful activity when compared to their normal counter-

parts.  However, no significant differences were revealed by their scores on
                                                             720
the Stanford-Binet and Goodenough intelligence tests.  Sulzer    reported
                                                                        251,253
similar conclusions from a different group of psychologic tests.  Others

have been unable to detect any differences in the performance of psychologic

tests in iron-deficient and normal patients even when patients served as
                   251
their own controls.

     An unusual behavioral disturbance of unknown etiology seen in iron

deficiency is pica, or the habitual desire for unnatural articles of food.

A long list of specific cravings has been recorded, including earth or clay,
                                                                264
laundry starch, ice, olives, writing chalk, and cigarette ashes.     Pica in

adults usually disappears soon after treatment with iron and may recur repeat-

edly with the return of iron-deficiency anemia.  How iron deficiency is related

to this craving is completely unknown.  A similar behavioral disturbance,
                                                                   .374
breath-holding compulsions,  has been attributed to iron deficiency,     but no

direct cause and effect relationship has been proved.
     Observations of decreased monoamine oxidase activity in iron-deficient
    725,726           137,837
rats        and humans        raise the possibility that this enzyme may

be responsible for some of the behavior disorders of iron deficiency.

Lack of monoamine oxidase in the brain of an iron-deficient patient might

simulate  the effect of drugs given to inhibit monoamine oxidase and produce

side effects of restlessness and irritability.  To date, these relationships

are completely conjectural.


                                186

-------
     Reproduction.  Because iron-deficiency anemia is especially common in

pregnant women, it is surprising that little information is available about

the effects of iron deficiency on fetuses or on a mother's ability to carry

the pregnancy to term successfully.  No available evidence exists that iron

deficiency in a population has a major effect on the ability to conceive
                          633
or on early fetal wastage.     The suggestion that iron deficiency in the

mother leads to iron deficiency in the infant resulting from depleted
                    714                  224,284,445,717,827
iron stores at birth    has been refuted.

     The fetus apparently has first option on the maternal plasma iron.

Placeatas from severely iron-deficient mothers weigh more than those from
                                                  53
women who received adequate iron during pregnancy.    Such hypertrophy

might result in greater blood flow to the fetal placenta, but this propo-

sition has not been proved.

     Susceptibility to infection.  The relationship between human iron

status and susceptibility to infection is controversial and has been

                                 722
thoughtfully reviewed by Sussman.     Almost all pathogenic microorganisms

have an absolute requirement for iron and their interaction with a human

host may be affected by the host's iron status.  Evidence can be presented

that iron deficiency both increases the risk of infection and that it protects
                                                    722
against it; the case for neither view is compelling.     Considerable addi-

tional information is required about the effects of iron on bacterial growth

and virulence, host defense, and the interaction between iron and its phy-

siologic binding proteins, transferrin and lactoferrin, before the connection

between iron and susceptibility to infection is clarified.*

     Potential for absorbing other metals.  Iron deficiency in the rat

enhances absorption of a variety of metals.  Manganese absorption is
                             230,529,621,736
increased in iron deficiency,                and neither metal affects the
*See also Chapter 2.
                                    187

-------
                              736
rat's absorption of the other.     However, when manganese was perfused into

the duodenums of human subjects with varying iron stores, its absorption

increased in those with iron deficiency.  Conversely, absorption was inhibited in
                                     736
individuals with adequate iron stores.    Retention of a tracer dose of

manganese-54 10 days after absorption by an iron-deficient patient was no
                                                  736
greater than by a patient with normal iron stores.
                                            621,738,740,768
     Absorption of cobalt by rats and humans                is increased in

the presence of iron-deficiency anemia.  Both metals are absorbed predomi-
                                                  737
nantly in the proximal small intestine of the rat.     In both species,

increasing the concentrations of either metal inhibits the absorption of
          668,669,739
the other,            suggesting that iron and cobalt share at least part

of a common absorptive pathway.  These features have been used in a cobalt
                                      769
test for human iron-deficiency anemia.

     Absorption of cadmium by iron-deficient mice was four- to  sixfold
                                  328
greater than by iron-replete mice.     Moreover, cadmium toxicosis produces

                                            133,624a
iron-deficiency anemia in the mouse and rat,         which can be corrected

by additional iron.  Whether or not human beings with iron deficiency absorb

greater amounts of cadmium is not known.

     Absorption of lead increases in rats fed iron-deficient diets, as  does
                               426,695
the toxic quality of the metal.         It is unknown if a similar increase

in lead absorption occurs in iron-deficiency anemia of children.  The fre-

quent association of iron deficiency and lead poisoning in children aged 1-3
                       166,727
yr in low-income groups        raises the possibility of increased lead

absorption as a secondary effect of iron deficiency.  However, pica asso-

ciated with iron deficiency often leads to ingestion of lead from paint and

plaster of older buildings in which these children live and clouds the ques-

tion of whether increased lead absorption or increased availability is at

issue.


                                 188

-------
     Absorption of cesium, magnesium,  mercury,  calcium,  and copper was not
                                 621
increased in iron-deficient rats.
                                   189

-------
                                 CHAPTER 6

                     ACUTE TOXICITY OF INGESTED IRON


     Acute toxicity of ingested iron appears to be related strictly to

exposure to therapeutic iron preparations.   It is not associated with inges-

tion of naturally occurring or other commercially produced substances.  In

the United States, an estimated 2,000 cases of accidental iron poisoning
                                             201,806
occur annually, with approximately 12 deaths.        During an 18-mo-period,  in whicH

1,645 children who had ingested toxic agents were seen, ferrous sulfate was

the agent in 6.2% of the incidents.

     In 1947 Forbes pointed out that small  children were accidentally poisoning
                                                                      276
themselves by swallowing oral iron preparations prescribed for adults.

Before that report, only sporadic comments  were published on acute, potentially

lethal poisoning from ingested iron.  Numerous case reports and reviews of acute
                                        14,152,257,263,274,279,449,814,815
iron toxicity have been published since.


EPIDEMIOLOGY

     Acute iron toxicity is found most commonly in children aged less than
                                            14,201
5 yr, predominately in the 1-2 yr age group.        Poisoning results when a

child encounters the parent's iron medication and, attracted by the superficial

resemblance to candy and the taste of the sugar-containing coat, it ingests many

tablets before discovery.  In adults, cases of acute iron toxicity occur almost
                                       152,257,279,449
exclusively because of suicidal intent.

     Acute iron toxicity caused by sources  other than medicinal iron prepara-

tions does not happen because of the large  quantities of iron that must be

ingested to produce major poisoning.  The average human lethal dose is about
                               201
200-250 mg iron/kg body weight.     Thus the average adult male must swallow

14 g of elemental iron (234 ferrous sulfate tablets).  For the average 2-yr-old

child, 3 g of elemental iron (50 ferrous sulfate tablets) is the average lethal

dose.

                                    190

-------
     However, fatalities have been reported from much smaller doses.   The

ingestion of 1.2-2.4 g ferrous sulfate led to death in a 17-mo-old child;

but the death was precipitated by peritonitis contracted following surgery
                                                    656
to relieve ferrous sulfate-induced pyloric stenosis.      Yet a 17-mo-old
                                                               143
child recovered after swallowing a dose as large as 4.2 g iron.     In one

review, the range of fatal amounts of ingested iron was 0.96-3.6 g elemental
                                                14
iron, whereas for nonfatal cases it was 0.3-3 g.    Although a few adults will

develop mild symptoms of gastrointestinal irritation following ingestion of

60-180 mg of medicinal iron, this quantity, the usual therapeutic dose, usually

produces no symptoms of overload or acute adverse effects.


CLINICAL STAGES
                                                                  14,806
     The features of acute iron poisoning have been well described       and

may be divided into four phases.

     •  Approximately 30 min after ingestion of the iron, vomiting—which

        may be bloody—begins in 80% of the patients.  More than half the

        patients become drowsy or lethargic.  Diarrhea, often bloody,

        develops in over 40% of the victims.  The pallor, tachycardia,

        and rapid respirations that often develop may indicate impending

        shock.  A patient may die during this phase, which extends over

        6-12 h.

     •  During the second phase, improvement is noted, stupor clears,

        and the vomiting and diarrhea subside.  This phase lasts 10-12 h

        and may be followed by complete recovery or sudden relapse.

     •  Many patients will relapse after the phase of improvement ends.

        Fever, pneumonitis, shock, coma, convulsions, and death may

        follow.  This phase occurs approximately 20 h after ingesting

        the iron.
                                   191

-------
     •  In those surviving for 3-4 days after iron ingestion, recovery

        is rapid.  In a small number of patients, gastric obstruction

        related to pyloric stenosis and stricture formation will develop.

     Laboratory findings during the first three phases may include marked

leukocytosis, evidence of metabolic acidosis, hyperbilirubinemia, and deranged
                  263,814
blood coagulation.         Serum iron levels vary, yet do have prognostic value.

In one report, shock or coma occurred in only 9 of 112 patients (8%) with an

initial serum iron value less than 500 ug/100 ml, but in 37% of those with an
                                         806
initial value in excess of 500 ug/100 ml.     The onset of coma or shock indi-
                          806,815
cated a serious prognosis.         Without treatment, all patients with these

signs died.  If neither coma nor shock were present, all patients survived.
                                                              806,819
With current modes of treatment, the mortality rate is 11-17%.


PATHOLOGY
                                                          651
     In addition to the removal of tissues during surgery,   several autopsies have
                                    '                                    14,152,279
been reported in children and adults who died from acute iron poisoning.

     In the stomachs examined, hemorrhagic necrosis and sloughing of areas of

mucosa with extension into the submucosa were common.  The necrotic surfaces

were covered with iron-positive pigment.  The mucosa was diffusely congested and

infiltrated with polymorphonuclear leukocytes and mononuclear cells.  Basement

membranes of lymphatics, capillaries and venules of submucosa and serosa stained

iron-positive.  Platelet thrombi were numerous in the submucosal capillaries and

veins.  Changes in the small intestine were the same as described for the stomach

but were less severe.  However, the most striking changes were in the proximal small
                                                   651
intestine.  Infarctions, particularly of the ileum,     were found to be associated

with thrombosis of the mesenteric veins. Autopsies revealed hemorrhagic necrosis of
                                            502
the periportal portion of the liver lobules.    Iron stains  of these areas showed
                                    192

-------
finely dispersed iron deposits in portal vein endothelium, Kupffer cells, and



periportal reticulum.  Other organs exhibited only general conditions of edema,



cloudy swelling, and areas of hemorrhage.





PATHOPHYSIOLOGY



Local Events



     Hemorrhagic necrosis of the gastrointestinal tract is a direct effect of



the interaction of iron and gastric hydrochloric acid.  This strongly acid



solution containing ferrous and ferric ions, chloride, and usually sulfate,



produces coagulation of protein and extensive corrosion.  Although the injury



to the intestinal tract may be severe enough to lead to scarring, the notion



that loss of blood and fluid from the gastrointestinal tract is a major cause



of shock and death has been disproved.



Systemic Events



     Most of the systemic effects of poisoning are associated with absorption



of excessive amounts of iron, resulting in the appearance of plasma iron not



bound to transferrin.  In experimental situations employing intravenous injec-



tions of ionic iron, exceeding the iron-binding capacity has produced such



striking symptoms and events in humans as sneezing, nausea, cutaneous flushing,



tachycardia, and hypotension.    '     In dogs, excessive iron is rapidly



absorbed across intact intestine, where it is responsible for profound meta-


                                    644 645
bolic consequences leading to death.   '     Thus, intestinal mucosal damage



need not be invoked to explain the excessive absorption of iron or the follow-



ing systemic events.



     Coagulation defects.  No systematic study of clotting and iron poisoning



has been reported in humans.  Studies in animals    suggest that alterations



in coagulation are secondary to a direct effect of iron on the various proteins
                                   193

-------
involved in blood coagulation.  These reactions can be simulated by a direct



addition of ferrous sulfate to human plasma.



     Metabolic acidosis.  Acidosis is a prominent and consistent feature of



experimental iron poisoning in dogs and a common development in iron poison-



ing in children.  Three factors are involved with the onset of acidosis.



Hydrogen ions are released from the conversion of ferrous iron to the ferric



form in the circulatory system, i.e.,




               1/4 02 + Fe"*"*" + 5/2H20 	> Fe(OH)3 + 2 H+.




Organic acids such as citric and lactic acids are accumulated prior to circu-


               644
latory failure.     Electron microscopic studies of ferrous sulfate-induced



liver damage show mitochondrial injury that constitutes an anatomic basis for


                              824
the organic acid accumulation.     In addition to the reasons just discussed,



lactic and citric acids accumulate as a result of the anaerobic metabolism



associated with shock.



     Shock.  Shock can be an early feature even of severe nonfatal iron poison-



ing.  Therefore, it is more than a terminal event.  The peripheral circulatory


                                    644
failure is induced by absorbed iron.      Although how the process works has not



been characterized fully, it appears that iron directly affects the peripheral



vasculature.  In dogs given fatal amounts of iron, six alterations were con-



sistently found;  reduced cardiac output; increased total peripheral resistance;



decreased plasma volume; hemoconcentration; decreased total blood volume; and


                       644
lowered blood pressure.     Using this and other information, Whitten and



Brough presented a hypothetical sequence of the pathophysiologic events of acute



iron poisoning, but the data on which they based their sequence were insufficient



to substantiate the hypothesis.
                                  194

-------
TREATMENT AND PREVENTION



     Details of treatment are beyond the scope of this report.  It is based




on efforts to rid the victim of unabsorbed iron by using gastric lavage and




cathartics.  In patients suspected of having ingested near-lethal amounts




of iron and in those with serum-iron concentrations in excess of the binding




capacity, the physician attempts to promote urinary excretion of the extra




plasma iron by binding it with a chelating agent, deferoxamine mesylate.


                                  143 274 705a
Early treatment is most important.   '   '      The shorter the interval




between ingestion of the iron compound and the initiation of therapy, the




better the prognosis.



     Prevention is the key to this public health problem.  Millions of pre-



scriptions are written each year in the United States for iron-containing



medications.  Moreover, iron compounds are available for purchase without pre-




scription.  Because many of the users are women with young children, consider-




able opportunity exists for iron ingest ion and poisoning to occur in that




susceptible group.  Public education and the necessity for physicians and



pharmacists to make their patients aware of the hazard are therefore of the




utmost importance.



     In summary, no known natural or dietary sources such as water, food, or




any beverage are likely to result in acute iron toxicity.  Acute poisoning




from ingested iron comes from accidental ingestion of medicinal iron, mainly




by children one to two years of age.  Iron poisoning may be fatal and early




treatment is crucial.  Prevention of exposure is the most important aspect



of the problem.
                                   195

-------
                                  CHAPTER 7
                            CHRONIC IRON TOXICITY


      In  relation  to  the  total body iron content,  the  amount  of  exchange  that
 occurs between humans  and  their  environment  is  small.   Disturbances  of iron
 balance  are  common,  but  they are almost invariably such that a  reduction in
 the  total body iron  content results, as discussed in  Chapter 5.  A major pur-
 pose  of  this chapter is  to explore if iron overload does, indeed, result in
 toxicity.

     As the quantities of storage iron  (soluble ferritin and insoluble hemosi-
derin) in the body rise,  the ratio of hemosiderin to  ferritin increases.   Thus
an increase in storage iron is called hemosiderosis  or siderosis, and it  may be
relative or absolute.  A relative siderosis is a part of any anemia not caused
by blood loss,  and merely represents an internal redistribution of the body's
iron, with less in the red cell mass and more in stores.  Siderosis may also be
localized to certain organs; in idiopathic pulmonary  siderosis,  it is the lungs
and in paroxysmal nocturnal hemoglobinuria, the kidneys.  Such conditions are
examples of focal siderosis.  When the total iron content of the body is  increased,
the extra iron is laid down in storage compounds, and it is  called absolute sidero-
sis or iron overload.  The term hemochromatosis is applied when the organs con-
taining grossly excessive amounts of storage iron show pathologic evidence of
damage, usually fibrosis.
IRON OVERLOAD
Pathogenesis
     When the total body iron content is increased,  the extra iron enters either
through the gastrointestinal tract or by the parenteral route.  In some circum-
stances,  both mechanisms  may be operative.

Oral Iron Overload
     Excessive dietary iron intake.  Although the absorptive mechanism normally
adapts itself to the body's need for iron, nevertheless the  actual quantities
                                  196

-------
                                                                       89
absorbed increase progressively as the intake of absorbable iron rises.     Sub-

stantial iron overload theoretically can be a function of large amounts  of absorb-

able iron, but the condition is rare except in southern Africa, where alcoholic
                                              159
beverages are prepared in iron drums and pots;     The mean iron content of such

brews has been found to vary between 40-80 mg/1 and the pH of the beverages is

low.  In Western countries, an analog sometimes is encountered in wine-drinking
           506
alcoholics.     However, the iron content of wines is considerably lower than

that of the beer consumed by South African blacks; red wines usually contain
                                                   26,506,610
between 5-6 mg/1 and white wines lesser quantities.            Finally,  there

have been occasional reports of patients who have misguidedly continued to consume
                                  157
iron tonics over extended periods.

     There is good reason to question the traditional assumption that dietary

iron overload is simply a function of the increased quantities of the metal in-

gested.  Only when the dietary iron is available for absorption is the overload

likely to occur.  The importance of availability is well illustrated by the situ-

ation prevailing among South African blacks.  On the basis of stool analyses, the
                                                            92,779
daily intake of many men was found to be between 100-200 mg.        Most of this

contaminating iron is consumed in maize porridge cooked in iron pots, or in beer

made from maize and sorghum fermented in iron drums.  Only very small quantities

of the iron in the porridge are absorbed.  In contrast, the absorption of iron in

the beer is good, presumably as a result of the formation of soluble, easily
                                                   87
absorbed complexes during the fermentation process.    Teff, a grain consumed in

Ethiopia, is heavily contaminated with iron but iron overload is not a problem,
                                                                 371
since the extra iron is present in an unabsorbable form or forms.

     Excessive absorption from diets of normal iron content.  Oral iron overload

is possible even when the dietary iron content is not inordinate.  In individuals

with idiopathic hemochromatosis, amounts of iron in excess of body requirements
                                    197

-------
                                                                              89
are regularly absorbed, and iron slowly builds up in the body over many years.

The metabolic defect responsible for the inappropriate absorption is unknown.

Indeed it seems likely that what is called "idiopathic hemochromatosis" may

include several separate diseases affecting at different sites the processes by
                                   199
which iron absorption is regulated.

     Regardless of the nature of the defect(s) responsible for increased iron

absorption, there is good evidence that idiopathic hemochromatosis is inherited.

Between 25-50% of first-degree relatives exhibit iron overload of varying degree,
                                            157
and siblings are the most commonly affected.     The majority show only a modest

increase in body iron stores, but some are heavily siderotic.  It is rare to find

severe iron overload in successive generations, but families exist in which it
               303
has been noted.     Because different patterns of expression of the disorder have
                                                                    670
been described, debate on the mode of inheritance has been spirited.     However,

such questions become irrelevant if it is accepted that the condition is not a

homogeneous one, and that any of several defects may be associated with excessive

absorption of iron from the gut.  The most common form of inheritance appears to

be an intermediate one in which the heterozygote manifests minor derangements of

iron metabolism, such as a raised plasma iron concentration and a moderate

increase in iron stores, and the homozygous state eventually leads to the accumu-
                                  670,818
lation of massive amounts of iron.          In contrast, in those families in

which successive generations have been affected with fully developed hemochroma-
                                                      43,88,219
tosis, the mode of inheritance appears to be dominant.           In addition,

there are reports of patients with the disease who were the offspring of con-

sanguineous marriages, in which neither parent showed any abnormality; presumably
                                                        218,219,587
the genetic defect in these circumstances was recessive.

     Resolution of the uncertainty is further compounded by the fact that the

ultimate expression of the gene in an individual can also be influenced by
                                     198

-------
                 157
external factors,    such as the amount of iron in the diet and physiologic

and pathologic blood losses.    Severe iron overload thus would be expected

to occur less frequently in affected females and to be uncommon in countries

such as India, in which the quantities of available iron in the diet are small.

Conversely, hepatic dysfunction in the iron-loaded liver might be more widespread
                                                                     273
and severer in individuals exposed to a hepatotoxin, such as alcohol.     Whereas

a small proportion of patients with alcoholic cirrhosis certainly do show severe
          305                                                   499,501
siderosis,    iron stores are usually normal in alcohol abusers.         Iron

overload is rare enough in alcoholics outside of southern Africa to suspect

that consumption of alcoholic beverages is not the only factor in such subjects.

It seems reasonable to suppose that some of them might be heterozygous for a

hemochromatosis gene and, had they not been exposed to alcohol, would never have

developed a recognizable clinical syndrome.

     Parenteral iron overload.  In several refractory anemias, life is sustainable
                                                      81,84,89,144,427,506,593,676,694
over long periods through repeated blood transfusions.

Because each 500-ml blood transfusion contains more than 200 trig iron, very large

quantities can be introduced via this route. In the absence of bleeding, the body's

capacity to excrete the extra iron is extremely limited—a few milligrams daily at
                              308
the most (see also Chapter 4).    As a result, massive quantities of storage iron

accumulate.  Although iron overload is a serious problem in some patients with

acquired aplastic anemias, the majority succumb to other complications of bone

marrow failure before the stage of massive overload has been reached.  Patho-

logic changes are more common in thalassemia major and other conditions asso-

ciated with defects in hemoglobin synthesis, including refractory sideroblastic

anemia, pyridoxine-responsive anemia, and refractory normoblastic anemia, and they

are occasional manifestations of a variety of chronic hemolytic states. There are
                                   199

-------
two reasons for parenteral iron overload.  First, the realization that the

quality of life in affected individuals is improved by maintaining the hemo-

globin at levels closer to normal has encouraged the regular use of transfusion
                          540
therapy from an early age.     Secondly, in anemias such as thalassemia major in

which erythroid activity is markedly increased, yet ineffective in delivering
                                                                                 44,
viable red cells into the circulation, iron absorption from the gut is increased.
97,258,688
            The iron overload that develops is thus not simply the result of

repeated transfusions, but also includes an oral component.  Hence the useful

term "iron-loading anemias" coined by Crosby.

     Although parenteral overload also could theoretically result from repeated

treatment with injectable iron compounds, no published evidence shows that it

represents a practical problem.


Prevalence

     The prevalence of iron overload in different populations varies not only

with the frequency with which  certain genes occur in different populations, but

with the composition of the average diet.  It is not only  the dietary iron con-

tent that is important, but also its availability for absorption.  Meat is a

good source of absorable iron, whereas cereals and other vegetables are poor
        518,752
sources.         Therefore, it is not surprising that no reports of clinically

manifest oral iron overload have come from India.  Differences in prevalence

among  certain Western countries may represent a subtle interplay between the

frequency of genes associated with a tendency to absorb too much iron and the

content of easily absorbable iron in the diet.  For example, the greater pre-

valence of clinical hemochromatosis in Australia as compared to the United

Kingdom may relate to the greater per capita consumption of meat in Australia.
                                                                                506
In France, Germany and Italy,  the consumption of red wines could be responsible.
                                  200

-------
     A real difficulty in assessing the prevalence of iron overload in different

countries is that quantitative methods for determining the amount of storage iron

in tissues have come into general use only recently.  The quantitative approach

has required either the chemical estimation of the iron concentrations in

affected tissues or the measurement of the urinary excretion of iron after admin-

istering specific chelating agents.  Such a distinction has considerable relevance.
                                                                       506
Histologically visible iron often is noted in organs such as the liver,    and if

cirrhosis is present the condition can be labelled hemochromatosis when the quan-

tities of iron are merely moderate.  With these reservations, current U.S. statis-

tics indicate a prevalence of 1 in 20,000 hospital admissions and 1 in 7,000
                             273                                        264
hospital deaths in one study,    and 4 in 100,000 necropsies in another.     Much

higher, although misleading, figures would be obtained if statistics were collected

in diabetic or liver clinics, because the presenting clinical features are so com-

monly related to diabetes or hepatic dysfunction.  Perhaps the most objective figures

available are those from a study in which nearly 4,000 liver specimens, the

majority obtained from hospital necropsies in 18 different countries, were ana-
                          160
lyzed chemically for iron.     In only 3 samples were the concentrations more

than 5 times normal and none was anywhere near the figure of 20-50 times normal

found in hemochromatosis.  Therefore, severe iron overload, at least as judged

by necropsy figures, is a relatively uncommon phenomenon in most countries.

     As noted, the exceptions are confined to southern Africa, where many popu-

lations consume home-brewed drinks high in iron.  There have been two attempts
                                        129,159
to define the prevalence in this region.         In one study carried out in

Johannesburg, 70% of black adult males and 25% of black adult females dying in
                                                                       90
hospital  were found to have excessive quantities of iron in the liver.    Both

prevalence and severity increased with age.  Although the degree of siderosis

was mild or moderate in most subjects, 20% of middle-aged males had concentrations
                                    201

-------
above 2% dry weight, comparable to amounts described in Western countries for

subjects with idiopathic hemochromatosis.   As exposure to inordinate quanti-

ties of iron in South African blacks is largely confined to adults, iron stores

are not abnormally high in infants and children; indeed, iron deficiency is not
                         158
uncommon in these groups.

     The prevalence of those anemias most commonly associated with severe iron

overload is not known with any certainty.  Aplastic anemia is a rare condition,

and has a median survival of only about two years; but thalassemia major and

several other chronic refractory anemias are compatible with survival for a number

of years.  These individuals are attracting the attention of investigators in-
                                                                   540
terested in the possible long-term effects of severe iron overload.     Thalassemia

major reaches its greatest prevalence (1 in 400) in Mediterranean peoples,   yet most

of the current knowledge concerning the effects of the accompanying iron overload

has been obtained from studies of small numbers of affected individuals whose

parents migrated to the United States, the United Kingdom, or Australia. Such

patients have usually been studied more systematically and they tend to have been

treated with more intensive regimens of blood transfusions.  The prevalence in

these countries varies, but it is very low.   In the United Kingdom just over 200

patients are thalassemics, most  of them inmigrants  from Cyprus  and  the
                        540
northwest part of India.

Rate of Iron Accumulation

     Most patients with idiopathic hemochromatosis develop clinical symptoms in
                                                        273,689
middle age, when body iron content ranges from 20-40 g.         This deposit can

result from an average positive iron balance of about  2-3 mg daily, but it is

doubtful if the accretion is uniformly spread out over the preceding years.

The increased stores tend  to depress iron absorption,  so  that by the time clinical
                                                                           817
manifestations appear  the  absorption rate is often within the normal range,

which is, of course, inappropriate to the iron-overloaded state.   After stores


                                  202

-------
have been removed by phlebotomy, the absorption rate rises to between 5-7 mg

daily, suggesting that it is initially high and then gradually declines as the
                   71,817
stores accumulate.            That the clinical manifestations are 10 times

more common in men than in women is probably caused by lower iron intakes and
                          273
greater losses in females.

     In dietary siderosis as it occurs in black South Africans, the body iron con-

tent builds up from late adolescence until the fifth decade.  The quantities of

iron in the body of the severely affected individuals are similar to those found
                              387
in idiopathic hemochromatosis.     Limited isotopic studies suggest a positive

daily balance of up to 3 mg between ages 20 and 50 yr.  However, the rate of
                                92
accumulation may not be uniform.

     Transfusional hemochromatosis is of concern in aplastic and hyperplastic
        89
anemias.    With aplastic anemia, the rate of iron deposition is directly related

to the frequency and the number of blood transfusions, although the quantity of

iron in the body may be much less than was present in the transfused blood, as

thrombocytopenia with resultant blood loss is a frequent complication in such

patients.  In hyperplastic anemias such as thalassemia major, good correlation

exists between the amount of iron administered in donor blood and the size of the

iron stores, but there is also an appreciable contribution from the gut, especially

when the subjects are very anemic.  As a result, the rate of iron accumulation is

greater than in the aplastic group, although the additional load from the excessive
                                                                               540
absorption varies according to the level at which the hemoglobin is maintained.


Distribution of Iron

     In idiopathic hemochromatosis the highest storage iron concentrations are
                                                                            689
found in the liver and pancreas, where they are between 50-100 times normal.

In the thyroid the concentration is usually about 25 times normal*and in the heart,
                                   203

-------
pituitary, and adrenals, between 10-15 times normal.  Lower concentrations of

about 5 times normal are found in the skin, spleen, kidney, and stomach.

     The striking histologic feature of the disorder is the widespread presence

of the golden-yellow iron storage compound, hemosiderin, in the parenchymal cells
               235,689
of many organs.         Deposition of hemosiderin is heavy in hepatocytes and

bile duct epithelium, with lesser amounts in Kupffer cells.  In the pancreas,

hemosiderin deposits are most prominent in the exocrine cells, but are also

found in the islets of Langerhans.  Epithelial cells of the thyroid, the parathy-

roid, and the anterior pituitary glands are all affected, whereas in the adrenals

the iron deposits usually are confined to the zona glomerulosa.  Hemosiderin

granules generally are scarce in the testes.  Cardiac involvement is a cardinal

feature, and deposits are heaviest in the perinuclear region of the muscle fibers
                  131,689
of the ventricles.         Smaller amounts are present in the atria and conducting

system.  Considerable attention has been devoted to investigating the deposition
                                                                                  240,
of iron in the joints.  The synovial lining may be laden heavily with hemosiderin.
329,689
          Electron microscropy has shown that type B cells are involved rather
                                       675
than the macrophage-like type A cells.

     Unlike the massive iron deposits in parenchymal cells throughout the body,

reticuloendothelial involvement is not prominent; splenic concentrations of iron
                         156,689
are only modestly raised,        and iron in the reticuloendothelial cells of the
                                             102
bone marrow is often within the normal range.

     The distribution of iron in the dietary iron overload of South African blacks

has been studied at all stages of its development, since the degree of siderosis

in autopsied subjects ranges from mild to massive.  The pattern is very different

from that observed in idiopathic hemochromatosis.  In the earliest stages, an

increase in hemosiderin granules is detectable in the hepatic parenchymal and
              86
 *•
Kupffer cells.    With hepatic concentrations of 5-10 times normal, the deposits
                                   204

-------
in these cells get denser, and the portal tract macrophages become involved. When

iron concentrations reach 20 times normal, heavy deposits are observable in all

three sites.  Hemosiderin is visible in the spleen from an early stage, and with

more severe degrees of iron overload, deposits may even exceed those in the
      86
liver.    Reticuloendothelial involvement is also apparent in the bone marrow,
                                               288
which may contain as much as 10 g storage iron.     There is moderately close
                                                                            102
correlation between the amounts of storage iron in the marrow and the liver.

Iron deposits elsewhere in the body are relatively scanty in most subjects.

Thus siderosis is largely confined to the hepatocytes and the reticuloendothelial
      362,387
cells.

     This pathologic picture is the usual one in black South Africans with

iron overload, but in one set of circumstances a different pattern occurs.

About 19% of severely siderotic individuals also exhibit micronodular
          96 387
cirrhosis.   '     In such subjects notable deposits are also found in the

parenchymal cells of a number of their organs, including the pancreas, thyroid,

adrenals, pituitary and myocardium.  The concentrations of iron are comparable

to those described in idiopathic hemochromatosis.


     The distribution of iron in subjects with parenteral iron overload is

influenced by the nature of the underlying anemia and the length of time over

which blood transfusions are administered.  When the marrow is hypoplastic, the

major impact is on the reticuloendothelial cells of the spleen, liver, and bone
       89
marrow.    This occurs because donor blood is eventually broken down in these

sites, and since the plasma iron turnover is reduced, there is little tendency

for it to leave.  If, however, the condition persists over many years, the iron

is slowly redistributed and marked parenchymal loading in tissues may result.

     In the second group of anemias, the marrow is hyperplastic rather than hypo-

plastic, which means that external and internal iron kinetics are different from
                                   205

-------
those prevailing in the hypoplastic anemias.  This variation affects the distri-

bution of the excess iron.  A large proportion of the red cells produced by the

overactive erythroid marrow are so defective that they do not even enter the cir-

culation, and those that do have a shortened life span.  As a result, erythropoietic

activity is great, and the plasma iron turnover markedly increases. Much of the

iron released from the catabolized hemoglobin of both these cells and the transfused

blood is rapidly returned to the plasma. In addition, iron absorption in such anemias

is typically increased despite the developing iron overload. A major contribution

from the gut is particularly likely in those conditions in which hemoglobin synthe-

sis is defective, since the red cells superficially resemble those found in iron

deficiency anemia.  As a result, oral iron therapy may mistakenly be given over

extended periods, as illustrated by reports of severe iron overload in subjects

with refractory anemias and hypercellular marrows who have received few or no
             89
transfusions.

     Although the ultimate iron distribution in subjects with refractory anemias and

hypercellular bone marrows may be influenced by numerous extrinsic factors, wide-

spread parenchymal deposits of iron generally are inherent in the disorders themselves.

Presumably deposition is a function of the rapid iron turnover through the plasma,

with subsequent unloading of iron onto the parenchymal cells of various organs.

Unloading is almost certainly facilitated by the iron-saturated transferrin

bathing the tissues; at least for the liver, iron uptake by hepatocytes is greatly

increased under such circumstances. Why the concentrations of iron are greater in

some organs than others is not known, but the rate of blood flow through the organ

and the relative number of transferrin receptors could well be responsible.

Reticuloendothelial deposits are also prominent in refractory anemias with hyper-
                 649
cellular marrows.     Such a feature is not surprising, since iron turnover through

these cells is several times normal.  That splenectomy in thalassemia major is
                                    206

-------
followed by some redistribution of iron in the liver and an increased amount in
                                         61,649,826
hepatocytes is a subject of some dispute.

     It is worth considering iron distribution in the liver in more

detail, because the one central denominator of all forms of iron overload is

involvement of the hepatocytes.  Hepatocyte uptake of plasma iron is facilitated
                                             182,408
by high-            saturation of transferrin,        a factor found in all forms

of iron overload, except for the dietary form occurring in black South Africans.

Another factor in those conditions associated with increased absorption from the

gut  (i.e., idiopathic hemochromatosis, dietary iron overload, and refractory

anemias with increased but ineffective erythroid marrow activity) is that some

of the absorbed iron may be deposited directly in hepatocytes from the portal

venous blood.  Normally such iron is transferrin-bound, but if the percentage

saturation of the metal is high and a large amount is absorbed, a good proportion
           809
may not be.     Any unbound iron is deposited immediately in the hepatic parenchy-

mal cells.  In addition, in any anemia characterized by an appreciable degree of

intravascular hemolysis, hemoglobin in the plasma, whether bound to haptoglobin
                                         351
or hemopexin, is taken up by hepatocytes;   a proportion of such iron might be

expected to remain there.  Finally, it is possible that some iron deposited in

hepatocytes may be derived from the small amount of iron circulating in the plasma

as ferritin or bound to a protein other than transferrin.


.Associated Pathologic Findings

     When reviewing the pathologic findings associated with severe iron overload,

it is customary to consider idiopathic hemochromatosis as the prototype.  The

major feature of the disorder is the presence of fibrosis in heavily siderotic

organs, particularly the liver.  A fine monolobular cirrhosis is almost invari-
                                                                            627,689
ably found, the lobules typically separated by wide bands of fibrous tissue.
                                                                               420
However, sometimes a multinodular, postnecrotic type of cirrhosis  may develop.
                                   207

-------
 In young subjects the fibrosis can be less severe.  Carcinoma of the liver is
                                                                   273
 an important late complication, occurring in about 15% of patients.     It is
                                                             235
 usually a hepatoma, but cholangiomas also have been reported.     In the pancreas

 the number of islets is often reduced, and fibrosis and degenerative changes
                        689
 develop in acinar cells.     Some fibrosis may also occur in the pituitary,

 adrenals and thyroid, but it is not prominent.  Testicular atrophy is manifested

 in the majority of subjects.  Dilation and hypertrophy appear in the myocardium;
                                                                         131
 occasionally degeneration of myocardial fibers and fibrosis are reported.

 The large joints may show calcium pyrophosphate deposition in hyaline and fibro-
          240
 cartilage.     Villous hyperplasia and fibrous thickening of the synovium have
                                                          780
 been noted, as have degenerative changes in the cartilage.     These pathologic

 changes are seen once the clinical features have become manifest; little is known

 of what occurs during that long latent period, extending over half a lifetime,

 during which the iron content of the body is slowly increasing.  It is clear

 from studies of affected siblings that hepatic siderosis precedes the onset of
         89
 fibrosis.

     In the dietary iron overload that occurs so commonly in South African blacks

 a clearer idea of the pathogenesis of iron overload has been obtained, particularly

 in regard to hepatic changes.  The results of several studies indicate that a defi-

nite association exists between severe iron overload and significant portal fibrosis
             86,90
 or cirrhosis.        This association cannot be explained by the prominence of

 severe siderosis in older subjects, because marked portal fibrosis or micronodular

 cirrhosis occurs in only 13% of black adults over age 40 in whom the siderosis

 is minimal, compared to 71% in severely siderotic individuals (iron concentra-
                         90
 tion, > 2.0% dry weight).    Iron loading in the parenchymal cells of other

 organs is uncommon.  However, if deposition occurs, it is usually not accompanied

 by fibrosis.  The only other common pathologic finding in these subjects is
                                  208

-------
osteoporosis, which is especially marked in the spine and may lead to collapse
                                                681
of vertebrae, particularly in the lumbar region.     A link between this form

of bone disease and severe iron overload has been demonstrated, and chemical

analysis of necropsy specimens has shown an inverse correlation between hepatic
                                                               503
iron concentration and the mineral density of iliac crest bone.

     Associated pathologic changes in subjects with transfusional hemochromatosis

are influenced by the nature of the underlying anemia and the age at which it

occurs.  In an analysis of 20 patients with hypoplastic anemia who had received

more than 47 1 of blood (equivalent to 25 g iron), 12 showed increased portal
                                                          89
fibrosis, but only 1 person had fully developed cirrhosis.    In contrast,

cirrhosis was much more prevalent in 31 patients with refractory anemias asso-

ciated with hypercellular erythroid marrows.  Cirrhosis was diagnosed in 26 and

increased fibrosis was noted in another 4.  Although these data suggest that iron
    i
overload involving the reticuloendothelial system is better tolerated when

parenchymal deposits are not present from an early stage, different investigators

have used different pathologic criteria, so that any judgments can be only tenta-

tive.

     More concrete information relating to thalassemia major has become available.

Iron concentrations in the organs of five subjects were observed to be in the range
                                        826
described in idiopathic hemochromatosis.     All were cirrhotic and three showed

frank pancreatic fibrosis.    Cardiac hypertrophy and hyaline degeneration

of myocardial fibers were found, but very little fibrosis.  In a study of 41

patients, 39 of whom had thalassemia major, congestive cardiac failure developed
                             255
in 26 and pericarditis in 19.   Necropsies were performed on 11 subjects. Iron

deposition was widespread, as was tissue fibrosis, especially in the liver, pan-

creas, gonads, thyroid, pituitary, and adrenal glands.  The heart was dilated and

hypertrophied and microscopy revealed large amounts of iron in muscle cells and
                                    209

-------
histiocytes.  Myocardial fibers showed focal degeneration and widespread fibro-

sis.  In a study in which 32 liver biopsies were performed on children with

thalassemia major, the severity of the fibrosis correlated closely with both
                                                 649
the patient's age and hepatic iron concentration.     The rate of fibrotic pro-

gression was a function of the liver iron concentration; when the concentration

was less than 3% dry weight, progression was relatively slow, whereas it accelerated

at higher concentrations.  Although the severity of the fibrosis correlated with

the degree of parenchymal siderosis, its relationship to the reticuloendothelial

involvement was closer.  An analysis of 207 patients showed that most of the 37

who had died had succumbed at an early age because of inadequate transfusion
                             540
therapy and/or hypersplenism.      The 8 older subjects had received an average

of 130.2 1 of blood (range, 68.2-188.9), and death was due to cardiac failure in

all of them.  Despite its presence in the liver and exocrine pancreas, there

was no fibrosis in the heart. A child who had received 65.8 1 of blood for con-

genital aplastic anemia died of myocardial insufficiency, and her heart was
                                665
examined by electron microscopy.     In addition to intracytoplasmic deposits,

some iron was consistently present within the nuclei and mitochondria of the

myocardium.  The mitochondria were swollen and disrupted, and the myofibrils

were reduced.


Clinical Features

     The clinical presentation in patients with idiopathic hemochromatosis has
                     89,264,273,689
been fully described.                Therefore, this discussion will examine

the relationship between the iron deposits and the manifestations of the disease.

Attention will be paid first to the functional disorders that affect those organs

in which the highest concentrations of iron occur.  Then the possible connection

between excessive quantities of iron in the body and certain secondary and in-

direct effects on body function will be considered.


                                   210

-------
     Direct associations between iron overload and clinical manifestations in

idiopathic hemochromatosis.   The major manifestations are skin pigmentation,

hepatomegaly, diabetes, cardiac abnormalities, endocrine changes and arthropathy.

Bronzed skin pigmentation is common in most subjects.  Excessive melanin is an

invariable finding, but hemosiderin deposits are seen  in about 50% of patients,

as well; in the latter group, the skin is a peculiar slate gray.

     Symptoms and signs associated with involvement of the liver are nonspecific, and

include cachexia, weight loss, and weakness.  The liver is almost always enlarged

and firm, and signs of hepatic dysfunction, such as palmar erythema, spider angio-

mata, loss of body hair, and testicular atrophy, may appear.  About 50% of patients

have splenomegaly, but ascites is rare.  The liver disease usually runs
                               629
a protracted and benign course,    except when alcohol abuse complicates matters.

Hepatoma eventually supervenes in about 15% of patients, its development suggested
                                                                                 273
by unexplained weight loss,  fever, nodular enlargement of the liver, and ascites.

     Symptoms related to the onset of diabetes are observed in more than half the
                                  239,712
subjects at the time of diagnosis,        and administration of insulin is usually

necessary.  It was once thought that the diabetes was ascribable to islet cell

failure resulting from local iron deposits, but recent studies have indicated that

its etiology is more complex: at least three factors may be involved.  Insulin

deficiency is undoubtedly one of them, but insulin resistance brought on by the
                     630
cirrhosis is another.     Carbohydrate intolerance is more severe in hemochroma-

totics than it is in subjects with alcoholic cirrhosis, and this difference is
                                                   712
associated with a notably smaller insulin response.     In addition, the preva-

lence of diabetes mellitus unassociated with marked siderosis is high in relatives

of subjects with idiopathic  hemochromatosis, suggesting that inheritance of
                                                                         40,239
traits for diabetes is independent of the inheritance of hemochromatosis.

     Cardiac complications are the presenting feature in about 15% of subjects,

but as the disease progresses the percentage rises, and they represent perhaps


                                   211

-------
the commonest cause of death.  Supraventricular and ventricular arrhythmias of

various types have been described, and both left- and right-sided heart failure

may develop.  The manifestations are usually those of a congestive cardiomyopathy.

In subjects under age 40, cardiac manifestations are more frequently the pre-

senting feature and almost always the cause of death, which usually follows

within a year of diagnosis unless specific venesection therapy is instituted.

     Endocrine changes are common in idiopathic hemochromatosis.  The most

striking are loss of libido, impotence, amenorrhea, and other evidence of hypo-
         100,273
gonadism.         For example, 70% of 115 patients studied exhibited loss of
                                 781
body hair and testicular atrophy.     Although it has been customary to ascribe

these changes to hepatic dysfunction, recent hormonal assays indicate that often
                                                         692
both the anterior pituitary and target organ      fails        Luteinizing hormone

levels were found to be low in most hemochromatotic subjects with testicular

atrophy, but normal in individuals with other types of cirrhosis and a similar
                          711
degree of gonadal atrophy.     It has been established that the low levels of

luteinizing and follicle-stimulating hormones were the result of pituitary
                                     70
rather than hypothalamic disturbance.    Testosterone concentrations have also

been found to be below normal, and they do not rise after specific trophic

stimulation.  These results are compatible with a primary disturbance of pitui-

tary gonadotrophic function followed by secondary and irreversible failure of

gonadal function.  Disturbances of other pituitary trophic hormones, however,

are uncommon.

     Although arthropathy was recognized only recently as a complication of
                                                           674
idiopathic hemochromatosis, its symptoms are often present.     In half of those
                                                                 240
individuals with such symptoms, chondrocalcinosis has been noted.     Acute

attacks of inflammatory synovitis are characteristic of the condition, but other
                                        227
clinical syndromes have been recognized.
                                   212

-------
     Iron overload from fermented beverages.   In studies  of this  condition in black
South
/Africans, attention has been directed to features characteristic of idiopathic

 hemochromatosis.  Yet it must be emphasized that the degree of siderosis found

 in the majority of subjects is not as great as in the idiopathic disease, and

 that even in heavily siderotic individuals the iron is normally confined to the

 liver and the reticuloendothelial system.  Marked hepatic siderosis occurs in a
                                                                           96,387
 very high proportion of those subjects who develop micronodular cirrhosis.

 The prognosis in such individuals is poor.  The majority succumb within a year

 to the effects of liver failure and/or portal hypertension.  As noted, the

 iron distribution in many siderotic individuals with micronodular cirrhosis is

 different from that found in the absence of cirrhosis.  In cirrhotic subjects,

 iron is heavily deposited in the pancreas and other organs; one set of autop-

 sies revealed that as many as 20% had suffered from diabetes mellitus during
      387
 life.     In another study, 7% of black males attending a diabetic clinic had
                                             680
 severe siderosis and micronodular cirrhosis.      The clinical presentation in

 such individuals was characteristic: affected patients were usually male,

 middle-aged, and underweight.  Their livers were firm and enlarged, and insulin

 was needed to control the diabetes.  Other features characteristic of idiopathic

 hemochromatosis have not been reported in this variety of dietary iron overload.

 Specifically, neither cardiac failure nor an endocrinopathy have been demon-

 strated to be associated with the siderosis.   However, certain other manifesta-

 tions, including ascorbic acid deficiency and osteoporosis, which appear to be

 indirect consequences of the increased body iron stores, do occur (see below).

      Transfusional siderosis.  Clinical manifestations similar to those of idio-

 pathic hemochromatosis have been noted, especially when the affected individuals

 have survived for extended periods.  As mentioned, cirrhosis is rare in patients

 with aplastic anemia, but it develops almost invariably in those with
                                  213

-------
refractory anemias and hypercellular bone marrows who live long enough.   Dia-



betes has been noted in both groups.  In 20 patients with aplastic anemia who



had received more than 47 1 of blood, 20% had overt diabetes and in another 25%,

                               89

glucose tolerance was impaired.    Overt diabetes mellitus also is frequent



among older patients with thalassemia major; and in one study,  insulin response

                                                                            446

to glucose was delayed and/or diminished in 6 out of 8 thalassemic subjects.



Two of these had insulin-dependent diabetes, as did 3 out of 8  patients  studied

          540

elsewhere.




     Other manifestations of disturbed endocrine function in patients with



thalassemia major are particularly prominent in those patients  who survive


               45                                      411
into the teens.    Growth is almost uniformly retarded,    but  not from



growth-hormone deficiency, as growth-hormone responses to a number of stimuli



remain normal.  It is probable that chronic anemia is the major factor,



because growth rates have improved in subjects who have been transfused  more


            414 570
intensively.   '     Although clinical evidence of hypogonadism is usual,



its cause is uncertain.  Levels of gonadotrophic hormones have  been reported



as being appropriate to the ages of the patients.   '   '     In contrast,



levels of luteinizing hormone were low in 4 of 5 adults tested  despite low


                                 447
estrogen and testosterone levels.     Thyroid and adrenal function are usually



normal.  Hypoparathyroidism is thought to be a preterminal complication of the


,.      540
disease.



     Possible indirect sequelae of iron overload.  Certain disorders of a



more general biochemical nature than those just discussed may occur in iron



overload.  Observations were initially made in black South Africans, but such



malfunctions may also occur in other varieties.



     The first condition that has been noted is ascorbic acid deficiency,



which appears in almost all South African blacks with substantial iron over-


                                          681
load.   Frank scurvy will develop in some.      Ascorbic acid deficiency is




                                   214

-------
most prominent in dietary iron overload, but it has also been noted in some

subjects with idiopathic hemochromatosis and transfusional siderosis, particu-
                                         541,591,790
larly in patients with thalassemia major.             Moreover, when guinea

pigs were  made siderotic by  repeated  injections of  iron dextran, they became

scorbutic  even when their diet contained enough of  the vitamin  to  be adequate

                  791
for the controls.

     South African  blacks with massive iron overload are not only  ascorbic-
                                           cno AQ1
acid depleted, but  frequently osteoporotic.    '     Some conception of the

frequency and strength of the association may  be gained from the results of

a radiologic survey of 110 asymptomatic middle-aged manual laborers.

Seventeen were found to be osteoporotic, as evidenced by vertebral body
                                                    503
deformity and other signs of decreased bone density.     In another combined

clinical and histologic investigation, the association between  the bone

disease and severe  iron overload was  found to  be highly significant, a con-

clusion further underlined by the finding in a necropsy study of a negative

                                                                        503
correlation between hepatic iron concentration and mineral bone density.

Statistical analysis suggested that age alone was not responsible  for the

correlation. In addition, a striking  correlation exists between clinical scurvy
                                    681
and osteoporosis  in this population.      Tissue stores of ascorbic acid have

been shown to be  extremely low in osteoporotic individuals, even when they
                            £ Q-l
do not exhibit frank scurvy.

     The incidence of osteoporosis in other forms  of iron overload has  not been

accurately documented,  but it has been observed in individuals  with idiopathic
                223,237
hemochromatosis.          Of special interest is a  report  that beef cattle in an

area of New Zealand where the iron content  of the  water is very high develop a

form of siderosis very similar to that seen in black South Africans.  In many

severely affected animals,  osteoporosis of  the vertebrae,  sternum,  and  ribs is
          335
prominent.
                                    215

-------
 Relationship  to Tissue Damage



      The  particular  toxic  role  of  iron  in inducing  specific manifestations



 of  organ  failure  associated with severe iron  overload will be analyzed  in



 this  section,



      Iron overload and the liver.  Hepatic fibrosis is common to all forms



of  iron overload, and the evidence suggests that the relationship is a causal



one.  Almost all subjects with idiopathic hemochromatosis have cirrhosis


                                                      ?73 689
by  the time the condition becomes clinically manifest, '  '    and in affected


                                                                       88 818
but preclinical siblings iron overload precedes the onset of cirrhosis.   '



In dietary iron overload the facts are even more persuasive,  because the incidence



of  serious portal fibrosis or cirrhosis correlates directly with hepatic iron


               90
concentrations.    This correlation is illustrated in Figure 7-1.  Similarly,



in patients with transfusional hemochromatosis, especially those with chronic



refractory anemias associated with increased but ineffective erythroid



activity,  the fibrous response appears to accelerate once iron concentration


                                   649
in the liver reaches 3% dry weight.     The relative rarity of cirrhosis in



subjects with aplastic anemias who have been transfused repeatedly may be a


                        89
function of two factors„    The iron derived from transfused blood primarily



goes  to the reticuloendothelial system.  Because of the reduced marrow activity,



relocation of this iron to other sites is slow.  Evidently reticuloendothelial



iron, even when it is within the liver, is less likely to produce cirrhosis



than  iron in hepatocytes.  The second potentially important factor is time.



Most  patients with aplastic anemia succumb within a relatively short time,



whereas subjects with other forms of refractory anemia may survive for decades.



     The evidence that iron is a fibrogenic agent and can cause hepatic



cirrhosis appears overwhelming.  Yet several questions still remain unresolved



concerning the influence of extraneous factors on the genesis of hepatic



fibrosis in siderotic subjects.  In transfusional siderosis,  they may include




                                  216

-------
g>
1   0-0.19
£  0.2-0.49
g  0.5-1.99
DC
Z
111
o
I
o
DC
2.0+
   0
20
40
60
80
          PREVALENCE OF MODERATELY SEVERE PORTAL FIBROSIS
          AND PORTAL CIRRHOSIS, %
FIGURE 7-1   Correlation between  increasing concentrations of
             hepatic storage  iron and portal fibrosis in black
             subjects with  dietary iron overload.  Replotted
             from data of Bothwell and Isaacson.
                           217

-------
 serum hepatitis and possibly  even  chronic  anoxia.  In  the other  forms  of  iron


 overload the most  important factor is alcohol:  at least 25% of  the  cases of

                                                                            0*7*^
 idiopathic hemochromatosis have a  history  of excessive alcohol consumption.    '


 In  the dietary iron overload  of South African blacks,  alcohol is of  even


 greater importance because the source of the excess iron is a fermented

                   92  779
 alcoholic beverage.   '     Subjects with the greatest  concentrations of


 liver iron thus tend to be the heaviest drinkers, and  the ones whose diets


 are the least nutritionally adequate.  Analysis is further bedevilled  by  the


 development of secondary iron overload in  a small number of subjects with

                                                         O "I f\ £OQ QOfl
 alcoholic cirrhosis who absorb excessive amounts of iron.   '    '
        /

      Despite these complications,  the ubiquity of hepatic fibresis in  iron


 overload regardless of pathogenesis supports the notion that iron is central


 in  its production.  However,  iron  probably is only a low-grade fibrogenic


 agent that requires high local concentrations and protracted exposure  to  exert


 its effects.  Such an interpretation .would be compatible with a  necropsy


 study of dietary iron overload in  which 30% of the bodies with hepatic iron
                                                                         90
 concentrations above 2% dry weight  showed no substantive portal  fibrosis.


 It would also fit with the experimental observation that the siderotic liver
                                                                        298
 is particularly vulnerable to nutritional,  metabolic,  and toxic  insults.


      Iron overload and the pancreas.  Diabetes is one of the commonest clinical


 manifestations of  idiopathic hemochromatosis:  about 80% of idiopathic hemochroma


 tosis patients have diabetes.   Its occurrence in cases of dietary iron overload


 and transfusional hemochromatosis strengthens  the concept that iron overload


plays a part in its pathogenesis.   Evidently heavy iron deposits in the islet


 cells of the pancreas lead to  reduced insulin output and hence diabetes.   How-


ever, such a conclusion is not as straightforward as it would first appear.


Not only are the iron deposits much less marked in the endocrine than in the
                                     218

-------
               235,689

exocrine cells,        but insulin levels are only diminished in some diabetics

                                239,712

with idiopathic hemochromatosis.         In addition, familial glucose intoler-

                                                                       40,239

ance and cirrhosis also contribute to the pathogenesis of the diabetes.



Insulin levels have not been measured in South Africans with dietary iron over-



load, but they have been found to be reduced in patients with thalassemia.  In



summary, insulin lack, secondary to damage of islet cells by iron, may be one of



the factors responsible for diabetes in some patients with iron overload.




     Iron overload and the endocrine disorders.   Hypogonadism is common in


                                         89 711
subjects with idiopathic hemochromatosis.  '      Iron deposits are prominent



in the anterior pituitary, and hormonal studies have demonstrated low levels



of luteinizing and follicle-stimulating hormones    unresponsive to hypothalamic-



releasing hormones.    Testosterone levels are reduced  and do not rise after



administration of trophic hormones, although testicular iron deposits are



scanty.   '     Hypogonadism is found in siderotic thalassemics.  Its pattern



is similar to that in idiopathic hemochromatosis, although it is unknown if


                                                                 540
the primary defect is at the pituitary or the hypothalamic level.     The few



published results suggest that iron deposits in the pituitary may have a



selective effect on function in some patients.  The sex trophic hormones are


                     447
the most susceptible,    because the levels of other trophic hormones are



usually normal.   '     Interpretation of results is complicated by the sparse-


                                                   540
ness of normal standards for childhood and puberty.     The testicular failure



should be at least partly ascribable to lack of trophic stimulation.



     Iron overload and the heart.  The most direct evidence of iron toxicity


                                                    273
has been obtained for disorders of cardiac function.     Cardiac failure is



suggested as the commonest cause of death in idiopathic hemochromatosis, as is



true for thalassendc subjects whose lives have been prolonged by maintaining


                                                              255 540
an adequate hemoglobin level with repeated blood transfusions.   '     In
                                  219

-------
idiopathic hemochromatosis the cardiopathy is more marked when the clinical


                              89
disease materializes in youth.    Why this should be so is not known, but the



phenomenon may relate to the rate of iron accumulation.  Fatal cardiopathies




in young adults and adolescent subjects with thalassemla who have been



heavily transfused are particularly noteworthy, because the age factor tends



to discredit the role of variables such as excessive alcohol intake and



anemia in the genesis of the cardiopathy.  Cardiac complications have not been


                                   159
described in dietary iron overload,    and two possible explantations exist



for this apparent anomaly.  Few subjects accumulate appreciable amounts of



iron in the myocardium even when they are heavily overloaded with iron, and



those who do, succumb rapidly to hepatic failure.




     Iron overload and arthropathy.  The frequent occurrence of chondrocal-



cinosis in idiopathic hemochromatosis has led to speculation as to how the



iron deposits might contribute to its genesis.  It has been suggested that


                                                                    329
the excessive iron in the joint tissues may inhibit pyrophosphatase.     Since



this enzyme normally hydrolyzes  pyrophosphate to the more soluble ortho-



phosphate, its inhibition promotes pyrophosphate deposition.  Complications



in joints have not been described in other forms of iron overload.



     Iron overload, ascorbic acid deficiency, and osteoporosis.   The diet of



black South Africans with iron overload certainly does not contain optimal



amounts of ascorbic acid, but that alone will not explain the severe tissue


                                        fifti
depletion of ascorbic  acid  they  exhibit.      Ascorbic  acid nutrition is adequate



in nonsiderotic subjects whose diet is similar except for the beer, and much


                                                                               504
of the available ascorbic acid is rapidly catabolized in siderotic individuals.



When large doses of ascorbic acid are given to such subjects, only small



quantities of the vitamin appear in the urine.  Instead, more of an end oxida-



tion product of ascorbic acid, i.e., oxalic acid, is excreted.  Presumably
                                  220

-------
the rise results from the large deposits of ferric iron, as a similar phenomenon


                                                  790
has been observed in other forms of iron overload.     Abnormal ascorbic acid



metabolism also has been demonstrated in dietary iron overload by tracing the


                              330
radioactively labeled vitamin.     Under such circumstances, the ascorbic



acid is also rapidly catabolized, although not to oxalic acid; a large fraction



of the carbon-14 label is excreted as carbon dioxide.



     The ascorbic acid deficiency accompanying dietary iron overload has



several consequences, one of which is interference with the release of iron


                               477
from reticuloendothelial cells.     Such inhibition provides at least a partial



explanation for the prominent reticuloendothelial deposits in those subjects.



It undoubtedly accounts for their frequently normal plasma iron concentrations,



because administering ascorbic acid produces a sharp rise to more appropriate


       789
levels.     The relatively low degree of transferrin saturation may well be the



reason why the unloading of iron onto parenchymal tissues is so rare in this



form of iron overload.



     That the ascorbic acid deficiency of dietary iron overload is a manifesta-



tion of deranged vitamin metabolism and not merely the result of a poor dietary



intake has been confirmed by the finding of reduced tissue ascorbic acid con-



centrations in well-nourished subjects with other forms of iron overload,


                             790
especially thalassemia major.     In addition, administration of large doses of



ascorbic acid to patients with an adequate diet is also followed by elevated



oxalic acid excretion.  Finally, it has been possible to reproduce a similar



condition in animals.  Guinea pigs made siderotic by injecting iron dextran



develop severe ascorbic acid depletion, even when their diets contain adequate


                       791
amounts of the vitamin.



     There is little doubt that iron overload is connected with ascorbic acid



deficiency, but the relationship between iron overload and osteoporosis is
                                   221

-------
not so well defined,,  At least one of the links between the two conditions is



ascorbic acid deficiency.  Scurvy is known to be associated with osteoporosis



 in  children  and experimental animals, which has been ascribed to the necessity



 of  ascorbic  acid  for  ostepgenesis,  including collagen synthesis, osteoblast


                                  772
 maturation,  and osteoid  formation.     Bone resorption may be increased in



 ascorbic acid deficiency; semiquantitative microradiography of  the osteo-



 porotic bones of  ascorbic acid-depleted guinea pigs has revealed not only the



 expected diminution in the bone formation surface, but an increase in the bone


                    791
 resorption surface.      A similar enlargement of the bone resorption surface



 has been observed in  osteoporotic black South Africans with severe dietary iron


         791
 overload.     In  addition, the results of experiments with radioactive calcium



 suggest that decreased bone formation and increased bone resorption may be



 present simultaneously.     Finally, repletion with ascorbic acid may lead to



 decreased urinary calcium excretion in such subjects.



     Although all this evidence is compatible with the thesis that ascorbic



 acid deficiency induced by iron overload is responsible for the osteoporosis,



 especially when coupled with the occurrence of osteoporosis in other forms of



 iron overload, it is not conclusive.  Other factors—such as alcoholism,  malnu-



 trition, and associated liver disease—may all play a part.   However,  their



 specific roles remain to be elucidated.



     Iron overload and porphyria cutanea tarda.  Porphyria cutanea tarda is


                                        440
 common in siderotic South African blacks.     Other sources indicate that mild



 to moderate iron overload is frequent.   in porphyria cutanea tarda patients



 in other countries.    '      The majority of such patients have a history of



alcohol abuse, and hepatic fibrosis or cirrhosis are often found.



     Possible reasons for the association include the large iron content of



some alcoholic drinks,    the enhancement of iron absorption that alcohol



 itself may cause,     and the increased iron absorption sometimes observed in
                                 222

-------
                              820

alcohol-induced liver disease.     Indeed, greater than normal rates of iron

                                                                    753

absorption have been found in subjects with porphyria cutanea tarda.



     However, the association could be causal in the opposite direction, that



is,  increased iron deposits may affect the pathogenesis of the porphyria. For



example, rats given large doses of iron dextran become porphyric when given

                                                       729

hexachlorobenzene more readily than do control animals.       Of possible



relevance, too, is the finding that the rate of in vitro uroporphyrin synthesis


                                                             439
increases when ferritin and cysteine are added to the system.     Although



increased amounts of iron may potentiate the metabolic defect in porphyria



cutanea tarda, it is not an essential trigger, as the condition has been



observed in subjects with normal storage iron concentrations,,



     Effects of therapy in iron overload.  If increased iron deposits are



capable of damaging tissues and impairing their function, then removal of the



iron would be expected to reverse or at least arrest those changes.  In the



belief that iron is_ noxious, the treatment of idiopathic hemochromatosis by


                                                    269
venesection therapy was started almost 30 years ago.     The rationale was a



simple one.  It was argued that the deficit created (+ 400 mg iron/1 blood)



in Lhe red cell hemoglobin would have to be made good from the iron stores.



It was soon confirmed that patients with idiopathic hemochromatosis were



able to tolerate the removal of large amounts of blood for protracted periods,



so that it was possible to return the body iron content to normal within a

            89

year or two.     Subsequent studies assessed the effects of such treatment



on the function of the most severely affected organs and investigated



whether or not survival was prolonged.    It has not been possible



to institute venesection treatment  in siderotic black South Africans,  who are



typically among the less sophisticated members of the community and unwilling
                                 223

-------
 to see large amounts of blood removed.   No judgment can thus be made on its



 therapeutic effects in dietary iron overload.   For obvious reasons,  venesection



 has not been practical in subjects with transfusional hemochromatosis.



 However, attempts have been made to reduce body iron content in such subjects



 by using specific iron chelating agents such as desferrioxamine and  diethylene-

                                 25

 triaminepentaacetic acid (DTPA).



     ±t should be stressed that no controlled clinical trial has been per-



formed in which two carefully matched groups of patients with idiopathic hemo-



chromatosis have been monitored,  with one being venesected and the other not.



Despite this drawback, the uniform consensus nevertheless has been that  remov-



ing iron is of definite value to the patient.  In general, treated patients



live longer, and the longer survival is  often accompanied by weight gain,


                                                                          80 819
reduced skin pigmentation and hepatomegaly, and improved cardiac function.  '




     Gross disturbances in hepatic function are unusual at the time of



diagnosis, and therapy usually corrects  minor abnormalities.   isolated  reports



exist of 6 patients in whom hepatic fibrosis or cirrhosis was judged  on  the


                                                          305
basis of liver biopsy findings to have reverted to normal.     In a more



systematic study of 85 patients,  5 who were initially cirrhotic were judged to

                                           QQ

exhibit only hepatic fibrosis 3-9 yr later.    However, difficulties in inter-



preting liver histology on the basis of  needle biopsy are well known. There-



fore, no final judgment is possible on the degree to which cirrhosis in



idiopathic hemochromatosis is reversible by iron removal.  There is no evidence


                                                             80
that portal hypertension is benefited by venesection therapy.    The incidence



of hepatoma does not appear to be reduced by treatment.  In fact, the disorder



may even be greater in treated subjects, although the incidence could be a



function of their longer survival rate.   Of 45 patients dying from the disease

                                             on

in one study, 13 (29%) succumbed to hepatoma.    Ten of the 13 completed a
                                  224

-------
course of venesection therapy, while 3 died within 6 mo of commencing therapy.
Three patients died within 3 yr and the remaining 7 died 4-15 yr after the
initial venesections.

     The degree to which glucose intolerance can be modified by therapy has
been carefully documented by Williams and his coworkers, who studied 85 hemo-
chromatotic patients, about two-thirds of whom were diabetic at the time of
             80
presentation.    The effects of venesection therapy were modest at best.
Improved glucose tolerance, designated as a reduction in insulin requirements
of at least 12 units/day, or a conversion to oral hypoglycemic therapy, occurred
in only 28% of the patients.  Fifty-eight percent showed no change in require-
ments for antidiabetic agents, and 14% needed more insulin after venesection
therapy.
     Cardiac function often improves after removing excess iron, and venesection
treatment undoubtedly has reduced the number of deaths ascribable to the
                     OAO OA1 ^ 1 *^ S99 ftQA
specific cardiopathy.   '   '   '   '     Many patients have not required any
maintenance cardiac therapy after the iron has been removed, which further
validates the efficacy of such therapy.  Timely treatment may also prevent onset
                                                                      819
of clinical manifestations of cardiac disease in affected individuals.
     No relief of hypogonadism or arthropathy has been recorded after
                      80
venesection treatment.
     The most complete statistics on the effects of therapy on survival are
                               80
those of Bomford and coworkers.    They have updated and extended their earlier
study, in which they found that the 5-yr mortality in 40 phlebotomized patients
                                                                      819
was only 11%, compared to 67% in a retrospective group of 18 patients.
The newer figures are based on 85 phlebotomized patients and a control group
of 26 subjects made up of the original 18 patients plus another 8 who had
refused therapy, died soon after diagnosis, or in whom diagnoses were made at
                                   225

-------
necropsy.  The survival at 5 yr was 64% in the treated group and 15% in the




untreated one.  Patients who had died from both groups were matched by age and




by  the presence of complications, including diabetes, cirrhosis, hepatic





failure, and hepatoma.  The analysis again yielded a highly significant




difference, shown in Figure 7-2.  The mean log survival in 45 dead treated




patients was 4.3 yr, compared to 2.9 yr in the untreated ones.  Twenty-two




percent of the treated group succumbed to neoplasms other than hepatoma,




whereas none did in the untreated group.  In another study, retrospective




analysis also revealed an improved survival rate in venesected patients, with



3 deaths in 19 treated individuals as compared to 4 deaths in the untreated




group of 7 subjects.



     Data on the effects of chelation therapy in thalassemia major are limited




but encouraging.  Two groups of children were followed from 5-8 yr.  The




treated group was given 0.5 g desferrioxamine intramuscularly 6 days a wk


                                             45 649
as well as 2 g DTPA/unit of transfused blood,  '     In both groups, hemoglobin




concentrations were maintained between 8-15 g/100 ml, and at the start of the




trial, the 9 children in each group were matched according to age, amount of




blood transfusedj and hepatic iron concentrations.  At the end of 6 yr,



significant differences were measured in the concentrations of iron in the




liver; in the group given chelating agents, mean values were 2.6% dry weight,



as compared to 4.27o in the control group.  The extent of hepatic fibrosis was




also carefully assessed, using an index obtained from camera lucida drawings



of liver biopsy specimens.  Initially the fibrosis indices overlapped widely,




but by the end of the study a highly significant difference had emerged;  the



fibrosis index had become more marked in the untreated group, yet had remained




the same in the chelated group.
                                  226

-------
                             YEARS AFTER DIAGNOSIS
FIGURE 7-2   Cumulative survival  after diagnosis in 45 treated
             and 18 untreated patients with idiopathic hemo-
             chromatosis.  The vertical lines at each interval
             represent +1.standard error.  Reproduced from
             Bomford et~al.80
                              227

-------
      These differences might  have been more dramatic  had  more  effective  chela-


 tion therapy been available.   Desferrioxamine  is  effective  only after  a  certain


 degree of siderosis  has been  reached,  equivalent  to 37.6-47  1  of transfused

       540
 blood.      Therefore,  it only is  possible to contain  the  iron  overload within


limits well within the range found in untreated idiopathic hemochromatosis.


Despite the limitations, the arrest of the hepatic fibrosis by such therapy  .


provides strong indirect confirmation of the fibrogenic potential of heavy


iron deposits in the liver.  Other noteworthy clinical observations also  came


out of this investigation.  Puberty was delayed in 4  out of 5 control children,


but in only 1 out of 4 treated patients, and a  growth spurt  had occurred  only


in those subjects in whom puberty was not delayed. The incidence of diabetes


and its clinical manifestations were similar in each  group.   In another study


in which a larger dose of desferrioxamine was given,  some  improvement in


clinical symptoms and signs was noted—the electrocardiogram and liver  function
                                    684
tests showed a return toward normal.


     Mild to moderate siderosis commonly accompanies  porphyria cutanea tardaa


Clinical manifestations  subside and porphyrin synthesis decreases when iron

                                           639
stores are depleted by venesection therapy.      It is uncertain if only the


removal of  iron is important for  improvement.  The subject who embarks on a


venesection program may  well also stop drinking, which may contribute to the


improvement.  Iron nevertheless appears to play some key role, since urinary

                                                             9 Sfi
porphyrin excretion has  diminished during venesection therapy     even in

                                     497
patients  who have continued to drink.     Biochemical relapse has been induced

                                498
by replenishing the iron stores.


     Animal models of iron overload.  The discussion thus far  is  based on


analyses  of human disease and  its treatment.  For a closer examination of the


chronic  toxicity induced by the  metal, an acceptable animal model is neces-


sary.  Unfortunately, attempts to create such a model have been unsuccessful.



                                 228

-------
     Much work and many  ingenious approaches have not produced hemochromatosis



 by dietary manipulation.  Various species have been fed diets supplemented



 with iron, and a number  of methods have been used to try and increase  iron



absorption.   They have included reducing the phosphate content  of the diet,



cyclic starvation and feeding, ligation of the pancreatic duct,  splenectomy



and administration of such substances as DL-e£hionine,  polysorbate 20 and


           89
D-sorbitol.     Despite such maneuvers, body iron concentrations  usually have



stayed much lower than those found in idiopathic hemochromatosis.  In most



instances,  the iron has been confined to the liver and reticuloendothelial



system, although minimal deposits have sometimes been noted in the pancreas,



thyroid, and myocardium,,   Tissue damage such as hepatic cirrhosis or pancreatic



fibrosis has never been produced by simple dietary iron overload.  Success



was, however, claimed by workers who gave a choline-deficient,  high-fat diet

                                                CQQ

and excessive iron to rats for more than a year.     The animals developed



fatty livers and progressively increasing cirrhosis;  the changes could be



prevented by folic acid.      However, the hepatic iron concentrations were



relatively modest (only 6 times normal), so that their relevance to hemo-



chromatosis  is questionable.   Similar doubt must be cast on studies in which



cirrhosis has been produced in rats by feeding iron supplements  plus carbon

              421

 tetrachloride,   although these experiments do support the contention of Golberg


         298
and Smith    that iron overload makes the liver more susceptible to other



noxious agents.   Their view has not gone unchallenged,  however,  as other



workers have been unable to demonstrate such difference when they used

                                   OOQ onc

ethionine as the second supplement.   '     The discrepancies may be explain-



able by the finding that iron loading does not potentiate the early "hepatitic"


                                                                       598
stage of ethionine injury, but rather the second "cirrhotogenic" stage.



This second stage involves interaction between excess iron and the cell membrane,


                                             421 824
but the exact mechanism has not been defined.    '



                                229

-------
     Iron has also been administered parenterally to animals as colloidal iron,
saccharated iron oxide, iron ascorbate gelatin, iron dextran, and red blood
cells.  It is usually injected intravenously, but the intraperitoneal and
intramuscular routes also have been used.  The final body iron content has
varied between 0.1-3.3 g/kg body weight, and animals were tested anytime
                        89
from 4 wk to 7 yr later.    When animals are given very large doses of iron,
certain pathologic changes are produced, but they have been unlike those found
in human hemochromatosis.  In guinea pigs, for example, subcutaneous injections
of inorganic iron to a total dosage of 1 g/kg body weight caused hemorrhagic
phenomena in the lungs and adrenals and patchy parenchymal damage of the
      585
liver.     However, neither cirrhosis nor pancreatic fibrosis was induced.
The only morphologic changes in rats subjected to huge doses of parenteral iron
were proliferation and hypertrophy of the hepatocytes, ascribed to the induc-
                          232
tion of protein synthesis.     In another experiment, massive iron overload
(2.5-3.'i g/kg body weight) was produced in dogs.   It caused anorexia,  apathy,
and weight loss with death after 5-10 mo, but autopsies revealed little morpho-
                               106                                         479
logic evidence of liver injury.     Greater success was achieved by Lisboa,
who produced cirrhosis in dogs after 4 yr by injecting iron intravenously as
iron dextran to an accumulated dosage of 3.5-5.8 g/kg body weight.    It
should be noted that the tissue iron concentrations in these studies were
several times' greater than those encountered in idiopathic hemochromatosis,
and that no pathologic changes appeared even after extended periods during
another study in which the iron levels were comparable.     Tests for hepatic
and cardiac function and for glucose tolerance in these dogs were normal.
Serum proteins,  liver function as measured by sulfobromophthalin sodium, and
glucose tolerance have also been found to be normal in iron-loaded rats and
rabbits.     Animal experiments in which siderosis has been accompanied by
alcohol have also been negative.   When rhesus monkeys were given iron dextran
intravenously (0.5 g/kg) and were then exposed to alcohol, or alcohol plus a
                                 230

-------
low-protein diet, or carbon tetrachloride,  no hepatic fibrosis was apparent


                                       564
in any of the groups even after 110 wk.



     The reasons for the almost uniform failure of parenteral iron-loading



experiments are several.  The first is undoubtedly the predominant reticulo-



endothelial localization of the injected iron complexes.   Although some iron



may be redistributed eventually, the amounts taken up by parenchymal cells



are usually small.  In one study in dogs, for example, almost no iron moved



out of the reticuloendothelial cells even after the passage of 7 yr.



Therefore, the pattern is totally dissimilar to that found in hemochromatosis.



If these experiments prove anything, it is  the capacity of the reticulo-



endothelial system to tolerate huge amounts of iron.  Another drawback is the



necessarily short duration of most such studies, compared to the length of



exposure in the human diseases.  Finally, species may well differ in their



susceptibility to the noxious effects of excessive tissue iron.



     Mechanisms involved in iron toxicity .  Despite the largely negative



results of attempts to produce experimental hemochromatosis in animals, the



cumulative experience in human subjects suffering from iron overload of



diverse etiologies strongly suggests that iron i£ noxious to tissues.  For it



to exert these effects, two criteria must be satisfied:  the metal must be



present in sufficient concentrations for long periods of time, and it must be



present in parenchymal calls.  In contrast, the reticuloendothelial cells



appear do be admirably equipped to store excessive quantities of the metal.



(In the dietary overload  encountered in South African blacks, the massive



quantities of iron in the spleen and reticuloendothelial cells of the bone



marrow do not appear to exert any direct pathologic effects.)  Why this strik-



ing difference in cellular susceptibility exists is not known.  Knowledge of



the mechanisms by which iron does damage tissues is itself largely speculative.
                                   231

-------
     Iron has been shown to accumulate in liver cell lysosomes of iron-loaded

rats,    and liver biopsies from humans with hemochromatosis have also revealed

dense deposits of ferritin and hemosiderin in structures tentatively diagnosed
             /"O oi Q
as lysosomes.  '     The activity of hepatic lysosomal enzymes has been shown

to be markedly increased in patients with various types of iron overload and their

                                                   61 9
lysosomes were strikingly more fragile than normal.     Perhaps iron accumula-

tion damages the  lysosomal membrane, which then releases acid hydrolases into

the cytoplasm and initiates cell damage.       These abnormalities return to

normal after removing, excessive iron from the liver, which suggests that the iron
            612
causes them.   However, the mechanism by which it injures the lysosomes is not

known.  The iron  deposits may  catalyze  the formation of  free radicals  and

they, in turn, may damage lysosomes and  other subcellular organelles through
                   /••t r\
lipid peroxidation.

     The other metabolic abnormality  induced by  iron overload  is depletion  of
                                            159
ascorbic acid levels in tissues (see above).     Whether chronic subclinical

depletion of the vitamin has effects other than scurvy is not known.

Lack  of its nonspecific antioxidant action may damage  susceptible  tissues.

Serum concentrations of vitamin E  (also an antioxidant)  have been  found  to  be

                               384
decreased in thalassemia major.



CARClNOGENESIS, MUTAGKNESIS. AND TERATOGENESIS

      Carcinogenesis needs to be considered as a  complication of iron deficiency,

of  iron overload  states, and of exposure of specific tissues to iron.  In connec-

tion with epithelial changes in the esophagus of patients with chronic iron

deficiency, a  few reports have appeared in the older literature of  carcinoma of
                  382
the mid-esophagus.   Althoughthis condition has  been assumed to be  related  to

iron  deficiency,  the causal relationship is poorly  established.  A  more  definite

association has been found between carcinoma of  the  liver and  parenchymal iron
                                   232

-------
overload, described in approximately 15% of patients with idiopathic hemochrom-
       273
atosis.     The malignancy may become manifest years after the removal of excess
     819
iron;    presumably the changes have a long gestation period not reversible by

iron removal.  A remote possibility exists that the malignancy is caused by

some other substance that is also absorbed in excessive amounts by hemochroma-

totics and which is not removed by phlebotomy.  Neoplasms of other organs also
                                                819
are more frequent in idiopathic hemochromatosis.     There is no suspicion of

local malignancy accountable to gastrointestinal ingestion of inordinate amounts

of iron.

     Intramuscular injections of iron dextran into animals have produced sar-
     298,496
coma.         This finding has implications for the therapeutic use of this

substance in treating iron-deficient patients who cannot respond to oral iron.

Isolated case reports of local malignant change in such patients give some sub-
               309
stance to this.     However, iron dextran is now given largely intravenously,

which should circumvent the possibility of risk.

     Iron has not been reported to be mutagenic or teratogenic.
                                233

-------
                                CHAPTER  8



                            INHALATION OF  IRON
SOURCES OF EXPOSURE



For Animals



        The exposure of animals to inhaled iron is almost exclusively a



laboratory phenomenon.  Syrian golden hamsters, guinea pigs, mice, and rats


                                                       97a
are the most commonly used animals in such experiments.     Laboratory




exposure  usually  is  through  natural  inhalation or  through  intratracheal  injec-



tions of iron oxide particles of 5 urn or less and  in single doses as high as



7 mg/m3 or a 2% solution.21'3483



        Although some reports   '    suggest that  pulmonary fibrosis can



occur after inhalation of iron oxide, others studying tumor formation   '   '



failed to note fibrosis in the lungs when iron oxide served as a vehicle for



benze/faj^pyrene and other polycyclic hydrocarbons found in coal tars.    '
        The lungs of animals exposed to ambient iron oxide near steel mills



and mines have not been examined for the effects of such exposure per se .



Cattle exposed to airborne dust



particularly high in fluorides near steel plants have been autopsied.  Their



lungs showed anthracosis, along with slight to moderate involvement of the



regional lymph nodes with carbonaceous material.  No evidence of any fibro-



sis or deposition of iron oxide was found.  These reactions decreased as the



distance from the plant increased.



FOT Humans



        Human exposure to inhaled iron generally comes from an occupational



source.  The following occupations present risks of inhalation of dust and
                                    234

-------
fumes of iron and its alloys and compounds:  iron-ore mining, arc welding,



                                                                379
metal (iron) grinding, iron and silver polishing, metal working,    sintering,




scarfing (personal communication, M. Bundy), and pigment manufacturing and




rubber manufacturing.


                                       173
     The Industrial Hygiene Association    issued an ambient air quality guide




for iron oxide, Fe 0~.  It stated that no potential acute or chronic hazards




were observed in exposed humans and animals even at the highest air concentra-


                                 833
tions reportedo  Iron oxide alone    is "biologically inert" in that it




produces no direct tissue injury in experimental animals or humans as a result




of exposures in the work place.




EFFECT AND FATE OF INHALED IRON



In Animals




     Pulmonary clearance.  When inhaled, iron oxide particles penetrate into



                           O (if.

the lung parenchyma of mice    and may be retained for more than 100 days.  The




particles are observed inside macrophages collected around terminal bronchioles


                             275
and lymphoid tissue.  Fisher,    using Freund's complete adjuvant (FCA) to




study clearance of iron-59 oxide in the rat,  found that clearance was increased



when iron oxide was administered 1-3 days after the use of FCA, because then the




response of free macrophages was the greatest.   The amount of iron cleared in




the first 24 h as well as the rate of clearance during the later phases were




increased compared to control animals 40 days after exposure.  These findings




suggest that phagocytic action is important in both early and late clearance



phases.




     Saffiotti et al.    reported that a small portion of the iron-oxide dust



introduced into the lungs of Syrian golden hamsters remained within the lungs



and the tracheobronchial lymph nodes throughout the lifespan of the animals.
                                    235

-------
Macrophages engulf the iron particles that reach   the alveoli, penetrate



the alveolar wall with this burden, and ultimately transport the particles



through this means.  Some of the particles are carried to the trachio-



bronchial lymph nodes.  This process of deposition, retention, and clearance



is not unique to iron oxide—it is common for all particles Syum or smaller



that enter the alveolar spaces.  Alveolar macrophages appeared to affect


                                                                    461
the clearance of radioactive iron dust from cats, rats, and monkeys.     The



bronchi of all species were almost completely cleared in less than 2 days,



whereas the alveoli took 16-28 days in the rata and cats and 280-300 days in



the monkey.


                                                                           626
     Pulmonary toxicity.  Using a scanning electron microscope, Port et al.



found acute changes in the surface morphology of hamster tracheobronchial



epithelium following thrice-weekly intratracheal administration of ferric



oxide in 10 doses of 5 mg each.  These multiple infusions of ferric oxide



particles smaller than 5 yim suspended in an 0.9% salt solution produced



a loss of ciliated cells and broad areas of abnormal, enlarged, unciliated



cells with roughened or wrinkled surfaces.  The abnormal cells were thought



to be areas of epithelial hyperplasia but their significance remains unknown.



This is the first and so far the only report of epithelial changes induced



from administration of relatively large doses of ferric oxide.  Whether lower



doses given over a longer exposure time would produce the same response or if



changes would be found in the lower airways of the bronchial tree is not known.



The scanning microscope seems to be a useful tool for exploring reactions to



what was previously considered an "inert" dust particle.



     Iron as a vehicle.  Iron oxide can serve as a vehicle to transport


                                                                   21
pollutants into the body via the air passages.  Amdur and Underbill   exposed



guinea pigs to open-hearth dust (90% iron oxide) alone or in combination with
                                   236

-------
several concentrations of sulfur dioxide.  Open hearth dust alone produced


no detectable respiratory response even when given at a concentration as high

         o
as 7 mg/m .  Neither did it affect the response to any of the sulfur dioxide,


which was administered in doses of 1.6-2.6 ^ig/ml.  The findings were similar

                             21
for an aerosol of iron oxide.    Soluble iron salts,  however, do potentiate

                            21
responses to sulfur dioxide.


In Humans


     Pulmonary clearance.  The fate of iron particles that enter the human


lung through occupational exposure has been studied.   Lung biopsies of seven


welders with siderosis have shown that some iron remains free in the alveoli


and bronchioles, but most of it is taken up in macrophages and transported


into the lymphatic channels.     Much of the incoming dust is collected on the


mucus blanket and is expectorated; the dust is recognized easily by the worker

                          327
because of its rust color.

                                       47? ASA A«S 64^
     Injury to the lung.  Investigators   '''    who studied the


mortality patterns of 59,000 steelworkers for 13 yr (1953-1966) were unable


to find any toxic effect that might be attributable to the inhalation of iron


oxide.

                                            •k
     In their study of 4,588 crane operators  in steel plants where exposure


                                          472
to ambient iron oxide occurs, Lerer et al.    reported that  the total mortality


experience, as reflected  in  a relative risk of  1.00, was the same as the


mortality experience of workers who had never operated cranes.  When analyzed


by specific causes, the Caucasian craneman had  a 10% deficit mortality risk


for  lung cancer.  That  is, 51.3 deaths were predicted and 47 were observed.
ic
 Crane operators were selected because of reports in the literature that they

stood an excess risk of dying from lung cancer.
                                    237

-------
The relative risk from other respiratory diseases was 0.81 (24 deaths



reported versus 28.8 expected), a decrease when compared to other occupations



in the same plant.  In a long-term study of 9,655 open hearth workers,



mortality from lung cancer was usually less than 1.00 and- mortality from non-



malignant respiratory diseases was similar to that of other steelworkers in


                                        643
the same plant, a relative risk of 1.00.



     Caucasian crane operators showed an overall excess risk of dying from


                                                                472
cardiovascular and renal diseases, with a relative risk of 1.09.     The excess



risk seemed to be related to particulate exposure,with the greater risk in



areas of greater particulate exposure.  The mortality sample of non-white crane



operators was too small to determine relative risks for cardiovascular disease.



Open-hearth workers showed a deficit in mortality from cardiovascular diseases



with a relative risk of 0.90, which is statistically significant at the 1%



level.  If particulates, i.e« iron oxide, are related to this phenomena, as



proposed in Lerer and coworkers* discussion of crane operators, then it is



exceedingly difficult to explain the deficit in cardiovascular disease mortality


                                                    643
in the open hearth workers studied by Redmond et al.



     The open-hearth workers had an excess mortality risk (17.6 deaths pre-



dicted versus 31 found)- from diseases of the digestive system, particularly



ulcers.  The relative risk was 1.22, which was statistically significant at the



5% level.  Heat stress and physical exertion, rather than particulates, may



have been at the root of the disorder.


         489
     Lowe    did not note the influence of iron oxide in inducing chronic



bronchitis and emphysema in steelworkers he examined in regard to exposure to



atmospheric dusts in the workplace.  The disorders were related only to the



smoking habits of the steelworkers.
                                   238

-------
                       732
     Teculescu and Albu    performed several lung-function studies on



14 workers with a mean age of 43 and who had been exposed to pure iron oxide



dust for an average of 10 yr.  They did not find abnormalities compatible



with pulmonary fibrosis, although chest radiographs showed nodular opacities.



These opacities are symptomatic of pulmonary siderosis.



     Siderosis,  the accumulation of iron oxide in the lungs, is a form of



pneumoconiosis.  Detailed information on the X-ray appearance, pathologic


                                                                       327 552
and physiologic responses of the disorder has been published elsewhere.   '



If exposure to pure iron oxide ceases, the dramatic nodulations on the chest



X-ray disappear.   '     They would not subside if fibrosis were a feature of



the pneumoconiotic nodule of siderosis.



     Although the term pulmonary siderosis is reserved for the accumulation of



pure iron oxide,    it is probable that inhaling pure iron oxide never leads to



fibrotic pulmonary changes, whereas inhaling iron oxide plus certain other



substances most certainly does.  Reports of pulmonary fibrosis            in



hematite miners   '    indicate that prolonged exposure, usually over 20-30 yr,



is required, although some signs have been noted after 10-yr exposures.  It



should be noted that the ore to which the miners were exposed contained



approximately 10% free silica, a known fibrogenic agent.  When pulmonary fibrosis



seems to be associated with occupational exposure to iron oxide, careful



investigation usually will uncover a simultaneous exposure to free silica.



Thus many exposures are to a mixture of materials and these additional particu-



lates (silica) may indeed produce fibrotic changes in the lungs.  The resulting



disease is called silicosis.
*
 In this chapter, siderosis only refers to pneumoconiosis resulting from the

inhalation of iron particles, and not to the cases of excess iron in the

blood or tissues discussed in Chapter 7.



                                  239

-------
     Some iron compounds are injurious when inhaled.     However, the danger-


ous properties of an iron compound are a function of the ligand with which the


iron is coupled.  Thus ferric arsenate and ferric arsenite possess the poison-


ous property of the arsenical ligand.  Similarly, iron pentacarbonyl, Fe(CO),.,


known as iron carbonyl, is one of the most dangerous of the metal carbonyls,


having toxic and flammable properties.


     This metal carbonyl is a yellow-brown liquid at ambient conditions and is

                  lOOa
highly flammable.      Its presence may be suspected whenever high partial


pressures of carbon monoxide come in contact with iron or steel vessels.  Iron


carbonyl is formed in gas-manufacturing processes, and it (along with nickel


carbonyl) must be removed to eliminate soot formation at the ultimate burning


point.  Iron carbonyl may be formed in illuminating gas that has passed through


iron pipes, in water gas, in coal gas stored underground, and in gases coritain.-


ing carbon monoxide held under pressure in steel cylinders.  Traces of this


carbonyl have been found in gases produced by gasification of oil or refinery


gas with steam over nickel catalysts at atmospheric pressure.  Liquid iron


carbonyl is used as an antiknock agent in some gasolines.


     The signs of iron carbonyl poisoning in animals are similar to those of


other metal carbonyls and include respiratory distress, cyanosis, tremors, and


paralysis of the extremities.  Death may be immediate or delayed for a few


days.  Human exposures similarly produce signs and symptoms similar for the


common metal carbonyls.  Immediate exposure produces giddiness and headache,


and they are occasionally accompanied by dyspnea and vomiting.  The signs and


symptoms may be relieved if the individual is removed to fresh air; however,


dyspnea returns 12-36 h later, with the addition of cyanosis, fever, and cough.
                                  240

-------
If the exposure has been to lethal concentrations, death follows in 4-11 days.



Pathologic changes include pulmonary hepatization, vascular changes, and



degeneration of structure in the central nervous system.




CARCINOGENESIS*


                94a
     Boyd et al.    studied the mortality of Cumberland iron-ore miners through



death certificates of 5,811 men from 2 mining communities and who died between



1948 and 1967.  Comparison of the iron miners' experience with that of other



local men and the relevant national experience led to suspicions of an occupa-



tional hazard of lung cancer associated with hematite mining.  Earlier,



Faulds and Stewart   a had reported an increased risk in Cumberland iron-ore



miners after studying post-mortem findings.  During the 20-yr period studied by



Boyd, among all iron miners (underground and surface workers) the 42 lung



cancer deaths observed were 50% higher than the number expected (28), an



increment statistically significant at the 5% level.  However, the number of



deaths from other cancers (74) was very close to what was expected (71).  Com-



paring underground and surface workers, the excess of lung cancer was confined



to the underground group (observed 36, expected 20.58); there were 6 deaths



among the surface workers, and 7.13 had been expected.  The higher incidence



of lung cancer mortality in the underground miners persisted when tested by



the same method against the proportionate mortality determined from the



national standards.  Post-mortem examinations revealed a high proportion (37%)



of oat-cell tumors.  This percentage was similar to the figure (43%) found by



Saccomanno et al.     for neoplasms among uranium miners in Colorado.  Data



on radon concentrations in the Cumberland mines were available for 1968, the
 See Cole and Goldman's review of occupational carcinogenesis for a com-

 parison of risks from iron and other metallic dusts and fumes.170a
                                  241

-------
year after the completion of the Boyd et^ al. study.   In 3 of the 4 mines
tested, none of the levels—ranging from 30 pCi/1 to 300 pCi/l--were below
30 pCi/1, the maximum permissible level recommended by the International
Commission on Radiologic Protection.  The lung cancer risk in the British
miners was much less than that estimated for radium mines in Schneeberg,
Germany, Jacymov, Newfoundland, and Colorado.  The radon concentration in
these latter mines was many times that found in the Cumberland mines.  The
average was 100 pCi/1 at Cumberland compared to 2,900 pCi/1 in Schneeberg
(even after considerable improvements were made); the highest radon concentra-
tion measured at the Cumberland mines (320 pCi/1) was far below the respective
measurements of 25,000 pCi/1 and 59,000 pCi/1 for the Newfoundland and
Colorado mines.  It may be, therefore, that the limit of a twofold increase
in lung cancer estimated for the West Cumberland iron miners is compatible
with their lesser exposure to radon in comparison with that of the other groups
of miners, for whom risk estimates extend upwards of 10 times normal for the
Colorado miners.  The exposed rock in the Cumberland has not revealed unusual
radioactivity, and it is speculated that the radioactivity is carried by the
mine water, which is believed to be the source of it in the Newfoundland
                94a
fluorspar mines.     Others have thought radon to be the probable causative
                                                                       552
agent in the development of lung cancer in hematite underground miners.
                     *5 OT"U
     Ishinishi e± al.     repeatedly instilled iron dusts collected from an
open-hearth furnace as well as dusts containing benzo^a/pyrene into the
tracheas of albino rats.  The suspensions injected were 7.5 mg iron dust,
7.5 mg iron dust plus 1 mg benzo^a/pyrene, and 1 mg benzo/a/pyrene suspended
in 0.2 ml distilled water.  The iron dusts and benzo^a/pyrene particles were
no larger than 5 ^um.  The iron dusts contained 52% iron and less than 1% nickel,
                                   242

-------
chromium, and arsenic.  No difference in the death rate occurred between the



experimental groups (8 out of 90) , and the control group (4 out of 30) .  Both



groups had been followed throughout their entire life.  After 15 weekly instal-



lations spread over about 4 months, one male rat out of 14 had a malignant



tumor and 3 female rats out of 15 had tumors in the iron-dust group.  In the



iron dust plus benzo/a.7pyrene group, 8 out of 17 males and 3 out of 9 females



had tumor So  In the benzo/q/pyrene group, 3 males out of 13 and 3 females out



of 14 had tumors.  There were no lung tumors in the control group comprised of



12 males and 14 females.  Several sarcomas of organs or tissues were detected,



but they were also found in the controls.  The lung tumor rates were about



twice as high in the iron dust plus benzo/aj^pyrene group as in either alone.



Because the dust from the iron furnace contained a small amount of nickel,



chromium, and arsenic (known carcinogens) ,  it may not be possible to conclude



with certainty that iron itself has a carcinogenic potency.  (This work also



reviewed Saffiotti's experiments,    which showed that ferric oxide alone did not



produce respiratory tumors.)



     A link with carcinoma of the lung has not been reported in surface mining
           o o£                                      Ofl"7 A 1 A

operations.     Although questions have been raised,   '   '    it has been



generally concluded that neither iron ore dust (hematite) nor ferric hydroxide



predispose to malignancy in humans.



     Ferric oxide given to hamsters, mice, and guinea pigs by inhalation or by



intratracheal route has not been found to be carcinogenic.     Ferric oxide



particles can serve as a vehicle for transporting known carcinogenic agents



(e.g., benzo/q7pyrene) into the lungs of animals and can produce squamous



metaplasia as well as lung cancer under such circumstances.   '   '   '   a>


683a,705b  _.        £J1        „            •,*,•,,            , 626
           The use of scanning electron microscope as described by Port et al.



may prove to be extremely useful in determining if neoplastic changes in



bronchial epithelium occur with exposure to iron.





                                   243

-------
                                 CHAPTER 9




                                  SUMMARY







IRON IN THE ENVIRONMENT




     Iron comprises about 5.4% of the continental crust.  Its concentration




in soils ranges from 0.7-4.2%, where it exists primarily as ferric oxyhydroxides




and in clay minerals.  Crude ores have from 20-69% iron, usually in the form




of iron oxide minerals.  Iron mobility in the regolith generally increases at




lower oxidation potentials and pH's in flooded soils and subaqueous sediments




where ferrous iron can exist in solution, at lower pH's and highly moist




aerated soils, in the presence of complexing organic compounds dissolved in




soil solutions, and with decreasing concentrations of iron-precipitating anions




such as sulfide, carbonate, and phosphate.  Bacteria also influence the trans-




port of iron by increasing the rate of iron oxidation, producing complexing




organic substances or inorganic acids through their metabolism, or oxidizing




or utilizing organic parts of chelates to release iron that can then accumulate




as ferrie oxyhydroxides.




     At chemical equilibrium in a well-aerated river, the dominant forms of




iron usually are ferric,.  Unless organic or inorganic complexing agents are




present, the solubility of ferric iron cannot be more than a few micrograras




per liter at pH's greater than 4, and the solubility at neutral pH is much lower.




River water generally contains some particulate ferric oxyhydroxide, however,




which may be in the colloidal size range.  Iron concentrations in sea water




range from a few tenths of a microgram per liter up to about 3 jug/1.  In mildly




reducing environments, dissolved ferrous species are dominant and iron




solubility at equilibrium may exceed 50 mg/1 at pH 6.  Ground waters in many
                                    244

-------
areas of the United States contain appreciable amounts of ferrous iron.  Con-


centrations between 0.5-10 mg/1 are common in such areas and larger concentra-


tions are occasionally measured.  When these waters are aerated, the iron is


oxidized and precipitated as ferric oxyhydroxide.  In strongly reduced systems,


iron may be retained by ferrous sulfides, which are highly insoluble until the


sulfur is oxidized.  Metallic iron is chemically unstable in water and many


water supplies contain iron that has been derived from corrosion of pipe and


other metal surfaces.  Organic complexes of ferrous and ferric iron are found


in surface and ground waters, and may increase the solubility of iron or


influence its reactivity in other ways.  Redox and precipitation reactions of


iron generally are fast at near neutral pH levels, and chemical equilibrium is


attainable.  Many of the processes are mediated by microorganisms.


     Iron in the atmosphere in remote areas of the United States and United Kingdom

                                                                  o
is present in year-average concentrations as low as 0.05-0.09 Aig/m ,whereas in a U. So

                                                                       3
iron and steel center, it has reached a 1-yr maximum average of 12>ig/m  and a 3-mo

                                        3
maximum average concentration of 16/ig/m .  Atmospheric iron in remote areas is large-


ly soil-derived.  The 5-yr   (1970-1974) median concentration at remote sites in

                                           3                          3
the United States was found to be 0.26/ag/m ; at urban sites, 1.3yug/m , or


5 times as high.  Maximum concentrations tend to occur in the windy season,


i.e., in late winter and early spring.  Lowest urban iron concentrations are


found in coastal states..  The iron fraction of soil-derived particulates is


roughly 1 part in 100 or 200; the global rate of entry of iron into the atmos-


phere is of the order of 1,000 metric tons/day.


     World production of pig iron and raw steel within the last 5 years has


reached 500 million and 700 million metric tons and continues to increase.  The


U. S. contribution, more than half of which comes from Pennsylvania, Ohio, and


Indiana, amounts to 17-19% of the world total.  Total particulate emission



                                    245

-------
factors for individual manufacturing operations are known to be affected


markedly by the type and degree of pollution control employed.  Estimation


of atmospheric contaminant concentrations resulting from given emission rates


is common practice.  Estimation of soil quantities raised by the wind is less


common, but practicable, and some measure of control through cultivation


methods is available.  The overall annual average concentration of total air-

                                                                 3
borne particles in the United States decreased from 80 to 66/ig/m  between 1970


and 1974.


     The mining and processing of iron or certain materials containing iron


minerals affect the environment in several ways.  Waste accumulated from iron


ore mining presents a problem only when sulfides are present to produce acidic


drainage.  Extraction of aluminum from bauxite results in wastes containing


1-25% ferric oxide, and these residues amount to about 5.6 million metric tons


annually.  Most of the wastes are deposited in adjacent lakes.  If uncontrolled,


emissions from steel mills can contribute appreciable quantities of iron oxides


to the air but the only problem demonstrated to date has been soiling.  The


carrier properties of these particles have not been adequately investigated,


and may be of concern.  Wastewaters high in iron produced by the steel industry


generally are sufficiently treated.  A major environmental problem relating to


iron-containing minerals involves the decomposition of pyrite and marcasite in


coal mines, which can result in acid drainages with high dissolved iron concen-


trations seeping into streams.   Some 17,600 km of streams have been affected by


drainage pollution from both active and exhausted underground and surface mines.


An additional problem lies in disposing of the voluminous amounts of sludge,


primarily ferric oxyhydroxides, that result from the treatment of acidic


drainage.
                                    246

-------
MICROORGANISMS AND IRON




     Iron-containing enzymes are ubiquitous in microorganisms (bacteria, yeast,




molds, and the microalgae),  where they catalyze essential cellular reactions




and promote the grand chemical cycles of the biosphere.  Aerobic and facultative




anaerobic microorganisms are equipped with diverse mechanisms for acquiring the




ferric ion.  The high-affinity uptake pathway, which involves organic carriers




(siderophores) and their corresponding membrane receptors, is presently the best




understood.  Both siderophore and cognate receptor are under powerful repression




by iron.  Certain enteric bacteria, viz., Escherichia coli and Salmonella




typhimurium, form receptors for exogenous siderophores, such as the ferrichromes,




which they do not themselves synthesize.  Citrate may be viewed as a rudimentary




siderophore:  E_. coli has an inducible uptake system for ferric citrate,




whereas 55. typhimurium lacks this permease.




     The formation of siderophores and matching receptors in pathogenic species




may endow the microorganism with a competitive advantage in iron-poor environ-




ments, for example, in the tissues of the host.  Similarly, in a mixed culture,




those species capable of elaborating siderophores may sequester the available




iron and thus achieve superiority over organisms lacking this ability—especially




if the latter also are unable to transport the siderophore made by its




competitor.  Metabolically available iron will repress synthesis of surface




receptors for siderophores which, in turn, will render the organism resistant




to certain bacteriocins and bacteriophages.




     Little is known about the possible role of siderophores in conditioning the




soil and making iron available to plants.  The same is true of the pharmacology




of siderophores and their affect upon mammalian iron transport.  Introducing




increased amounts of dietary iron may change the microflora of the human large




intestine.
                                    247

-------
      Rhodotorulic acid and 2,3-dihydroxybenzoic acid are two microbial products




 that are being evaluated clinically as agents for removing excess body deposits




 of iron.  Desferal"',  the trade name of desferrioxamine B,  a siderophore from




 Streptomyces pilosus, is the drug of choice for treating acute,  accidental iron




 poisoning in young children.  Its utility to patients with chronic iron over-




 load is being investigated.




 IRON AND PLANTS




      Plants require a continuous supply of iron for growth because it is an




 essential component of many heme and nonheme enzymes and carriers.  Iron con-




 centration is not usually deficient in soils.  However,  25-30% of the world's




 land is calcareous in the surface horizon, and availability of trivalent iron in




 aqueous solutions of  most calcareous soils is inadequate for plant growth.  Iron




 poisoning may also be observed in plants growing on acid soil.  For more than




 50 years agronomists  have tried to change the soil to fit  the plant, but no




 economical way has been found to supply iron to plants on problem soils.  It has




 been appreciated that some plants are able to use iron in calcareous soil (iron-




 efficient plants) and some cannot (iron-inefficient plants).  Iron-efficient




plants use nitrate, iron from ferric phosphate, and ferric chelates, and they




 tolerate other heavy  metals  better than iron-inefficient plants.   This affinity




 is genetically controlled and it is possible to  avoid  iron chlorosis by




 genetically tailoring the plant to fit a problem soil»




 IRON METABOLISM IN HUMANS AND ANIMALS




      Iron in the human body is largely in the form of hemoglobin within




 circulating erythrocytes, which serve to transport oxygen.  In the plasma, iron




 is bound to transferrin, which distributes the iron among body tissues accord-




 ing to their needs.   Myoglobin, cytochromes, and various other iron-dependent
                                    248

-------
tissue enzymes perform vital functions in all body cells, yet constitute a




small fraction of body iron.  Surplus iron is deposited as ferritin and hemo-




siderin, predominantly in reticuloendothelial cells, liver parenchyma, and




muscles.




     A unique feature of human iron metabolism is the extremely limited external




iron exchange.  Recent studies indicate a marked limitation in the availability




of food iron for absorption.  Available iron varies from < 1 to about 5 mg/day,




according to the composition of the food ingested.  Heme iron is well absorbed,




whereas the absorption of nonheme iron is highly variable, depending on the




presence and amount of blocking and enhancing substances in the diet.  The




intestinal mucosa also, modifies absorption according to individual needs, and




in men this regulation is effective.  However, the amount of dietary iron avail-




able to women and infants is so low as to threaten iron balance even when




absorption is maximal.  The iron requirements of pregnancy are so great as to




exceed the amount of iron that can be absorbed from an unsupplemented diet.




     Despite the many improvements in modern nutrition, iron balance in females




and infants remains precarious.  Reduction in caloric intake and diminished




iron content of food because of decreased contamination during procurement or




preparation undoubtedly contribute to the problem.  Perhaps the most important




cause, however, is that most dietary iron is now derived from vegetable and




other nonanimal sources and thus is of very low availability.




     Iron nutrition in animals presents few problems because of their higher




intake and more efficient absorption.  However, the pig is an exception; its




rapid growth makes fortification or supplementation necessary.




IRON DEFICIENCY




     Iron deficiency in the United States and worldwide is a major health




problem.  In sheer numbers of people affected, it is one of the most prominent






                                   249

-------
nutritional deficiencies.  Iron deficiency is most common in children, in women



during their reproductive years, and particularly in the pregnant woman.  In




adult men, nutritional iron deficiency is rare, and bleeding is the more important




cause of the deficiency.  The early stages of iron deficiency before the develop-




ment of overt anemia were once difficult to measure.  Accurate tests, including




serum ferritin concentration, transferrin saturation, and red cell proto-




porphyrin content are now available for evaluating the iron status of populations.




     Because iron is essential to oxidative metabolism of all body cells as well




as a key element in many enzymatic functions, it is not surprising that iron




deficiency produces many structural and functional abnormalities.  The abnormal-




ities in red blood cells, gastrointestinal tract, and integument are rather




easily demonstrated.  Defects within the cell itself (except for altered mito-




chbndrial morphology) are more subtle.  The effects of iron deficiency on




behavior, mental and physical performance, growth, reproduction, and suscepti-




bility to infection are poorly defined and difficult to separate from the




consequences of anemia and other factors.  Animal studies indicate that iron




deficiency may predispose to increased absorption of other heavy metals and




potentiate their toxicity.




ACUTE TOXICITY OF INGESTED IRON




     Acute poisoning from ingested iron is unlikely to be encountered from any




source other than medicinal iron.  No known natural or dietary sources--such as




water, food, or any beverage--are likely to cause acute iron toxicity.




CHRONIC IRON TOXICITY




     Dietary iron overload rarely, if ever, occurs, but overload may be produced




by large volumes of alcoholic beverage high in iron, as happens in black South




Africans.  Otherwise, iron overload is caused by abnormal mucosal regulation




that permits excessive absorption from a diet of normal iron content, or by
                                    250

-------
the parenteral administration of iron, usually in the form of transfused red




cells.  Conspicuous differences exist in iron distribution in different types




of overload; in idiopathic hemochromatosis, deposits are almost exclusively




in the parenchymal tissues, particularly in the liver, whereas with transfusions




the iron is deposited predominantly in the reticuloendothelial cell.  These




differences are important, because it is the parenchymal location of iron which




is harmful.




     The two genetic disorders most commonly associated with iron overload are




thalassemia and idiopathic hemochromatosis.  Thalassemia major is relatively rare




in the United States, affecting only about 5,000 persons.  Current U. S. figures




suggest that the prevalence of clinically manifest idiopathic hemochromatosis is




of similar magnitude (1 in 20,000 hospital admissions and 1 in 25,000 hospital




necropsies).  Asymptomatic siblings with increased iron are considerably more




frequent; it remains speculative whether such individuals would accumulate




sufficient iron to produce full-blown clinical hemochromatosis if exposed to a




high dietary intake of iron or excess alcohol.




     Chronic parenchymal iron overload produces characteristic clinical mani-




festations, including hepatomegaly and liver dysfunction, diabetes, cardiac




failure, endocrine abnormalities, arthritis, and abnormal skin pigmentation.




These symptoms can be arrested or reversed in the individual with idiopathic




hemochromatosis through the removal of excess iron by phlebotomy.  In the iron




overload of thalassemia, chelates have been used to remove excess iron, and




evidence has been obtained that this treatment can arrest tissue damage.




     Carcinoma of the liver is a complication of idiopathic hemochromatosis and




a causal relationship between iron and this malignancy is thought to exist.




The intramuscular injection of iron dextran in animals has produced sarcoma, and




the association has been reported in a few instances in humans.
                                    251

-------
INHALATION OF IRON




     Ambient air contaminated with iron oxide has not been responsible for a




discernible fibrotic response in humans or animals.  Laboratory exposure of




animals (mice, rats, guinea pigs, and hamsters) to iron oxide produces only an




accumulation of iron oxide in the pulmonary parenchyma without any evidence of




fibrosis.  Eipthelial hyperplasia in the tracheobronchial tree of the hamster




has been the only finding noted.




     Occupational exposure to iron oxide in the work place has been found to




produce pulmonary siderosis; however, it is not accompanied by fibrosis.   No




other documented response has been noted.   Pulmonary siderosis itself is  a




benign disorder.




     Suspicions of an occupational hazard of lung cancer associated with  hematite




mining have not been conclusively proved.   Whenever an increased risk has been




suspected, another more logical cause can be proposed.  The probable cause of




the increased risk is the presence of radon in the work environment.
                                   252

-------
                                 CHAPTER 10




                              RECOMMENDATIONS






1.    Additional research on the aqueous chemistry of iron and precipitated




      iron compounds is needed to improve the treatment of high-iron waters




      such as the products of acid mine drainage.




          The chemical mechanisms and thermodynamics or the precipitation




          and aging of ferric oxyhydroxides should be studied, as well as




          the effects of other elements and organic ligands.




20    Because coal mining can be expected to increase, we should continue to




      seek the most practical means of preventing or diminishing the forma-




      tion of acid mine drainage.





3.   More information is needed on airborne iron concentrations  in relation




     to the iron content of regional soils  and  dusts deposited in the vicinity




     of industrial centers.




          As a contribution to background data  on environmental  exposure to




          chemical species,  more information is needed on airborne iron




          concentrations as they relate to  the  iron content of regional




          soils, and also to iron deposited on  the earth's surface in the




          vicinity of industrial centers.




4.    Basic research should  be encouraged on the adsorption and  catalytic




      properties of particulate iron compounds  in the atmosphere to determine




      their importance in carrying substances or promoting reactions that




      affect biologic systems.




          Emphasis should be placed on the  surface properties  of ferric




          oxides, as well as the effect of  their mode of formation and




          history on these properties.
                                   253

-------
 5.   Studies of the relative solubilities of iron compounds in soils should




     be pursued to aid in determining the influence of different soils on iron




     availability to plants.




          In particular, the solubilities of iron incorporated in different




          clays, metal oxides, organic chelates, and the effect of incorporat-




          ing other elements on the solubility of ferric oxyhydroxides in




          soils should both be investigated,,





6.   Basic biochemical research on iron should be promoted.




          In order to achieve the objectives of Recommendations 6-8,  research




          on molecular and cellular aspects of iron metabolism must be con-




          t inued.




7.   More information is needed on the basic mechanism and  regulation of iron




     transport in microorganisms.




     •    Induction and derepression mechanisms and microbiologic iron-




          transport systems need to be examined in relation to microbial




          virulence and resistance to infection.




     •    It must  be determined if siderophores supply or deny iron to




          higher organisms, and the pharmacologic impact  of these sub-




          stances  on animals needs to be  ascertained.




     •    Siderophore ligands,  although virtually ferric-specific,  may form




          complexes with toxic  metal ions such as plutonium and move  such




          substances through the food chain.   The possibility  of this problem




          needs  to be recognized and studied.
                                    254

-------
 8.   Clarification of the molecular processes of absorption and transport




      and their regulation in plants is needed.




      •    The manner in which an iron-deficient environment induces




           metabolic changes that enhance iron uptake should be determined„




           The subjects to be investigated should include the source of




           hydrogen ions and the source and nature of reductants released




           by roots.




      •    The movement of ferrous iron into roots and its oxidation should




           be studied.  The internal pathway of iron, including its oxida-




           tion to ferric citrate and ultimate utilization in plant tops,




           also should be scrutinized.




      •    How other metals and chemical compounds affect the iron metabolism




           of plants should be pursued in greater detail.




 9.   The matching of plant species with local soil conditions should be




      pursued.




           Regional soil and plant tissue testing laboratories could be




   !        established for this purpose.




10.   Iron nutrition should be investigated further.




      •    Methods have been devised to measure iron availability in food,




           but studies of enhancing and inhibiting factors are required to




           understand the intricacies of availability.  Investigating the




           availability of iron in water and its effect on the iron balance




           of populations consuming water high in dissolved iron also would




           be helpful.




      •    The effect of iron deficiency on absorption of other metals,




           particularly lead, should be evaluated.
                                     255

-------
11.   The prevalence of iron deficiency in the United  States  should  be better




      defined.




      •    Large-scale population surveys employing a  battery of sensitive




           screening tests for iron deficiency in statistically valid




           samples of U. S. subjects are needed.




      •    The relation between iron deficiency and diet  should be examined,




12.   The effects of iron deficiency other than anemia should be defined.




      •    Particular attention should be given to the importance of the




           depletion of tissue-dependent tissue enzymes.




      •    More discriminating tests of functional abnormalities produced  by




           iron deficiency should be designed.  Such tests  should include




           observations of behavior as well as examinations of subcellular and




           organ functions,




      •    The interaction between iron and its physiologic binding  proteins,




           transferrin and lactoferrin, needs to be elucidated before  the




           connection between iron and susceptibility  to  infection is  clarified.




13.   Improved diagnostic and therapeutic approaches to chronic iron toxicity




      should be found.




      •    The mechanism by which excess iron produces tissue damage should




           be determined and its carcinogenicity defined.   This task would be




           simplified by the development of a suitable animal model.




      •    A better means of detecting individuals with excessive iron absorp-




           tion should be developed.




      •    More efficient chelating agents and methods of chelate administra-




           tion should be sought to improve the treatment of  patients  with




           iron overload who cannot be phlebotomized.
                                    256

-------
                                APPENDIX






        ANALYSIS OF IRON IN ENVIRONMENTAL AND BIOLOGIC  SAMPLES
     Many .analytic techniques have been developed  to measure  iron  in  a




variety of materials.  More than 1,200 methods exist for  the  microdeter-




mination of iron in air, water, waste, and biologic samples.   Listed  in




descending order of popularity, the methods were reported  as  follows:




spectrophotometry, 42%; electrochemical, 17%; atomic absorption  spectro-




photometry, 11%; emission spectrography, 10%; complexometric  or  volumetric,




10%; radiochemical, 4%; fluorescence, 2%; chromatography,  1.6%;  catalytic




or kinetic, 1.1%; mass spectrography, 0.8%; microprobe, 0.5%;  and  chemi-




luminescence, 0.2%.




ENVIRONMENTAL SAMPLES





     Although atomic absorption spectrophotometry  (AAS) is a  relatively




recent development,  it has become the method of  choice  for a  large portion




of published research during  the past 20 years.  Many  of  these less popular




methods were created for specific problems and as  such  are only  applicable




to such situations.  At present, AAS appears to  be the  preferred technique




for     determining  iron in air, water, and waste  samples  because  of  its




specificity, speed,  and absence of  interferences.




     Many  colorimetric reagents have been proposed for•the spectrophoto-




metric determination of iron, but the most common  reagents remain  phenan-




throline and its derivatives  and bipyridyl and tripyridyl.  Most




Environmental Protection Agency (EPA) and Association  of  Official  Agricul-




tural Chemists  (AOAC) official procedures employ one of these reagents.
                                   257

-------
     Optical emission  and X-ray  spectrography  have  been used to determine

iron in environmental  samples, but  few  laboratories  have  elected  to use

these techniques routinely.   Both  techniques are  less  accurate  than the
                                                           433
atomic absorption and  spectrophototnetric  procedures.   Kopp    has  reported

a coefficient of variation of  7% at  a concentration  of 6A rig iron/ml

for his optical emission technique.  An  interesting  application of X-ray

spectrography was reported in  which  water was  filtered through  a  resin-

coated filter paper to retain  the metals  that  were  then analyzed  by energy-
                               699
dispersive X-ray fluorescence.

     Because of their .wide usage at  this  time, only  the atomic  absorption

and coloritnetric procedures will be  given a general  review.   Detailed  in-
                                                                       24,34,
structions may be found by consulting the references for  this  section.
104,245,419,433,566,609,663,699,731,733,735,765,771,766
                                                          Reliable  results

only can be obtained from good representative  samples  properly  prepared for

analysis.   Again, detailed instructions  for sample preparation  are beyond

the scope  of this report, but  some special procedures  will  be  discussed to

provide guidance in selecting  a sample preparation  technique.

Sample Preparation

     Water samples.  Ferrous  and ferric  iron are found in water as dissolved

solids, colloidal solids, and  suspended  solids.   Under reducing conditions,

iron is ferrous, but upon exposure to air or oxidizing materials it will con-

vert to ferric  iron.   Ferric  iron  may hydrolize  and form insoluble, hydrated

ferric oxide.  The  ferric oxide may adhere  to  the container wall unless  the

pH  of the  sample is low.  In  general, the word "iron"  is  used to denote  a

total of both the ferrous and  ferric states.
                                  258

-------
     If ferrous and ferric  iron are  to be  determined  separately,  a portion

of the water sample must be filtered and acidified  immediately after col-

lection.  Extreme care must be exercised to  prevent  the  oxidation of the

ferrous iron and hydrolysis of the ferric  ion  during  filtration.

     Colloidal iron tends to adhere  to the walls  of  plastic  containers.   If

suspended iron is to be measured separately, the  sample  should be shaken

vigorously to ensure homogeneity just before removing a  portion for analysis.

Samples should be analyzed  as soon after they  are  collected  as possible  to

avoid changes in the sample from bacterial growth  or  oxidation.

     If only dissolved iron is to be determined,  the  water  can be filtered
                                                       731,766
through a membrane filter with a pore size of  0.45^im.          The filtrate

is used to determine the dissolved iron.   However,  the pore  size  of the
                                                                     419
filter will affect the amount of soluble iron  found.   Kennedy et  al.

have demonstrated the variations in  the amount of  soluble metals  in water

samples according to the pore size of the  filter  being used.   The suspended

material retained on the filter can  be measured for  suspended iron, or the

total iron can be measured  on the sample as  received  and the  suspended iron

calculated by difference.

     Two books may be consulted for  specific steps  in collecting  and pre-
                     104,731
paring water samples.

     Particulate air samples.  It is standard  to  collect ambient  air samples

on 20 x 25 cm fiberglass filters and follow  the general  procedure of the

National Air Sampling Network.  Environmental  samples are collected on a mem-

brane or a fiberglass filter.  The collected particulate can be leached  from

the fiberglass filters by repeated treatments  of  a hot,  1:1  hydrochloric

acid;nitric acid mixture.   The leaching solutions  are combined and taken to
                                    259

-------
 dryness before solubilizing  for  analysis.  The  blank filter should  always


 be checked for iron contamination.



     Most  membrane  filters  can  be destroyed  in  the  hot acid mixture.  How-



 ever,  ester-type  filters  may  require  repeated ashing with  nitric acid to



 eliminate  all  the carbon  residue  of  the  filter.   One type  of polyvinyl



 chloride membrane filter  on the market must be  given an additional treat-



 ment with  perchloric acid to  complete  the  destruction of Che filter.



     Organic membrane  filters can be  employed for most sampling except when



 hot metal  sparks may strike the filter and burn  holes in it.  If this is



 a possibility, a  fiberglass filter should  be used.



     Because of the rigorous  acid treatment necessary,  it  is not possibly



 to differentiate between  ferrous  and  ferric iron in  air samples.



     Waste material samples.    It is very difficult  to collect  a  representa-



tive sample of waste materials  because they generally are  a  mixture  of solids



and liquids.  Rigorous ashing with nitric acid is usually  required to destroy



all organic materials and solubilize the iron,,   Sometimes  an extra ashing with



hydrogen peroxide or perchloric acid is necessary.  Perchloric  acid  is safe to



use if a small amount of  sulfuric acid is present and if most of the organic



material already has been destroyed with nitric  acid.



 BIOLOGIC SAMPLES


Feeds and Fertilizers



     Problems unique to analysis  of feeds  and fertilizers  are usually those



 of collection and preparation of  representative  samples.   The actual analysis


 by atomic absorption and/or the various spectrophotometric procedures is



 straightforward and differs little from procedures used for other  materials.



Animal byproducts used as feeds are analyzable by techniques used  for the
*
 Biologic samples are discussed below.
                                  260

-------
appropriate animal tissues; accordingly, this discussion of feed analysis



will concentrate on plant materials.  Much of the material presented here


                                                        27a 312a
has been excerpted from methods reported in two sources.   '



     Sampling.  The AOAC method recommended for sampling bagged fertilizers



is to use a slotted tube with a solid cone tip at one end to remove a core



of fertilizer diagonally from end to end.  For lots of 10 or more bags, at



least 10 bags should be sampled.  For lots of less than 10 bags, at least



10 samples should be taken, at least one from each bag.  For bulk fertilizers,



at least 10 cores should be taken from different regions.  The samples should



be made composite, thoroughly mixed, and reduced to required size by repeated



quartering.



     Preparation of samples.  The AOAC method for preparing inorganic fertil-



izers for mineral analysis is as follows.  Dissolve 1 g of well-ground sample



in 10 ml hydrochloric acid in a 150-ml beaker.  Boil and evaporate solution



to near dryness on a hotplate; do not bake,,  Redissolve residue in 20 ml of



0.5 N hydrochloric acid, boiling gently if necessary.  Filter through fast



paper into a 100-ml volumetric flask, washing paper thorougly with distilled



water.  Dilue to volume with distilled water.  If further dilutions are neces-



sary, they can be made with 0.1 N hydrochloric acid.



     The various organic materials that comprise fertilizers can be prepared



for analysis by employing the appropriate dry ashing or wet digestion



techniques used for plant samples.  With both inorganic and organic fertil-



izers, precautions should be taken to obtain representative samples and to



minimize contamination during all phases of preparation.



Plants



     Collection.  Plant samples for iron analysis can be collected in paper



or polyethylene bags if they have not been treated with talc or other



                                    261

-------
materials used to prevent the bags from sticking together.  Aluminum containers




are generally not satisfactory for fresh material because they may be corroded




by plant exudates.  Selecting representative plant samples in the field is




critical, especially if mixed forages are being collected.  One way of achiev-




ing this is to mark off  small areas  randomly,  collect  all plant material  in




each area, and make them into one or more composite samples.  Sampling of feed




grains is best accomplished in sampling tubes; however, tubes made of noncon-




taminating materials should be used.




     Cleaning and drying.  Samples should be cleaned and dried as soon as




possible after collection.  Contamination of plant samples with soil can be



a serious problem and all adherent materials should be removed.  Plants can




be cleaned by rinsing thoroughly with 0.2 N hydrochloric acid or by washing




them in detergent solution followed by distilled water rinse.  If the same




sample is to be used for multiple mineral analyses, care should be taken to




avoid leaching losses of the more soluble elements.  After thorough cleaning,




samples are usually air-dried.   The most common procedure  is to spread out




the plant material in a thin layer and dry it overnight in a forced air oven




at 70  C.  Drying at higher temperatures may result in decomposition and loss




of dry matter.




     Grinding and subsampling.   Grinding can be a serious source of iron con-




tamination for plant materials.   Grinding sometimes can be avoided by analyz-




ing an entire plant; however,  if this is not possible,  special precautions




must be taken to avoid contamination.  Using a mortar and pestle or a ball mill




made of agate are satisfactory methods.   Fragile materials such as dried leaves




can often be crushed by hand in a plastic bag.  The use of metal-containing




mills is often a source of serious but inconsistent iron contamination.  Porce-



lain, flint, or mullite instruments also may contribute to iron contamination.
                                   262

-------
      Ground  samples  should be dried overnight prior to analysis.  After drying,



 they  can be  stored in  glass  or plastic  containers with plastic  lids.  The



 sample containers should not be more than half-full to allow for thorough mixing



 before subsamples are  taken.  Even finely ground materials may  segregate



 because  of differences in particle size and density;  i.e., stems and  leaves



 often grind  differently and  thorough mixing is required to assure representa-



 tive  samples.



      Dry ashing.  Most dry ashing procedures in use have been modified to meet


                                                                      312a
 the needs of the individual users.  The following methods are typical.



      Samples (usually  1 or 2 g) are weighed into appropriate containers and




 placed in a  cool muffle furnace.  The temperature is  raised to  500 +  50° C



for 4-5 h or overnight.  Any  unashed  organic materials can be  removed  by  wetting



the sample with 2 ml  of 5 N nitric acid, slowly evaporating to dryness, then



returning the sample  to a muffle furnace and heating  to 400° C for  15  min.   The



muffle furnace should be cool « 200   C) when  the acid-treated samples are



introduced,  or violent  decomposition  may occur.   Nitric acid treatment generally



is not required.  After the sample is thoroughly ashed, it is  moistened with



a small amount of distilled water and 2 ml of  concentrated hydrochloric acid



 is added.  The liquid  is evaporated  to dryness on a  steam plate and allowed



to bake 1 h to dehydrate the  silica.   The residue is  then dissolved  in 2.5  ml



of 2 N nitric acid and  transferred to a volumetric flask.   The container  should



be rinsed several times with  hot distilled water;  add washings to the  volumetric



flask.  After the sample has  cooled and been diluted  to volume, allow  any



silica to settle out  or remove it by  centrifugation before analysis.



     If plant materials that  contain  large amounts of silica are being ashed,



the foregoing procedure sometimes is  modified  to eliminate potential adsorp-



tion of minerals to the silica.   Some disagreement exists as to whether or  not
                                  263

-------
 this problem is serious and the modified procedure may be necessary only when




 there are large amounts of silica and analysis of the solution for minerals




 that are present in very small concentrations is anticipated.   in this modified




 method, the samples are treated in the manner previously outlined, including




 the step of ashing in a muffle furnace.  After ashing, the samples are moistened




 with distilled water, and 5 ml of 48% hydrofluoric acid and 5 drops of con-




 centrated sulfuric acid are added.  The samples  are then heated gently on a




 hotplate until the hydrofluoric acid evaporates.   The residue is dissolved in




 5 ml of 2 N nitric acid and transferred to a volumetric flask with the aid of




 a stream of hot water.




     Wet digetion.  In many laboratories, wet digestion  of samples with nitric




 acid and perchloric acids is preferred  to dry ashing.  Both methods appear to




 be satisfactory for determining iron.   In a typical procedure, a 2-g sample of




 plant material is weighed into a 250-ml beaker, allowed to stand overnight in




 20-31 ml of concentrated nitric acid, heated gently until the initial vigorous




 reactions have subsided, and slowly  evaporated to near dryness.  After the




 sample has cooled, 10 ml concentrated nitric acid, 2 ml concentrated sulfuric




 acid, and 10 ml of 72% perchloric acid  are added„  The sample is then returned




 to the hotplate, covered with a watchglass and heated until all organic matter




 is dissolved and the solution is colorless.  The coverglass is then removed and




 the sample is evaporated to near dryness at just below the boiling point.  The




 residue is dissolved in 5 ml of 2 N  nitric acid and transferred to a volumetric




 flasko  Silica generally presents few problems when wet digestion is employed.




     The use of perchloric acid to digest samples necessitates a number of




precautions.  The removal of easily oxidized material by pretreatment with




nitric acid is essential, as is the addition of sulfuric acid to prevent exces-




 sive heating during the evaporation  step.  Samples should not be allowed to




 char when evaporating the nitric acid; if charring occurs, more nitric acid





                                   264

-------
 should  be added  and  the sample heated  until  all  charred  material  dissolves.



 If charred material  appears  in the  sample  after  perchloric  acid has  been  added,



 the sample must  be cooled  and more  nitric  and  perchloric acids added.   Heat-



 ing charred samples  in the presence of perchlorates will almost inevitably



 result  in an explosion. Additionally, perchloric  acid fumes  should  only  be



 discharged in hoods  designed for perchloric  acid use.  Alternatively,  samples



 can be  digested  in Kjeldahl  flasks  and the acid  fumes exhausted via  a  glass



 manifold connected to a water aspirator.   The  problem of perchlorate can  be


                                                               605a
 avoided by digestion with  sulfuric  acid and  hydrogen peroxide.



METHOD OF ANALYSIS



Atomic Absorption  Spectrophotometry



     Atomic  absorption  Spectrophotometry offers  the fastest analytic method



for determining  iron  in environmental  and  biologic samples.   (Most plant



materials  contain  enough iron so  that  no special techniques are required.)



It  is specific, with  generally little  or no  interference  from the  other metals



in  the sample matrix.   Four  lines in the iron  spectrum may be used to analyze



solutions, depending  upon  the iron concentration in the solution  to be analyzed.



The wavelengths and sensitivities of the four  lines are listed in Table A-l.




     Although the detection limit (defined  as the concentration at which the



signals from the analyte is twice the background signal)  will vary from



instrument to instrument, it  is  approximately O.OOS^ug/ml at 248.3 nm when



an air-acetylene flame is used.   Citric acid  has been reported to  suppress



the absorbance by up  to 50% at  a  concentration of ZOO^ug/ml.    The  nitrous



oxide-acetylene flame is supposed to eliminate most interferences.
                                  265

-------
                                Table A-l




      Wavelengths and Sensitivities of the Iron Spectra Used in AAS—
Wavelength,
run
248.3
372.0
386,0
392.0
Band Pass,
run
0.2
0.2
0.2
0.2
Optimal
Range,
2.5
25
50
800
Working Typical Sensitivity
jig/ml
10
- 100
- 200
- 3200
ug/ml
0.062
0.55
0.90
17.0
-From Varian Techtron.





—Sensitivity is the concentration in an aqueous solution expressed in ug/ml




that absorbs 1% of the radiation passing through a cloud of atoms being




determined.
                                   266

-------
     With the development of the graphite furnace, carbon rod, and tantalum



ribbon  accessories  for atomic absorption spectrophotometers, the detection



limit for iron has  been reduced by  several orders of magnitude.  However, it



will still vary depending upon the  equipment, sample matrix, and specific



operating conditions.  For  instance, Ediger et_ al.., using the HGA-2000 graphite



furnace, reported  the detection limit  for iron in saline water as




0.00002 mg/ml,    whereas Bagliano £t al.,  using the HGA-70 graphite furnace,

                                                                    O /
reported a detection limit of 0.001 mg/ml for iron in uranium oxide.



     The samples to be analyzed must be free of all suspended particulate to



prevent the aspirator system from clogging.   The matrix of the calibration



standards should be similar to that of the samples to be analyzed.   If the



sample matrix is very complex,  analysis can be accomplished by the method of



addition, which will reduce the matrix effects between standards  and samples.



The iron in the sample can be concentrated  and matrix effects reduced by



chelating and extracting the iron with ammonium pyrrolidine dithiocarbamate



and methyl isobutyl ketone.



     Operating procedures are described in the manufacturer's methods manuals



and will vary slightly from instrument to instrument.   Specific procedures



for AAS analysis have been well characterized.24'104'566'609'731'735'765'771



Spectrophotometric Procedures



     Review of the literature indicates that many colorimetric procedures



have been proposed.  However, at  this time,  either thed,aL'- or ?: > ?)'-



bipyridyls and the phenanthrolines appear to be the reagents of choice.   '    '


731 733
   '     In general, both of these colorimetric reagents have good  sensitivity



and color stability, obey Beer's  law, and can be used  over a wide  range.   The



procedures are relatively simple and direct.   Both systems react with ferrous



iron so that it is possible to determine the ferrous iron directly and the





                                   267

-------
 total  iron after  reduction to  the  ferrous  states,,  The  ferric  iron can be



 calculated by difference.




      For accurate results, the iron in the sample must be, available readily



 and not bound up in stable complexes such as those that form with pyro-



 phosphates.  Since the reagents  react in acidic solutions, the precipitation



 of hydroxides and phosphates can be prevented.



 Conclusions



     Colorimetric and atomic absorption spectrophotometry offer satisfactory



 procedures  for measuring iron  in a variety of  environmental and biologic



 samples.  The specific choice  of method must be chosen by the  analyst,  based



 on  the type of sample, analysis desired, equipment available for performing



 the analysis,  and any special  requirements of  the federal agency having juris-



 diction.   If  total iron is  to  be determined, atomic absorption spectrophoto-



 metry is probably the best  method.  If ferrous and ferric-iron need to  be
       • : .  .       •          •< .  :      •  . .      "


 measured separately, a colorimetric procedure must be used.




HUMAN AMD ANIMAL TISSUES



     Assays of interest include measurement of blood iron compounds, primarily



 used to identify  iron deficiency; measurements of iron stores, of utility in



 evaluating  iron deficiency  and iron overload;  and isotopic measurements, for



 determining absorption, loss,  and internal iron exchange.



     The blood of humans and other vertebrates have three important iron



 compounds.  Hemoglobin constitutes most of blood iron and usually  is measured



 colorimetrically by the cyanmethemoglobin method.     Plasma transferrin



 iron, is the compound of especial interest  because it represents transport iron



 and reflects  the  amount of  the metal available to body tissues.  As a rule,


                                              474
 plasma iron is      measured colorimetrically.     It is not influenced by



 the presence  of hemoglobin, whereas free iron, large amounts of ferritin, and





                                   268

-------
 parenteral iron do affect plasma iron.   Measuring transferrin-binding capacity
 is important because the availability of transferrin iron is influenced by the
 degree of saturation of the iron-binding protein.  Colorimetric and radio-
                                    180
 active methods have been described.      Serum ferritin may be assayed radio-
                  535
 immunometrically;    the protein is  useful in detecting iron deficiency and
overload.  Although not an iron compound itself, red cell protoporphyrin
increases when the iron supply to the marrow becomes inadequate and therefore
                                                 441
serves as an additional gauge of iron deficiency.     However, erythrocyte
protoporphyrin also rises with lead poisoning and certain unusual  intrinsic
disorders of porphyrin metabolism.
     Reticuloendothelial iron stores can be determined by examining a marrow
                                                                         35
aspirate for particulate hemosiderin within the reticuloendothelial cells   or
by determining the amount of;iron in marrow particles.     Hepatic iron stores
may be evaluated by the intramuscular  injection of desferrioxamine and sub-
sequent analysis of the urine excreted during the next 24 h.  This technique is
                                                                   333
particularly effective when parenchymal iron overload is suspected.     Serum
ferritin is useful in evaluating reticuloendothelial and parenchymal iron stores
                                                                           402
since its concentration usually parallels the size of tissue iron deposits.
Nonheme hepatic iron may be analyzed directly in tissue obtained by biopsy and
         160
necropsy.
     Much of the information concerning iron turnover has been made possible
through the use of iron isotopes.  Iron-59, with a half-life of 44 days, is the
most convenient and its /^-radiation is detectable by crystal counting in vitro,
                                                89
by surface counting, and by total body counting,,    Iron-55 has a half-life of
4 yr; its X-rays are of such low energy that samples must be processed by wet
ashing for liquid scintillation counting.  Iron-55 and -59 can be counted
                                                               242
differentially  in the same sample with minimal cross-counting.     A crude

                                   269

-------
method for evaluating absorption is determination of the degree of elevation in
                                                                 247
plasma iron after orally administering iron to a fasting subject.     More
accurate determinations can be made by measuring red cell or total-body activity
                                  89
2 wk after ingestion of radioiron.     Absorption of nonheme and heme iron from
complex meals is ascertainable through extrinsic radioiron tags added to the
meal.77'187
                                  270

-------
                              REFERENCES


 1.  Abei, H. , and H. Suter.  Acatalasemia, pp. 1710-1729.  In J. B. Stanbury,

          J.~ B.~ Wyngaarden, and D.~ S.~ Fred ricks on, Eds.  The Metabolic Basis of


          Inherited Disease.  (3rd ed.)  New York:  McGraw-Hill Book Co., 1972.

 2.  Adams,  J.  F.   The clinical  and  metabolic  consequences of total  gastrectomy.

                                                     g
          II."  Anaemia:   Metabolism  of  iron, vitamin  12 and folic acid.   Scand.


          J." Gastroenterol.   3:145-15,  1968.

 3. Adamson, J. W. , and C. A. Finch.  Hemoglobin function, oxygen affinity, and


         erythropoietin.  Ann. Rev.  Physiol.   37:351-369, 1975.

 4.   Agarwala,  S.  C.,  C.  P.  Sharma, P. N.  Sharma,  and B.  D.  Nautiyal.   Suscep-

          tibility of some  high yielding varieties  of wheat to deficiency of

          micronutrients  in sand  culture,  pp.  1047-1064.   In International


          Symposium on  Soil Fertility  Evaluation.   Proceedings.   Vol.  1.


          New  Delhi:   Indian Society of Soil  Science,  Indian Agricultural


          Research Institute, 1971.

 5. Ahlbom, H. E.  Simple achlorhydric anaemia, Plutnmer-Vinson syndrome, and


        carcinoma of the mouth,  pharynx,  and oesophagus in women:  Observations


        at Radiumhemmet, Stockholm.   Brit.  Med. J.  2:331-333, 1936.

 6.   Aisen,  P.   The transferring (siderophilins),  pp. 280-305.  In G. L.    \


          Eichhorn, Ed.   Inorganic Biochemistry.  Vol. 1.  New York:

          Elsevier Scientific Publishing  Company,  1973.

 7.   Aisen,  P.,  R. Aasa, and A.  G. Redfield.   The  chromium, manganese,  and cobalt


          complexes of transferrin.   J.~ Biol.  Chem.   244:4628-4633,  1969.


8. Aisen, P., and  E."  B.  Brown.  Structure  and  function of  transferrin.  Prog.

        Hematol.   9:25-56,  1975.


9. Aisen,  P., and A. Leibman. The stability constants of the Fe34 conalbumin com-


        plexes.   Biochem. Biophys. Res. Comm.   30:407-413, 1968.
                                 271

-------
 10.  Ajl, S.  J., S. Kadis, and T.  C.  Montie,  Eds.  Microbial Toxins.  Vol.  I.


           Bacterial Protein Toxins.   New York:   Academic Press, 1970.  517 pp.


 11.  Ajl, S.  J., S. Kadis, and T.  C.  Montie,  Eds.  Microbial Toxins.  Vol. II A.


           Bacterial Protein Toxins.   New York:   Academic Press, 1970.  412 pp.


 12«   Ajl,  S.  J.,  S.  Kadis,  and T. C. Montie, Eds.  Microbial  Toxins.   Vol.  III.


            Bacterial  Protein Toxins.  New York:   Academic Press,  1970.   548  pp.

     o       *
 13. Akeson, A., G. Biorck,  and R. Simon.  On the content of myoglobin  in  human


          muscles.  Acta Med. Scand.   183:307-316, 1968.


 14.  Aldrich,  R. A.  Acute iron toxicity,  pp.  93-104.   In R. 0. Wallerstein and


           S.  R.  Mettier,  Eds.   Iron  in  Clinical Medicine.   Berkeley;  University


           of  California Press,  1958.


 15.  Alexanian, R.  Urinary  excretion of erythropoietin  in normal men and  women.


          Blood  28:344-353, 1966.

 16.  Allgood, J. W.,  and E.  B. Brown.  The relationship  between duodenal mucosal


          iron concentration and iron absorption  in human subjects.   Scand.  J.


          Haematol.   4:217-229, 1967.


 17.  Ambler, J.  E., and J.  C. Brown.   Cause  of  differential  susceptibility to zinc


          deficiency in two varieties  of navy  beans  (Phaseolus  vulgaris L.)


          Agron.  J.  61:41-43,  1969.


 18. Ambler, J. E., and J.  C. Brown.  Iron-stress response in mixed and monocultures


         of soybean cultivars.  Plant Physiol.  50:675-678,  1972.


 19. Ambler, J.  E.,  and J.  C. Brown.  Iron supply in soybean seedlings.  Agron. J.


         66:476-478, 1974.


 20. Ambler, J. E., J.  C. Brown, and H. G. Gauch.  Sites  of iron reduction  in soybean


         plants.   Agron. J.  63:95-97, 1971.


21.  Amdur,  M.  0.,  and D.  W. Underbill.   Response of guinea  pigs to a combination of


          sulfur dioxide and open hearth dust.   J." Air  Pollut.  Control Assoc.  20:


          31-34,  1970.


                                    272

-------
  22.   American Chemical Society.   Cleaning our Environment.   The Chemical Basis for




            Action.   Washington,  D.C.:   American Chemical Society, 1969.   249 pp.




  23.   American Iron and Steel  Institute.   Annual Statistical  Report.   _/  1975_7




            Washington,  D.  C. :  American Iron and Steel  Institute, 1976.   98 pp.




  24.   Lewis, L. L.   Interlaboratory testing  programs-for the chemical  analyis




            of  metals.   Standardization  News  4(9):19-23, 1976.



  25.   Anderson, W.  F.,  and M. C.  Killer, Eds.  Proceedings of a  Symposium.




            Development  of Iron Chelators for Clinical Use.  DREW  Publ. (NIH)




            76-994.   Bethesda, Md.:  U.  S.  Department of Health,  Education,




            and Welfare, National Institutes  of Health, 1976.  277 pp.



 26.  Anguissola,  A. B.  The  nutritional value  of wine as regards its  iron content,




           pp.  71-74.  In L.  Hallberg, H.-G.  Harwerth, and A."  Vannotti, Eds.   Iron




           Deficiency:   Pathogenesis, Clinical  Aspects, Therapy.   New  York:




           Academic  Press,  1970.




 27.    Aristovskaya, T. V., and G. A.  Zavarzin.  Biochemistry of iron in soil,




            pp. 385-408.  In A. D." McLaren and J. Skujins,  Eds.   Soil Biochemistry.




            Vol. 2.   New York:  Marcel Dekker,  1971.



 27a.   Association of Official  Analytical Chemists.   Official  Methods  of  Analysis.




            (12th  ed.)   Washington, D. C.:   Association  of Official  Analytical




            Chemists, 1975.   1094  pp.




 28.    Atkin, C. L.,  and J. B. Neilands.  Rhodotorulic acid, a diketo-piperazine




            dihydroxamic acid with growth-factor  activity.  I.  Isolation  and




            characterization.  Biochemistry  7:3734-3739, 1968.




29.    Aung-Than-Batu, U.Hla-Pe, Thein-Than,  and  Khin-Kyi-Nyunt.   Iron-deficiency in




            Burmese  population groups.   Amer. J."  Clin. Nutr.  25:210-217,  1972.




30.    Awai, M., and  E.  B. Brown.   Examination of  the role of xanthine  oxidase  in




            iron absorption by the rat.  J. Lab.  Clin. Med.  73:366-378, 1969.






                                      273

-------
  31.   Badenoch, J., and S. T. Callender.  Effect of corticosteroids and gluten-free
             diet on absorption of iron in idiopathic steatorrhea and coeliac disease.
             Lancet  1:192-194, i960.
  32.   Badenoch,  J., and S. T. Callender.  Iron metabolism  in  steatorrhea.  The  use  of
             radioactive  iron  in  studies  of absorption and utilization.   Blood  9:
             123-133, 1954.
  33.  Badenoch, J., J. R.  Evans, and W.  C. D. Richards.  The stomach in  hypochromic
            anaemia.  Brit. J. Raematol.  3:175-185, 1957.
  34.    Bagliano,  G. , F. Benischek, and  I. Huber.  Application of graphite  furnace
              atomic absorption to the determination of impurities in uranium
              oxides without preliminary  chemical separation.  Atom. Absorpt. Newslett.
              14:45-48, 1975.
 35.    Bainton, D. F.,  and C.  A.  Finch.   The diagnosis of iron deficiency anemia.
             Amer.  J." Med.  37:62-70, 1964.
 36.     Baird,  I.  M., E.  K.. Blackburn,  and  G.  M.  Wilson.   The pathogenesis of  anaemia
              after partial  gastrectomy.   I~  Development  of  anaemia  in relation to
              time  after  operation,  blood  loss,  and  diet.   0.7-1." Med.   28:21-34, 1959.
 37.     Baird,  I.  M., 0.  G. Dodge,  F. J.  Palmer,  and R.  J. Wawman.  The  tongue and
              oesophagus in iron-deficiency anaemia and the effect of iron therapy.
              J. Clin. Path.  14:603-609,  1961.
 38.     Baird, I. M. , and G. M." Wilson.   The pathogenesis  of anaemia after partial
             gastrectomy.   H~  iron absorption after partial gastrectomy.  Q." J.~
             Med.  28:35-41, 1959.
39.     Baker,  S.   Discussion, pp.  142-144.  In H.  Kief,  Ed.  Iron Metabolism  and
              Its Disorders.  New  York:  American Elsevier Publishing Co.  Inc.,  1975.
40.      Balcerzak,  S.  P.,  D. H. Mintz,  and M.  P. Westerman.  Diabetes mellitus  and
             idiopathic  hemochromatosis.   Amer.  J.' Med.  Sci.   255:53-62, 1968.
                                    274

-------
 41.  Balcerzak,  S.  P.,  W.  W.  Peternel,  and  E. W.  Heinle.   Iron  absorption in
           chronic  pancreatitis.   Gastroenterology  53:257-264,  1967.
 42.  Balcerzak,  S. P.,  J. W. Vester,  and A. P.  Doyle.   Effect of iron deficiency
           and red cell age on human erythrocyte catalase activity.   J. Lab. Clin.
           Med.   67:742-756, 1966.
42a. Balcerzak, S.  P., M. P. Westerman,  E. W. Heinle,  and F. H.  Taylor.  Measurement
          of iron stores using deferoxamine.  Ann. Intern: Med.   68:518-525, 1968.
43.  Balcerzak, S. P., M. P.  Westerman,  R. E. Lee, and A.  P. Doyle.  Idiopathic
         haemochromatosis:   A  study of  three families.  Amer. J. Med.  40:857-873,
         1966.
44.  Bannerman, R. M., S. T.  Callender,  R. M. Hardisty, and R. S. Smith.   Iron
         absorption  in  thalassaemia.  Brit. J. Haematol.  10:490-495,  1964.
45. Barry,  M., D. M. Flynn, E. A. Letsky, and R. A.  Risdon.  Long-term chelation
         therapy in thalassaemia major:   Effect on liver iron concentration,
         liver histology, and clinical progress.  Brit.  Med. J.   2:16-20, 1974.
 46.  Barshad,  I.   Chemistry of soil  development,  pp. 1-70.   In F. E. Bear,
         Ed.   Chemistry of the Soil.   (2nd  ed.)   American Chemical  Society
         Monograph Series  No.  160.  New York:   Reinhold  Publishing  Corp.,  1964.
 47.  Beadle,  G. W.   Yellow stripe -  a  factor for chlorophyll deficiency in maize
          located in the P_r ££ chromosome.  Amer.  Natur.   63:189-192,  1929.
 48.  Beal,  V.  A., A.  J.  Meyers, and  R. W. McCammon.  Iron intake, hemoglobin, and
         physical  growth during the first two years of  life.  Pediatrics  30:
         518-539,  1962.
 49.  Beaton, G. H.   Epidemiology of iron deficiency, pp.  477-528.  In A.  Jacobs
          and M.  Worwood, Eds.  Iron in Biochemistry and Medicine.   New York:
          Academic Press, 1974.
50.  Beaton,  G. H., Myo  Thein,  H.  Milne,  and M.  J. Veen.   Iron requirements  of
         menstruating women.   Amer. J."  Clin. Nutr.  23:275-283,  1970.
51. Becking, G. C.   Influence  of dietary  iron levels on hepatic drug metabolism  in
         vivo and in vitro in the rat.   Biochem. Pharmacol.   21:1585-1593,  1972.
                                    275

-------
  52.  Beevers, L.,  and R." H. Hageman.  Nitrate reduction  in higher plants.  Ann.  Rev.
            Plant Physiol.  20:495-522, 1969.
  53.  Beischer, N.  A., R. Sivasamboo, S. Vohra, S. Silpisornkosal, and  S. Reid.
           Placental hypertrophy in severe pregnancy anaemia.  J. Obstet. Gynaecol.
           Brit. Commw.   77:398-409, 1970.
  54.  Bell, W.' D.,  L. Bogorad, and W. J." Mcllrath.  Response o£ the yellow-stripe
            maize mutant  (y_s,) to ferrous and ferric iron.  Bot. Gaz.   120:36-39,  1958.
  55.  Bender,  M."  L.   Manganese nodules,  pp.  673-677.   In R.  W.  Fairbridge,  Ed.
           The Encyclopedia of Geochemistry and  Environmental Sciences.   New York:
           Van Nostrand  Reinhold  Company,  1972.
 56.  Benjamin, B.  I., S. Cortell, and M. E. Conrad.   Bicarbonate-induced iron com-
           plexes and iron absorption:  One effect of pancreatic secretions.
           Gastroenterology  53:389-396, 1967.
 57.   Bernat,  I.   Ozaena:  A Manifestation  of  Iron Deficiency.   New York:  Permagon
           Press,  1965.   116 pp.
 58.   Bernat,  I.,  and J4  Vallo.   Ozaena:  The  causes  of its  familial  occurrence.
           Acta Med. Acad. Sci. Hung.   20:89-105,  1964.
 59.   Berner,  R.   Iron -  abundance in natural  waters,  Section 26-1,  2 pp.   In
           K.  H.  Wedepohl, Ed.   Handbook of Geochemistry.  Vol. II/2.  New
           York:   Springer-Verlag,  1970.
 60.   Berner, R. A. Principles of Chemical Sedimentology.  New York:   McGraw-Hill
           Book Company,  1971.  240 pp.
 61.   Berry,  C. L., and W. C. Marshall.  Iron distribution  in  the liver of  patients
           with thalassaemia major.  Lancet  1:1031-1033, 1967.
62.   Berry, J. A., and H. M. Reisenauer.  The influence  of molybdenum  on iron
           nutrition of tomato.  Plant Soil  27:303-313,  1967.
63.   BessiSj M., and J.  Caroli.   A comparative study of hemochromatosis by
           electron microscopy.  Gastroenterology  37:538-549, 1959.
                                     276

-------
 64.     Beutler, E.  Iron enzymes in iron deficiency.  VI.  Aconitase activity and
              citrate metabolism.  J. Clin. Invest.  38:1605-1616, 1959.
 65.     Beutler, E.,  and R.  K.  Blaisdell.  Iron enzymes in iron deficiency.   III.
              Catalase in rat red cells and liver with some further observations on
              cytochrome c.   J.  Lab.  clin. Med.   52:694-699,  1958.
 66.     Beutler, E.,  and R.  K.  Blaisdell.  Iron enzymes in iron deficiency.   V.
              Succinic dehyrogenase  in rat liver,  kidney and  heart.   Blood  15:
              30-35,  1960.
 67.     Beutler, E.,  W.  Drennan, and M. Block.  The bone marrow  and  liver  in  iron-
             deficiency  anemia:  A histopathologic study of  sections with  special
             reference  to stainable  iron  content.  J. Lab. Clin.  Med.  43:427-439,
             1954.
 68.     Beutler, E.,  S. E. Larsh, and c. W. Gurney.  Iron therapy in chronically
             fatigued, nonanemic women:  A double-blind study.  Ann. Intern. Med.
             52:378-394, 1960.
 69€      Beveridge, B. R., R. M.  Bannerman, J. M.  Evanson,  and L.  J. Witts.  Hypo-
             chromic  anaemia:  A retrospective  study and follow-up of 378  in-
             patients.   Q. J. Med.   34:145-166, 1965.
 70.      Bezwoda, W. R.,  T. H. Bothwell, L. A. van der Walt,  S. Kronheim,  and  B. L.
             Pimstone.   An investigation  into gonadal dysfunction in patients with
             idiopathic  haemochromatosis.  Clin.  Endocrinol.  6:379-385,  1977.
 71.     Bezwoda, W. R., P.  B. Disler, S. R. Lynch, R.  W. Charlton, J. D.  Torrance,
             D.  Derraan, T.  H. Bothwell, R. B.  Walker,  and F.  Mayet.   Patterns of
             food iron absorption in iron-deficient white and Indian subjects and
             in venesected haemochromatotic patients.   Brit.  J.  Haematol.   33:425-
             436, 1976.
72.     Biddulph,  0.   Translocation of minerals  in plants,  pp. 261-275.  In E.
             Truog, Ed.   Mineral  Nutrition of Plants.  Madison:   University of
             Wisconsin Press, 1961.

                                   277

-------
 73.   Bidstrup, P. L.  Other  industrial  dusts, pp.  156-172.   In  D.  C.  F.  Muir,

           Ed.  Clinical Aspects of  Inhaled Particles.  London:  William  Heinemann

           Medical Books Limited,  1972.

 74.  Biesecker, J. E., and J. R. George.  Stream Quality in Appalachia as Related

           to Coal-Mine Drainage, 1965.  U.S. Geological Survey Circular  526.


           Washington, D.C.:  U.S. Government Printing Office, 1966.   27  pp.

 75.   Bjorn-Rasmussen, E.   Iron absorption from wheat bread:   Influence of various

           amounts  of bran.   Nutr. Metab.   16:101-110,  1974.

 76.   Bjorn-Rasmussen, E., and  L.  Hallberg.   Iron  absorption  from  maize:   Effect of

           ascorbic  acid  on  iron absorption from maize  supplemented with  ferrous


           sulphate.   Nutr. Metab.   16:94-100,  1974.

 77.   Bjorn-Rasmussen, E., L. Hallberg,  B. Isaksson,  and B. Arvidsson.  Food  iron

           absorption  in man:  Applications of the  two-pool extrinsic  tag method

           to measure  heme and nonheme iron absorption  from the whole  diet.   J.


           Clin. Invest.  53:247-255, 1974.
        it
78.   Bjorn-Rasmussen, E. , L. Hallberg,  and R. B. Walker.  Food  iron absorption

           in man.  I.  Isotopic exchange between  food  iron and  inorganic iron

           salt added  to food:   Studies  on maize, wheat, and  eggs.  Amer.  J.  Clin.

           Nutr.  25:317-323, 1972.

79.  Blaxter, K. L., G.  A. M. Sharman, and A. M. MacDonald.   Iron-deficiency

           anaemia in calves.   Brit.  J. Nutr.   11:234-246,  1957.

80.  Bomford, A.,  R. J.  Walker, and R.  Williams.  Treatment  of iron overload

           including results  in a personal series of 85 patients  with idiopathic

          haemochromatosis,  pp. 324-331.  In H.  Kief, Ed.   Iron  Metabolism and

          Its Disorders.   New York:   American Elsevier Publishing Co.  Inc.,  1975.

8i. Bomford, R. R.,  and C. P. Rhoads.   Refractory  anaemia.   I.  Clinical and

         pathological aspects.   Q.  J. Med.  10:175-234,  1941.
                                 278

-------
 82.   Bond,  J.,  and T.  T.  Jones.   Iron  chelates  of  polyaminocarboxylic  acids.
           Trans.  Faraday  Soc.   55:1310-1318,  1959.
 83. Bothwell, T. H.  Iron overload in  the Bantu, pp. 362-375.   In F. Gross, Ed.   Iron
         Metabolism:  An  International Symposium Sponsored by CIBA, Aix-en-Provence,
         lst-5th July,  1963.  Berlin:  Springer-Verlag,  1964.
 84.  Bothwell, T. H.  The relationship of transfusional haemosiderosis to idiopathic
          haemochromatosis.   South Afr. J. Clin. Sci.  4:53-70,  1953.
 85. Bothwell, T. H., and  T. Alper.  The cardiac complications of haemochromatosis:
         Report of a case with a  review of the literature.   South Afr. J. Clin.  Sci.
         2:226-238, 1951.
 86. Bothwell, T. H., and  B. A. Bradlow.   Siderosis  in  the Bantu:  A combined
         histopathological and chemical study.  Arch.  Path.  70:279-292,  1960.
 87.  Bothwell, T.  H., and R.  W.  Charlton.  Dietary  iron overload, pp. 221-229.
          In H. Kief, Ed.  Iron Metabolism and Its  Disorders.   New York:
          American  Elsevier Publishing Co. Inc., 1975.
 88. Bothwell, T. H., I.  Cohen, 0.  L. Abrahams, and  S. M. Perold.  A familial study
         in  idiopathic hemochroniatosis.  Amer. J. Med.   27:730-738, 1959.
 89. Bothwell, T. H., and  C. A. Finch.  Iron Metabolism.  Boston:  Little, Brown,
         and Company,  1962.  440  pp.
 90.Bothwell, T. H., and C. Isaacson.   Siderosis in  the Bantu:  A comparison of
        the incidence in  males and females.  Brit.  Med. J.    1:522-524, 1962.
 91. Bothwell, T. H., G. Pirzio-Biroli,  and C.  A. Finch.  Iron absorption.  I.
         Factors influencing absorption.   J~ Lab.  Clin. Med.   51:24-36, 1958.
92.Bothwell, T.  H., H.  Seftel, P.  Jacobs,  J. D.  Torrance, and N. Baumslag.  Iron
        overload in Bantu subjects:   Studies  on the availability of iron in  Bantu
        beer.   Amer. J.  Clin.  Nutr.   14:47-51,  1964.
93. Bothwell, T.  H.,  B. van Lingen, T. Alper,  and  M. L.  du Preez.  The cardiac
         complications  of hemochromatosis:  Report  of  a  case including radio-
         iron studies  and a note  on etiology.   Amer. Heart J.   43:333-340,  1952.

                                   279

-------
94.  Bowen,  H. J. M.   Trace Elements in Biochemistry.   New York:  Academic Press,




          1966.  241 pp.



94a.   Boyd,  J.  T.,  R.  Doll,  J.  S.  Faulds,  and  J.  Leiper.   Cancer of the lung in




          iron ore (haematite)  miners.   Brit.  J.  Ind.  Med.   27:97-105, 1970.




95.   Boyer, J.  F., and V. E. Gleason.   Coal and  coal mine drainage.   J.  Water Pollut.




          Control  Fed.  48:1284-1287,  1976.




96.   Bradlow,  B. A.,  J. A.  Dunn,  and  J. Higginson.  The  effect  of  cirrhosis on




          iron storage.   Amer.  J.  Path.  39:221-237, 1961.




97.   Brain, M.  C.,  and A. Herdan.   Tissue iron stores  in sideroblastic anaemia.




          Brit.  J. Haematol.   11:107-115,  1965.




97a.   Brain, J. D., P. A.  Valberg, S. P. Sorokin, and W.  C.  Hinds.  An iron




           oxide aerosol  suitable  for animal exposures.  Environ. Res.   7:




           13-26, 1974.




98.   Braude, R., A.  G. Chamberlain, M. Kotarbinska, and K. G. Mitchell.  The




           metabolism of  iron in piglets given labelled  iron either orally  or  by




           injection.  Brit. J. Nutr.   16:427-449, 1962.



98a.   Bramer, H. C.   Water Pollution Control in the Carbon and Alloy  Steel  Indus-




           tries.  (Prepared for the U. S. Environmental Protection Agency)   EPA-




           600/2-76-193.  Carnegie, Penn.:  Datagraphics, Inc.,  1976.  274pp.



99.    Bremner, I., and  A. C.  Dalgarno.  Iron  metabolism in  the veal calf.  The




            availability of  different iron compounds.   Brit.  J.  Nutr.   29:229-243,




            1973.




^00.   Bricaire, H., J. Guillen, J. Leprat, and E. Modigliani-Kourilsky.   Le syndrome




            endocrinien des  hemochromatoses idiopathiques.   (A propos  de  88




           observations).   Bull.  Soc.  Med. Hop. Paris  117:909-928, 1966.



lOla.   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.





                                    280

-------
 lOOb. Brief,  R.  S.,  J.  W.  Blanchard,  R.  A.  Scala,  and J.  H.  Blacker.   Metal
            carbonyls in the petroleum industry.   Arch.  Environ.  Health  23:
            373-384,  1971.
 101.  Briggs, G. A.   Plume Rise. AEC Critical Review Series.  TID-25075.   Oak
            Ridge, Tenn.:  U. S. Atomic Energy Commission, 1969.  81 pp.
 102.  Brink,  B., P.  Disler, S.  Lynch, P. Jacobs, R. Charlton, and T.  Bothwell.
            Patterns  of iron storage in dietary iron overload and idiopathic hemo-
            chromatosis.  J. Lab. Clin. Med. 88:725-731, 1976.
 103.  Brise, H., and L. Hallberg.  A method for comparative studies on iron absorp-
            tion in  man using two radioiron isotopes.   Acta  Med.  Scand.   171(Suppl.):
            7-22, 1962.
 104.   Brown, E. , M. W.  Skougstad,  and M. J. Fishman.   Techniques of Water-Resource*
            Investigations  of the United States  Geological Survey.  Book 5. Chapter
            Al.  Methods for Collection and Analysis of Water Samples  for Dissolved
            Minerals and Gases.  Washington, D.  C.:  U. S. Government  Printing
            Office,  1970.   160 pp.
 105.   Brown,  E.  B.,  Jr., R. Dubach, D. E.  Saith, C. Reynafarje, and C. V. Moore.
            Studies in iron transportation and metabolism.  X. Long-term  iron
            overload in dogs.  J. Lab. Clin. Med. 50:862-893, 1957.
 106.   Brown, E. B., Jr.,  D. E.  Smith, R. Dubach,  and C.  V.  Moore.  Lethal iron over-
            load in  dogs.   J. Lab.  Clin. Med.  53:591-606, 1959.
 107.   Brown, J. C.  An evaluation  of bicarbonate  induced iron chlorosis.   Soil Sci.
            89:246-247,  1960.
 108.   Brown, J. C.  Iron  chlorosis in plants.   Adv. Agron.   13:329-369,  1961.

109.   Brown, J. C.  Competition between phosphates and the  plant for  Fe from Fe24
            ferrozine.   Agron.  J.   64:240-243, 1972.
                                     281

-------
 HO. Brown, J. C., and J. E. Ambler.  Iron-stress response in tomato (Lycopersicon


           esculentum)    1.   Sites of Fe reduction, absorption and transport.


           Physiol. Plant.  31:221-224,  1974.


      Brown, J. C., and J. E. Ambler.  "Reductants" released by roots of Fe-deficient
 ill*

           soybeans.  Agron. J.  65:311-314, 1973.
       Brown, J. C.,  J. E. Ambler, R. 1.  Chaney, and C.  D.  Foy.  Differential


            responses of plant genotypes  to micronutrients,  pp. 389-418.  In


            J.  J.  Mortvedt,  P. M.  Giordano, and W.  L.  Lindsay,  Eds.  Micronutri-


            ents in  Agriculture.   Proceedings of a  Symposium,  1971.  Madison,


            Wise.:   Soil Science  Society  of America,  Inc.,  1972.


 ,,-    Brown, J. C., and R.  L. Chaney.  Effect of iron on the transport of citrate


            into the xylem of soybean and tomatoes.  Plant  Physiol.  47:836-840,


            1971.


 „,,    Brown, J. C.,  R. L. Chaney, and J." E. Ambler.  A new tomato mutant inefficient
 114.

            in the transport of iron.  Physiol. Plant,  25:48-53, 1971.
       Brown,  J.  C.,  and S.  B.  Hendricks.   Enzymatic activities as indications of
J. J.D •

            copper and  iron  deficiencies  in plants.   Plant Physiol.   27:651-660,  1952.


116.   Brown,  J.  C.,  R. S. Holmes,  and L.  0. Tiffin.  Iron chlorosis in soybeans  as


            related to  the genotype of rootstock.   Soil Sci.   86:75-82, 1958.


117.   Brown,  J.  C., R. S.  Holmes,  and L.  0. Tiffin.  Iron chlorosis in soybeans as


            related to  the genotype of rootstalk:   3.  Chlorosis susceptibility and


            reductive capacity at the root.  Soil  Sci.  91:127-132, 1961.


118.    Brown,  J.  C., and W.  E. Jones.  A technique to determine iron efficiency in


            plants.  Soil Sci. Soc. Amer.   40:398-405, 1976.


119.    Brown,  J.  C., and W.  E. Jones.  Heavy-metal toxicity in plants.  1.   A


            crisis in embryo.  Commun. Soil Sci. Plant Anal.   6:421-438, 1975.
                                     282

-------
 120.  Brown,  J.  C.,  and  W.  E.  Jones.   pH changes  associated  with iron-stress  response.



           Physiol.  Plant.   30:148-152,  1974.




 121.  Brown, J. C., and W. E. Jones.  Phosphorus efficiency as related to iron




           efficiency in sorghum.  Agron. J.  67:468-472, 1975.




 122.  Brown, J. C., 0. R. Lunt, R. H. Holmes, and L. 0. Tiffin.  The bicarbonate ion




           as an indirect cause of iron chlorosis.  Soil Sci.  88:260-266, 1959.



 123.  Brown,  J.  C.,  and  L.  0.  Tiffin.  Iron  stress as related  to the  iron  and  citrate




           occurring in  stem exudate.  Plant Physiol.  40:395-400,  1965.
 124. Brown, J. C., L. 0. Tiffin, and R. S. Holmes.  Competition between chelating




          agents and roots as factor affecting absorption of iron and other  ions by




          plant species.  Plant Physiol.  35:878-886, 1960.




 125. Brown,  J.  C.,  L. 0.  Tiffin, A.  W.  Specht,  and  J.  W.  Resnicky.   Iron absorption




          by roots as affected by  plant species  and  concentration of chelating agent.




          Agron. J.  53:81-85, 1961.




 126. Brumfitt, W.  Primary iron-deficiency anaemia in young men.   Q. J. Med.




          29:1-18, 1960.
127. Bryant, P. M.  Methods of Estimation of the Dispersion of Windborne Material



          and Data to Assist  in Their Application.  Report AHSB(RP)-R-42.  Harwell,




          Didcot, Berkshire:  United Kingdom Atomic Energy Authority,  1964.   36  pp.




128.  Bryson, R. A., and D. A. Baerreis.  Possibilities of major climatic modifi-




           cations and their implications: Northwest India, a case for  study.  Bull.




           Amer. Meteorol. Soc.  48:136-142, 1967.




129.  Buchanan,  W. M.   Bantu siderosis.  -- A review.   Central Afr.  J.  Med.   15:105-113,



           1969.



130.  Budowsky, E. I.  The mechanism of  the mutagenic action of hydroxylamines.




           Prog. Nucleic Acid Res. Molec. Biol. 16:125-188, 1976.






                                    28S

-------
 131.   Buja,  L.  M.,  and  W.  C.  Roberts.   Iron in the heart.   Etiology and  clinical
             significance.   Amer.  J.""Med.   51:209-221,  1971.
 132.  Bullen, J. J., H. J. Rogers, and E. Griffiths.  Bacterial iron metabolism in
            infection and immunity, pp.  518-551.  In J. B. Neilands, Ed.  Microbial
            Iron Metabolism:  A Comprehensive Treatise.  New York:  Academic Press,
            1974.
 132a.   Bulman,  R.  A.  A critique  of the chemistry of plutonium and the transuranics
             in  the biosphere.   Struct.  Bond.   (in press)
 133.    Bunn, C. R., and G. Matrone.  In vivo interactions of cadmium, copper, zinc
             and iron in the mouse and rat.  J. Nutr.  90:395-399, 1966.
 134.    Burks,  J.  M.,  M.  A.  Siimes,  W.  C.  Mentzer,  and  P.  R.  Dallman.   Iron deficiency
             in an Eskimo village.   The value of serum  ferritin in assessing iron
             nutrition before and after a  three-month period  of iron supplementation.
             J. Pediatr.   88:224-228,  1976.
135.    Burman, D.  Haemoglobin levels in normal infants aged 3 to 24 months and the
             effect of iron.  Arch. Dis. Child.  47:261-271, 1972.
136.    Byers, B. R., and C. E. Lankford.  Regulation of synthesis of 2,3-dihydroxy-
             benzoic acid in Bacillus subtilis by iron and a biological secondary
             hydroxamate.  Biochim. Biophys. Acta  165:563-566, 1968.
137.    Callender, S., D. G. Grahame-Smith, H. F. Woods, and M. B. H. Youdim.  Reduction
             of platelet monoamine oxidase activity in iron deficiency anaemia.  Brit. J.
             Pharmacol.  52:447P-448P, 1974.
138.     Callender,  S, T.   Iron deficiency due  to malabsorption of  food  iron,  pp.
              168-175.   In  H. Kief, Ed.  Iron Metabolism and Its Disorders.  Pro-
              ceedings °f tne Third Workshop Conference Hoechst, Schloss Reisensburg,
              6-9 April, 1975.  International Congress Series 366.  New York:
              American Elsevier Publishing Co., Inc., 1975.

                                      284

-------
 139.  Callender, S. T., S. R. Marney, Jr., and G. T. Warner.  Eggs and iron absorp-



           tion.  Brit. J. Haematol.  19:657-665, 1970.



 140.  Callender, S. T., L. J. Witts, G. T. Warner, and R. Oliver.  The use of a simple




           whole-body counter for haematological investigations.  Brit. J. Haematol.
 141.   Cammock, E. E., A. Turnbull, C. L. Fausnaugh, F. J. Cleton, L. M. Nyhus,




            C. A. Finch, and H. N. Harkins.  Iron absorption from food after subtotal




            gastrectomy in the dog.  Surg. Forum  12:299-301, 1961.



 142.   Canale, V. C., P. Steinherz, M. New, and M. Erlandsort.  Endocrine function  in




            thalassemia major.  Ann. N."" Y7 Acad. Sci.   232:333-345,  1974.




 143.   Cann, H. M. , and H.  L. Verhulst.  Accidental poisoning in young children.




            The hazards of iron medication.  A.M. A. J. Dis. Child.  99:688-691, 1960.



 144.   Cappell, D. F., H. E. Hutchison, and M. Jowett.  Transfusional siderosis:




            The effects of excessive iron deposits on the tissues.  J. Path.




            Bacteriol.  74:245-264, 1957.




 145.   Capriles,  L. F.  Intracranial hypertension and iron-deficiency anemia:  Report




            of four cases.   Arch.  Neurol.  9:147-153, 1963.




 146.   Card, R. T., and M.  C. Brain.  The "anemia" of childhood.   Evidence for a




            physiologic response to hyperphosphatemia.  New Engl. J. Med.  288:




            388-392, 1973.




147 _   Carlson, T. N. , and J. M. Prospero.  The large-scale movement of Saharan




            air outbreaks over the northern equatorial Atlantic.   J. Appl.




            Meteorol.  11:283-297, 1972.
148.    Carr, J. G., C. V. Cutting, and G. C. Whiting, Eds.  Lactic Acid Bacteria




            in Beverages and Foods.  Proceedings of a Symposium, University of




            Bristol, 1973.  New York:  Academic Press, 1975. 415 pp.
                                    285

-------
 149.    Carroll,  D.   Role of clay minerals  in the transportation of iron.
             Geochitn.  Cosmochim.  Acta  14:1-28,  1958.
 150.    Cavill, I.,  and C. Ricketts.   The kinetics of iron metabolism, pp. 613-
             647.  In A. Jacobs and M. Worwood,  Eds.   Iron in Biochemistry and
             Medicine.  New York:  Academic Press, 1974.
 151.    Cawse, P. A.  A Survey of Atmospheric Trace Elements in the U.K.  (1972-73).
             Report No. AERE-R-7669.   Harwell:  U. K.  Atomic Energy Research Estab-
             lishment, 1974.  96 pp.
 152.    c'ernelc,  M., V. Kusar,  and S. Jeretin.  Fatal  peroral iron poisoning in a
             young woman.  Acta Haematol.  40:90-94,  1968.
 153.   Chaberek,  S.,  and  A. E. Martell.  Metal chelates  in  biological  systems,
            pp. 416-504.   In Organic  Sequestering Agents.   New  York:   John
            Wiley and  Sons, 1959.
 153a.  Chandra,  R.  K.  Iron status  in  human  response and  susceptibility to infec-
            tion, pp.        .  In  A.  Jacobs,  Ed.  Iron Metabolism.  Ciba  Founda-
            tion  Symposium No. 51.   (in  press)
 154.   Chaney, R. L.,  J. C. Brown, and L. 0. Tiffin.   Obligatory reduction of ferric
            chelates in iron uptake by soybeans.   Plant Physiol.  50:208-213, 1972.
155.   Chang,  L.  L.   Tissue  storage  iron in Singapore.   Amer.  J.  Clin.  Nutr.   26:952-
            957,  1973.
156.   Charlton,  R.  W.,  C.  Abrahams,  and T.' H.  Bothwell.   Idiopathic hemochromatosis
            in young subjects:   Clinical,  pathological,  and chemical findings in four
  	       patients.  Arch.  Path.  83:132-140,  196;.
157.   Charlton,  R.  W.,  and T. H. Bothwell.  Hemochromatosis:  Dietary and genetic
            aspects.  Prog. Hematol.   5:298-323, 1966.
                                  286

-------
 158.  Charlton, R. W., T. H. Bothwell, F. G." H. Mayet, C. J. Uys, and I. W. Simson.
            Liver  iron stores in different population groups in South Africa.   South
            Afr. Med. J.  45:524-529, 1971.
 159.  Charlton, R. W., T. H. Bothwell, and H.  C. Seftel.  Dietary iron over-
            load.  Clin. Haematol.  2:383-403, 1973.
 160.Charlton,  R. W.,  T)'.  M."  Hawkins,  W.  0.  Mavor, and T.  H.  Bothwell.   Hepatic
          storage iron concentrations in different population groups.   Amer.  J.
          Clin. Nutr.   23:358-371,  1970.
 161.Charlton,  R. W.,  P.  Jacobs,  H.  Seftel, and  T.  H. Bothwell.   Effect of alcohol
          on iron absorption.   Brit.  Med.  J.  2:1427-1429,  1964.
 162. Cheney,  B. A.,  K. Lothe, E. H. Morgan, S.  K. Sood, and C. A. Finch.  Inter-
            nal iron exchange  in  the  rat.   Amer.  J. Physiol.  212:376-380,  1967.
 163. Chisholm, M. The  association  between  webs,  iron  and  post-cricoid  carcinoma.
            Postgrad.  Med. J.   50:215-219,  1974.
 164.  Chisholm, M., G.  M. Ardran,  S. T.  Callender, and  R. Wright.  A  follow-up
            study of patients  with  post-cricoid  webs.  Q. J. Med. 40:409-420, 1971.
 165.    Chisholm, M.,  G. M.  Ardran,  S.  T. Callender,  and R.  Wright.  Iron
             deficiency and  autoimmunity in post-cricoid webs.  Q.  J.  Med.
             40:421-433, 1971.
 166. Chisolm, J. J.,  Jr.  Poisoning due to heavy metals.   Pediatr.  Clin.  North Amer.
          17:591-615,  1970.
 167.Chvapil, M., J. Hurych, and E. Ehrlichova.    The  effect of iron deficiency on  the
          synthesis of collagenous and non-collagenous proteins in wound granulation
          tissue and in the heart of  rats.  Exp.  Med.  Surg.  26:52-60, 1968.
168. Clark, Rf  B., and J." C.' Brown.   Internal root control of iron uptake and
          utilization  in maize genotypes.   Plant Soil  40:669-677,  1974.
169. Clark,  R.  B.. L. 0. Tiffin, and J. C. Brown.  Organic  acids and  iron  trans-
           location in maize genotypes.  Plant Physiol.  52:147-150, 1973.
                                   287

-------
 170.  Clarke, F. W., Ed.  The Data of Geochemistry.  (5th ed.)  U.S. Geological




           Survey Bulletin 770.  Washington, D.C.:  U.S. Government Printing Office,




           1924.  841 pp.



 170a.  Cole,  P.,  and M.  B.  Goldman.   Occupation,  pp.  167-184.   In J.  F.  Fraumeni,




            Jr.,  Ed.  Persons  at High Risk of Cancer.   An Approach to Cancer



           Etiology and Control.  New York:   Academic  Press,  Inc.,  1975.




 171.   Cole,  S.  K.,  W. Z. Billewicz,  and A. M. Thomson.  Sources of variation in




            mentrual blood loss.  J.  Obstet.  Gynaecol.  Brit. Commonw.  78:933-939,




            1971.




 172.    Cole,  S. K., A.  M.  Thomson, W. Z.  Billewicz,  and A.  E.  Black.  Haemato-




             logical characteristics  and menstrual  blood losses.   J.  Obstet.




             Gynaecol.  Brit. Commonw.  79:994-1001,  1972.




 173.    Community Air Quality Guides.  Iron oxide.   Amer. Ind. Hyg. Assoc. J.




             29:1-6, 1968.




 174.   Connor, J. J., and H.  T. Shacklette.  Background Geochemistry of Some Rocks,




            Soils, Plants,  and Vegetables in the Conterminous United States.  U.S.




            Geological Survey Professional Paper 574-F.  Washington, B.C.:  U.S.




            Government Printing Office,  1975.  168 pp.
175.   Conrad, M.  E.   Factors  affecting  iron  absorption,, pp.  87-120.   In I.  Hallberg,




             H.-G.  Harwerth,  and  A. Vannotti,  Eds.   Iron  Deficiency.   Pathogenesis,




             Clinical Aspects,  Therapy.   New York:   Academic Press,  1970.
176.   Conrad, M. E., B. I. Benjamin, H. L. Williams, and A. L. Foy.  Human absorption




            of hemoglobin-iron.  Gastroenterology  53:5-10, 1967.




177.   Conrad, M. E., L. R. Weintraub, and W. H. Crosby.  Iron metabolism  in rats




            with phenylhydrazine-induced hemolytic disease.  Blood  25:990-998,




            1965.
                                   288

-------
 178.  Conrad, M. E., L. R. Weintraub, and W. H. Crosby.  The role of the intestine


           in iron kinetics.  J. Clin. Invest.  43:963-974, 1964.


 179.  Conrad, M. E., Jr., and W. H. Crosby.  Intestinal mucosal mechanisms controlling


           iron absorption.  Blood  22:406-415, 1963.


 180.   Cook,  J.  D.   An evaluation of adsorption methods for measurement of  plasma


            iron-binding capacity.   J. Lab.  Clin.  Mcd.   76:497-506,  1970.


 181.   Cook, J. D., J. Alvarado, A. Gutnisky, M. Jamra, J. Labardini, M. Layrisse,

                               /                                             /•
            J. Linares, A. Loria, V. Maspes, A. Restrepo, C. Reynafarje, L. Sanchez-


            Medal, H.~ Velez, and F. Viteri.  Nutritional deficiency and anemia in


            Latin America:  A collaborative study.  Blood  38:591-603, 1971.


 182.    Cook,  J. D.,  W.  E.  Barry,  C. Hershko,  G." Fillet,  and C. A. Finch.   Iron kinetics


            with  emphasis  on iron overload.   Amer. J." Path.   72:337-343, 1973.


 183.    Cook,  J. D.,  G.  M.  Brown,  and  L.  S.  Valberg.   The effect  of  achylia gastrica


             on iron absorption.   J.  Clin.  Invest. 43:1185-1191,  1964.


 184.    Cook,  J. D.,  C.  A.  Finch,  and  N.  J.  Smith. "Evaluation of  the iron  status of a


             population.  Blood  48:449-455, 1976.
185.    Cook,  J.  D.,  C.  Hershko,  and  C.  A.  Finch.   Storage  iron kinetics.   V.   Iron


             exchange in the  rat.   Brit.  J.  Haematol.   25:695-706,  1973.
186.    Cook,  J.  D.,  M.  Layrisse,  and  C. A. Finch.  The  measurement  of  iron  absorption.


             Blood  33:421-429,  1969.


187.    Cook, J.  D.,  M.  Layrisse,  C.  Martinez-Torres,  R. Walker,  E.  Monsen,  and C. A.


             Finch.   Food iron absorption  measured  by  an extrinsic tag.  J.  Clin.


             Invest.   51:805-815,  1972.


188.   Cook,  J. D., D. A.  Lipschitz,  L.  E. M.  Miles, and C.  A.  Finch.   Serum ferritin


            as a  measure of iron stores  in normal subjects.   Amer. J. Clin.  Nutr.


            27:681-687,  1974.
                                       289

-------
 189.    Cook, J. D., G. Marsaglia, J. W. Eschbach, D. D. Funk, and C. A. Finch.



             Ferrokinetics:  A biologic model for plasma iron exchange in man.




             J. Clin.  Invest.  49:197-205, 1970.
 190.    Cook, J. D., V. Minnich, C. V. Moore, A. Rasmussen, W. B. Bradley, and C. A.



             Finch.  Absorption of fortification iron in bread.  Amer. J. Clin. Nutr.




             26:861-872, 1973.
 191.   Cook,  J.  D.,  and  E.  R.  Monsen.   Food  iron absorption in human subjects.   V.   Tb.6




            effect  of  egg albumin  and  its  amino acid  components.  Amer. J. Clin. Nutr.




            (in  press)
192.  Cook, J. D., and E. R. Monsen.  Food iron absorption  in man.   II.  The




           effect of EDTA on absorption of dietary non-heme  iron.  Amer. J.




           Clin. Nutr.  29:614-620, 1976.




193.  Corn, M.  Aerosols and the primary air pollutants -- nonviable particles.




           Their occurrence, properties, and effects, pp. 77-168.  In A. C. Stern,




           Ed.  Air Pollution.   (3rd ed.)  Vol. 1.  New York:  Academic Press, 1976.
194.   Cotes,  J.  E.,  J. M. Dabbs, P.  C. Elwood,  A. M.  Hall,  A. McDonald,  and  M.  J.




            Saunders.  The response to  submaximal exercise  in adult  females;  rela-




            tion  to haemoglobin  concentration.   J. Physiol.  (Lond.)   203:79P-80P,



            1969.
195.   Council  on Foods  and  Nutrition.   Committee  on Iron Deficiency.   Iron




            deficiency in the  United  States.   J.A.M.A.  203:407-412,  1968.




196.  Cowan, J. W.,  M. Esfahani, J. P.  Salji, and  S. A. Azzam.  Effect of phytate




           on iron absorption in the rat.   J. Nutr.   90:423-427, 1966.



196a.   Creasia,  D. A., and P. Nettesheim.  Respiratory cocarcinogenesis studies




           with  ferric oxide:  A text case of current experimental models, pp.




           234-245.   In E. Karbe and J. F. Park, Eds.  Experimental Lung Cancer.




           Carcinogenesis and Bioassays.  New York:   Springer-Verlag, 1974.





                                    290

-------
 197.  Crichton, R. R., H. Huebers, E. Huebers, D. Collet-Cassart, and  Y. Ponce.
           Comparative studies on ferritin, pp.  193-200.  In R." R. Crichton,  Ed.
           Proteins of Iron Storage and Transport in Biochemistry and  Medicine.
           New York:  American Elsevier Publishing Company,  Inc.,  1975.
 198.  Cronan,  D.  S.   Authigcnic  minerals  in deep-sea  sediments,  pp.  491-525.
            In  E.  D.  Goldberg,  Ed.   The  Sea.   Vol.  5.   Marine Chemistry.   New
            York:   Wiley-Interscience,  1974.
 199.  Crosby, W. H.  Serum ferritin fails to indicate hemochromatosis - nothing
             gold  can  stay.  New  Engl. J. Med.  294:333-334,  1976.  (editorial)
 200.   Crosby, W.  H., J.  I. Munn,  and F. W.  Furth.  Standardizing  a  method for clin-
             ical  hemoglobinometry.  U.  S.  Armed  Forces Med.  J.   5:693-703, 1954.
 201.  Crotty, J. J.  Acute iron poisoning in children.  Clin. Toxicol.  4:615-619,
             1971.
 202.    Curtis, C. D., and D.  A.  Spears.  The formation of sedimentary iron minerals.
             Econ. Geol.   63:257-270,  1968.
 203.    Dacie,  J.  V., and J. C. White.   Erythropoiesis with particular reference to
             its  study by biopsy of human bone marrow:   A review.  J. Clin. Path.
             2:1-32,  1949.
 203a.  Dagg, J. H., A. Goldberg,  J. R.  Anderson,  J.  S.  Beck,  and  K. G. Gray.  Auto-
            immunity in iron-deficiency anaemia.   Brit.  Med.  J.   1:1349-1350, 1964.
204.   Dagg> J. H., J. M. Jackson, B. Curry, and A. Goldberg.  Cytochrome  oxidase  in
            latent iron deficiency (sideropenia).  Brit. J. Haematol.   12:331-333,
            1966.
205.    Dagg,  J.  H.,  J.  A.  Smith,  and A.  Goldberg.  Urinary excretion of iron.   Clin.
             Sci.   30:495-503,  1966.
                                       291

-------
  206.   Dallman, P. R.   Iron restriction in the nursing rat:   Early effects  upon tissue




             heme proteins,  hemoglobin and liver iron.   J.  Nutr.   97:475-480,  1969.




  207.   Dallman, P. R.   Tissue  effects  of iron  deficiency,  pp. 437-375.   In  A.




  208.   Dallman, P. R.,  and  J.  R.  Goodman.  Enlargement of  the mitochondrial compartment




             in iron and copper deficiency.   Blood   35:496-505,  1970.




  209.   Dallman, P. R.,  and  J.  R.  Goodman.  The effects of  iron deficiency on the




             hepatocyte;  A  biochemical and  ultrastructural study.   J.  Cell  Biol.




             48:79-90,  1971.



 210.    Dallman,  P.  R., and  H.  C.  Schwartz.  Distribution  of  cytochrome  c and  myo-




              globin in  rats  with  dietary iron  deficiency.  Pediatrics  35:677-686,




              1965.




 211.   Dallman, P. R., and H.  C.  Schwartz.  Myoglobin and  cytochrome response during




             repair of iron deficiency in the rat.   J.  Clin.  Invest.  44:1631-1638,




             1965.
 212.    Dallman, P. R., M. A. Siimes, and E. C. Manies.  Brain iron:  Persistent




             deficiency following short-term iron deprivation in the young rat.  Brit.




             J. Haematol.  31:209-215, 1975.




 2l2a.   Dallman, P. R., P. Sunshine, and Y. Leonard.  Intestinal cytochrome response




             with repair of iron deficiency.  Pediatrics  39:863-870, 1967.




 213.    David,  C.  N.   Ferritin and iron metabolism in Phycomyces,  pp.  149-158.   In J.  B.




             Neilands, Ed.   Microbial Iron Metabolism:   A Comprehensive Treatise.   New




             York:  Academic Press,  1974.
214.    Davidson, W. M. B., and J. L. Markson.  The gastric mucosa in iron




              deficiency anaemia.  Lancet  2:639-643, 1955.
                                   292

-------
 215. Davies,  C.  T.  M.,  A.  C.  Chukweumeka,  and J. P. M. van Haaren.   Iron-deficiency
            anaemia;   Its effect  on maximum  aerobic  power  and responses  to  exercise in
            African males aged  17-40 years.  Clin. Sci. Mol.  Med.  44:555-562,  1973.
 216. Davis, P. N., L. C. Norris, and F. H. Kratzer.  Iron deficiency studies in
            chicks.  J. Nutr.  84:93-94, 1964.   (letter)
 217. Davis, P. N., L. C. Norris, and  F.  H.  Kratzer.  Iron utilization and  metabolism
           in the chick.   J. Nutr.   94:407-417,  1968.
 218. Davison, R. H.   Inheritance of haemochromatosis:   A report on a family with
           consanguinity.  Brit.  Med.  J.   2:1262-1263,  1961.
 219. DebreC R., J-C. Dreyfus,  J. Frezal,  D. Labie,  M.  Lamy, P. Maroteaux,
           F. Schapira, and G.  Schapira.   Genetics of haemochromatosis.   Ann. Hum.
           Genet.  23:16-30, 1958.
 219a. Deichmann,  W.  B.,  and W. H. Gerarde.  Toxicology of Drugs and Chemicals.
            New York:  Academic Press,  1969.  805 pp.
 220.  De Kimpe,  C. R., and  Y. A. Martel.  Effects of vegetation on the  distribution
            of  carbon,  iron, and  aluminum  in the B horizons  of Northern  Appalachian
            Spodosols.  Soil Sci. Soc.  Amer. J.  40:77-80, 1976.
 221.  DeKock, P. C.   Fundamental  aspects  of  iron nutrition of plants,  pp. 41-44.
           In Ministry of Agriculture,  Fisheries and Food, Technical  Bulletin
           21.   Trace Elements  in Soils and  Crops.   Proceedings of a  Conference,
           1966.  London:  Her  Majesty's  Stationery  Office,  1971.
 222.  DeKock,  P.  C., and R. I. Morrison.  The metabolism of  chlorotic leaves.
            2.  Organic acids.  Biochem. J.  70:272-277, 1958.
 223.  Delbarre, F. L'osteoporose des  hemochromatoses.   Sem.  Hop.  Paris   36:
           3279-3294, 1960.
224.  de Leeuw, N. K. M., L. Lowenstein,  and Y-S. Hsieh.  Iron deficiency  and
            hydremia  in normal  pregnancy.  Medicine  (Baltimore)  45:291-315,  1966.
                                    293

-------
  225.    Deller,  D.  J.   Iron59  absorption measurements  by whole-body counting:

              Studies  in alcoholic  cirrhosis, hemochromatosis,  and  pancreatitis.

              Amer.  J.  Dig.  Dis.  10:249-258, 1965.

  226.   deMooy, C.  J.   Iron Deficiency in Soybeans - What Can Be Done About It?

              Pm-531  Cooperative Extension Service.  Ames:   Iowa State University,

              1972.  2 pp.

  227.   de Seze, S., J. Solnica, D. Mitrovic, L. Miravet,  and H. Dartman.  Joint and

             bone disorders and hypoparathyroidism in hemochromatosis.  Semin.

             Arthritis Rheum.  2:71-94, 1972.

 227a.   Desy, D. H.  Iron and steel scrap, pp.  699-711.  In U. S. Bureau of Mines.

             Minerals Yearbook 1974.  Vol. 1.  Metals, Minerals, and Fuels.

             Washington, D. C.:  U. S. Government Printing Office,  1976.

 228.    de Villiers, J. M.  Pedosesquioxides - composition and colloidal interactions

              in soil genesis during the Quaternary.  Soil Sci.   107:454-461, 1969.

 229.    Diekmann,  H.  Siderochromes/Iron(III)-trihydroxamates7,  pp.  449-457.   In

             A.  I. Laskin and H.  A.  Lechvalier,  Eds.   CRC Handbook  of Microbiology.

             (Vol.  3)   Cleveland:   CRC Press, 1973.

 230.    Diez-Ewald, M., L. R. Weintraub,  and  W.  H.  Crosby.   Interrelationship of iron

             and manganese metabolism. Proc. Soc.  Exp. Biol.  Med.   129:448-451, 1969.

 231.   Dialer, P." B., S." R. Lynch,  R. W.  Charlton, J." D. Torrance,  T. H. Bothwell,

            R. B.~ Walker, and F. Mayet.  The effect of tea on iron absorption.  Gut

            16:193-200, 1975.

232.   Domellof,  L.  Effects of Parenteral Iron  Overload on the Rat Liver.   Goteborg:

            Elanders Boktryckeri  Aktieboleg,  1972. 83  PP-                    ,

233.  Drysdale, J. W., P'. Arosio, T. Adelman, J. T. Hazard, and D.  Brooks.  'Iso-

           ferritins in normal and diseased states, pp. 359-366.  In R. R.

           Crichton, Ed.  Proteins of Iron Storage and Transport in Biochemistry

           and Medicine.  New York:  American Elsevier Publishing Company, Inc.,

           1975.
                                      294

-------
234.    Dubach,  R.,  C.  V.  Moore,  and  S.  Callender.   Studies  in iron transportation

             and metabolism.   IX.   The excretion of  iron as  measured by the isotope
             technique.   J. Lab.  Clin. Med.   45:599-615, 1955.
235.  Dubin,  I.  N.   Idiopathic hemochromatosis  and  transfusion siderosis:  A review.

           Amer. .J." Clin.  Path.  25:514-542,  1955.
236.    Duce,  R.  A.,  P.  I. Parker,  (Eds.)  and C.  S.  Giam.  Pollutant Transfer to
             the Marine  Environment.  Deliberations  and  Recommendations of NSF/IDOE
             (National  Science Foundation, International Decade of  Ocean Exploration)
             Pollutant  Transfer Workshop,  held  in Port  Aransas, Texas, Jan. 11-12,
             1974.   Kingston:  _/ University of Rhode  Island Press_/,  1974.  55 pp.
237.    du Lac,  Y.,  G.  Deloux, and  R. Deuil.  Arthropathies  et chondrocalcinoses au
             cours  des  hemochromatoses.   Rev. Rhum.   34:758-769,  1967.  (summary in
             English)
238.    Dunn, W. L.  Iron-loading,  fibrosis,  and hepatic carcinogenesis.  Arch. Path.
             83:258-266, 1967.
239.   Dymock, I. W., J. Cassar,  D. A. Pyke,  W.  G. Oakley, and  R.' Williams.  Observa-

            tions on the pathogenesis, complications and treatment  of diabetes in
            115  cases of haemochromatosis.  Amer. J.~Med.  52:203-210, 1972.
240.  Dymock,  I." W. ,  E."  BT D.~ Hamilton,  J." W.~ Laws,  and  R.~ Williams.   Arthropathy  of
           hemochromatosis:   Clinical and  radiological  analysis of  63 patients with
           iron  overload.  Ann. Rheum. Dis.   29:469-476,  1970.
241.   Dzubay, T. G., and R.  K. Stevens.  Ambient air analysis with  dichotomous
            sampler and x-ray  fluorescence spectrometer.  Environ.  Sci. Technol.
            9:663-668,   1975.
242.  Eakins, J. D., and D. A. Brown.  An improved method for  the simultaneous de^
           termination  of iron-55  and iron-59 in blood by liquid scintillation
           counting.  Int. J.  Appl.  Radiat.  Isot.   17:391-397, 1966.
                                      295

-------
243.   Easley,  R.  M.,  Jr.,  B.  F.  Schriener,  Jr.,  and P.  N.  Yu.   Reversible-cardio-
            myopathy  associated with hemochromatosis.   New Engl.  J.  Med.   287:
            866-867,  1972.
244    Edgerton,  V. R.,  S.  L.  Bryant, C.  A.  Gillespie,  and  G.  W.  Gardner.   Iron
            deficiency anemia  and physical performance  and  activity of rats.   J.
            Nutfr.   102:381-399,  1972.
245.   Ediger,  R.  D.,  G.  E. Peterson,  and  J.  D. Kerber.   Application of the
            graphite  furnace to saline water analysis.   Atom. Absorpt.  Newslett.
            13:61-64,  1974.
246.   Eisenbud, M.   The  primary  air pollutants—radioactive.   Their occurrence,
            sources,  and  effects,  pp.  197-231.  In  A. C.  Stern,  Ed.   Air Pollution.
            (3rd ed.)  Vol. 1.  New  York:  Academic Press,  1976.
247.   Ekenved, G.  Iron absorption studies:  Studies on oral  iron  preparations
            using serum iron and  different  radioiron isotope techniques.  Scand.
            J. Haematol.   (Suppl. 28):1-30,  1976.
248.   Ekenved,  G.,  I.  Halvorsen,  and  L.  Solvell.   Influence  of  a  liquid antacid
            on the absorption  of different  iron  salts.   Scand. J.  Haematol.
            (Suppl.  28):65-77,  1976.
249.   Ellis,  B. G., and B.  D. Knezek.  Adsorption reactions  of  micronutrients
            in soils,  pp.  59-78.   In J.  J.  Mortvedt,  P.  M.  Giordano,  and W.  L.
            Lindsay, Eds.  Micronutrients in Agriculture.   Madison, Wis.:
            Soil Science Society of America,  1972.
250.  El  Wakeel,  S.  K.,  and  J.  p.  Riley.  Chemical  and mineralogical  studies  of
          deep-sea  sediments.  Geochim.  Cosmochim. Acta 25:110-146-,  1961.
251. Elwood, P.  C.,  and D. Hughes.   Clinical  trial  of iron  therapy on psychomotor
          function in  anaemic  women.   Brit. Med. J.   3:254-255,  1970.
252.  Elwo°d>  P.  C., A.  Jacobs, R.  G.  Pitmin,  and  C.  C.  Entwistle.  Epidemiology
          of  the Paterson-Kelly syndrome.  Lancet  2:716-720, 1964.

                                   296

-------
253.   Elwood, P. C.,  W. E. Waters,  W.  J. W.  Greene,  P.  Sweetnam,  and M. M. Wood.



            Symptoms and circulating haemoglobin level.    J.  Chron.  Dis.  21:




            615-628, 1969.



254    Emery, T.  Biosynthesis and mechanism of action of hydroxamate-type  sidero-




            chromes, pp. 107-123.  In J." B.' Neilands, Ed.  Microbial  Iron Metabolism:




            A Comprehensive Treatise.  New York:  Academic Press, 1974.




255.   Engle,  M. E., I. G.  Erlandson, and C.  H. Smith.   Late  cardiac complications of




            chronic, severe, refractory anemia with hemochromatosis.  Circulation




            30:698-705, 1964.




256.   Epstein,  J. H.,  and A. G. Redeker.  Porphyria cutanea tarda:   A  study of  the




            effect of  phlebotomy.  New Engl. J. Med.  279:1301-1304,  1968.





257.   Eriksson, F.,  S. V.  Johansson,  H." Mellstedt,  0."  Strandberg,  and  P.'0.   Wester.




            Iron intoxication in two adult patients.  Acta Med. Scand.   196:231-236,




            1974.




258.   Erlandson, M. E., B. Walden,  G.  Stern,  M. W. Hilgartner, J. Wehman, and C. H.




            Smith.  Studies on congenital hemolytic syndromes.   IV.   Gastrointestinal




            absorption of iron.   Blood  19:359-378, 1962.
258a.  Eugster,  H.  P.,  and I.-M.  Chou.   The depositional environments of




            Precambrian banded iron-formations.   Econ.  Geol.   68:1144-1168,  1973.





259.   Evans, H. J.  Role of molybdenum in plant nutrition.  Soil Sci.  81:199-208,




            1956.




260.   Evans, H. J.  The biochemical role of iron in plant metabolism, pp. 89-110.




            In Mineral Nutrition of Trees.   A Symposium.  Duke University School




            of Forestry Bulletin 15.  Durham, N.  C.:   Duke University, 1959.




261.   Evans,  J.  Treatment of heart failure in haemochromatosis.   Brit.  Med.  J.   1:



            1075-1078,  1959.






                                     297

-------
 262.   Eyster, E.   Traumatic hemolysis with hemoglobinuria due to ball variance.




             Blood   33:391-395,  1969.




 263.   Fairbanks, V. F., J. L. Fahey,  and E. Beutler.   Acute iron poisoning,




             pp. 359-374.  In Clinical Disorders of Iron Metabolism.  (2nd ed.)




             New York:  Grune & Stratton, 1971.




 264.   Fairbanks, V. F., J. L. Fahey, and E.  Beutler.  Clinical Disorders of  Iron




             Metabolism.  (2nd ed.)  New York:  Grune & Stratton, 1971.  486 pp.




 265     Faulds, J.  S.  Haematite pneumoconiosis in Cumberland Miners.  J. Clin.




             Path.  10:187-199, Aug. 1957.




 265a   Faulds, J. S., and M. J. Stewart.  Carcinoma of the lung in haematite




             miners.  J. Path. Bacteriol.  72:353-366, 1956.





 266.      Fillet,  G., J.  D.  Cook,  and  C.  A.  Finch.   Storage iron kinetics.   VII.   A




               biologic  model for  reticuloendothelial  iron transport.   J. Clin. Invest.




               53:1527-1533,  1974.



 267.     Fillet, G.,  and  G. Marsaglia.   Idiopathic  hemochromatosis. (IH).   Abnormality in




              RBC  transport  ot iron by  the reticuloendothelial system (RES).   Blood




              46:1007,  1975.   (abstract)



 268.     Finch, C. A.  Body iron exchange in man.  J. Clin. Invest.  38:392-396, 1959.
269.     Finch,  C. A.   Iron metabolism in hemochromatosis.   J.  Clin.  Invest.   28:




               780-781,  1949.




270.     Finch, C. A., K. Deubelbeiss, J. D. Cook, J. W. Eschbach, L. A. Harker, D. D.




              Funk, G. Marsaglia, R. S. Hillman, S. Slichter, J. W. Adamson, A. Ganzoni,




              and E. R. Giblett.  Ferrokinetics in man.  Medicine (Baltimore)  49:17-53,




              1970.




271.    Finch, C. A., F. Hosain, E. H. Morgan, G. Marsaglia, E. Giblett, and R. S.




             Hillman.  The  ferrokinetic approach to anemia.  Series Haematol.  6:



             30-40, 1965.




                                    298

-------
272.    Finch, C. A., I. R. Miller, A. R. Inamdar, R. Person, K. Seiler, and B.
             Mackler.  Iron deficiency in the ratt  Physiological and biochemical
             studies of muscle dysfunction.  J. Clin. Invest.  58:447-453, 1976.
273.    Finch, S. C., and C. A. Finch.  Idiopathic hcmochromatosis, an iron storage
             disease.  A.  Iron metabolism in homocliromuLoais.  Medicine (Baltimore)
             34:381-430, 1955.
274.  Fischer, D. S., R. Parkman, and S. C. Finch.  Acute iron poisoning in children:
           The problem of appropriate therapy.  J.A.M.A.  218:1179-1184, 1971.
275.   Fisher, M. W., P. E. Morrow, and C. L. Yuile.  Effect  of Freund's complete
            adjuvant upon the clearance of iron-59 oxide from rat lungs.  J.
            Reticuloendothel.  Soc.   13:536-556,  1973.
276.  Forbes,  G.  Poisoning with a preparation of iron, copper, and manganese.  Brit.
           Med. J.  1:367-370, 1947.
277.   Ford, H. W.  Sludges and Associated Problems Involving Agricultural Drains in
            Florida Wetlands.   Paper Presented at 1970 Specialty Conference,  American
            Society of Agricultural Engineers and American Society of Civil Engineers,
            Miami, Florida, Nov.  4-6,  1970.
278.    Forth,  W., and W. Rummel.   Iron absorption.   Physiol.  Rev.   53:724-792, 1973.

279.    Foucar, F. H.,  B. S.  Gordon, and S. Kaye.   Death following ingestion  of ferrous
             sulfate.  Amer.  J.  Clin.  Path.   18:971-973, 1948.
280.    Foy, H., and A. Kondi.  Anaemias of the tropics:  Relation to iron intake,
             absorption and losses during growth, pregnancy and lactation.  J. Trop.
             Med. Hyg.  60:105-118, 1957.
280a.    F°y» H-» A-  Kondi, and W.  H.  Austin.  Effect of dietary phytate on fecal
             absorption of radioactive ferric chloride.  Nature  183:691-692, 1959.
                                      299

-------
281.   Frenkiel,  F.  N.,  and  R. E. Munn,  Eds.   Turbulent  Diffusion in Environmental
            Pollution.   Proceedings  of  a Symposium held  at  Charlottesville,  Vir-
            ginia April  8-14,  1973.  Advances  in  Geophysics Vol.  18B.   New York:
            Academic  Press,  1974.  389  pp.
282.   Frick,  P.  G.   Iron  deficiency in blood  donors,  pp. 511-517.   In I.  Hallberg,
            H.-G.  Harwerth,  and  A. Vannotti, Eds.  Iron  Deficiency:   Pathogenesis,
            Clinical  Aspects,  Therapy.   New York:  Academic Press, 1970.
282a.  Fritz,  J.  C.,  G.  W. Pla,  T. Roberts,  J. W.  Boehne, and  E.  L.  Hove.
            Biological availability  in  animals of iron from common dietary
            sources.   J. Agric.  Food Chem.   18:647-651,  1970.
283.  Frost, A. E.,  H. H.  Freedman,  S.  J. Westerback,  and A.  E. Martell.  Chelating
           tendencies of N,N'-ethylenebis-[2-(o-hydroxyphenyl)J -glycine.  J. Amer.
           Chem.  Soc.  80:530-536,  1958.
283a. Fuhr, I., and H. Steenbock.  The effect of dietary calcium, phosphorus, and
           vitamin D on the utilization of iron.   I.  The effect of phytic acid
           on the availability of iron.  II.  The effect of vitamin D on body iron
           and hemoglobin production.   III.   The relation of rickets to anemia.
           J. Biol.  Chem.   147:59-75,  1943.
284. Fullerton, H. W.  The  iron-deficiency  anaemia  of late infancy.   Arch.  Dis.
          Child.  12:91-110,  1937.
285. Furugouri, K.  Characteristic aspects of iron metabolism in piglets.  Jap.
          Agric. Res. Q. (JARQ)   9:171-176,  1975.
286. Fuwa,  K., W.  E.  C.  Wacker,  R. Druyan, A.  F.  Bartholomay,  and B. L. Vallee.
          Nucleic acids  and  metals,  II:   Transition  metals  as  determinants of
          the conformation of  ribonucleic acids.   Proc.  Nat.  Acad. Sci.  46:
          1298-1307,  1960.
287. Gaffuri,  E.   Conoscenze attuali sui tumori professionali dell'apparato
          respiratorio.  Lotta Tuberc.   40:281-302,  1970.

                                  300

-------
 288.    Gale, E., J. Torrance,  and T. Bothwell.  The quantitative estimation of total
             iron stores in human bone marrow.   J. Clin.  Invest.   42:1076-1082, 1963.
 289.   Gandra, Y. R., and R. B. Bradfield.  Energy expenditure and oxygen handling
            efficiency of anemic school children.  Amer. J. Clin. Nutr.  2'-:
            1451-1456, 1971.
 290. .  Garby, L., L. Irnell, and I. Werner.   Iron deficiency in women of fertile age
            in a  Swedish community.  Acta Med.  Scand.   185:113-117,  1969.
 291.  Garibaldi,  J. A.  Influence of temperature on the biosynthesis of iron transport
           compounds by Salmonella typhimurium.  J. Bacteriol.  110:262-265, 1972.
 292.  Garibaldi,  J. A.,  and J.  B. Neilands.   Formation of iron-binding compounds by
           micro-organisms.  Nature  177:526-527, 1956.
 293. Gardner,  G." W., V.' R.' Edgerton, R.  J." Barnard,  and  E.  M. Bernausr. Cardio-
           respiratory,  hematological and physical performance responses of anemic
           subjects to iron treatment.  Amer.  J. Clin.  Nutr.  28:982-988,  1975.
 294.  Garrels,  R. M., and C. L.  Christ.  Solutions, Minerals,  and Equilibria.   New
           York:   Harper and Row,  1965.  450 pp.
 295. Garrels, R.  M., and F. T.  Mackenzie.  Evolution of Sedimentary Rocks.   New York:
          W. W. Norton and Company,  1971.  397  pp.
 296.  Giblett,  E. R.  F.  transferrin, pp.  555-568.  In W.  S. Root and N. I.
           Berlin, Eds.    Physiological Pharmacology.   Vol.  V. Blood.   New
           York:   Academic Press,  1974.
 296a. Gitlow, S.  E., and M. R.  Beyers.  Metabolism of iron.  I.  Intravenous
           iron tolerance tests in normal subjects and patients with hemochroma-
           tosis.  J. Lab. Clin. Med.  39:337-346, 1952.
297. Glover, J.,  and A.  Jacobs.  Activity pattern of iron-deficient rats.   Brit.  Med.
           J.   2:627-628,  1972.
298. Golberg, L., L. E.  Martin,  and J.  P.  Smith.  Iron overloading  phenomena
          in animals. Toxicol.  Appl. Pharmacol.  2:683-707, 1960.
                                    301

-------
  299.   Golberg, L.,  and J. P4  Smith.   Iron overloading and hepatic vulnerability.
             Amer.  J. path.  36:125-149,  1960.
  300.  Golberg, L., J. P. Smith, and L. E. Martin.  The effects of intensive and
            prolonged administration of iron parenterally in animals.  Brit. J.
            Exp. Path.  38:297-311, 1957.                                         	
  30-^   Goldberg,  E.  C.   Chemistry --  the oceans as a chemical system,  pp. 3-25.
             In M.  N. Hill, Ed.   The Sea.   Vol.  2.  Composition of Sea-Water,
             Comparative and Descriptive  Oceanography.   New York:   Interscience
             Publishers, 1963.
 302.   Goodman, J.  R.,  J. B. Warshaw, and P.  R. Dallman.  Cardiac hypertrophy in
             rats with iron and  copper deficiency:  Quantitative contribution of
             mitochondrial enlargement.  Pediatr. Res.   4:244-256, 1970.
 303.  Goossens, J.  P.   Idiopathic  haemochromatosis:   Juvenile and  familial type -
             endocrine aspects.   Neth.  J.~ Med.   18:161-169,  1975.
 304.  Goya, N., S. Miyazaki, S. Kodate,  and B. Ushio.  A family of congenital atrans-
            ferrinemia.   Blood  40:239-245, 1972.
 3Q5    Grace, N. D.,  and L. W.  Powell.  Iron storage disorders of the liver.
            Gastroenterology  67:1257-1283, 1974.
306.   Grady, R. W., J. H. Graziano, H. A. Akers, and A. Cerami.  The development of
            new iron-chelating drugs.  J. Pharmacol. Exp. Ther.  196:478-495, 1976.
307.   Jones, J. G.,  and C. G. Warner.  Chronic exposure to  iron oxide, chromium
            oxide,  and nickel oxide  fumes of metal dressers  in a steel-works.
            Brit. J.  Ind. Med.  29:169-177, 1972.
308.   Green, R., R. Charlton, H. Seftel, T. Bothwell, F. Mayet, B. Adams, C. Finch,
            and M. Layrisse.  Body iron excretion in man:  A collaborative study.
            Amer. J. Med.  45:336-353, 1968.
                                   302

-------
 309.     Greenberg,  G.   Sarcoma after-  intramuscular iron injection.   Brit. Med.  J.
              1:1508-1509,  1976.
 310.    Greenberg, M. S., G. Strohmeyer, G. J. Hine,  W. H. Keene, G. Curtis, and T. C,
             Chalmers.  Studies in iron absorption.   III.   Body radioactivity measure-
             ments of patients with liver disease.  Gastroenterology  46:651-661,
             1964.
 311.   Greengard, J., and  J. T. McEnery.   Iron poisoning  in children.   GP  37(2):
            88-93,  1968.
 312.   Greenwald, I.  The  effect of phosphate on the solubility of calcium carbonate
            and of bicarbonate on the solubility of calcium and magnesium phosphates.
            J. Biol. Chem.  161:697-704, 1945.
 312a.   Greweling, T.  The Chemical Analysis of Plant Tissue.   Agronomy Mimeo
             66-22.   Ithaca, N.  Y. :   Cornell University,  1966.   82 pp.
313.   Grosberg, S. J.  Hemochromatosis and heart failure:  Presentation of a case
            with survival after three years' treatment by repeated venesection.  Ann.
            Intern. Med.  54:550-566, 1961.
314.   Guha, D. K.,  B.  N. Walia, B. N. Tandon, M. G. Deo, and 0. P. Ghai.   Small  bowel
           changes  in  iron-deficiency anaemia of childhood.  Arch. Dis. Child.   43:
           239-244, 1968.
315.   Haematite  and iron oxide, pp. 29-36.  In IARC Monographs  on the Evalua-
            tion of Carcinogenic Risk of Chemicals to Man.  Vol. 1.  Lyon, France:
            World Health Organization, 1972.
316<   Hahn, P. F.,  E.  Jones, R. C. Lowe,  G. R. Meneely,  and W. Peacock.  The  relative
           absorption  and  utilization of  ferrous and  ferric iron  in anemia  as
           determined  with the radioactive isotope.   Amer. J. Physiol.   143:191-197,
           1945.
                                       303

-------
317.   Halitsky, J.   Gas diffusion near buildings, pp. 221-225.  In D. H. Slade,




            Ed.   Meteorology and Atomic Energy  1968.  TID-24190.  Oak Ridge,




            Tenn.:   U.  S.  Atomic Energy Commission,  Division of Technical Infor-




            mation,  1968.



317a.  Hall,  D.  0.   Copper and  iron metalloproteins.   Trends  Biochem.  Sci.   2:




            N86,  1977.   (letter)




318.   Hall,  D.  0.,  R.  Cammack,  and K.  K. Rao.  Non-haem  iron proteins, pp.  279-334.




            In A. Jacobs and M.  Worwood, Eds.  Iron  in Biochemistry and Medicine.




            New  York:   Academic  Press,  1974.




319.   Hall,  G. J. L.,  and A. E. Davis.  Inhibition of  iron absorption by magnesium




           trisilicate.  Med. J. Austral.  2:95-96,  1969.




320.  Hallberg,  L.,  and E. Bjorn-Rasmussen.   Determination of iron absorption from




           whole diet:   A  new  two-pool  model  using two radioiron isotopes given as




           haem  and  non-haem iron.   Scand.  J. Haematol.  9:193-197,  1972.




321.  Hallberg, L.,  A.-M. Hogdahl, L. Nilsson, and G.  Rybo.   Menstrual blood loss  --



           a  population study.   Variation at  different  ages and  attempts  to  define




           normality.   Acta Obstet. Gynecol.  Scand.   45:320-351,  1966.



322.  Hallberg,  L., and L. Nilsson.  Constancy of individual mentrual blood  loss.



           Acta Obstet. Gynecol. Scand.  43:352-359, 1964.




323.   Hallberg, L., and L. Solvell.  Iron absorption studies.  Acta Med. Scand.




           168(Suppl.  358):1-108, 1960.




324.  Hallberg,  L., L. Solvell, and B. Zederfeldt.   Iron absorption after partial




           gastrectomy:  A comparative study on the absorption from ferrous  sulphate




           and hemoglobin.  Acta Med. Scand.   179(Suppl. 445):269-275, 1966.




325.  Hallgren,  B.,  and P. Sourander.  The effect  of  age  on the non-haemin iron in the




           human  brain.  J. Neurochem.  3:41-51, 1958.
                                       304

-------
 326.    Hamilton, A.,  and  H. L. Hardy.  Industrial Toxicology.   (3rd  ed.)
            ^Acton, Mass.:  Publishing Sciences Group, Inc., 1974.  575 pp.
 327.   Hamilton, A.,  and H.  I.  Hardy.  Metal fumes,  pp.  195-196; benign dusts,
             pp. 197-201; and mixed dusts,  pp.  443-455.   In Industrial Toxicology.
             (3rd ed.)   Acton, Mass.:   Publishing  Sciences  Group, Inc.,  1974.
 328.  Hamilton, D. t., and L. S. Valberg.  Relationship between cadmium and  iron
            absorption.  Amer. J7 Physio1.  227:1033-1037, 1974.
 329.  Hamilton, E., R. Williams, K. A. Barlow, and P. M. Smith.  The arthropathy of
            idiopathic haemochromatosis.  Q.  J. Med.   37:171-182, 1968.
 330. Hankes, L. V., C. R. Jansen,  and M.  Schmaeler.  Ascorbic  acid catabolism in Bantu
           with hemosiderosis (scurvy).  Biochem. Med.  9:244-255, 1974.
 331.   Hansen,  H. A.,  and  A.  Weinfeld.  Hemosiderin  estimations and  sideroblast
             counts  in  the  differential  diagnosis  of  iron deficiency  and other
             anemias.   Acta Med.  Scand.   165:333-356,  1959.
 332.   Hard, S.  Non-anemic iron deficiency as an etiologic factor in diffuse loss of
            hair of the scalp in women.  Acta Derm. Venereol. 43:562-569,  1963.
 333.  Barker, L. A., D. D. Funk, and C. A. Finch.  Evaluation of  storage  iron by
           chelates.  Amer. J. Med.  45:105-115, 1968.
 334.   Harrison,  P. M.,  R.  J.  Hoare, T.  G.  Hoy,  and I.  G.  Macara.   Ferritin
            and haemosiderin:   Structure and  function,  pp.  73-114.  In A.  Jacobs
            and M. Worwood,  Eds.   Iron  in Biochemistry  and Medicine.   New  York:
            Academic  Press,  1974.
335.   Hartley, W. J., J. Mullins,  and  B. M. Lawson.  Nutritional  siderosis  in
           the bovine.  N. Z. Vet. J.  7:99-105,  1959.
336.   Harvard"University.   Oxygenation of Ferrous  Iron.   Water Pollution  Control
            Research  Series 14010-06/69.  Washington,  D.  C.:   U. S. Government
            Printing  Office,  1970.   /~201  pp._7
                                       305

-------
 337.    Hasegawa,  T.,  and A.  Sugimae.   Studies on heavy metals in atmospheric
             particulates  -  on the size distributions.   Kuki Seijo ( Jap. Air
              Clean.  Assoc.) 9(1):1-9,  1971.   (in Japanese,  summary in English)
 338.    Hasegawa,  T.,  A.  Sugimae,  J.  Fujii,  K.  Matuso,  and  Y.  Okuyama.   Trace
              metals  in suspended  particulates  (III).  Taiki Osen  Kenkyu (J.  Jap.
              Soc.  Air  Pollut.)  5:210,  1970.   (in Japanese)
 339^   Haskins,  D.,  A. R.  Stevens,  Jr.,  S. Finch, and C.  A. Finch.   Iron metabolism.
             Iron stores in  man  as  measured by  phlebotomy.   J.  Clin.  Invest.   31:
             543-547, 1952.                                                          _
 340.   Hathorn,  M. K.  S.   The influence of hypoxia on  iron  absorption in the rat.
             Gastroenterology   60:76-81,  1971.
 341.   Heck, W.  W.,, and  L. F.  Bailey.  Chelation of trace  metals in nutrient
             solutions. Plant Physiol.  25:573-582, 1950.
 342.   Hegenauer,  J.,  and P.  Saltman.  Iron  and  susceptibility to infectious  disease.
             Science  188:1038-1039, 1975.
 343.   Hegsted,  D. M., C. A.  Finch, and T. D.  Kinney.   The  influence of diet on iron
             absorption.  II."  The interrelation of iron and phosphorus.  J.  Exp.  Med.
             90:147-156, 1949.
 344.   Heinrich, H. C.  Intestinal  iron absorption in man - methods of measurement,
            dose relationship, diagnostic and therapeutic applications,  pp. 213-296.
             In L.  Hallberg, H.-G. Harwerth,  and  A. Vannotti, Eds.  Iron  Deficiency:
 345.  Hem, J. D.   Chemical factors that influence the availability of iron and
            manganese in aqueous systems.   Geol.  Soc.  Amer.  Bull.   83:443-450, 1972.
345a.  Hem., J. D.  Study and Interpretation of the Chemical Characteristics  of
            Natural Water.   (2nd ed.)  Geological Survey Water-Supply Paper  1473.
            Washington, D. C., 1970.  363 pp.
                                       306

-------
 346.    Hem, J.  D.   Role of hydrous metal oxides in the transport of heavy metals




             in the environment (G. F.  Lee).   Prog. Water Technol.  7:149-153,




             1975.  (Discussion)



 347<    Hem, J.  D., and W.  H.  Cropper.   Survey of Ferrous-Ferric  Chemical Equilibria




             and Redox Potentials.   U.S.  Geological Survey Water-Supply Paper 1459-A.




             Washington,  D.C.:   U.S. Government Printing Office,  1959.   31 pp.



 348>  Hemmaplardh, D., S.  G. Kailis, and E. H. Morgan.  The effects of  inhibitors of




            microtubule and microfilament function  on transferrin and iron uptake by




            rabbit  reticulocytes and bone marrow.   Brit. J. Haematol.   28:53-65, 1974.




 348a.  Henry, M.  C.,  and  D.  G.  Kaufman.   Clearance of benzofajpyrene  from hamster




            lungs after administration  on coated particles.  J. Nat. Cancer Inst.




            51:1961-1964,  1973.




 349.   Henry, M.  C.,  C.  D.  Port,  and D. G.  Kaufman. Importance of  physical properties




            of benzo(a)pyrene-ferric oxide  mixtures in  lung tumor induction.   Cancer




            Res.  35:207-217,  1975.



 350.   Henry, M.  C.,  C.  D.  Port,  and D.  G.  Kaufman. Role of particles  in respir-




            atory carcinogenesis  bioassay,  pp.  173-185.   In E. Karbe  and J. F.




            Park, Eds.   Experimental Lung Cancer :   Carcinogenesis  and  Bioassays.




            International Symposium Held  at the Battelle Seattle  Research Center.




            Seattle,  WA,  USA,  June  23-26, 1974.   New York:   Springer-Verlag,  1974.



351.    Hershko,  C.,  J. D.  Cook, and C. A.  Finch.   Storage iron kinetics.   II.  The




             uptake of hemoglobin iron  by hepatic parenchymal cells.  J.  Lab. Clin.




             Med.   80:624-634,  1972.
352.   Hershko, C., J..D. Cook, and C. A. Finch.  Storage iron kinetics.   III.




             Study  of desferrioxamine action by selective radioiron  labels  of RE




             and parenchymal cells.  J. Lab. Clin. Med.  81:876-886, 1973.
                                    307

-------
 353.   Hershko, C., J. D. Cook, and C. A. Finch.  Storage iron kinetics.  VI.  The
            effect of inflammation on iron exchange in the rat.  Brit. J. Haematol.
            28:67-75, 1974.
 354.   Hershko, Ch.,  A.  Karsai, L. Eylon, and G. Izak.   The effect of chronic iron
            deficiency on some biochemical functions of the human hemopoietic tissue.
            Blood  36:321-329, 1970.
 355>    Heslop-Harrison, J.   Development, differentiation and yield,  pp. 291-321.
             In J.  D.  Eastin, F.  A.  Haskins,  C.  Y.  Sullivan,  and C.  H.  M.  Van Bavel,
             Eds.   Physiological  Aspects  of  Crop Yield.   Madison, Wise.:   Ameri-
             can Society of Agronomy and  Crop Science Society of America,  1969.
 356.    Hesse, P. R.   A Textbook of Soil Chemical Analysis.  New York:  Chemical
             Publishing Co.,  1971.  520 pp.
 357.    Hesseltine, C.  W.,  C. Pidacks, A.  R.  Whitehall, N. Bohonos,  B.  L. Hutchings,
              and J. H.  Williams.  Coprogen,  a new growth  factor  for  coprophilic
              fungi.  J. Amer.  Chem. Soc.   74:1362,  1952.
 358.    Hewitt, E. J.   The  essential nutrient elements:   Requirements  and  interactions
              in plants, pp.  137-360.  In F.  C.  Steward, Ed.  Plant Physiology:   A
              Treatise.  Vol.  3.   Inorganic Nutrition of Plants.  New York:   Academic
              Press, 1963.
 359.   Hewitt, E. J.  Trace elements in  plants:  Biochemical aspects,  pp.  21-34.
             In Ministry of  Agriculture,  Fisheries and  Food, Technical  Bulletin
             21.  Trace  Elements  in Soils and  Crops.  Proceedings of  a  Conference,
             1966.  London:   Her Majesty's  Stationery Office, 1971.
360.   Hewitt, E. J.,  and S. C. Agarwala.  Reduction of triphenyltetrazolium chloride
            by plant  tissues and its relationship to molybdenum status.  Nature  169:
            545-546,  1952.
                                      308

-------
 361.    Hidy,  G.  M..  and  J.  R.  Brock.   An assessment  of the global sources of
             tropospheric aerosols,  pp.  1088-1097.  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,  1971.
 362.   Higginson, J., B. G. Grobbelaar, and A. R. P.  Walker.  Hepatic fibrosis and
            cirrhosis in man in relation to malnutrition.  Amer. J. Path.  33:
            29-53, 1957.
 363     Hill,  C.  H.  Studies on  the  mutual  interaction  of  iron and  manganese  on
             absorption from chick duodenal segments  in situ.  Fed.  Proc.   30:236,
             1971.   (abstract)
 364.    Hill,  C.  H.,  and  G. Matrone.  Studies on copper and iron deficiencies in grow-
             ing  chickens.   J.  Nutr.  73:425-431, 1961.
365     Hillman,  R.  S.,   and C.  A.  Finch.   Red Cell Manual.   (4th ed.)   Philadelphia:
             F.  A.  Davis  Company,  1974.   84 pp.
366.    Ho,  W.,  and A.  Furst.  Intratracheal instillation method for mouse lungs.
             Oncology  27:385-393,  1973.
367.    Hoag, M. S., R. 0. Wallerstein, and M. Pollycove.  Occult blood loss in iron
             deficiency anemia of infancy.  Pediatrics  27:199-203, 1961.
368.    Hobbs,  J.  R.   Iron deficiency after partial gastrectomy.   Gut   2:141-149,  1961
369.    Hoffbrand, A. V.,  and S.  A.  Broitman.   Effect of chronic nutritional iron
             deficiency on the small intestinal disaccharidase activities of growing
             dogs.  Proc.  Soc. Exp.  Biol.  Med.   130:595-598,  1969.
370.    Hoffbrand,  A.  V. ,  K.  Ganeshaguru,  J. W.  1.  Hooton,  and M.  H. N.  Tatersall.
             Effect of iron deficiency and desferrioxamine  on DNA synthesis in
             human cells.   Brit.  J.  Haematol.   33:517-526,  1976.
                                     309

-------
 371.   Hofvander,  Y.   Haematological investigations in Ethiopia with special
             to a high iron intake.   Acta Med.  Scand.  Suppl.   494:1-74,  1968.
 372.   Hogan,  G. R.,  and B.  Jones.   The relationship  of koilonychia and iron
             deficiency in infants.   J.  Pediatr.   77:1054-1057,  1970.
 373.  Holmes,  R.  S., and J. C. Brown.  Chelates as correctives for chlorosis.  Soil
            Sci.  80:167-179, 1955.
 374.  Holowach, J.,  and D. L.  Thurston.  Breath-holding spells  and anemia.   New Engl.
            J.  Med.   268:21-23, 1963.
 375. Horowitz, N.  H., G. Charlang,  G. Horn, and N. P. Williams.  Isolation and
           identification of the conidial germination factor of Neurospora crassa.
           J. Bacteriol.  127:135-140, 1976.
 376.  Howell,  D.  Significance of iron  deficiencies.  Consequences  of mild
           deficiency  in children, pp.  65-69.  In National Research  Council,
           Food and Nutrition  Board.  Extent and Meanings of  Iron Deficiency
           in  the U. S.  Proceedings  of Workshop, March 8-9,  1971.  Washington,
           D.  C.:   National Academy  of  Sciences, 1971.
377.   Hubers, H., E. Hubers, W. Forth, G. Leopold, and W. Rummel.  Binding of iron
            and other metals in brush borders of jejunem and ileum of the rat in
            vitro.  Acta Pharmacol. Toxicol.  29(Suppl. 4);22, 1971.
378.   Hubers, H. ,  E.  Hubers,  W.  Forth,  and W.  Rummel.   Iron absorption and
            cellular transfer  protein in the intestinal mucosa of mice with
            hereditary anemia.  Naunyn-Schmiedeberg's Arch.  Pharmacol. 274:R56,
            1972.   (abstract)
379.   Hueper,  W.  C.   A Quest into the Environmental Causes of Cancer of the Lung.
            Public Health Monograph No.  36.   Washington,  D. C. :   U.  S.  Government
            Printing Office,  1955.  54 pp.
380.  Hurtado,  A., C. Merino, E.  Delgado.   Influence of anoxemia on the hemopoietic
           activity.   Arch.  Intern. Med.  75:284-323, 1945.
                                     310

-------
 380a.   Hussain, R.,  and V.  N.  Patwardhan.   The influence of phytate on  the
   	    absorption of iron.   Indian J.  Med.  Res.   47:676-682,  1959.
 381.   Hussain, R., V. N. Patwardhan, and S. Sriramachari.  Dermal  loss of  iron in
            healthy Indian men.  Indian J. Med. Res.  48:235-242, 1960.
 382.   Hutton, C. F.   Plummer Vinson syndrome.   Brit.  J. Radiol.  29:81-85, 1956.
 383.   Hyde, B. B., A. J. Hodge, A. Kahn, and M. L. Birnstiel.  Studies on phyto-
            ferrin.  I."  Identification and localization.  J." Ultrastruct. Res.
            9:248-258, 1963.                	
 384.   Hyman, C.  B.,  B. Landing,  R. Alfin-Slater, L.' Kozak, J. Weitzman,  and  J.  A.
            Ortega,   dl-^£-  Tocopherol,  iron,  and lipofuscin  in  thalassemia.   Ann.
            N.  Y.  Acad. Sci.   232:211-220,  1974.
 385.   Iljift. W.  S.'  Metabolism of  plants  affected  with  lime-induced chlorosis
            (calciose).   I.   Nitrogen metabolism.   Plant Soil 3:239-256,  1951.
 386.   Iljin, W. S.  Metabolism of plants affected with lime-induced chlorosis
            (calciose).  III.  Mineral elements.  Plant Soil  4:11-28, 1952.
386a.   International Committee for Standardization in Hematology.  Recommended
             methods for radioisotope red cell survival studies.  Blood  38:378-
             386, 1971.
387.    Isaacson, C., H. C. Seftel, K. J. Keeley, and T. H. Bothwell.   Siderosis
             in the Bantu:  The relationship between iron overload and cirrhosis.
             J. Lab. Clin. Med.  58:845-853, 1961.
387a.   Isied,  S.  S., G.  Kuo,  and K.  N.  Raymond.   Coordination isomers of biologi-
             cal iron transport compounds.   V.   The preparation and chirality of
             the chromium(III) enterobactin complex and model  tris(catechol)chrom-
             ium(III) analogues.   J.  Amer.  Chem.  Soc.   98:1763-1767,  1976.
                                     311

-------
 387b.   Ishinishi,  N.,  Y.  Kodama,  E.  Kunitake,  K.  Nobutomo,  And Y.  Fukushima.
             The carcinogenicity of dusts  collected from an  open-hearth furnace
             for the smelting of iron:   A  preliminary experimental  study,  pp.  480-
             488.   In G.  F.  Nordberg,  Ed.   Effects and Dose-Response Relationships
             of Toxic  Mctnls.   New York:   Elsevier Scientific  Publishing Compnny,
             1976.
 388.   Ito> T., and J. B. Neilands.  Products of  "low-iron fermentation" with  Bacillus
            subtilis;  Isolation, characterization and synthesis of 2,3-dihydroxy-
   	benzoylglycine.  J. Amer. Chem. Soc.  80:4645-4647, 1958.
  _     Iverson, W. P.  Microbial corrosion of Iron, pp. 475-513.  In J. B. Neilands,
            Ed.  Microbial Iron Metabolism:  A Comprehensive Treatise.  New York:
            Academic Press, 1974.
 390.    Jackson, M. L.   Chemical composition of soils, pp.  71-141.   In F.  E.  Bear,
             Ed.  Chemistry of the Soil.   (2nd  ed.)  American  Chemical Society
             Monograph Series No.  160.   New York:   Reinhold  Publishing Corp.,  1964.
 391.   Jackson, T.  L., J. Hay,  and D. P. Moore.   The  effect of Zn on  yield and chemical
            composition of  sweet  corn in the Willamette  Valley.  Amer. Soc. Hort.  Sci.
            91:462-471, 1967.
392.   Jacobs,  A.   Anaemia and post-cricoid carcinoma.   Brit. J. Cancer  15:736-744,
  	1961.
393.   Jacobs, A.  Iron-containing  enzymes  in the buccal  epithelium.   Lancet   2:
            1331-1333, 1961.
394.   Jacobs, A.  Iron deficiency:  Effects on tissues and  enzymes, pp.  135-
            141.   In H. Kief, Ed,   Iron Metabolism and Its Disorders.  New York:
            American Elsevier Publishing Co. Inc., 1975.
395.   Jacobs,  A.   Oral cornification in anaemic patients.  J. Clin. Path.  12:
            235-237,  1959.

                                     312

-------
  396.    Jacobs, A.  The buccal mucosa in anaemia.   J.  Clin.  Path.   13:463-468, 1960.
  397.    Jacobs, A.  Tissue changes in iron deficiency.  Brit. J. Haematol.  16:1-4,
              1969.   (annotation)
  398,    Jacobs, A., and G. S. Kilpatrick.  The Paterson-Kelly syndrome.  Brit. Med. J.
              2:79-82, 1964.
 399.    Jacobs, A., J.  II. Lawric, C. C.  Entwistle, and H. Campbell.  Gastric  acid
              secretion  in chronic iron-deficiency anaemia.  Lancet   2:190-192,  1966.
 400.   Jacobs, A., and P. M. Miles.   Role of gastric secretion in iron absorption.
             Gut  10:226-229, 1969.
 401.   Jacobs,  A., F.  Miller, M. Worwood, M. R.  Beamish, and C. A. Wardrop.   Ferritin
             in the serum of  normal subjects  and  patients with  iron deficiency and iron
             overload.   Brit. Med.  J.  4:206-208,  1972.
 402.    Jacobs,  A., and M. Worwood. Ferritin in  serum:  Clinical  and  biochemical
             implications. New  Engl.  J.  Med.   292:951-956,  1975.
403.   Jacobs, P., T. Bothwell, and R. W. Charlton.  Role of hydrochloric acid in iron
            absorption.  J. Appl. Physiol.  19:187-188, 1964.
404.   Jacobs, P., and C. A. Finch.  Iron for erythropoiesis.  Blood  37:220-230, 1971.
405.    Jacobson,  L.  Maintenance  of  iron  supply  in  nutrient  solutions  by  a  single
             addition of  ferric potassium  ethylenediamine  tetraacetate.  Plant  Physiol.
    	     26:411-413,  1951.
406.    Jalili, M. A., and S. Al-Kassab.  Koilonychia and cystine content of nails.
             Lancet  2:108-110, 1959.
        James, H. L.   Chemistry of the Iron-Rich Sedimentary Rocks.   U.S.  Geological
             Survey Professional  Paper 440-W.   Washington, D.C.:   U.S.  Government
             Printing Office, 1966.   61 pp.
                                      313

-------
 408.   Jandl, J. H., J. K. Intnan, R. L. Simmons, and D. W. Allen.  Transfer  of  iron
            from serum iron-binding protein to human reticulocytes.  J. Clin. Invest.
            38:161-185, 1959.                                                      	
 409    Jarvis,  J.  H., and A.  Jacobs.   Morphological abnormalities  in lymphocyte
            mitochondria associated with  iron-deficiency  anaemia.  J.  Clin.  path.
            27:973-979,  1974.	
409a.  Jenkins, D.  J.  A., M.  S.  Hill,  and J.  H.  Cummings.   Effect of wheat fiber
            on blood lipids,  fecal steroid excretion and serum iron.   Amer. J.
            Clin.  Nutr.   28:1408-1411,  1975.
 410.  Jenne,  E. A.  Controls on Mn,  Fe, Co,  Ni, Cu, and Zn concentrations in
            soils and water:   The significant role of hydrous Mn and Fe oxides.
            Adv. Chem. Ser.  73:337-387, 1968.
 411.  Johnston, R. E., and W. M. Krogman.  Patterns of growth in children with
            thalassemia major.  Ann. N.Y. Acad. Sci.  119:667-679, 1964.
 412.  joske,  R. A., E.  S. Finckh, and  I.  J.  Wood.   Gastric biopsy:   A study of 1,000
            consecutive  successful gastric biopsies.   Q. J. Med.  24:269-294, 1955.
412a.  Judisch,  J.  M.,  J.  L.  Naiman,  and F. A.  Oski.   The fallacy of a fat iron-
            deficient child.   Pediatrics  37:987-990,  1966.
 413,  Kailis,  S. G., and E. H. Morgan.  Transferrin and iron uptake by rabbit bone
            marrow cells in vitro.  Brit. J.  Haematol.  28:37-52, 1974.
 414.  Kattamis, C., N.  Touliatos, S.  Haidas, and N. Matsaniotis.  Growth of children
            with thalassaemia;  Effect of different transfusion regimens.  Arch. Dis.
            Child.   45:502-505,  1970.
 415.   Kavin,  H.,  R. W.  Charlton, P. Jacobs,  R.  Green,  J.  D.  Torrance, and T.  H.
            Bothwell.   Effect of  the exocrine pancreatic  secretions  on iron
            absorption.  Gut   8:556-564,  1967.
                                      314

-------
416.   Kawai, M. ,  T. Matsuyama, A.  Sakai, and T.  Yoshitsugu.  Iron and steel, pp.
            742-745.  In Encyclopaedia of Occupational Safety and Health.  Vol. 1.
            Geneva:  International  Labour Office, 1971.
4j7    Keller-Schierlein, W.,  V.  Prelog,  and H.  Zahner.  Siderochrome.   (Naturliche
            Eisen(III)-trihydroxamat-Komplexe.)   Fortschr.  Chem. Org. Naturst.
            22:279-322,  1964.
418.   Kelly, A. B.  Spasm of the entrance to the oesophagus.  J. Laryngol.  34:
            285-289, 1919.
419.   Kennedy,  V.  C., G.  W. Zellweger, and  B. F.  Jones.  Filter pore-size effects on
            the  analysis  of Al, Fe,  Mn, and  Ti in water.  Water Resources Res.   10:
            785-790, 1974.
420.  Kent, G., and H. Popper.  Liver biopsy in diagnosis of hemochromatosis.  Amer.
           J.'Med.  44:837-841, 1968.  (editorial)
421.  Kent, G.,  F. I.' Volini,  0.  T, Minick,  E.  Orfei, and J. de la Huerga.   Effect
           of iron loading upon the formation of collagen in the hepatic injury
           induced by carbon tetrachloride.   Amer. J. Path.   45:129-155, 1964.
422.  Kim, Y. S.,  and D. F. Martin.  On the effect of iron-citrate on the growth
           of the Florida red tide organism, Gymnodinium breve.  Environ. Lett.
           9:55-58, 1975.
423.  Kimber, C.,  and L. R. Weintraub.  Malabsorption  of iron  secondary  to  iron
           deficiency.  New Engl. J. Med.   279:453-459, 1968.
424.   Kincaid,  R.  L., and  B.  L. O'Dell.   Enterochelin  and unrecognized growth
            stimulants for  guinea pigs.  Fed. Proc.   33:666,  1974.
425.   Kirkby, E.  A.,  and K. Mengel.  Ionic  balance in different tissues of the tomato
            plant  in relation  to  nitrate,  urea,  or ammonium nutrition.  Plant Physiol.
            42:6-14, 1967.
                                    315

-------
  426.   Klauder, D. S., and H.  G.  Petering.   Protective value of dietary copper and iron



              against some toxic effects of lead in rats.   Environ.  Health Perspect.



              12:77-80,  1975.




  427.    Kleckner, M.  S.,  Jr., A. H.  Baggenstoss,  and j. F. Weir.  Iron-storage  diseases.




              Amer.  J. Clin. Path.  25:915-931,  1955.




  428.    Klemic, H., H. L. James, and G. D. Eberlein.  Iron,  pp. 291-306.  In D. A.




              Brobst and W. P. Pratt, Eds.  United States Mineral Resources.  (U.S.




              Geological Survey Professional Paper 820).  Washington, D.C.:  U. S.




              Government Printing Office, 1973.




 429.     Klinger, F.  I.  Iron ore.   Preprint from Bureau  of  Mines  Bulletin 667.  A




               Chapter from Facts and Problems,  1975 Edition.   Washington, D.  C.:




               U.  S. Government  Printing Office, 1975.   21 pp.




 430.     Klinger, F.  L.   Iron ore, pp.  651-675.   In Bureau of  Mines.  Minerals  Year-




               book  1974.   Vol.  1.  Metals, Minerals and Fuels.  Washington, D.  C.:




               U. S.  Government  Printing Office,  1976.
431.     Knizley, H., Jr., and W. D. Noyes.  Iron deficiency anemia, papilledema,



               thrombocytosis, and transient hemiparesis.  Arch. Intern Med.  129: •




              483-486, 1972.



432.    Kolb, E.  The metabolism of iron  in farm animals under normal and  pathologic




             conditions.  Adv. Vet. Sci   8:49-114,  1963.
433.     Kopp, J. F.  Suggested method for spectrochemical analysis of water by the




              rotating disk technique using an optical emission spectrometer, pp. 1010-




              1015.  In Methods for Emission Spectrochemical Analysis.  (6th ed.)




              Philadelphia:  American Society for Testing and Materials, 1971.
                                       316

-------
 434.    Krauskopf,  K.  B.   Geochemistry  of micronutrients,  pp.  7-40.   In J.  J.




             Mortvedt,  P.  M.  Giordano,  and W.  L.  Lindsay,  Eds.   Micronutrients




             in Agriculture.   Proceedings of  a Symposium.   Madison, Wise.:




             Soil  Science  Society  of  America,  Inc.,  1972.



 435.    Kuhn, J. N.,  M. Layrisse,  M.  Roche, C. Martinez-Torres,  and  R.  B.  Walker.




             Observations  on the mechanism  of iron absorption.   Amer. J. Clin.  Nutr.




             21:1184-1188, 1968.




 436.   . Kuo,  B., E. Zaino, and  M. S. Roginsky.   Endocrine  function in  thalassemia




              major.  J. Clin. Endocrinol. Metab.   28:805-808, 1968.




 437.    Kurtz,  H. F.   Bauxite,  pp. 211-223.   In  1970 Bureau of  Mines  Minerals




              Yearbook.  Vol. 1.  Metals, Minerals, and Fuels.  Washington, D. C.:




              U. S. Government Printing Office, 1972.




 438.   Kurtz,  H. F.  Bauxite, pp.  189-204.   In 1972 Bureau of Mines  Minerals




            Yearbook.  Vol.  1.  Metals, Minerals, and Fuels.  Washington,  D. C.,




            U.  S. Government Printing Office, 1974.
439.   Kushner, J. P., G. R. Lee, and S. Nacht.  The role of iron in the pathogenesis




            of  porphyria cutanea tarda:  An in vitro model.  J. Clin. Invest.  51:




            3044-3051, 1972.



440.   Lament, N. McE., and M. Hathorn.  Increased plasma iron and liver pathology in




            Africans with porphyria.  South Afr. Med. J.  34:279-281, 1960.




441.  Langer, E. E., R. G. Raining, R. F. Labbe, P. Jacobs, E. F. Crosby, and C. A.




           Finch.  Erythrocyte protoporphyrin.   Blood  40:112-128, 1972.





442.   Langmuir, D.  Particle size effect on the reaction goethite  = hematite




            water.  Amer. J. Sci. 271:147-156,  1971.
                                     317

-------
 443. Langmuir, D.  The Gibbs free energies of substances in the system Fe-09-H90-C0
                                                                            £•  £.


           at 25°C, pp. 180B-184B.  In Geological Survey Research, 1969.   (U.S. Geo-



           logical Survey Professional Paper 650-B).  Washington, D.C.:  U.S. Govern


           ment Printing Office, 1969.


 444.  Langmuir,  D.,  and D.  0. Whittemore.   Variations  in  the  stability of



            precipitated ferric  oxyhydroxides.  Adv.  Chem. Ser.  106:209-234,  1971.



 445.  Lanzkowsky,  P.   The  influence  of  maternal  iron-deficiency anaemia  on the



            hemoglobin of the  infant.  Arch. Dis.  Child.   36:205-209,  1961.



 446.  Lassman, M.  N., M. Genel, J." R. Wise, R. Handler,  and P.  Pelig.  Carbohydrate



            homeostasis and pancreatic islet cell function in  thalassemia.  Ann.



            Intern. Med. 80:65-69, 1974.



 447.  Lassman, M.  N.,  R. T. O'Brien,  M. A. Pearson,  J. K. Wise, R. R.  Donabedian,



           P'"„' Felig,  and Mi Genel.  Endocrine evaluation  in thalassemia major.



           Ann. NTY'fAcad. Sci.   232:226-237, 1974.


 448.  Latimer, W.  M.   Oxidation Potentials.  The  Oxidation  States, of  Elements


            and Their  Potentials in Aqueous  Solutions.  (2nd ed.)  Englewood



           Cliffs, N.  J.:   Prentice-Hall, Inc.,  1952.  392  pp.
449.  Lavender, S., and J. A. Bell.  Iron intoxication in an adult.  Brit. Med. J.



           2:406, 1970.


450.  Lawlor, M. J., W. H. Smith, and W. M. Beeson.  Iron requirement of the growing



           lamb.  J. Anim. Sci.  24:742-747, 1965.



451.  Layrisse, M.  Dietary iron absorption, pp. 25-33.  In H. Kief, Ed.  Iron



           Metabolism and Its Disorders.  Proceedings of the Third Workshop


           Conference Hoechst, Schloss Reisensburh, 6-9 April, 1975.  International



           Congress Series 366.  New York:  American Elsevier Publishing Company,


           Inc., 1975.
                                      318

-------
                                       X
452. Layrisse, M. , J.  D.  Cook, C. Martinez-Torres, M. Roche, I. N. Kuhn,  R.  B.

          Walker,  and  C.  A.  Finch.  Food iron absorption:  A comparison of vege-


          table  and  animal  foods.  Blood  33:430-443, 1969.

453. Layrisse, M., and C. Martinez-Torres.   Food iron absorption:   Iron supplementa-

          tion of food.  Prog. Hematol.   7:137-160, 1971.


454. Layrisse, M., and CT Martinez-Torres.  Model  for measuring dietary absorption


           of heme iron:  Test with a  complete meal.  Amer. J7 Clin.  Nutr.  25:


           401-411, 1972.
                          X
455. Layrisse, M., C. Martinez-Torres, J. D. Cook,  R. Walker,  and  C.  A. Finch.  Iron

          fortification of food;   Its  measurement by the extrinsic tag method.

          Blood  41:333-352, 1973.

                           '                 .      /
456.  Layrisse, M., C. Martinez-Torres,  and M.  Gonzalez.  Measurement of the total

           daily dietary iron absorption by the  extrinsic tag model.   Amer. J. Clin.

           Nutr.   27:152-162, 1974.


457. Layrisse, M., C.  Martinez-Torres,  M.  Renzy,  and I. Leets.  Ferritin  iron

           absorption in man.  Blood   45:689-698,  1975.
                            X
458.  Layrisse, M., C. Martinez-Torres,  and M.  Roche.  Effect of interaction of various

           foods  on iron absorption.   Amer.  J.  Clin.  Nutr.  21:1175-1183,  1968.

459.  Layrisse, M., A. Paz,  N. Blumenfeld,  and M. Roche.  Hookworm anemia:  Iron

           metabolism and erythrokinetics.   Blood  18:61-72,  1961.

460.   Layrisse,  M.,  and M.  Roche. The  relationship  between  anemia and hookworm

            infection:   Results of surveys  of rural Venezuelan  population.  Amer.

            J. Hyg.  79:279-301,  1964.

461.   Le Bouffant, L., J.-P.  Henin, J.-C. Martin,  and H.  Da Niel.  Etude  experi-

            mentale de  1'epuration pulmonaire.   Action de 1'empoussierage  sur la

                      , '                     ^
            capacite  d'epuration.  Application a 1'homme.  Lille Med.  17:1091-

            1101, 1972.


                                    319

-------
  462.   Lee, D. n. R.  Biological effects of ingested asbestos:  Report and commentary.
             Environ. Health Perspect.  9:113-122, 1975.
  463.    Lee,  R.  E. ,  Jr., J. S.  Caldwell,  and G.  B. Morgan.   The evaluation of
              methods for measuring suspended particulates  in air.   Atmos.  Environ.
              6:593-622,  1972.
  464.    Lee,  R.  E.,  Jr., and R. K. Patterson.  Size determination of atmospheric
              phosphate,  nitrate, chloride, and ammonium particulate in several urban
              areas.   Atmos. Environ.  3:249-255,  1969.
 465.    Lee,  R.  E., Jr., R. K. Patterson,  and J.  Wagman.  Particle-size distribution
               of  metal components  in urban  air.   Environ. Sci. Technol.  2:288-290,
               1968.
 466.   Lees, F., and F. D. Rosenthal.  Gastric mucosal lesions before and after treat-
             ment in iron deficiency anaemia.  Q.  J. Mted.   27:19-26,  1"958.
 467.    Leonard, B. J.  Gastric acid in iron-deficiency anaemia.  Lancet  2:440-441,
               1966.
 468.    Leonard, B. J.  Hypochromic anaemia in R.A.F.  recruits.   Lancet   1:899-902,
               1954.
 469.   Leonard,  C. D.,  arid I.  Stewart.  An available source of iron for plants.   Proc.
             Amer. Soc.  Hort. Sci.   62:103-110, 1953.
 470.    Leong, J., J. B.  Neilands, and K. N. Raymond.  Coordination isomers of bio-
             logical iron transport compounds III.  (1)  Transport of ^-cis-chromic
             deferriferrichrome by Ustilago sphaerogena.  Biochem.  Biophys. Res. Commun.
             60:1066-1071, 1974.
471.    Lepp,  H., Ed.  Geochemistry of Iron.  Stroudsburg,  Penn. .>   Dowden,  Hutchinson,
             and Ross,  1975.  464 pp.
472.     Lerer, T. J. ,  C. K.  Redmond,  P. P.  Breslin,  L.  Salvin, and H. W. Rush.
              Long-term mortality study of  steelx-jorkers.  VII.  Mortality patterns
              among crane operators.   J. Occup. Med.  16:608-614, 1974.
                                      320

-------
 473.  Lerner, A.  B.  On  the  etiology of vitiligo and gray hair.  Amer.  J. Med.   51:
             141-147,  1971.   (editorial)
 474.   Lewis, S.  M.  International Committee for Standardization in  Hematology:
              Proposed recommendations for measurement of serum  iron  in  human blood.
              Amer.  J. Clin. Path.  56:543-545, 1971.
 475.   light, P.  A., and R.  A.  Clegg.  Metabolism in iron-limited growth, pp.
              35-64.   In J.  B. Neilands, Ed.  Microbial Iron Metabolism.   New
              York:  Academic  Press, 1974.
 476.   Lintzel, W.,  J. Rechenberger, and E. Schairer.  Uber  den Eisenstoffwechsel
              des Neugeborenen und  des Sauglings.  Z. Gesamte  Exp. Med.   113:591-
       >      612,  1944.
 477. Lipschitz, D. A.', T. H. Bothwell,  H. C. Seftel, A. A. Wapnick,  and  R.  W. Charlton.
           The role of ascorbic  acid in the metabolism of storage iron.   Brit. J.
           Haematol.  20:155-163,  1971.
 478.     Lipschitz, D. A., J.  D. Cook, and C.  A.  Finch.  A  clinical  evaluation of
              serum ferritin as  an index of iron stores.  New Engl.  J.~ Med.  290:
              1213-1216,  1974.
 479.   Lisboa, P.  E.  Experimental hepatic cirrhosis in dogs  caused  by  chronic massive
             iron overload.  Gut   12:363-368, 1971.
 480.   Livingston, D." A."   Chemical Composition  of Rivers and  Lakes.   U.S. Geological
             Survey Professional Paper 440-C.  Washington,  D ,~C .":  U.'S." Government
             Printing  Office,  1963. 64  pp.
 481.     Llinas,  M. Metal-polypeptide interactions:  The conformational state of
              iron  proteins.   Struct.  Bond.   17:135-220, 1973.
482.   Llinas,  M., M. P.  Klein, and J.'  B.  Neilands.   Solution conformation of the
             ferrichromes.  III.~  A comparative  proton  magnetic resonance study of
   5         glycine-  and  serine-containing ferrichromes.   J.  Mol.  Biol.  68:265-284,
             1972.
                                    321

-------
 483.   Llinas,  M.,  D.  M.  Wilson,  and  J.  B.  Neilands.   Effect of metal binding on the




            conformation  of enterobactin.   A proton and carbon-13 nuclear magnetic




            resonance study.  Biochemistry  12:3836-3843, 1973.



 484.    Lloyd,  J. W. ,  and  A.  Ciocco.   Long-term mortality study of steelworkers.




             I.  Methodology.   J.  Occup. Med.  11:299-310,  1969.



 485.    Lloyd, J. W.,  F. E, Lundin, Jr., C. K. Redmond, and P.  B.  Geiser.  Long-




             term mortality  study  of steelworkers.  IV.   Mortality by work area.




             J. Occup. Med.  12:151-157,  1970.



 486.  Lochhead, A. G.  Two new species of arthrobacter requiring  respectively




           vitamin B,~ and the terregens factor.  Arch. Mikrobiol.  31:163-170,  1958.





 487.    Lock, L. F.   Iron  Deficiency  in  Plants:   How  to Control it in Yards  and




             Gardens.  Home  and Garden Bulletin No. 102.   Washington, D.  C. :   U.  S.




             Department of Agriculture,  1976.  8  pp.




 488.    Lovenberg,  W., Ed.   Iron-Sulfur  Proteins:   Vol.  1.   Biological  Properties.




             New York: Academic  Press,  1973.  385  pp.



 489.    Lowe, C.  R.,  H.  Campbell, and T. Kbosla.   Bronchitis in two integrated




           •  steel  works.   III.  Respiratory symptoms and ventilatory capacity




             related to  atmospheric  pollution.  Brit. J.  Ind. Med.  27:121-129,  1970.




 490.    Lucas, D. H.   Choosing  chimney heights in the presence  of buildings,  pp.




             47-52.  In Proceedings of the International  Clean  Air Conference,




             Melbourne University, May,  1972.  Parkville,  Vic.,  Australia:   Clean




             Air Society of Australia  and New  Zealand,  1972.



491.    Lucas,  R.  E.,  and  B.  D. Knezek.   Climatic and soil conditions promoting




             micronutrient deficiencies  in  plants,  pp.  265-288.  In J.  J. Mortvedt,




             P. M.  Giodano,  and W. L.  Lindsay, Eds.   Micronutrients in  Agriculture.




             Madison,  Wis.:   Soil Science  Society of  America,  1972.
                                    322

-------
 492. Luckey, M., R. Wayne, and J. B. Neilands.  In vitro competition between ferri-
           chrome and phage for the outer membrane T5 receptor complex of Escherichia
           coli.  Biochem.  Biophys. Res.  Commun.   64:687-693,  1975.
 493.  Lundgren,  D.  A.   Atmospheric  aerosol  composition  and  concentration as  a
            function of particle  size and  of time.   J. Air Pollut. Control Assoc.
            20:603-608,  1970.
 494.  Lundgren,  D.  A.   Determination of particulate  composition, concentration
            and  size distribution changes with  time.  Atmos. Environ.   5:645-651,
            1971.
 495.   Lundgren,  D.  G.  Microbiological  problems  in  strip mine  areas:   Relation-
            ship  to  the metabolism  of  Thiobacillus   ferrooxidans.  Ohio J. Sci.
            75:280-287,  1975.
 496.   Lundin,  P.  M.  The  carcinogenic action  of  complex iron preparations.   Brit.
            J.  Cancer  15:838-847,  1961.
 497.   Lundvall,  0.  The  effect of  phlebotomy  therapy  in  porphyria cutanea tarda:
            Its relation  to the phlebotomy-induced reduction of iron stores.   Acta
            Med.  Scand.   189-33-49,  1971.
 498.   Lundvall,  0.  The  effect of  replenishment  of  iron stores after phlebotomy
            therapy  in porphyria cutanea tarda.   Acta  Med.  Scand.   189:51-63, 1971.
 499.   Lundvall,  0., and  A. Weinfeld.  Iron stores in  alcohol abusers.   II.   As
            measured with the  desferrioxamine  test.  Acta Med.  Scand.   185:271-277,
            1969.
500.   Lundvall, 0., A. Weinfeld, and  p. Lundin.  Iron storage  in  porphyria cutanea
            tarda.   Acta Med.  Scand.   188:37-53,  1970.
501.  Lundvall, 0.,  A." Weinfeld, and ?~  Lundin.   Iron  stores in alcohol abusers.
           I.   Liver iron.  Acta Med.  Scand.   185:259-269,  1969.
                                   323

-------
  502.   Luongo, M.' A., and S.~ S.  Bjornson.   The liver  in ferrous sulfafce poisoning;   A
             report of three fatal cases ir.  children and an experimental study.   New
             Engl.  J. Med.  251:995-999, 1   4.
  503.  Lynch, S. R.,  I. Berelowitz,  H.  C.  Seftel,  G. B.  Miller,  P.  Krawitz,  R. W. .
            Charlton, and T. H.  Bothwell.  Osteoporosis  in  Johannesburg Bantu males;
            Its relationship to  siderosis and ascorbic acid deficiency.   Amer. J.
            Clin. Nutr.  20:799-807,  1967-
 504.  Lynch,  S. R.,  H.  C.  Seftel, J. D. Torrance, R.  w. Charlton, and T. H. Bothwell.
             Accelerated  oxidative catabolism of ascorbic acid in siderotic Bantu.
             Amer.  J.~ Clin.  Nutr.  20:641-647, 1967.
 505.   Lynch,  S.  R.",  H.  C.  Seftel, A.  A. Wapnick, R.  W. Charlton, and T. H. Bothwell.
              Some  aspects of calcium metabolism in normal  and osteoporotic Bantu sub-
              jects with  special reference to the  effects of iron overload and ascorbic
i              acid  depletion.  South Afr. J. Med.  Sci.   35:45-56, 1970.
 506.      MacDonald,  R. A.  Hemochromatosis  and Hemosiderosis.   Springfield,  Illinois:
               Charles C  Thomas, 1964.   374  pp.
507.    MacDonald, R. A., R. S. Jones,  and  G.  S.  Pechet.    Folic acid  deficiency and
              hemochromatosis.  Arch. Path.   80:153-160,  1965.
508.      MacDonald,  R. A.,  and G.  S. Pechet.   Experimental hemochromatosis  in rats.
               Amer.  J.~ Path. 46:85-109, 1965.
509.      Macdougall,  L.  G.   Red cell metabolism in  iron deficiency anemia.   III.  The
               relationship  between  glutathione peroxidase,  catalase, serum vitamin  E,
               and  susceptibility  of  iron deficient  red  cells to oxidative hemolysis.
               J. Pediatr.   80:775-782,  1972.
510.      Macon, W. L., and  W.  J.  Pories.  The  effect  of iron deficiency anemia  on
               wound  healing.  Surgery   69:792-796,  1971.
                                       324

-------
  511.  Magnusson, E. E. 0.  Iron absorption after antrectomy with gastroduodenostomy.
            Studies on the absorption from food and from iron salt using a double
            radioiron isotope technique and whole-body counting.  Scan4. J. Haematol.
            (Suppl. 26):1-111, 1976.
  512.  Mailliard,  J.  A.   Iron deficiency following gastric  resection.   Amer.  J.
            Gastroentero].    45:109-113,  1966.
  513.  Markson, J. L., and J. M.  Moore.   Autoimmunity in pernicious anaemia and
            iron-deficiency anaemia:  A complement-fixation test using human
            gastric mucosa.  Lancet  2:1240^1243,  1962.
 514.  Martell, A. E.  The chemistry of metal chelates in plant nutrition.   Soil Sci.
            84:13-26, 1957.
 515.  Martin, D.  F.   Coordination chemistry of the oceans.  Adv. Chem. Ser.  67:
            255-269,  1967.
 516. .  Martin, E.  J., and R.  D.  Hill.  Mine drainage research program of the
            Federal Water Pollution Control Administration, pp. 46-63.  In
            Second Symposium on Coal Mine Drainage Research, Mellon Institute,
            1968.  Monroeville, Penn.:  Bituminous Coal Research, Inc., / 1968_/.
 517.    Martinez-Torres,  C.,  and M. Layrisse.  Iron absorption from veal muscle.
             Amer. J. Clin. Nutr.  24:531-540, 1971.
 518.    Martinez-Torres,  C.,  and M. Layrisse.  Nutritional factors in iron
             deficiency:   Food iron absorption.   Clin. Haematol. 2:339-352,  1973.
 519.   Martinez-Torres, C., I." Leets, and M."  Layrisse.  Iron absorption by humans  from
            fish.  Arch. Latinoamer. Nutr.   25:199-210, 1975.
            s
 519a.  Martinez-Torres, C.,  I. Leets, M.  Renzi, and M. Layrisse.   Iron absorption
            by humans  from veal  liver.  J. Nutr.   104:983-933,  1974.
520.   Matrone, G., C. Conley, G."  H.~ Wise, and R." K.  Waugh.  A  study  of iron  and
            copper requirements  of dairy  calves.  J.~  Dairy Sci.  40:1437-1447,  1957.
521.   Matrone, G., E. L. Thomason, Jr., and C. R. Bunn.   Requirements and utilization
            of iron by the baby pig.  j. Nutr.  72:459-465, 1960.

                                     325

-------
  522.   McAllen,  P.  M.,  N.  F.  Coghill,  and M.  Lubran.   The treatment of hacmochromatcsis:
             With particular reference  to the  removal  of iron from the body by repeated
             venesection.   Q.  J.  Med.   26:251-276,  1957.
  523.  McCall, M. G., G. E. Newman, J.  R. P. O'Brien,  and L. J. Witts.  Studies in
            iron metabolism.   2.  The  effects of experimental iron deficiency in the
            growing rat.  Brit.  J. Nutr.   16:305-323,  1962.
  524.   McCance,  R.  A.,  and E. M. Widdowson.  Absorption and excretion of iron.
             Lancet  2:680-684,  1937.
 525.   McCormick, R. A.,  and G.  C. Holzworth.  Air pollution climatology, pp. 643-700.
             In A." C. Stern, Ed.   Air Pollution.  (3rd ed.)  Vol.  1.   New York:
             Academic Press, 1976.
 526.   McCoy,  B.  A.',  R. E.  Bleiler, and  M.  A. Ohlson.  Iron content of  intact
             placentas and  cords.   Amer.  J.~ Clin. Nutr.   9:613-615,  1961.
 527.   Mclntosh, M. A., and C.  F. Earhart.  Effect of iron on the relative
             abundance of two large polypeptides of the Eschefichia coli outer
             membrane.  Biochem.  Biopbys. Res. Comroun.  70:315-322, 1976.
 528.    McMullen, T.  B. , R. B. Faoro,  and  G.  B. Morgan.   Profile  of pollutant ....>.
              fractions  in nonurban suspended  particulate  matter.   J.  Air Pollut.
             Control  Assoc.   20:369-372,  1970.
 529.    Mena,  T., K.  Horiuchi, K.  Burke,  and  G. C. Cotzias.   Chronic manganese
             poisoning;  Individual susceptibility and absorption of iron.   Neurology
             19:1000-1006,  1969.
 530.    Mendel,  G.  A.  Studies  on iron absorption:   I.   The relationships between
              rate of erythropoiesis, hypoxia  and  iron absorption.  Blood  18:727-736,
              1961.
531.    Mendel, G. A., R. J. Weiler,  and  A.  Mangalik.   Studies on  iron absorption.
             II.   The absorption  of iron  in experimental  anemias of diverse etiology.
             Blood  22:450-458,  1973.
                                       326

-------
 532.  Meyer-Bertenrath, J. G., and W. Domschke.  I.  Mitt.:  Reversible Ablosung

            eisenhaltiger chromoproteide von Rattenleber-Ribosomen.  Z. Naturforsch.


            258:61-66, 1970.

 533.  Mikesell, M. E., G. M. Paulsen, R. Ellis, Jr., and A. J. Casady.   Iron utiliza-


           tion by efficient and  inefficient  sorghum lines.  Agron.  J.   65:77-80,  1973,


 534.  Miles, A. A., and P.  L. Khimji.   Enterobacterial chelators  of  iron:  Their


           occurrence, detection, and relation to  pathogenicity.   J. Med. Microbiol.


   	     8:477-490,  1975.


 535.  Miles, L. E. M., D. A. Lipschitz, C. P.  Bieber, and J. D. Cook.  Measurement


           of serum ferritin by a 2-site immunoradiometric assay.  Anal. Biochem.


    	61:209-224, 1974.

 536.  Miller, D.  F.  (National Research Council Committee on Feed Composition


           of the Agricultural Board). Composition of Cereal Grains and Forages.


           Publication 585.  Washington, D. C.:  National Academy  of Sciences,


           1958.   663 pp.

 537.  Miller, M.  W.   The  Pfizer Handbook of Microbial Metabolites.  New  York:


  	McGraw-Hill,  1961.  772  pp.

 538.  Minnich, V., A. Okc^uoglu, Y. Tarcon, A.  Arcasoy, S. Cin, 0.  Yorukoglu, F. Renda,
                        \j
           and B. Demirag.  Pica  in Turkey.   II."   Effect of clay  upon  iron absorption.


   	Amer.  J.~ Clin. Nutr.   21:78-86, 1968.

 539.  Mitchell, J. H., and  G.  Blomqvist.   Maximal  oxygen uptake.   New  Engl.  J.  Med.


            284:1018-1022,  1971.

 540.  Modell, C.  B.  Transfusional haemochromatosis, pp. 230-240.  In  H. Kief,


           Ed.  Iron Metabolism and Its Disorders.  New York:  American Elsevier


           Publishing Co. Inc., 1975.

 541.   Modell, C. B., and J. Beck.  Long-term desferrioxamine therapy  in thalassemia.


            Ann.  N.'Y. Acad. Sci.   232:201-210, 1974.


542.  Moersch, H. J.,  and H. M. Conner.  Hysterical dysphagia.  Arch.  Otolaryngol.


           4:112-117,  1926.



                                      327

-------
 543.   Monsen,  E.  R.,  and J.  D.  Cook.   Food iron absorption in human subjects.
             IV.   The  effects  of  calcium and phosphate on the absorption of non-
             heme iron.   Amer.  J.  Clin.  Nutr.   29:1142-1148,  1976.
 544.  Monsen, E. R., I. N. Kuhn, and C. A. Finch.  Iron status of menstruating women.
            Amer. J. Clin. Nutr.   20:842-849, 1967.
 545.  Moore,  C.  V.   Iron nutrition  and  requirements.   Ser.  Haematol.   6:1-14,  1965.
 546.  Moore,  C.  V., and R. Dubach.   Observations on the absorption of iron from foods
            tagged  with  radioiron.   Trans.  Assoc. Amer.  Phys.  64:245-256,  1951.
 547.  Moore,  D.  J.   A comparison of the trajectories  of rising buoyant plumes
            with  theoretical/empirical models.   Atmos. Environ.  8:441-457, 1974.
 548.  Morgan, E. H.  Transferrin  and transferrin iron,  pp.  29-71.   In A. Jacobs
            and M.  Worwood, Eds.   Iron in Biochemistry and Medicine.   New York:
            Academic Press, 1974.
 549.   Morgan, E. H.,  G. Marsaglia,  E. R. Giblett,  and C. A.  Finch.  A method of
            investigating internal iron  exchange utilizing two  types of trans-
            ferrin.  J.  Lab. Clin. Med.   69:370-381, 1967.
 550.   Morgan, 0. S.,  P. B. B.  Gatenby,  D.  G.  Weir, and J. M. Scott.   Studies on an
            iron-binding component in human gastric juice.   Lancet  1:861-863,  1969.
 551.   Morgan, W.  K. C., and H. D. Kerr.  Pathologic and physiologic  studies of
            welders' siderosis.   Ann. Intern. Med.  58:293-304,  1963.
 552.   Morgan, W. K. C., and A.  Seaton.   Hematite pneumoconiosis (silicosiderosis.
            or mixed dust fibrosis), pp.  233-240; hematite,  pp. 377-378.  In
            Occupational Lung  Diseases.   Philadelphia:  W.  B. Saunders Company, 1975.
552a.  Morris, E. R. , and  R. Ellis.   Isolation of monoferric phytate  from wheat
            bran and its biological value as an iron source to the rat.  J. Nutr.
            106:753-760, 1976.
553.   Mortvedt,  J. J.,  P. M.  Giordano,  and W.  L. Lindsay,  Eds.  Micronutrients
            in Agriculture.  Proceedings of a Symposium, 1971.   Madison, Wise.:
            Soil Science Society of America, Inc., 1972,  666 pp.
                                     328

-------
554.  Mros, B.> G. Bruschke, and D. Voigt.  Eisenmangel als attologischer Faktor der
          Ozaena?  Deutsch. Gesundh.  21:2216-2217, 1966.
555  Muir, D. C. F., Ed.  Clinical Aspects of Inhaled Particles.  London:
          William Heinemann Medical Books Limited, 1972.  192 pp.
556.  Muirhead,  H.,  and J.  Greer.   Three-dimensional Fourier  synthesis  of human
          deoxyhaemoglobin at  3.5 A resolution.   Nature  228:516-519,  1970.
557.  Murphy,  J. R., A. M.  Pappenheitner,  Jr.,  and  S. T.  De Borms.   Synthesis  of
          diphtheria tox-gene  products  in Escherichia coli extracts.    Proc.
          Nat.   Acad.  Sci.,  U.S.A.  71:11-15,  1974.
558.  Murphy, L. S., and L. M.  Walsh.  Correction  of micronutrient  deficiencies with
          fertilizers, pp. 347-387.   In J. J. Mortvedt, P. M. Giordano, and  W. L.
          Lindsay,  Eds. Micronutrients  in Agriculture.  Madison,  Wis.:  Soil
          Science  Society  of America, Inc.,  1972.
559,  Murphy, T.  P., D. R.  S. Lean, and C.  Nalewajko.   Blue-green algae:   Their
          excretion of iron-selective chelators enables them to dominate other
          algae.   Science    192:900-902,  1976.
560.  Murray,  M.  J., and N.  Stein.  Does  the  pancreas  influence  iron absorption?  A
          cirtical review  of information to  date.  Gastroenterology 51:694-700,  1966.
561   Murray, M. J., and N.  Stein.  The  effect of  achylia gastrica in  rats on
           the  absorption  of dietary iron.   Proc.  Soc.  Exp.  Biol.  Med.   133:
           183-184,  1970.
562.   Naiman, J. L., F." A. Oski,  L.' K. Diamond,'G." F. Vawter, and H. Schwachman.
           The  gastrointestinal effects of iron-deficiency anemia.  Pediatrics
           33:83-99, 1964.
562a. Nakatnura,  F.  I.,  and H. H.  Mitchell.  The utilization for hemoglobin regen-
           eration of the iron  salts used in the enrichment of flour and bread.
           J. Nutr.  25:39-48,  1943.
563.   Nason»  A., and W. D." McElroy.  Modes of action of the essential mineral
           elements, pp. 451-536.   In F." C.~ Steward,  Ed.  Plant Physiology:   A
          Treatise.  Vol.  3.  Inorganic Nutrition of Plants.  New York:  Academic
          Press, 1963.

-------
564.   Nath,  L.,  S.  K.  Sood,  and  N.  C.  \-sysk.   Experimental sidc-rosis and liver
            injury in the rhesus  monkey.   J. Path.   106:103-111,  1972.
565.   National Academy of Sciences.   National  Academy of Engineering.  Environ-
            mental Studies Board.   Water Quality Criteria 1972.   A Report of the
            Committee on Water Quality Criteria.  Washington,  D.  C.:   U.  S.
            Government  Printing Office,  1974.   594  pp.
566.  National Institute for Occupational Safety and  Health.  Method  P&CAM 173,
           pp. 173-1--173-8.   In N10SH Manual  of Analytical Methods.   Cincinnati:
           U. S.  Department of Health, Education, and Welfare, 1974.
567.  National Research Council.  Agricultural Board.  Composition of Concentrate
           By-Product  Feeding Stuffs.  Publ. 449.  Washington, D. C.:  National
           Academy   of Sciences, 1956.  126 pp.
568.  National Research Council, Agricultural Board and  Canadian Department of
           Agriculture.  Atlas of Nutritional Data on United  States  and  Canadian
           Feeds.  Washington, D. C.:  National Academy  of Sciences,  1971.  722 pp.
569.  National Research Council, Committee on Animal  Nutrition, Agricultural
           Board and Department of Agriculture, Canada,  Research Branch,
           Committee on Feed Composition.  United States-Canadian Tables of
           Feed Composition.   Nutritional Data for United States and  Canadian
           Feeds.  Second Revision.  Publication 1684.  Washington, D. C.:
           National Academy of Sciences, 1969.   92 pp.
570. Necheles, T." P.", S." Chung, R.~ Sabbah, and D." Whittert.  Intensive transfusion
          therapy in thalassemia major:  An eight-year follow-up.  Ann.  N. Y.
          Acad. Sci.   232:179-185, 1974.
571.  Neilands, J.  B.  Microbial iron transport compounds (siderophores), pp.  5-
           44.  In W. F.  Anderson and M.  C.  Hiller, Eds.   Proceedings of a Sym-
           posium.   Development of Iron Chelators for Clinical Use.   DHEW Publ.
           (NIH)  76-994.   Bethesda,  Md.:   U.  S.  Department of Health, Education,
           and Welfare,  National Institutes of Health,  1976.

                                     330

-------
  572.   Neilands,  J.  B.   Iron and its  role in microbial physiolgy,  pp.  3-34.   In
             J.  B.  Neilands,  Ed.   Microbial Iron Metabolism:   A Comprehensive
             Treatise.   New York:   Academic Press,  1974.   597  pp.
'  573.  Neilands, J. B.  Microbial iron transport compounds (siderochromes), pp.
            167-202.  In G. L. Eichhorn, Ed.  Inorganic Biochemistry.  Vol. 1.
            New York:  Elsevier Scientific Publishing Company, 1973.
  574.  Neilands, J.  B.   Naturally occurring  non-porphyrin  iron compounds.  Struct.
             Bond.  1:59-108,  1966.
  575.  Neilands, J. B.   Siderophores:  Biochemical ecology and mechanism of iron
            transport in enteric  bacteria.   Adv. Chem. Ser.  162:3-32,  1977.


  576.  Neilands, J. B.,  and C. Ratledge.  Microbial iron transport compounds (sidero-
            phores), pp.       In A. I. Laskin and H.  A.  Lechevalier, Eds.   CRC Hand-
            book of Microbiology.  Cleveland:  CRC Press,        (in press)


 577.  Nettesheim,  P., D. A.  Creasia,  and  T.  J.  Mitchell.   Carcinogenic and cocarcino-
            genic effects of  inhaled synthetic smog and ferric oxide particles.   J.
            Nat. Cancer  Inst.   55:159-169,  1975.
 578. Newcomb, E. H.  Fine structure of protein-storing plastids in bean root tips.
            J. Cell Biol.  33:143-163, 1967.
 579.  Newton, M.,  L. M. Mosey, G. E.  Egli, W. B. Gifford,  and C.  T. Hull.   Blood loss
            during and immediately after delivery.   Obstet. Gynecol.  17:9-18, 1961.
 580.  Nicholas, D. J. D.,  and A. Nason.  Mechanism of action of nitrate reductase
            from Neurospora.   J.  Biol. Chem.  211:183-197,  1954.
 581.  Nicholls, D.  Iron,  pp. 979-1051; cobalt, pp. 1053-1107; and nickel, pp.
            1109-1161.   In J. C.   Bailar, Jr., H. J. Emeleus,  R. Nyholm, and A.  F.
            Trotman-Dickenson, Eds.  Comprehensive Inorganic  Chemistry.  Vol. 3.
            New York:  Pergamon Press, 1973.
 582,   Nicholls, P.,  and  W. B.  Elliott.  The  cytochromes, pp.  221-277.   In A.  Jacobs
            and M.  Worwood, Eds.   Iron  in  Biochemistry  and Medicine.  New York;
            Academic Press, 1974,
                                     331

-------
583.   Nicholson, W. J.  Analysis of amphibole asbestiform fibers in municipal water
           supplies.  Environ. Health Perspect.   9:165-172, 1974.
584.   Nilsson, L., and So'lvell, L.   Clinical studies on oral contraceptives - &
           randomized doubleblind,  crossover study of 4 different preparations

           (Anovlar® mite, Lyndieremite,  Ovulen® and Volidan*}.  Acta Obstet.

           Gynecol. Scand.  46(Suppl., 8) : 1-31,  1967.
585.   Nissim, J. A.  Experimental siderosis:  A study of the distribution,

   '        delayed effects and metabolism of massive amounts of various iron

           preparations.  J. Path.  Bacteriol. 66:185-204, 1953.
586.   Noyes, W. D., F. Hosain, and C. A. Finch.   Incorporation of radioiron into

           marrow heme.  J. Lab. Clin. Med.  64:574-580, 1964.

 587.  Nussbaumer, T., H. C. Pattner, and A. Rywlin.  Hemochromatose juvenile chez
           trois soeurs et un frere avec consanguinite des parents.  Etude anatomo-
           clinique et genetique du syndrome endocrino-hepato-myocardique.  J.

           Genet. Hum.  1(2):53-82, 1952.
 588.   Oades,  J.  M.   The nature and distribution  of iron  compounds  in soils.
            Soils Fertiliz.  26:69-80,  1963.
 589.   Oborn, E. T.  A survey of pertinent biochemical literature,  pp. 111-190,
            Paper 1459-F.  In Chemistry of Iron  in Natural Water.  U. S.  Geological
            Survey Water-Supply Paper 1459.   Washington,  D.  C.:   U.  S. Government
            Printing Office, 1960.
 590.   Obom,  E.  T. ,  and J.  D.  Hem.  Microbiologic  factors  in the solution and
            transport  of iron,  pp.  213-235,  Paper  1459-H.   In Chemistry of Iron
            in Natural Water.   U. S.  Geological Survey Water-Supply Paper  1459.
           Washington,  D. C.:  U.  S. Government Printing Office, 1961.
 591.  O'Brien, R.~ T.  Ascorbic acid enhancement of desferrioxamine-induced urinary
           iron excretion in thalassemia  major.   Ann.  N7 YV" Acad. Sci.  232:221-225,
           1974.
                                     332

-------
 592.   Oertli,  J.  C.,  and L.  Jacobson.   Some quantitative considerations in iron
            nutrition  of higher plants.   Plant  Physiol.   35:683-688,  1960.
 593.   Oliver,  R.  A..M.   Siderosis  following transfusions of  blood.   J.  Path.
            Bacteriol.   77:171-194,  1959.
 5944   Olsen, C.  Iron absorption in different  plant species  as a function of
            the pH value of the solution.  C. R.  Trav.  Lab.  Carlsberg  31:41-
            59, 1958.
 595.   Olsen, S. R.  Micronutrient  interactions,  pp. 243-264.  In J.  J.  Mortvedt,
            P.  M.  Giordano, and W.  L.  Lindsay,  Eds.   Micronutrients  in Agriculture.
            Madison, Wis.:  Soil Science Society  of  America,  1972.
 596.   Olsen, S. R.", F.  S."  Watanabe, and C.  V.  Cole.  Effect  of sodium bicarbonate  on
            the  solubility of phosphorus  in  calcareous soils.  Soil Sci.  89:288-291,
            1960.
 597.   Olsson, K.  S.  Iron  stores in normal men and male blood donors:  As measured by
            desferrioxamine and quantitative  phlebotomy.  Acta Med. Scand.  192:
            401-407, 1972.
 598.   Orfei, E., F. I.  Volini, F. Madera-Orsini, 0. T. Minick, and G. Kent.  Effect
             of  iron loading on the hepatic  injury induced by ethionine.  Amer. J.
            Path.  52:547-567, 1968.
 599.   Pakiser,  L. C.,  and R.  Robinson.   Composition of  the continental  crust  as
            estimated  from seismic  observations,  pp. 620-626.  In J." S.~ Steinhart
            and  T. J.  Smith,  Eds.   The Earth Beneath the Continents.   A  Volume in
            Honor  of Merle A.  Tuve.   (A.  G.  U.  Monograph 10)   Washington, D.C.:
            American Geophysical Union,  1967.
600.   parer, J. T.", W."  Di  Jones, and J. Metcalfe.  A quantitative comparison  of
            oxygen transport in sheep and human subjects.  Respir. Physiol.  2:
            196-206, 1967.
601.   Park, C.  F., Jr.,  and R." A. MacDiarmid.  Ore  Deposits.  (3rd ed.)  San
            Francisco:   W.~ H.~ Freeman and Company, 1975.  529 pp.
                                      333

-------
 602.   Parker,  R.  L.   Composition of the Earth's  Crust.   U.S.  Geological  Survey Pro-
             fessional Paper 440-D.   Washington, D.CT:  U.Sf Government Printing Office,
             1967.   19 pp.
 603.   Parks,  G. A.  The isoelectric points  of solid oxides,  solid  hydroxides,  and
             aqueous hydroxo complex systems.   Chem. Rev.   65:177-198,  1965.
 604.   Paterson, D."  Rl"  A  clinical  type of dysphagia.   J." Laryngol.  34:289-291,  1919.
 605.   Payne, g. M., and R. A. Finkelstein.   Pathogenesis, and immunology of exper"-
             mental gonococcal infection:  Role of iron in virulence.  Infect.
             Immun.   12:1313-1318, 1975.
 605a.   peacock, W. C., R.  D. Evans, J. W. Irvine, Jr., W. M. Good, A. F. Kip,  S.
             Weiss, and J.  G. Gibson.  The use of two radioactive isotopes of iron
             in trace studies of erythrocytes.  J. Clin.  Invest.  25:504-615, 1946.
 606.    Pearson, W. N.,  and M. B. Reich.  Studies of ferritin and a new iron-binding
             protein  found  in the intestinal mucosa of the rat.  J. Nutr.  99:
             137-140, 1969.
 607.     Peirson, D.  H.  La teneur de 1'air en metaux lourds.  Intertecnic
              1973(March):192-194.
 608.     Peirson, T).  H., P. A. Cawse, and R. S. Cambray.  Chemical uniformity of
              airborne particulate matter, and a  maritime effect.  Nature   251:
              675-679, 1974.
 609.     Perkin-Elmer Corporation.  Analytical Methods  for Atomic Absorption Spectre-
              photometry.  Norwalk,  Conn.:  Perkin-Elmer  Corporation, 1971.
 610.    Perman,  G.  Hemochromatosis and red  wine.  Acta Med.  Scand.   182:281-284,  1967.
 611.    Perutz,  M."  F., H." Muirhead,  J.  M. Cox,  and L.  C.  G.  Goaman.   Three-
              dimensional Fourier  synthesis of  horse oxyhaemoglobin at 2.8 A
              resolution:  The  atomic model.   Nature  219:131-139,  1968.
612.    Peters, T.  J., and C. A. Seymour.  Acid hydrolase  activities  and  lysosomal
             integrity in liver biopsies from patients  with  iron  overload.  Clin.  Sci.
             Mol. Med. 50:75-78,  1976.
                                      334

-------
 613.  Peters,  W.  J.,  and  R." A. J. Warren.   Itoic acid  synthesis  in  Bacillus  subtilis.
             J.  Bacteriol.   95:360-366, 1968.
 614.  pettijohn,  F.~ J.  Chemical Composition of Sandstones, Excluding Carbonate and
           Volcanic Sands.  U.S. Geological Survey Professional Paper 440-S.
           Washington, D.C.~:  U."S.~ Government Printing Office, 1963.  19  pp.
 615.  pettijohn,  F.~ J.   Sedimentary Rocks.   (3rd ed.)  New York:  Harper  and Row, ^
            1975.   628 pp.
 616.  Pirzio-Biroli,  G., T. H. Bothwell, and C. A.  Finch.  Iron absorption.   II.  The
           absorption of radioiron administered with a standard meal in man.  J. Lab.
           Clin.  Med.  51:37-48, 1958.
616a.  Pirzio-Biroli,  G.,  and  C. A. Finch.   Iron absorption.   III.   The  influence of
            iron stores  on iron absorption  in the normal  subject.  J. Lab.  Clin. Med.
            55:216-220,  1960.
 617.  Plunkett, E.  R.  Handbook  of  Industrial  Toxicology.   New York:   Chemical
            Publishing Co., Inc.,  1976.   552 pp.
 618.  Poldervaart,  A.  Chemistry of the  earth's  crust, pp. 119-144.  In A. Poldervaart,
            Ed.  Crust of the  Earth   (A  symposium).   Geological Society of America
            Special Paper 62.   Baltimore:   Waverly Press, 1955.
 619€   Pollack, J.  R.,.B. N.  Ames,  and J.  B. Neilands.   Iron  transport  in Salmonella
             typhitnurium;   Mutants blocked  in the biosynthesis of enterobactin.
             J.  Bacteriol.   104:635-639,  1970.
                        '">                                 59
 620.   Pollack, S., S.  P.  Balcerzak,  and W. H. Crosby.   Fe   absorption in human sub-
             jects using total-body  counting technic.   Blood  28:94-97,  1966.
 621.   Pollack, S., J." N. George,  R." C.  Reba,  R." Kaufman,  and W. H. Crosby.  The
             absorption of nonferrous metals in iron deficiency.   J.~ Clin. Invest.
             44:1470-1473, 1965.
 622.   Pollack,  S.,  and F. D. Lasky.  A new iron-binding protein isolated from
             intestinal mucosa.   J." Lab.  Clin.  Med.   87:670-679,  1976.
 623.   Pollitt,  E.,  and R. L. Leibel.  Iron deficiency  and behavior.  J.  Pediatr.
             88:372-381, 1976.
                                       335

-------
 624.   Pollycove, M.   Iron  metabolism  and  kinetics.   Serain.  Hematol.   3:235-298,  1966.
 624a.    Pond, W. G., and E. F. Walker, Jr.  Cadmium-induced anemia in growing rats:

              Prevention by oral or parenteral iron.  Nutr. Rep. Int.   5:365-370, 1972.
 625.   Pootrakul, P., A. Christensen, B. Josephson, and C. A. Finch.   The role of

             transferrin in determining internal iron distribution.  Blood  49:

              957-966,  1977.
 626.    Port,  C. D.,  M.  C.  Henry,  D.  G.  Kaufman,  C.  C.  Harris,  and K.  V.  Ketels.

              Acute changes  in the  surface  morphology of hamster tracheobronchial

              epithelium following  benzo(a)pyrene  and ferric  oxide administration.

              Cancer Res.  33:2498-2506, 1973.
 627.     Powell, L. W.  Changing concepts in haemochromatosis.  Postgrad. Med. J.

              46:200-209, 1970.
 628.     Powell, L.  W.   Normal human iron storage  and its relation to  ethanol con-

              sumption.   Australas.  Ann.  Med.   15:110-115,  1966.
 629.     Powell, L. W., R. Mortimer, and 0. D. Harris.   Cirrhosis  of the  liver:   A

              comparative study of  the four major aetiological groups.  Med. J.

              Austral.  1:941-950,  1971.
 630.    Pozza, G., and  A.  Ghidoni.   Studies on  the diabetic syndrome of idiopathic

             haemochromatosis.  Diabetologia  4:83-86, 1968.
 631.    Price, C. A.  Iron compounds and plant  nutrition. ^Ann. Rev. Plant Physiol.   19:

             239-248, 1968.
632.    Price,  C. A., H. E.  Clark,  and E. A.  Funkhouser.   Functions of micronutri-

             ents in plants, pp. 231-239.  In  J.  J. Mortvedt, P.  M. Giordano, and

             W.  L. Lindsay, Eds.  Micronutrients in  Agriculture.   Proceedings of

             a  Symposium, 1971.  Madison, Wise.:  Soil  Science Society  of America,

             Inc.,  1972.
                                      336

-------
633.  Pritchard, J.   Significance of iron deficiencies.   In reproductive per-
           formance,  pp.  81-85.   In National Research Council,  Food and Nutri-
           tion Board.   Extent and Meanings of Iron Deficiency  in the U. S.
           Proceedings  of Workshop,  March 8-9, 1971.   Washington, D.  C.:  N
           National Academy of Sciences,  1971.
634. Pritchard, J. A., and R. A. Mason.  Iron stores of normal adults  and replenish-
          ment with oral iron therapy.  J.A.M.A.  190:897-901, 1964.
635. Prockop, D. J.   Role of iron  in the  synthesis of collagen  in connective tissue.
          Fed. Proc.  30:984-990, 1971.
636. Puro, D. G., and G. w. Richter.  Perritin synthesis by free and membrane-bound
           (poly)ribosomes of rat  liver.  Proc. Soc. Exp. Med. Biol.  138:399-403,
           1971.
637.  Ragan,  H. A., S.  Nacht, G. R. Lee,  C. R. Bishop, and G.  E.  Cartwright.  Effect
           of ceruloplasmin on plasma iron in copper-deficient  swine.  Amer. J.
           Physiol.  217:1320-1323, 1969.
638.  Ramsay, W. N. M., and E. A. Campbell.  Iron metabolism in the laying hen.
           Biochem. J.   58:313-317, 1954.
639.  Ramsay, C. A., I.  A. Magnus, A. Tumbull, and H. Baker.   The treatment of
           porphyria cutanea tarda by venesection.  Q. J. Med.   43:1-24, 1974.
64o.  Rankin, G. L. S., N. Veall, R. G. Huntsman, and J.  Liddell.  Measurement with
             Cr of red-cell loss in menorrhagia.   Lancet   1:567-569, 1962.
641.   Rath,  C. E., and C. A. Finch.  Sternal marrow hemosiderin:  A method for the
            determination of available Iron stores in man.  J.  Lab. Clin. Med.  33:
            81-86, 1948.
642.   Raymond,  K.  N.  Kinetically inert  complexes of  the  siderophores  in studies
           of microbial  iron  transport.  Adv. Chem. Ser.   162:33-54, 1977.
643.   Redmond,  C.  K., J.  Gustin,  and E.  Kamon.   Long-term mortality  experience
           of  steelworkers.   VIII.   Mortality patterns of  open hearth  steel-
           workers  (A preliminary report).  J. Occup. Med.  17:40-43, 1975.
                                    337

-------
  643a.  Reinhold, J. G. , F. Ismail-Beigi, and B. Faradji.  Fibre vs. phytate as
              determinant of the availability of calcium, zinc and iron of bread-
              stuffs.  Nutr. Rep. Int.  12:75-85, 1975.
  644.  Reissmann, K. R., and T. J.  Coleman.  Acute  intestinal iron  intoxication.
              II.  Metabolic, respiratory  and  circulatory  effects of  absorbed  iron
              salts.  Blood  10:46-51,  1955.
  645.  Reissmann, K. R., T. J. Coleman,  B. S. Budai,  and L. R. Moriarty.  Acute
              intestinal  iron intoxication.  I.   Iron absorption, serum iron and
              autopsy findings.  Blood   10:35-45, 1955.
  645a.  Reno, H. T.  Iron and steel, pp.  677-697.  In  U. S. Bureau of Mines.
              Minerals Yearbook 1974.   Vol. 1.  Metals, Minerals, and Fuels.
              Washington, D. C.:  U.  S.  Government Printing Office, 1976.
  646.    Rhoads, W.  A.,  and A.  Wallace.  Possible involvement of  dark fixation of C02 in
              lime-induced chlorosis.  Soil Sci.  89:248-256,  1960.
 647.   Richmond, v.  S., M.  Worwood, and A. Jacobs.   The iron content of intestinal
             epithelial  cells  and its subcellular distribution:   Studies on normal,
             iron-overloaded,  and iron-deficient  rats.   Brit.  J.  Haematol.   23:605-614,
             1972.
 648.    Rios,  E., D.  A.  Lipachitz,  J. D. Cook, and N. J. Smith.  Relationship  of
            maternal and infant  iron stores as assessed by determination of plasma
            ferritin.   Pediatrics   55:694,  1975.
 649.    Risdon,  R. A., M. Barry, and  D.  M. Flynn.  Transfusional iron overload:  The
            relationship between tissue iron  concentration and hepatic fibrosis in
            the thalassaemia.  J. Path.   116:83-95,  1975.
650.    Robbins, E., and T. Pederson.   Iron:   Its intracellular localization and
            possible role in cell division.   Proc. Nat. Acad. Sci. U.S.AT  66:
            1244-1251,  1970.
651.   Roberts,  R. J., S.  Nayfield,  R.  Soper,  and T.  H. Kent.  Acute  iron intoxication
            with intestinal  infarction  managed in part  by  small bowel resection.  Clin.
            Toxicol.  8:3-12,  1975.
                                        338

-------
 652.    Roche,  M.,  and M.  Layrisse.   The nature and causes of "hookworm anemia."  Amer.
             J. Trop.  Med. Hyg.   15:1029-1100,  1966.
 653.    Rogers, C.  H., and J.  W.  Shive.   Factors affecting the distribution of iron in
             plants.   Plant Physiol.   7:227-252, 1932.
 654.    Ronov, A. B., and A. A. Yaroshevsky.  Earth's crust geochemistry, pp. 243-254.
             In R. W. Fairbridge, Ed.  Encyclopedia of Geochemistry and Environmental
             Sciences.  New York:  Van Nostrand Reinhold Co., 1969.
 655.    Rosier, H. J., and H. Unge.  Geochemical Tables.  (Rev. ed.)  New York:
             Elsevier Publishing Company, 1972.  468 pp.   (Translated  from the  German
             by H. Liebscher.)
 g56     Ross,  F.  G. M. Pyloric  stenosis and fibrous  stricture of the stomach due to
             ferrous  sulfate poisoning.   Brit.  Med.  J.   2:1200-1202,  1953.
 657.    Russell,  C.  S., and W.  J.  Vaughan.   Steel Production:  Processes, Products, and
             Residuals.   Baltimore:   Johns  Hopkins University Press,  1976.   328 pp.
 658.    Rybo,  G.  Menstrual blood loss  in relation to parity  and menstrual pattern.
             Acta Obstet. Gynecol. Scand.  45(Suppl. 7):1-45, 1966.
 659.    Rybo, G.   Menstrual loss of iron, pp. 163-171.   In L. Hallberg, H.-G. Harwerth,
             and A.  Vannotti,  Eds.   Iron Deficiency:   Pathogenesis, Clinical Aspects,
             Therapy.   London;   Academic Press, 1970.
 659a.   Saccomanno, G., V.  E. Archer, R. P.  Saunders, I.  A.  James, and P. A.
             Beckler.  Lung cancer of uranium  miners on the  Colorado plateau.
             Health Phys.   10:1195-1201, 1964.
660.    Saffiotti, U., F. Cefis, and L.  H.  Kob.  A method for the experimental  induction
             of bronchogenic carcinoma.   Cancer Res.  28:104-113, 1968.
660a.   Saffiotti, U., R.  Montesano, A.  R.  Sellakumar, and D. G. Kaufman.   Respir-
             atory tract  carcinogenesis  induced in hamsters  by  different dose
             levels of benzo/a__/pyrene  and  ferric oxide.  J.  Nat. Cancer Inst.
             49:1199-1204,  1972.
                                       339

-------
  661.  Sagone, A. L., and S. P. Balcerzak.  Activity of iron-containing enzymes in
            erythrocytes and granulocytes in thalassemia and iron deficiency.  Amer.
            J. Med. Sci.  259:350-357, 1970.
 662.   Samoilova,  L. M.,  and V.  I. Kireev.  Morphological changes in the  lungs
             of  animals  exposed to the effect of aerosols developing in the course
             of welding  and  cast-iron surfacing.  Gig. Tr. Prof. Zabol. 19(4):
             53-54, 1975.   (in Russian)
 663.   Sandell,  E. B.   Manganese.  F.   Bio  materials,  pp. 437-440.  In Colorimetric
             Determination of Traces  of  Metals.   (2nd  ed.)   New York:   Interscience
             Publishers, Inc.,  1950.
 664.   Sander,  0.  A.  Further  observations  on  lung changes  in electric arc
             welders.  J.  Ind.  Hyg. Toxicol. 26:79-85,  1974.
 665.   Sanyal,  S.  K.,  W.  Johnson,  B.  Jayalakshmamma,  and A.  A. Green.   Fatal "iron-
             heart" in an adolescent:   Biochemical  and ultrastructural  aspects of
             the heart.   Pediatrics   55:336-341,  1975.
 666.    Sayers,  M.  H.,  S.  R. Lynch,  R. W. Charlton, T. H.  Bothwell,  R.  B.  Walker, and
             F.  Mayet.   Iron absorption from rice meals cooked with  fortified salt
             containing ferrous sulfate  and  ascorbic acid.   Brit.  J. Nutr.  31:
 667.    Sayers, M. H.,  S. R. Lynch, R. W. Charlton, T. H. Bothwell, R. B. Walker,  and
             F. Mayet.  The effects of ascorbic acid supplementation on the  absorption
             of iron in maize, wheat and  soya.  Brit. J. Haematol.  24:209-218,  1973.
668.    Schade,  S.  G., B.  F. Felsher, G. M.  Bernier, and M.  E. Conrad.   Inter-
             relationship of cobalt and iron absorption.  J. Lab.  Clin. Med.
             75:435-441, 1970.
669.    Schade, S. G., B. F. Felsher, B. E. Glader, and M. E. Conrad.  Effect of
             cobalt upon iron absorption.  Proc. Soc. Exp. Biol. Med.  134:741-743,
             1970.
                                      340

-------
 670.   Scheinberg,  I.  H.  The  genetics of hemochromatosis.  Arch.  Intern. Med.
            132:126-128,  1973.
 671.   Schiffer,  L. M., D. C. Price, and E. P. Cronkite.  Iron absorption and anemia,
            J.  Lab. Clin. Med.   65:316-321, 1965.
 671a.  Schroeder, H.  A.   The Trace  Elements and Man.   Some  Positive and Negative
            Aspects.   Old Greenwich, Conn.:  The  Devin-Adair  Company,  1973.   180 pp.
 672.   Schueneman,  J.  J., M. D.  High,  and W. E. Bye.   Air Pollution Aspects  of
            the Iron  and  Steel Industry.  Public  Health  Service  Publication  No.
            999-AP-l.   Cincinnati,  Ohio:  U. S. Department  of Health,  Education,
            and Welfare,  1963.   129 pp.
 673.   Schviler, P.   Iron  alloys, compounds, pp. 740-741. In  Encyclopaedia  of
            Occupational  Health and Safety.  Vol.  1.   Geneva:  International
            Labour Office,  1971.
 674.   Schumacher, H. R., Jr.   Hemochromatosis and arthritis.  Arth. Rheum.  7:
            41-50, 1964.
 675.   Schumacher, H. R., Jr.   Ultrastructural characteristics of the synovial
            membrane in idiopathic haemochromatosis.   Ann.  Rheum. Dis.  31:465-
            473, 1972.
676.   Schwartz, S. 0., and S. A. Blumenthal.   Exogenous hemochromatosis result-
            ing from blood  transfusions.   Blood  3:617-640, 1948.
677.   Schwertmann, U.  Die Bildung von Eisenoxidmineralen.  Fortschr. Mineral.
            46:274-285, 1969.
678.   Sebesta, W.  Ferrous metallurgical processes, pp.  143-169.   In  A. C.  Stem,
            Ed.  Air  Pollution (2nd ed.).  Vol. 3.  New  York:  Academic  Press,  1968.
679.   Seckbach,  J.  iron content and ferritin in leaves of iron treated Xanthium
            pensylvanicum plants.   Plant Physiol.   44:816-820, 1969.
680.   Seftel,  H. C.,  K.  J. Keeley, C.  Isaacson,  and T.  H.  Bothwell.   Siderosis  in
            the Bantu:  The clinical incidence of hemochromatosis  in diabetic sub-
            jects.  J.  Lab. Clin. Med.  58:837-844, 1961.

                                     341

-------
 681.  Seftel, H. C., C. Malkin, A. Schmaman, C. Abrahams, S. R. Lynch, R. W. Charlton,




            and T. H. Bothwell.  Osteoporosis, scurvy, and siderosis in Johannesburg




            Bantu.  Brit. Med. J.  1:642-646, 1966.



 682.  Sehmel, G. A., and F. D. Lloyd.  Resuspension of Plutonium at Rocky Flats.




            Report BNWL-SA-5085.  (CONF-740921-11)   Richland,  Wash.:   Battelle




            Pacific Northwest Laboratories, 1974.  39 pp.



 683.   Seip,  M.,  and S.  Halvorsen.  Erythrocyte production  and  iron  stores  in premature




             infants  during  the first  months  of  life:  The anemia  of  prematurity  --




             etiology, pathogenesis, iron  requirement.  Acta Paediatr.  Scand.   45:




             600-617,  1956.




 683a.    Sellakumar,  A.,  and  P.  Shubik.  Carcinogenicity of  different polycyclic




              hydrocarbons in the respiratory  tract  of hamsters.   J.  Nat.  Cancer




              Inst.   53:1713-1719,  1974.
684.    Seshadri, R., J. H. Colebatch, P. Gordon, and H. Ekert.  Long-term administra-




             tion of desferrioxamine in thalassaemia major.  Arch. Dis. Child.  49:




             621-626, 1974.



685.    Shacklette,  H.  T.,  J.  C.  Hamilton,  J.  G.  Boerngen,  and  J.  M.  Bowles.   Elemental



             Composition of Surficial  Materials  in the  Conterminous United  States.   U.S.




             Geological Survey Professional Paper 574-D.  Washington,  D.C.:   U.S.  Govern-




             ment Printing  Office,  1971.   71 pp.



686.    Shacklette, H. T., H. I. Sauer, and A. T. Miesch.  Geochemical Environments and




             Cardiovascular Mortality Rates in Georgia.  'U.S. Geological Survey Pro-




             fessional Paper 574-C.  Washington, D.C.:   U.S. Government Printing Office,




             1970.  39 pp.




687.   Shah, B.,  and B. Belonje.  Liver storage  iron in Canadians.  Amer. J. Clin.



            Nutr.  29:66-69, 1976.
                                       342

-------
 688.  Shahid,  M.  J.,  and N. A.  Haydar.   Absorption of inorganic iron in thalassaemia.
            Brit.  J.  Haematol.   13:713-718,  1967.
688a.  Sharpe,  L.  M.,  W.  C.  Peacock,  R.  Cooke,  and R.  S.  Harris.   The effect of
            phytate and other food factors on iron absorption.  J.  Nutr.   41:433-
            446,  1950.
 689.  Sheldon, J.  H.   Haemochromatosis.  London:   Oxford University Press,  1935.
            382 pp.
 690.  Shutn, Y. S. , and W.  D. Loveland.  Atmospheric trace element  concentrations
            associated with agricultural field burning in the Willamette Valley of
            Oregon.  Atmos. Environ.  8:645-655,  1974.
690a.   Shupe, J.  L.   Effects of kish on animals,  p.  2.   In  0.  S.  Cannon,
            J. L.  Shupe,  and R.  E.  Lamborn.  The  Problems of Airborne Dust  and
            "Kish" in Utah  County.   Agricultural  Experiment  Station Research
            Report 2.   Logan:   Utah State University,  1972.
 691.   Siimes, M. A., J. E. Addiego, Jr.,  and P. R. Dallman.  Ferritin in  serum:
             Diagnosis of iron deficiency and iron overload  in infants and  children.
             Blood  43:581-590, 1974.
 692.   Simon,  M., P. Franchimont, N. Murie, B. Ferrand, H.  van Cauwenberge, and M.
             Bourel.  Study of somatotropic and gonadotropic pituitary function in
             idiopathic haemochromatosis (31 cases).  Europ. J. Clin. Invest.   2:
      	384-389,  1972.
 693.   Simpson,  F. B., and J.  B. Neilands.   Siderochromes in Cyanophyceae:   Isolation
             and  characteristics of schizokinen from Anabaena sp.    J. Phycol.  12:
     	   44-48, 1976.
 694.    Sinniah,  R.   Transfusional siderosis and  liver cirrhosis.  J. Clin. Path.
             22:567-572,  1969.
 695.    Six, K. M., and  R.  A. Goyer.  The influence of  iron  deficiency  on tissue content
             and  toxicity of ingested lead  in  the rat.   J. Lab. Clin. Med.   79:128-136,
             1972.
                                        343

-------
 696.   Skinner, C., and C. F. Kenmure.  Haemochromatosis presenting as congestive




            cariomyopathy and responding to venesection.  Brit. Heart J.  35:




            466-468, 1973.



 697.    Slade, D. H., Ed.  Meteorology and Atomic Energy 1968.  TID-24190.  Oak




             Ridge, Tenn.:  U. S. Atomic Energy Commission, Division of Technical




             Information, 1968.  445 pp.



 698.   Smith, D.  W.  Steel  industry wastes.  J.  Water Pollut.  Control Fed.   48:




            1287-1293,  1976.



 699.   Smith,  F.  C.,  Jr.,  J.  M.  Mathieson,  0. Masi,  and  T.  M.  Spittler.   Simultaneous



            analysis  of  dissolved metal  pollutants  in water.   Amer.  Lab.   7(12):41-47,




            1975.



 700.   Smith,  M.  D., and B.  Mallett.  Iron absorption before and after partial




            gastrectomy.  Clin. Sci.  16:23-34, 1957.



 701.   Smith,  N.  J.,  S.  Rosello, M.  B.  Say, and K.  Yeya.  Iron stores in the first




            five  years  of life.   Pediatrics  16:166-173, 1955.



 702.   Smith,  P.  F., and A.  W. Specht.  Heavy-metal nutrition in relation to iron



            chlorosis of citrus seedlings.  Proc. Fla.   State Hort. Soc.  65:101-108,




            1952.



 702a.    Spedding, D. J.   The interaction of gaseous pollutants with materials at




             the surface of the earth, pp. 213-241.  In J. O'M. Bockris, Ed.



             Environmental Chemistry.   New York:  Plenum Press, 1977.



 703.   Spiro,  Th. G., and P. Saltman.   Polynuclear complexes of iron and their bio-



            logical implications.  Struct. Bond.  6:116-156, 1969.



704.    Sprague,  G. F.   Germ plasm  manipulation in the future, pp.  375-395.  In




             J. D. Eastin, F. A. Haskins, C. Y. Sullivan, and C. H.  M.  Van Bavel,




             Eds.  Physiological Aspects of Crop Yield.   Madison, Wise.:  American




             Society of Agronomy and Crop Science Society of America, 1969.
                                     344

-------
 705.   Stamper, J. W., and H. F. Kurtz.  Aluminum.  Preprint  from Bureau of Mines




           Bulletin  667.  A Chapter  from Mineral Facts and Problems,  1975 edition.




           Washington, D.C.:  U.S. Government Printing Office,  1975.   30 pp.



"705a.   Stein, M.. D. Blayney, T. Feit, T.  G. Georgen, S. Micik,  and W.  L.  Nyhan.




            Acute iron poisoning in  children.  West. J. Med.   125:289-297,  1976.




 705b.   Stenback,  F., A.  Ferrero, R.  Montesano,  and p. Shubik.   Synergistic effect




            of  ferric oxide on  dimethylnitrosamine carcinogenesis in  the Syrian




            golden hamster.  Z. Krebsforsch.   79:31-38,  1973.




 706.    Steiner,  B. A.   Ferrous  metallurgical operations, pp. 890-929.  In A.  C.




            Stern, Ed.   Air Pollution.   (3rd ed.)  Vol.  4.   New York:  Academic



            Press, 1977.




 707.    Stevens, A." Rl',  Jr.,  D.  M.  Coleman, and C.  A." Finch.   Iron metabolism:   Clinical




             evaluation of iron stores.   Ann.  Intern. Med.   38:199-205, 1953.




 yog.    Stewart, G.  R.   The regulation of nitrite reductase  level in  Lemna  minor L.




             J.  Exp.  Bot.  23:171-183, 1972.




 709.    Stewart, M.  J.,  and J.  S.  Faulds.  The pulmonary fibrosis of  haematite




             miners.   J. Path.  Bacteriol. 39:233-253, 1934.




 710.    Stewart, W.  B., P.~ S. Vassar, and R.~ S. Stone.  Iron absorption in  dogs during




             anemia  due to acetylphenylhydrazine.  J. Clin.  Invest.   32:1225-1228,  1953.




 710a.    Stiles, I. W.,  J.  M.  Rivers, I.  R. Hackler,  and D. R.  Van Campen.  Altered



              mineral absorption in the rat due to dietary fiber.  Fed. Proc.   35:




              744, 1976.   (abstract)



 711.    Stocks,  A. E.,  and F.  I. R. Martin.  Pituitary function in haemochromatosis.




             Amer. J. Med.  45:839-845, 1968.
                                        345

-------
712.    Stocks, A. E., and L.  W.  Powell.   Carbohydrate intolerance in idiopathic
            haemochromatosis  and cirrhosis of the liver.    Q.  J.  Med.  42:733-749,
            1973.
713.    Stone,  W.  D.   Gastric  secretory response  to  iron therapy.   Gut   9:99-105,  1968.
714^    Strauss, M. B.  Anemia of infancy from maternal iron deficiency in pregnancy.
            J. Clin. Invest.   12:345-353, 1933.
715.   Strom, G. H.   Transport  and diffusion of stack  effluents, pp. 401-501.   In A.  C.
           Stern,  Ed.  Air Pollution.   (3rd ed.) Vol. 1.   New York:   Academic Press,
           1976.
716.    Stumm, W., and G. F. Lee.  Oxygenation of ferrous iron.   Ind. Eng. Chem.
            53:143-146,  1961.
717.   Sturgeon,  P.   Studies of  iron requirements of  infants.  III.  Influence of
            supplemental  iron  during normal pregnancy on mother and infant.   B.  The
            infant.   Brit. J.  Haematol.   5:45-55, 1959.
718.    Sturgeon,  P., and A. Shoden.  Total liver storage iron in normal populations
            of the  USA.  Amer.  J.~Clin.  Nutr.   24:469-474, 1971.
719.   Subtramanian,  K. N., G.  Padmanaban,  and P. S. Sarma.  Folin-Ciocalteu reagent
            for the  estimation of siderochromes.  Anal. Biochem.  12:106-112,  1965.
720.   Sulzer, J. L.  Significance of iron deficiencies.   Effects of iron
            deficiency on psychological  tests in children,  pp.  70-76.   In
            National Research Council,  Food and  Nutrition Board.   Extent and
            Meanings of Iron Deficiency in the U.  S.   Summary of Proceedings
            of Workshop, March 8-9,  1971.  Washington, D.  C.:   National
            Academy of Sciences, 1971.
 721.    Surgenor,  D.  M.,  B. A. Koechlin,  and L.  E.  Strong.  Chemical, clinical, and
            immunological studies on the products  of human plasma fractionation:
            XXXVII.  The metal-combining globulin  of human plasma.  J. Clin.  Invest.
            28:73-96, 1949.
                                      346

-------
722.  Sussman, M.  Iron and infection,  pp.  649-679.   In A.  Jacobs and M.
           Worwood, Eds.  Iron in Biochemistry and Medicine.   New York:
           Academic Press, 1974.
723.  Suzman, M. M.  Syndrome of  anemia,  glossitis and dysphagia:  Report of eight
           cases, with special reference  to observations at autopsy in one instance.
           Arch. Intern. Med.  51:1-21, 1933.
723a.  Sverdrup,  H. U.,  M. W.  Johnson,  and  R.  H. Fleming.  The Oceans.  Their
           Physics,  Chemistry, and General Biology.   New York:   Prentice-Hall,
           Inc.,  1942.   1087  pp.
724.  Swarup, S., S. K. Ghosh, and J. B.  Chatterjea.  Aconitase activity in iron
           deficiency.  Acta Haematol.   37:53-61,  1967.
725.   Symes,  A.  L.,  K.  Missala,  and T. L.  Sourkes.  Iron- and riboflavin-
           dependent metabolism  of a monoamine in the rat in vivo.   Science
           174:153-155, 1971.
726.   Symes,  A. L.,  T.  L. Sourkes, M. B. H. Youdim, G. Gregoriadis,  and H. Birnbaum.
           Decreased monoamine oxidase activity in liver of iron-deficient rats.
           Can. J. Biochem.  47:999-1002,  1969.
727.   Szold,  p.  D.   Plumbism and iron  deficiency.   New Engl. J.  Med.   290:520,  1974.
            (letter to Editor)
728.   Tait,  G. H.  The identification  and  biosynthesis of siderochromes formed by
            Micrococcus denitrificans.   Biochem. J.   146:191-204, 1975.
729.   Taljaard,  J. J. P., B. C.  Shanley, W. M. Deppe,  and S. M.  Joubert.  Porphyrin
           metabolism in  experimental hepatic siderosis in the rat.   II.  Combined
           effect  of iron overload and hexachlorobenzene.  Brit. J.  Haematol.
           23:513-519,  1972.
730.  Tan, A. T., and R. C. Woodworth.   Ultraviolet difference spectral studies of
           conalbumin complexes with transition metal ions.  Biochemistry  8:
           3711-3716, 1969.
                                     347

-------
 731.   Taras,  M.  J.,  A.  E.  Greenberg,  R.  D.  Hoak,  and  M.  C.  Rand,  Eds.   Standard
            Methods  for  the Examination of Water and Wastewater.   (13th  ed.)
            Washington,  D.  C.:   American Public  Health Association,  1971.   874  pp.
 732.   Teculescu, D., and A. Albu.   Pulmonary function in workers  inhaling iron
            oxide dust.   Int.  Arch.  Arbeitsmed.   31:163-170, 1973.
 733.   Tentative  method  of  analysis  for iron content of atmospheric  particulate
            matter  (bathophenanthroline method)  12126-01-70T, pp.  299-301;  and
            Tentative method of  analysis  for iron  content  of atmospheric particu-
            late  matter  (volumetric  method)  12126-02-70T,  pp. 302-304.   In
            Methods of Air  Sampling  and Analysis.  Washington, D.  C.:  American
            Public Health Association,  1972.
 734.  Theis, T. L., and  P.  C.  Singer.   Complexation of iron(II) by organic  matter  and
           its effect on iron(II) oxygenation.   Environ.  Sci. Technol.   8:569-573,
           1974.
 735.   Thompson,  R. J.,  G.  B.  Morgan,  and L. J.  Purdue.  Analysis  of selected
            elements  in  atmospheric  particulate  matter by atomic absorption.
            Atom. Absorpt.  Newslett.  9:53-57,  1970.
 736.   Thomson, A. B. R., D. Olatunbosun, and L. S.  Valberg.  Interrelation of
            intestinal transport system for manganese and iron.   J.  Lab. Clin.
            Med.   78:642-655, 1971.
 737.   Thomson, A. B. R., C. Shaver, D. J. Lee,  B. L.  Jones,  and L.  S. Valberg.
            Effect of varying  iron stores  on site  of intestinal absorption  of cobalt
            and iron.  Amer. J.  Physiol.   220:674-678,  1970.
 738.   Thomson, A. B.' R., and  L. S.  Valberg.  Intestinal  uptake of iron, cobalt, and
            manganese in the iron-deficient  rat.   Amer. J. Physiol.   223:1327-1329,
            1972.
 739<  Thomson, A. B.  R., and L.  S.  Valberg.   Kinetics  o£ intestinal  iron absorption
           in the rat:   Effect of  cobalt.  Amer. J.  Physiol.  220:1080-1085, 1970.
740.   Thomson, A. B. R., L. S.  Valberg,  and D.  G. Sinclair.  Competitive nature of
            the intestinal  transport mechanism  for cobalt  and iron in the rat.  J.
            Clin.  Invest.   50:2384-2394,  1971.
                                    348

-------
741.   Thome,  D.  W. ,  and H.  B.  Peterson.   Irrigated Soils.   Their Fertility and
            Management.   (2nd ed.)   New York:   The Blakiston Company, Inc., 1954.
            392 pp.
742.  Thome, D. W., F." B. Wann, and W. Robinson.  Hypotheses concerning lime-induced
           chlorosis.  Soil Sci. Soc. Aroer. Proc.  15:254-258, 1950.
743.   Tiffin, L.  0.   Iron  translocation.   I.  Plant  culture,  exudate sampling,  iron-
            citrate analysis.  Plant  Physiol.  41:510-514,  1966.
744.    Tiffin,  L. 0.   Iron translocation.   II.   Citrate/iron ratios  in plant  stem
             exudates.  Plant Physiol.   41:515-518,  1966.
745.   Tiffin, L.~  0.  Translocation of  iron citrate  and phosphorus  in xylem exudate
            of soybean.   Plant Physiol.  45:280-283,  1970.
746.   Tiffin,  L.  0.   Translocation of micronutrients in plants,  pp.  199-229.
            In J.  J.  Mortvedt, P.  M.  Giordano, and W. L.  Lindsay, Eds.  Micro-
            nutrients in Agriculture.  Proceedings of a Symposium, 1971.
            Madison,  Wise.:  Soil Science Society of America, Inc.,  1972.
747.  Tiffin, L. 0.,  and  J." C." Brown.  Absorption of iron from iron chelate by sun-
           flower  roots.   Science   130:274-275,  1959.
.748.   Tisdale, S."L., and W." L.~ Nelson.   Soil Fertility and Fertilizers.   (3rd  ed.)
            New York;  Macmillan Publishing Company, Inc.,  1975.   694 pp.
749.  Torrance,  J." D.", R.~ W."  Charlton,  A.  Schmaman,  S.  R. Lynch,  and T. H." Bothwell.
           Storage iron in "muscle."   J~ Clin. Path.  21:495-500, 1968.
750.   Toth,  S.  J. The physical chemistry  of  soils,  pp.  142-162.   In F.  E. Bear, Ed.
            Chemistry  of  the  Soil.  (2nd ed.)   (American  Chemical Society Monograph
            Series No.  160)   New York:   Reinhold  Publishing  Corp.,  1964.
751.  Trump, B. P., J. M. Valigorsky, A. U. Arstila, W. J. Mergner, and T. D. Kinnoy.
           The relationship of intracellular pathways of iron metabolism to cellular
           iron overload  and  the iron storage diseases.  Amer.  J. Path.  72:295-336,
           1973.
                                      349

-------
 752.   Turnbull,  A.   Iron  absorption, pp.  369-403.   In  A.  Jacobs  and M.  Worvood,
             Eds.   Iron  in  Biochemistry  and Medicine.  New  York:   Academic  Press,
             1974.
 753.  Turnbull, A., H.  Baker, B. Vernon-Roberts, and I. A. Magnus.   Iron metabolism
           in porphyria cutanea tarda and  in erythropoietic protoporphyria.    Q.
           J? Med.  42:341-355, 1973.
 754.   Turner, D.  B.  Workbook of Atmospheric Dispersion Estimates.   Office  of
             Air Programs Publ. AP-26.   Research Triangle Park, N.  C.:  U.  S.
             Environmental  Protection Agency, 1970.   88  pp.
 755. Twyman, E."  S.  The effect of iron  supply on the yield and  composition of leaves
           of tomato plants.   Plant  Soil  10:375-388,  1959.
 756. Ullrey, D.~ E.", E." R.~  Miller, 0." A. Thompson,  I." M." Ackertnann, D." A." Schmidt,
           J.~ A.~ Hoefer, and  R.~ W.~ Luecke.  The  requirement of the baby pig for
           orally administered iron.  J.~ Nutr.   70:189-192, 1960.
 757.  Underwood,  E. J.  The  Mineral Nutrition of Livestock.   Aberdeen:   The
           Central Press,  Ltd.,  1966.   237 pp.
 758.  Underwood,  E. J.  Trace Elements  in  Human and  Animal Nutrition.  (3rd  ed.)
           New  York:   Academic press, 1971.  543 pp.
 759.  United Kingdom Ministry of Housing and Local Government.  U. K. Clean Air
           Act 1956.  Memorandum on Chimney Heights.  (2nd ed.)  London:   H.  M.
           Stationery Office, 1967.
760.  U.  S.  Department  of  Health, Education  and Welfare.   Public  Health Service.
           Preliminary Findings  of  the  First Health  and Nutrition Examination
           Survey,  United  States, 1971-1972.  Dietary Intake  and  Biochemical
           Findings.   DHEW Publ.  (HRA)74-1219-1.  Rockville,  Md.:   U.  S.  Depart-
           ment  of Health, Education, and  Welfare, 1974.   183 pp.
                                     350

-------
761.  U. S. Department of Health, Education, and Welfare.  Ten-State  Nutrition




           Survey  1968-1970.  IV.  Biochemical.  DHEW Public.  No.  (HSM)  72-8132.




           Atlanta,  Georgia:  U. S. Department of Health,  Education,  and Welfare,




           Center  for Disease Control, j_  1972_/.  296 pp.




762.  u-  S. Department  of  the  Interior.   Federal Water Pollution Control Admin-




            istration.   The  Cost of Clean  Water.  Vol.  3.   Industrial  Waste




            Profiles No.  1  - Blast  Furnace and  Steel Mills.  FWPCA Publ.  I.W.P.-l.




           Washington,  D.  C.:   U.  S.  Government Printing Office, 1967.  100 pp.




763.  u-  s- Environmental Protection  Agency.   Air Pollution  Aspects of Emission




            Sources:   Iron  and  Steel Mills—A Bibliography with Abstracts.   Publ.




            AP-107.  Research Triangle Park,  N. C.:   U.  S.  Environmental  Protec-




            tion Agency,  1972.   84 pp.




764.   u<  s>  Environmental  Protection  Agency.  Compilation of Air Pollutant




            Emission Factors.   (2nd  ed.)   Publ. AP-42.   Research Triangle Park,




            N.  C.:   U.  S. Environmental Protection Agency, 1973.  _/~292 pp._7



765.   U<  S'  Envil*>nmetttal  Protection Agency.  Methods Development and Quality




            Assurance Research Laboratory.  Iron  (STORET NO.  Total 01045), pp.




            110-111.  In Methods for Chemical Analysis of Water and Wastes.  EPA-




            625-/6-74-003.   Washington, D. C.:   Office of Technology Transfer,




            U.  S. Environmental Protection Agency,  1974.



765a  ^'  **• Environmental Protection  Agency.   National  Atmospheric Data Bank.




            Data File on  Iron.   Research Triangle Park,  N.  C.:   Monitoring and




            Data Analysis Division, Office of Air Quality Planning and Standards,




           U.  S. Environmental  Protection Agency.   _/ Data are  continuously




           accumulated,, available  from EPA._/
                                     351

-------
766.  U. S. Environmental Protection Agency.  Office of Technology Transfer.




           Methods for Chemical Analysis of Water and Wastes.  EPA 625/6-74-003.




           Washington, D. C.:   U. S. Environmental Protection Agency, 1974.  298 pp.



767   U. S. Environmental Protection Agency.  Progress in the Prevention and




           Control of Air Pollution in  1975.  Annual Report  of the Administrator




           of the Environmental  Protection Agency to the Congress of the United




           States in Compliance with Section 313 of Public Law 91-604, The




           Clean Air Act, as Amended.   Senate Document No. 94-228.  Washington,




           D. C.:  U. S. Government Printing Office, 1976.   171 pp.



768   Valberg, L. S., J. Ludwig, and D. Olatunbosun.  Alterations in cobalt



           absorption in patients with disorders of iron metabolism.   Gastro-




           enterology  56:241-251, 1969.




769   Valberg, L." S., J." Sorbie, W. E.  N. Corbett, and J. Ludwig.  Cobalt  test  for




           the detection of iron deficiency anemia.  Ann. Intern. Med.  7,7:181-187,




           1972.



       Van Dyke, D. C., H. 0.  Anger, and Y. Yano.  Progress  in determining bone




            marrow distribution in vivo.  Prog. Atom. Med.   2:65-84,  1968.
771.   Varian Techtron.   Iron  section.   In Analytical Methods  for Flame Spectros-




            copy.  Palo Alto,  Calif.:  Varian Techtron,  1972.   18 pp.




772.    Vilter, R. W.  Ascorbic acid.  IX.  Effects of deficiency in human beings,




             pp. 348-380.   In W." H." Sebrell,  Jr., and R. S.~ Harris, Edsw  The




             Vitamins:  Chemistry, Physiology,  Pathology.  (Vol. 1)  New York:



             Academic Press, 1954.




773.   Vinogradov, A. P.   The  Geochemistry of Rare and  Dispersed Chemical Elements in




            Soils.   (2nd  ed.)   New York:  Consultants Bureau,  Inc.,  1959.   209 pp.




774.   Vinson, P. P.  Hysterical  dysphagia.  Minn. Med.  5:107-108,  1922.
                                      352

-------
775.   Viteri, F. E., and B. Torun.  Anaemia and physical work capacity.  Clin.

           Haematol. 3:609-626, 1974.

776   Waddell, J.,  H.  F.  Sassoon,  K.  D.  Fisher,  and C.  J. Carr.  A Review of

           the Significance of Dietary Iron on Iron Storage Phenomena.  Bethesda,

           Md.:   Federation of American Societies for Experimental Biology, Life

           Sciences Research Office,  1972.  81 pp.

777.  Wagman,  D. D., W. H.  Evans,  V.  B.  Parker,  I.  Halow,  S.  M.  Bailey,  and R.

           H.  Schumm.   Selected Values of  Chemical Thermodynamic Properties.

           Tables  for Elements  35  Through  53  in  the Standard  Order of Arrangement.

           National  Bureau  of  Standards  Technical  Note  270-4.   Washington,  D.  C.:

           U.  S. Government Printing  Office,  1969.   141  pp.
77g   Waid, J. S.  Hydroxamic acids in  soil systems, pp.  65-101.   in  E.  A.  Paul  and


           A. D. McLaren, Eds.  Soil Biochemistry.  Vol.  4.  New York:   Marcel

           Dekker, 1975.

779^  Walker,  A.  R. P.,  and  U.  B.  Arvidsson.   Iron "overload" in the South African


           Bantu.  Trans.  Roy.  Soc.  Trop.  Med.  Hyg.  47:536-548, 1953.
      •
7g0<  Walker,  R. J.,  I.  W. Dymock, I. D. Ansell,  E.  B. D.  Hamilton,  and R. Williams.

           Synovial biopsy in haemochromatosis  arthropathy:   Histological findings

           and iron deposition in  relation to total  body  iron overload.   Ann.  Rheum.

           Dis.  31:98-102,  1972.

781.  Walker, R. J., and R. Williams.  Haemochromatosis and  iron overload,  pp.

           589-612.   In A. Jacobs and M. Worwood, Eds.  Iron in Biochemistry  and

           Medicine.  New York:  Academic Press,  1974.
 782.  Wallace,  A.,  and  0.  R.  Lunt.   Iron chlorosis  in horticultural plants,  a review.

           Amer.  Soc. Hort.  Sci.   75:819-841,  1960.
                                    353

-------
783.  Wallace, A., R. T. Mueller, 0. R. Lunt, R. T. Ashcroft, and L. M. Shannon.




           Comparisons of five chelating agents in soils, in nutrient solutions, and




           in plant responses.  Soil Sci.   80:101-108, 1955.



784.  Wallerstein, R. 0., and P. M. Aggeler.  Diagnosis of anemia in childhood.




           J. Chron. Dis.  12:377-390, 1960.



785>  Walters, G. 0., F. M. Miller, and M. Worwood.  Serum ferritin concentration




           and iron stores in normal subjects.  J. Clin.  Path.   26:770-772, 1973.




786   Wann,  E.  V.,  and W.  A.  Hills.   The genetics  of  boron and  iron transport in the




            tomato.   J.  Hered.   64:370-371,  1973.



787.  Wann, F. B.  Control of Chlorosis in American Grapes.  Agricultural Experi-




           ment Station Bulletin 299.  Logan:  Utah State Agricultural College,




            1941.  27 pp.




   <  Wanta,  R.  C.   Meteorology and air pollution, pp. 187-226".   In A.  C.  Stem,




            Ed.   Air Pollution.   (2nd  ed.)  Vol. 1.  New  York:  Academic Press,




            1968.



      Wapnick, A.  A.,  T.  H.  Bothwell,  and H.  Seftel.  The relationship  between serum



            iron  levels  and  ascorbic acid  stores in siderotic Bantu.   Brit.  J.




            Haematol.   19:271-276,  1970.



790.  Wapnick, A. A.,  S. R. Lynch, P. Krawitz, H.  C. Seftel,  R.. W.  Charlton,  and  T.  H.




          Bothwell.  Effects  of iron  overload on  ascorbic acid metabolism.   Brit. Med.




          J.  3:704-707,  1968.




791. Wapnick, A. A., S. R. Lynch, H. C. Seftel, R. W. Charlton, T. H. Bothwell,  and




          J. Jowsey.  The effect of siderosis and ascorbic acid depletion on bone




          metabolism, with special reference to osteoporosis in the Bantu.  Brit. J.



          Nutr.  25:367-376, 1971.
                                     354

-------
792.  Warnock, R. E.  Micronutrient uptake and mobility within corn plants  (Zea mays
           L.)  in relation to phosphorus-induced zinc deficiency.  Soil Sci. Soc.
 	Amer. Proc.  34:765-769, 1970.___    	
793   Warren, R. A. J., and J. B. Neilands.  Mechanism of microbial catabolism of
           ferrichrome A.  J. Biol. Chem  240:2055-2058, 1965.
7g,   Watt,  B. K.,  and A.  Merrill.   Composition of Foods:   Raw,  Processed,  Prepared.
           Agriculture Handbook No.  8.  Washington,  D.C.":   U.S."  Government  Printing
           Office,  1963.   190 pp.    	
795>  Wawzkiewicz,  E.  J.,  H.  A.  Schneider,  B.  Starcher,  J.  Pollack,  and J.  B.
           Neilands.   Salmonellosis  pacifarin  activity of  enterobactin.  Proc.  Nat.
           Acad.  Sci.   U.S.A.   68:2870-2873,  1971.
      Wayne, R., K. Frick, and J. B. Neilands.  Siderophore protection  against
          colicins M, B, V, and la in Escherichia coli.  J. Bacteriol.   126:7-12,
          1976.
797.   Wayne,  R.,  and J.  B.  Neilands.   Evidence for common binding sites for
            ferrichrome compounds and  bacteriophage 80 in the cell envelope of
            Escherichia coli.  J.  Bacteriol.   121:497-503,  1975.
798.   Webb,  T.  E.,  and F.  A.  Oski.   Iron deficiency anemia and scholastic achieve-
            ment in  young adolescents.   J. Pediatr.   82:827-830,  1973.
799>   Weinberg, E.  D.   Iron and susceptibility to infectious disease.   Science
            184:952-956, 1974.
800.  Weinfeld,  A.  Iron stores, pp. 329-372.  In L. Hallberg, H. G. Harwerth,
           and A.' Vannotti, Eds.  Iron Deficiency:  Pathogenesis, Clinical Aspects,
           Therapy.   New York:   Academic Press, 1970.
801.   Weinfeld,  A.   Storage iron  in man.   Acta Med.  Scand.  177(Suppl.  427):1-
            155,  1964.
                                     355

-------
 802.  Weinstein,  L. H.,  and W.  R.  Robbing.  The  effect  o£  different  iron  and  man-
             ganese nutrient levels  on  the catalase and cytochrome oxidase  activi-
             ties of green and albino sunflower  leaf  tissues.  Plant Physiol.   30:
             27-32,  1955.
 803.  Weintraub,  L. R.,  M. E. Conrad, and W. H.  Crosby.  Regulation  of the
             intestinal absorption of iron by the  rate of erythropoiesis.   Brit. J.
             Haematol.  11:432-438,  1965.
 804.   Weintraub, L. R., M. E. Conrad, and W. H. Crosby.  The significance of  iron
            turnover in the control of iron absorption.  Blood  24:19-24,  1964.
 805.   Weiss, M. G.  Inheritance and physiology of efficiency in iron utilization in
             soybeans.  Genetics  28:253-268, 1943.
 805a.  Welch, R. M., and  D. R. Van  Campen.  Iron  availability to rats from
             soybeans.   J.  Nutr.  105:253-256, 1975.
 805b   Westberg, K., N. Cohen, and K. W. Wilson.  Carbon monoxide:  Its role in
             photochemical  smog.  Science  171:1013-1015, 1971.
 806.   Westlin, W.  F.   Deferoxamine in the treatment of  acute iron poisoning:  Clin-
             ical experiences with 172  children.   Clin. Pediatr.  5:531-535,  1966.
 807.   Wetzel,  R. G.  Limnology.  .Philadelphia:   W. B.  Saunders Company, 1975.
            743  pp.
 808.   Wheby, M. S., G. E. Suttle,  and K. T.  Ford.  Intestinal absorption  of hemo-
             globin  iron.   Gastroenterology  58:647-656, 1970.
 809.   Wheby, M. S., and  G. Umpierre.  Effect of  transferrin saturation on iron
             absorption  in man.  New Engl. J. Med.  271:1391-1395, 1964.
810.   Whelpdale,  D. M.,  and R. E.   Munn.  Global  sources, sinks, and  transport of
             air pollution, pp. 289-324.  In A. C. Stern, Ed.  Air Pollution.   (3rd
             ed.)   Vol. 1.  New York:  Academic Press, 1976.
                                     356

-------
811.   Whitehead,  J.  S.  W.,  and  R. M.  Bannerman.   Absorption  of  iron by gastrec-
            tomized rats.  Gut 5:38-44,  1965.
812. Whitson, T.  C.,  and E. E.  Peacock, Jr.  Effect of  alpha alpha  dipyridyl  on
           collagen synthesis in healing wounds.   Surg.  Gynecol. Obstet.   128:
           1061-1064,  1969.
813. Whittetnore,  D.  0., and D.  Langmuir.  The  solubility  of  ferric  oxyhydroxides  in
           natural waters.   Ground Water   13:360-365,  1975.
814.  Whitten,  C. P., and  A. J.  Brough.   The pathophysiology of acute iron poisoning.
           Clin. Toxicol.   4:585-595,  1971
815.  Whitten,  C. P., G. W.  Gibson,  M. H.  Good, J." F.  Goodwin, and A. J, Brough.
           Studies in actue  iron poisoning.   1.  Desferrioxamine in the treatment  of
           acute iron poisoning:  clinical observations, experimental studies, and
           theoretical considerations.  Pediatrics  36:322-335,  1965.
815a. Widdowson, E. M.,  and  R. A. McCance.   Iron exchanges of adults on white
           and  brown bread diets.  Lancet   1:588-591,  1942.
816. Widdowson, E. M.,  R. A. McCance, and C. M.  Spray.  The  chemical composition

           of the human  body.  Clin. Sci.  10:113-125, 1951.
817. Williams,  R., F. Manenti,  H. S. Williams,  and C. S. Pitcher.   Iron absorption
          in idiopathic haemochromatosis before, during, and after venesection
          therapy.  Brit.  Med.  J.  2:78-81, 1966.
818. Williams,  R., P. J. Scheuer, and S. Sherlock.  The inheritance  of  idiopathic
          haemochromatosis:  A  clinical and liver biopsy  of  16  families.  Q. j.
          Med.  31:249-265, 1962.
819. Williams,  R., P.'M. Smith,  E. J. F. Spicer, M. Barry, and S.  Sherlock.
          Venesection therapy in idiopathic haemochromatosis:  An analysis of
          40 treated and 18 untreated patients.   Q. J.  Med.  38:1-16,  1969.
820. Williams,  R., H. S. Williams, P.  j. Scheuer, C. S.  Pitcher, E.  Loiseau,  and
          S. Sherlock.  Iron absorption  and  siderosis  in chronic liver disease.
          Q. J. Med.  36:151-166, 1967.

                                    357

-------
 821.   Williams,  R.  E.   Waste Production and  Disposal  in Mining,  Milling,  and




             Metallurgical  Industries.   San Francisco:  Miller  Freeman Publica-




             tions,  Inc., 1975.   489  pp.



 822.  Wintrobe, M. M.  Clinical Hematology   (7th ed.)  Philadelphia;  Lea and




            Febiger, 1974.  1896 pp.



 823.  Witts, L. J.  Simple achlorhydric anaemia.  Guy's Hosp.  Rep.   80:253-296, 1930.





 824.  Witzleben,  C. L., and B. E.  Buck.  Iron overload hepatotoxicity:  A postulated




            pathogenesis.   Clin. Toxicol.  4:579-583,  1971.



 825.  Witzleben,  C. L., and N. J.  Chaffey.  Response of the iron-loaded liver to




            prolonged ethionine feeding.  Arch. Path.  80:447-452, 1965.




 826.  Witzleben, C. L., and  J. P. Wyatt.  The effect  of  long  survival on the




             pathology of thalassaemia major.  J. Path. Bacteriol.  82:1-12,  1961.



 827.   Woodruff, C. W.,  and E.  B. Bridgeforth.  Relationship between  the hemogram of




             the infant  and  that of the  mother during pregnancy.   Pediatrics   12:




             681-685, 1953.



 828.   Woodworth,  R. C., K. G. Morallee, and R.  J.  p.  Williams.  Perturbations of




            the proton magnetic resonance spectra of conalbumin and siderophilin




            as a result  of  binding  Ga3+ or Fe34".   Biochemistry   9:839-842,  1970.



 829.  World  Health  Organization.  Nutritional Anemias.   Report of a WHO  Group of



           Experts.  WHO Technical Report  Series No. 503.  Geneva:  World Health




           Organization, 1972.   29 pp.



830.  World Health Organization.  Nutritional Anemias.   Report  of a WHO Scientific




           Group.   WHO Technical Report Series No.  405.  Geneva:  World Health



           Organization, 1968.  37  pp.




831.  Worwood, M.,  A.  Edwards, and  A. Jacobs.   Non-ferritin iron compound in rat



           small intestinal mucosa  during iron absorption.  Nature  229:409-410,



           1971.
                                      358

-------
 832.  Wretlind, A.  Food iron supply, pp.  39-64.  In L. Hallberg,  H.-G. Harwerth,



           and A. Vannotti, Eds.   Iron Deficiency:  Pathogenesis,  Clinical Aspects,



           Therapy.  New York:  Academic Press, 1970.


 833.  Wright,  G.  W.   The  Effects  of  Inhaled  Iron  Oxide on  the  Health of Man.



            Paper 62-73  Presented  at  the 55th Annual  Meeting, Air  Pollution



            Control Association, Chicago,  111.,  May,  1962.   8 pp.


 no,   Wutscher,  H.  K.,  E.  0.  Olson,  A.  V.  Shull,  and A.  Peynado.   Leaf nutrient
 OO^f.


            levels,  chlorosis,  and growth  of  young grapefruit trees on 16 root-



            stocks grown on calcareous  soil.   J. Amer.  Soc.  Hort.  Sci.   95:259-



            261,  1970.


 835.  Wynder, E. L., S.  Hultberg,  F.  Jacobsson, and I. J. Bross.  Environmental



           factors in cancer of the upper alimentary tract:  A Swedish study with



           special reference to Plummer-Vinson (Paterson-Kelly) syndrome.  Cancer



           10:470-487; 1957.



836.  Yancey, R.  j.j  jr.}  5.  A. L.  Breeding,  and C.  E.  Lankford.   Enterochelin:



            A Virulence  Factor for Salmonella ryphimurium.   Paper  B 82 Presented



            at American  Society for Microbiology,  76th Annual Meeting,  Atlantic



            City,  N.  J.,  May 2-7,  1976.


837.  Youdim, M. B. H.,  H. F.  Woods, B. Mitchell, D. G. Grahame-Smith, and S.



            Callender.   Human platelet  monoamine oxidase activity  in iron-deficiency



            anaemia.  Clin. Sci. Mol. Med.  48:289-295, 1975.


838.   Zadeh, J. A., C.  D. Karabus, and J. Fielding.   Haemoglobin  concentration



            and other values in women using an intrauterine device or taking corti-



            costeroid contraceptive pills.  Brit.  Med. J.  4:708-711, 1967.


838a.   Zagainov, V. A.,  A. G. Sutugin,  I.  V.  Petryanov-Solokov, and A. A.



            Lushnikov.  Probability of adhesion of molecular aerosol particles



            to surfaces.  Dokl. Akad. Nauk SSSR  221:367-369, 1975.   (in Russian)



839.   Zajic, J. E.  Microbial Biogeochemistry.  New York;   Academic Press, 1969.



            345 pp.


                                      359

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/1-78-017
4. TITLE AND SUBTITLE
IRON
7. AUTHOR(S)
Subcommittee on Iron
2.
3. RECIPIENT'S ACCESSIOt*NO.
5. REPORT DATE
February 1978
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Committee on Medical and Biologic Effects of
Environmental Pollutants
National Academy of Sciences
Washington, D.C. 20418
12. SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory ..RTF,
Off ice 'of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
15. SUPPLEMENTARY NOTES

10. PROGRAM ELEMENT NO.
1AA601
11. CONTRACT/GRANT NO.
68-02-1226
13. TYPE OF REPORT AND PERIOD COVERED
NC
•'"••- - 14. SPONSORING AGENCY CODE 	
EPA-600/11

• • • .
16. ABSTRACT
This document surveys the effects of organic and inorganic iron that are relevant
to huma.ns and their environment. The biology and chemistry of iron are complex and
only pajtially understood. Iron participates in oxidation reduction processes that
not only affect Its geochemical mobility, but also its entrance into biologic .systems.
Hydrated ferric oxide surfaces have adsorbent properties and may act as reaction sites
and catalysts. In biologic systems, the iron atom is incorporated into several proteir
enzymes that participate in many oxygen and electron transport reactions.
The report addresses the quantity and form of iron in the environment, its move-
ment and the interaction between inorganic and organic forms of the metal. Some plants
have capabilities of retrieving iron from the soil; vertebrates in general appear to
he able to achieve satisfactory iron balance. Humans are the outstanding exception -
hundreds of millions of the world's peoples are iron-deficient because of inadequate
amounts of available iron the diet; deficiency may thus be the major iron-related
environmental health problem faced by humans. The presence of large deposits of ferri-
tin and hemosiderin in parenchymal tissues has been shown to result in damage to
several vital organs. Acute iron toxicity has been reported, but only with the ingest-
ion of large amounts of iron salts. Pulmonary inhalation of iron compounds from
industrial exposure has not been shown to be a hazard.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
iron
air pollution
toxicity
health
ecology
chemical reactions
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
b.lDENTIFIERS/OPEN ENDED TERMS

19. SECURITY CLASS (This Report)
UNCLASSIFIED
20. SECURITY CLASS (This page)
UNCLASSIFIED
c. COSATI Field/Group
06 F, H, T
21. NO. OF PAGES
366
22. PRICE
EPA Form 2220-1 (9-73)
360

-------