Considerations of iron metabolism in early infancy

C O N S I D E R A T I O N S OF IRON M E T A B O L I S M IN E A R L Y I N F A N C Y NATHAN J. SMITH, M.D. MADISON, W I s .

HE

occurrence of clinically apiron-deficiency anemia is uncommon in infants less than 6 months of age. The recognized instances are limited to p r e m a t u r e l y born infants, to infants who have lost significant amounts of blood v e r y early in life, and to infants who are the products of multiple births. 1 Recent advances in the u n d e r s t a n d i n g of the metabolism of iron suggest that derangements ill the supply of iron to the fetus and the newborn infant may contribute to the high incidence of "nutritional" anemias in infants older than 6 months of age. Likewise derangements in protein metabolism 2 with compromised production of hemoglobin, resembling that accompanying iron-deficiency states, have been recently studied in very young infants. This condition, called "transient dysproteinemia," has prompted renewed interest in the metabolism of iron in the fetus and in the very young infant. This brief review will summarize the current status of our knowledge of the normal iron demands of the fetus and infant, how these demands may be altered, and how they are satisfied most effectively. Such a discussion can serve to alert the physician to the instances in which supplemental iron }s indicated and contraindicated. In addition the deficiencies in knowledge of iron metabolism and the need for

T parent

F r o m the D e p a r t m e n t of Pediatrics, Univ e r s i t y of W i s c o n s i n School of Medicine. 654

f u r t h e r critical investigation in ~his area will be apparent. IRON META B O LIS ~ IN THE FETUS

D u r i n g normal intrauterine development, the fetus will demand approximately 300 rag. of iron2 The total iron content and body weight increase proportionately during gestation, the concentration of iron per unit of fetal weight remaining relatively constant throughout fetal development. 4 Studies of fetal weight and iron content have permitted the calculation of fetal iron demand at various periods of pregnancy. The demand is negligible during the first and second trimesters, being no more than a total of 8.0 mg. In the last trimester the rapidly growing fetus requires approximately 4.0 rag. per day, a total of approximately 300 rag. During the last half of pregnancy, the maternal hemoglobin mass will increase likewise to a degree demanding up to 500 rag. of iron2 Thus, it is not surprising that iron-deficiency anemia is a common finding late in pregnancy even in the primipara who has eaten a so-called "normal" diet. The combination of iron contained in blood lost during delivery, iron contained in the placenta, and the 300 rag. of iron required by the infant totals the amount of maternal iron loss during a pregnancy. This loss exceeds by more than 200 rag. the amount of iron conserved by the absence of

SMITH:

CONSIDERATIONS OF IRON METABOLISM

655

menses during pregnancy (250 to 300 the placenta is accomplished through rag.). This deficit may well be the the formation of ferritin, having dembasis of a more deficient iron supply onstrated that placental ferritin beavailable to the fetus during f u t u r e comes labeled with F e 59 from the maternal circulation. The mechanisms pregnancies. 6 The mechanism involved in the involved in this incorporation of matransfer of iron across the placenta to ternal iron in the placenta and its subthe fetus is not known. It has been sequent release to the fetus are not known. demonstrated that iron is transported As previously mentioned, the fetal to the placental site in the maternal circulation bound to a beta-1 globulin, demand for iron late in pregnancy transferrin. A similar protein in the usually totals 3 to 4 rag. daily. The fetal circulation is involved in the maternal iron " t u r n o v e r " in the proctransport of iron from the placenta to ess of normal hemoglobin production fetal tissues. The concentration of is between 25 and 30 rag. daily. Thus, transferrin (iron binding protein) has the fraction of the total iron " t u r n been found to be significantly higher over" which goes to the fetus is relain the maternal plasma than in the tively small. Nevertheless, the amount fetal plasma taken from the umbilical of iron destined for the fetus may be vessels. In addition the plasma iron significantly influenced by factors level is distinctly higher in cord blood which alter the maternal " t u r n o v e r " than in the maternal circulation. 7 I t of iron. Pribilla 1. has shown that abis therefore apparent that the transfer scess formation in animals decreases of iron across the placenta to the fetus the maternal iron turnover and likeis against a distinct gradient; the wise diminishes the amount of iron transfer of iron requiring the removal transferred to the fetus. When large of iron from relatively low concentra- amounts of iron were injected into tions in the maternal plasma to the these animals, iron was increased in higher concentrations in the fetal cir- fetal as well as maternal tissues. The culation. That this is a one-way trans- need for extended observations of this fer mechanism has been demonstrated kind is apparent as well as a clearer by the injection of Fe 59 into the um- understanding of the mechanisms of bilical circulation. Following such an placental transfer of iron. I t is not injection, radioactive iron failed to unreasonable to postulate that a variety of factors which influence the appear in the maternal circulation2 metabolism of iron in the mother may The dynamic processes involved in distinctly alter the transfer of iron to the cellular transport of iron across the fetus. Whether or not iron dethe placenta have not been defined. f i c i e n c y in the mother interferes with The rate of transfer is extremely rapid, the relatively small amount of iron occurring in a matter of a few minutes, normally destined each day for transand would appear to be governed prinfer to the fetus cannot be answered cipally by the fetal need for iron. 1~ It can no longer be attributed to the from direct observation. A detailed destruction of maternal erythrocytes study of infants born of iron-deficient in the placenta. WShler ~ has sug- mothers would suggest that this might gested that the transfer of iron across be the case.

656

THE

THE INFLUENCE

JOURNAL

OF M A T E R N A L I R O N

DEFICIENCY

The relationship of iron-deficiency anemia in the mother to the hemoglobin levels of the newborn infant has been a subject of considerable study Strauss ~2 concluded t h a t anemia during pregnancy would lead to anemia during the first year of life in the infant. Several years later Woodruff and Bridgeforth 13 could not confirm any correlation between deficits in hemoglobin in mothers and altered hemoglobin concentrations in their infants. The absorption of iron has been studied as an index of iron " d e m a n d " in mothers and infants. No correlation could be established by this method between the mothers' hemoglobin at the time of birth and the amount of iron absorbed by their infants. ~4 However, two independent studies have recently demonstrated that infants of low birth weight and infants of high birth order are more likely to develop clinically a p p a r e n t iron-deficiency anemia during early life. It is postulated that in these situations the amount of iron received from maternal sources was less than normal and that the supply of maternal iron influenced hemoglobin synthesis during infancy. 15, 16 Difficulties in specifically identifying iron deficiency in pregnant women and quantitating hemoglobin production in v e r y young infants have been in a large part responsible for the inability to satisfactorily define the relationship between hemoglobin production in the infant and iron deficiency in their mothers. Recently L u n d 17 has done much to clarify this relationship by using suitable biochemical techniques to determine the presence of iron deficiency and by determining

OF P E D I A T R I C S

total body hemoglobin as a measure of hemoglobin production. Using the total hemoglobin mass, the levels of free erythrocyte protoporphyrin, and, to a lesser extent, the serum iron levels as indicators of iron-deficiency anemia, these workers have found from 10 per cent to over 50 per cent of pregnant patients to be anemic. In a subsequent study of 66 pregnant women and their infants, it was found that infants born of anemic mothers had a reduction of 20 per cent of the total circulating hemoglobin mass when it was measured after these infants reached 4 to 5 days of age. The more severe the anemia in the mother the more profound the change in the newborn infant. Inasmuch as the principal site of "iron storage" in the newborn infant is that iron in the circulating hemoglobin, these low levels of total hemoglobin can logically be expected to influence the production of irondeficiency anemia later in infancy, is In an additional series of 38 infants Sisson and L u n d is attempted to evaluate the effect of early and late interruption of the umbilical circulation at the time of birth on blood volume and hemoglobin mass. Demarsh and coworkers 2~ and Duckman and his colleagues 21 studied the so-called "placental transfusion" effect of delayed ligation of the cord and concluded that as much as 60 c.c. of blood and its iron-containing hemoglobin could be salvaged for the infant whose cord was tied late. It is of interest that in these recent studies neither early nor late clamping of the cord with the infant held above or below the level of the placenta significantly altered the total blood volume or hemoglobin mass as measured on the fourth day of life. If the cord were manually milked, higher

SMITH:

CONSIDERATIONS

hemoglobin contents and plethora were obtained. The results of such studies give satisfying support to the practice of " m o d e r a t i o n " in the hitherto confused issue of when to i n t e r r u p t the cord circulation. IRON REQUIREMENT IN EARLY INFANCY

The iron requirement of any given infant during the first 6 months of life can be thought of as the difference between the total body iron present at birth and that present in hemoglobin and other body tissues at 6 months of age. Physiologie losses and the excretion of iron is thought to be too minute to be of significance in a n y consideration of iron needs. The iron requirement is determined principally by the total body iron present at birth and the rate of growth. U n d e r circumstanees considered " n o r m a l , " these two factors m a y be responsible for as much as a 300 per eent variation in the delnand for exogenous iron during the first year of life. a The iron demand in a normal inf a n t can be quite simply calculated as follows: One can begin with a "norreal" infant whose birth weight is 3.5 kilogranls and whose total blood volulne is 85 c.c. per kilogram, with a hemoglobill concentration at birth of 17 Gin. per cent. In each g r a m of hemoglobin there is 3.4 rag. of iron, which accounts for 173.0 rag. of total iron in the i n f a n t u n d e r consideration. Such an infant could have 6 rag. per kilogram of tissue iron and estimated iron stores of 35 tng. of iron. Thus, the total body iron present at birth would be 229 mg. I f such an infant enjoyed an expected growth experience and weighed 8.0 kilogranis at 6 months of age, his normal total body iron would have increased to a p p r o x i m a t e l y

OF 1RON M E T A B O L I S M

657

300 mg. The 70 lng. iron demand would have been met b y assimilation of iron f r o m exogenous dietary sources. I t is of interest to calculate iron demands in this fashion for infants whose iron content might be influeneed by commonly encountered variables experienced in early infancy. I n f a n t s born of iron-deficient mothers in the group studied b y Sisson and I m n d 1~ had a 20 per cent reduction in their hemoglobin mass at birth. This wouht represent an iron deficit of approximately 35 mg. in the hypothetical infant weighing 3.5 kilograms, considered above, and an increase of 50 pet' cent in the total demand for iron in the first 6 months of life. Any loss of infant blood p e r i n a t a l l y or d u r i n g early infancy would likewise add lo lifts demand. A helnorrhage of as little as 20.0 ml. of blood in the newborn period would add a p p r o x i m a t e l y 12.0 mg. to the iron requirement. Rates of growth in excess of tha| used in the above calculation will ot>viously be reflected in inereases in the iron requirement. This is encountered most strikingly in the p r e m a t u r e l y bm'n child. Calculated iron requirelnents in excess of 110.0 rag. for the first 6 months of life are not lmCOmmon in the l>rematurc infant.'-"-' A less striking' increase in iron demand assoeiated with increases in ~'rowlh r a w has recently been suggested by Woodruff *~ in comparing iron demands of male and female infants. A preponderance of male infants was found in a large group of patients suffering fronl iron-deficiency anemia. The hemoglobin concentration of male infants has been shown to be less than in the female, yet the growth rate of males is greater in the first year of life and their hemoglobin increment is

658

THE JOU'RNAL OF PEDIATRICS

greater. Therefore, there are greater iron requirements. I t is suggested that sex differences in iron metabolism might exist in infancy as they do in adults. Is it reasonable to assume that "normal" requirements for iron can be satisfied from dietary sources alone? Calculations based on the iron content of foods given in a conventional diet with the introduction of solid foods at 3 months of age indicate an average intake of iron of approximately 5.5 rag. daily. A total of somewhat more than 1.0 Gin. of iron would be ingested during the first 6 months of life. E i g h t y per cent of this iron is derived from iron-enriched infant cereals. Most of the remaining 20 per cent of the diet a r y iron is presented in the form of vegetables, fruit, meat, and eggs. The small amount of iron in milk makes an insignificant contribution (78 rag.) to the total iron intake in the first 6 months of life. 22 That such dietary iron can be assimilated from various dietary sources has been demonstrated recently. Quantitative studies have been made of the absorption of iron from infant foods labeled in vivo with radioactive iron. F r o m 10 to 20 per cent of the iron in milk, cereal, eggs, and liver was absorbed by infants under 6 months of age. Thus, if an infant ingests 1,000 rag. of iron in an optimal diet in the first 6 months and absorbs an average of 10 per cent of the dietary iron, his calculated demands of 70 rag. should be readily available to him from his diet. It must be emphasized that this situation presupposes an optimal diet which is particularly well supplied with iron-supplemented infant cereal, a normal growth rate, a normal corn-

plement of iron stores at birth, and the absence during this period of pathological states such as gastroenteritis, hemorrhage, or infection. The high incidence of iron-deficiency anemia in pregnant women in the United States has been mentioned. A variety of situations in which fetal and infant blood is lost have recently been described. 23 These situations are not uncommon, and the amount of blood lost by the infant need not be great to significantly disturb the balance in iron demand and supply in the first 6 months of life. Infections, particularly those involving the gastrointestinal tract, are not uncommon in infancy. Even in the face of increased iron demand, i.e., iron deficiency, infants with gastroenteritis are unable to assimilate iron from their diets. 22 Such studies indicate that there exists a small margin between the demands for iron and that amount available from a normal diet. Likewise there is a frequent occurrence of abnormal situations in very early life in which the demand for iron becomes greater than can be satisfied by the diet. Faced with this situation, consideration has been given to the desirability of supplementing the normal dietary intake of iron of all infants with some form of medicinal iron. Sturgeon 3 has administered iron intramuscularly to a group of "normal" infants. He evaluated r e s u l t a n t changes in hemoglobin levels, iron binding capacity of the serum, free erythrocyte protoporphyrin, s e r u m iron, and serum copper concentrations. Comparison of the results with those found in the study of adequate controls demonstrated that many of the chemical manifestations of apparent "iron deficiency" in normal infancy

SMITH:

CONSIDERATIONS

can be altered by the administration of i n t r a m u s c u l a r iron. I n studies of the absorption of iron greater absorption was noted in " n o r m a l " infants u n d e r 1 y e a r of age, being comparable to absorption f o u n d in iron-deficiency states in older infants or children. 22 This finding can be interpreted to indicate the presence of an increased need for iron in the early age group. These studies have been interpreted by some as f u r t h e r evidence demandm g the administration of medicinal iron p r e p a r a t i o n s to all infants. There is little doubt that the recent knowledge of the metabolism of iron has clearly demonstrated the need for iron supplementation in a v a r i e t y of situations where this need had been overlooked in the past. However, to att e m p t to satisfy the needs for iron in these infants b y the " r o u t i n e " administration of iron to all infants would a p p e a r to be an u n w a r r a n t e d departure f r o m critical and thoughtful pediatric practices. A more desirable approach to the assurance of optimal iron nutrition in infancy can be the physician's thoughtful, individual evaluation of the iron needs of the very r a p i d l y growing infant, the infant who has bled, or who has suffered f r o m gastrointestinal disturbances. ADMINISTRATIONOF IRON IN INFANCY W h e n administration of medicinal iron in infancy is limited to those infants whose prenatal or perinatal experience has resulted in the presence of a deficit in the body's iron content, full therapeutic doses of iron should be given. Indications for the use of lesser doses have not been defined in infancy. U n d e r 6 months of age, adequate doses of orally administered iron salts are in the range of from 45 to 60

OF I R O N M E T A B O L I S M

659

rag. of elemental iron. This iron is most readily administered as a n y one of several " d r o p " p r e p a r a t i o n s now available. I n such p r e p a r a t i o n s iron is usually present in a concentration of 25.0 rag. p e r cubic centimeter. Doses of f r o m 15 to 30 rag. of elemental iron (0.6 to 1.2 c.c.) can be given 3 to 4 times daily with e x p e c t a t i o n of optimal assimilation. Smaller doses will have to be given too f r e q u e n t l y throughout the d a y to be practical. W h e n larger amounts are given in a single dose, the percentage of the iron absorbed is reduced. 24 Oral iron medication is best given in a fasting state, one to two hours prior to feedings to avoid the formation of insoluble and unabsorbable iron complexes with phosphates and other constituents of milk. Gastrointestinal intolerance to oral iron is an extremely uncommon occurrence in early life. I t will usually be necessary to maintain administration of iron medication for several weeks to three to f o u r months if deficits are to be corrected and adequate replenishment of tissue iron components is to be realized. A v a r i e t y of additives have been combined with iron in an a t t e m p t to enhance its effectiveness. Molybdenum, copper, vitamin C, cobalt, and vitamin B12 are available in a variety of so-called "hematinics" in combination with iron. None of these additives are known to increase the effectiveness of the administered iron, and some of these materials are known to be toxic. 23 When iron is needed, the most effective, least expensive, and safest medication to use is a simple iron salt. An iron-dextran compound suitable for intramuscular administration is

660

THE

JOURNAL

OF P E D I A T R I C S

2-C 14. The incorporation of the isotope in each protein fraction was similar in the hypoproteinemic infants and in the controls. However, the isotope disappeared from the circulation at an increased rate in the infants with low plasma proteins. It was postulated that this increased rate of plasma protein degradation without compensatory increases in the rate of protein synthesis was responsible for the hypoproteinemia in these patients. The hypoferremia, hypocupremia, and DISTURBANCESIN IRON AND PROTEIN anemia were secondary to a primary 5IETABOLISM IN EARLY INFANCY abnormality in plasma protein metaboConsiderable interest has been evi- lis~u In spite of a high intake of denced recently in the metabolic inter- dietary iron in these infants, it was relationships of iron and plasma pro- not until medicinal iron was given that teins in the early months of life. In the anemia was corrected. Recovery ]956 Ulstrom, Smith, and Heimlich 2 from the hypoproteinemia occurred reported their experience with four in- spontaneously in each patient before 9 fants less than 7 months of age who months of age, unrelated to any had edema, pallor, and a striking de- changes in diet or administration of gree of irritability. Laboratory study any medication, including iron. The revealed the presence of severe hypo- hypocupremia and hypoferremia dischromic microcytic anemia, hypopro- appeared as the plasma protein conteinemia, hypocupremia, and hypofer- centration returned toward normal. remia. No other evidences of renal or At the same time Lahey and Shuhepatic disease were present. The hy- bert 23 and Sturgeon and Brubaker ~6 poproteinemia was characterized by a reported their experiences with similar uniform reduction of the concentra- infants manifesting hypoproteinemia, tion of albumin and of all of the globu- hypocupremia, h y p o f e r r e m i a , and lin components. Normal growth rates anemia. The onset of disease in both had been experienced by all of the in- these groups of patients was somewhat fants. Prenatal and postnatal abnor- later than in those studied by Ulstrom malities could not be found. Of con- and co-workers? Of considerable insiderab]e importance was the fact that terest was the fact that both Lahey's ill each instance these patients had and Sturgeon's patients were receiving been taking an entirely "normal" diet. cow's milk as their sole source of food Balance studies indicated that the intake. Because of this fact, both of over-all nitrogen balance was positive, these groups of workers postulated but a greater than expected excretion that the condition was the result of a of urinary nitrogen was demonstrated. "new" deficiency state, a conclusion Plasma protein turnover rates were not in keeping with the dietary intake determined in three of the infants by of the four infants reported by U1using tracer doses of phenylalanine- strom. The resulting question to be

available. Well-controlled studies indicate that the response to this medication is no more rapid and no greater than that to properly administered oral iron. Intramuscular iron is expensive, potentially toxic, and produces some local discomfort. In light of these considerations the use of this material is appropriately limited to those infants with gastrointestinal disease who cannot tolerate orally administered iron salts.

SMITH:

CONSIDERATIONS OF IRON METABOLISM

answered is whether the patients were suffering f r o m one and the same disorder or whether two or more similar clinical states exist with v a r y i n g etiology. C a r t w r i g h t has recently reported an experience with hypocupremic, hypoproteinemic infants and concludes t h a t at least two similar clinical states with differing causes can be recognized in this group of anemic and edematous infants. One group of patients who suffered f r o m a p r i m a r y plasma protein deficit was similar to those reported b y Ulstrom, and perhaps a larger group had a nutritional deficiency as the p r i m a r y etiologic factor. Lahey as well as Stnrgeon are in s y m p a t h y with this suggestion. However, until more definitive studies of plasma protein metabolism are available in affected infants, until the clinical findings can be corrected by some specific nutrient, and until a deficit in the body content of some nutrient or its metabolic products is demonstrated, the question of one or more than one etiologic factor must remain open. That iron and protein metabolism are closely related is shown in areas other than the syndrome of transient dysproteinenfia. I n a p p r o x i m a t e l y ]5 per cent of infants with nutritional iron-deficiency anemia, detectable abnormalities in plasma proteins are present. W h e t h e r these changes, most commonly hypoalbuminemia, are related to comprised rates of synthesis or rapid loss of the albumin f r o m the circulation remains to be determined. Doubtless the future must provide more precise information as to the relationships of iron and protein metabolism in the early months of life when the demands of growth of these two body constituents are great.

661

REFERENCES 1. Smith, N. ft., a n d Rosello, S.: Iron Deficiency in ~nfancy a n d Childhood, Am. J. Clin. N u t r i t i o n 1: 275, 1953. 2. Ulstrom, l~. A., Smith, N. J., and Heimlich, E. IV[.: T r a n s i e n t Dysproteinemia in I n f a n t s , A New Syndrome. I. Clinical Studies, A. M. A. J. Dis. Child. 92: 219, 1956. II. Studies of P r o t e i n Metabolism Using Amino Acid Isotopes, 93: 536, 1957. 3. Sturgeon, P.: Iron Metabolism. A Review, Pediatrics 18: 267, 1956. 4. Widdowson, E. M., and Spray, C. M.: Chenfical Development in Utero, Arch. Dis. Childhood 9.6: 205, 1951. 5. Pribilla, W., Bothwell, T. H., and Finch, C . A . : Iron T r a n s p o r t to the F e t u s in Man, in Iron in Clinical Medicine, University of California Press, 1958, pp. 58-64. 6. Moore, C . V . : The I m p o r t a n c e of Nut r i t i o n a l Factors in t h e Pathogenesis of Iron Deficiency Anemia, Am. J. Clin. N u t r i t i o n 3: 1, 1955. 7. Smith, C. H., Schulman, I., and Morganthau, J . E . : Iron ~Vfetabolism in I n f a n t s and Children. Serum Iron and Iron B i n d i n g Protein, Diagnostic and Therapeutic Implications, Advances Pediat. 5: 195, 1952. 8. Bothwell, T. H., Pribilla, W., and Finch, C.A.: T r a n s p l a e e n t a l Iron T r a n s p o r t to the R a b b i t (To Be Published). 9. Wfhler, F.: Zur Physiologic und Pathologie des Speichereisens; fiber den intermedi~ren Eisenstoffwechsel der Plazenta, Deutsche meal. Wchnschr. 80: 3(t, 1955. 10. Vosberg, G. J., and Flexner, L. B.: M a t e r n a l Plasma as a Source of Iron for the F e t a l Guinea Pig, Am. J. Physiol. 161: 202, 1950. II. Pribilla, W.: Tierexperimentelle Untersuchungen fiber den Eisenaustausch zweschen M u t t e r und Foet naeh intravenoser Eisengabe, Acta haemat. 12: 371, 1954. 12. Strauss, M. B.: A n e m i a of I n f a n c y From M a t e r n a l Iron Deficiency in Pregnancy, J. Clin. Invest. 12: 345, 1933. 13. Woodruff, C. A., and Bridgeforth, E. B.: Relationship B e t w e e n the Hemogram of the I n f a n t and T h a t of the l~[other During Pregnancy, P e d i a t r i c s 12: 681, 1953. 14. Oettinger, L., Jr., Mills, W. B., and Hahn, P . F . : Iron Absorption in Premature and Full-term Infants, .1. PEDIAT. 45: 302, 1954. 15. Woodruff. C. A.: Multiple Causes of Iron Deficiency in I n f a n t s , J. A. M. A. 167: 715, 1958. 16. Guest, G. W., and Brown, E. W.: E r y t h r o c y t e s and Hemoglobin of the Blood in I n f a n c y and Childhood. III. Factors in V a r i a b i l i t y , A. M. A. J. Dis. Child. 93: 486, 1957.

662

THE JOURNAL OF PEDIATRICS

17. Lurid, C. J.: The Iron Deficiency Anemia of Pregnancy, Am. J. Obst. & Gynec. 62: 947, 1951. 18. Sisson, T. R. C., and Lund, C . J . : The Influence of Maternal Iron Deficiency on t h e Newborn, Am. J. Clin. Nutrition 6: 377, 1958. 19. Whipple, G. A., Sisson, T. R. C., and Lund, C . J . : Delayed Ligation of the Umbilical Cord, Its Influence on the Blood Volume of the Newborn, Obst. & Gynec. 10: 603, 1957. 20. Demarsh, Q. B., Windle, W. F., and Alt, I-L L.: Blood Volume of the Newborn I n f a n t in Relation to E a r l y and L a t e Clamping of the Umbilical Cord, Am. J. Dis. Child. 63: 1123, 1942. 21. Duekman, S., Merk, It., Lehman, W. X., and ReRan, E.: The Importance of Gravity in Delayed Ligation of the Umbilical Cord, Am. J. Obst. & Gynec. 66: 1214, 1953.

22. Schulz, J., and Smith, N. J.: Quantit a t i v e Study of the Absorption of Food Iron in I n f a n t s and Children, A. M. A. J. Dis. Child. 95: 109, 1958. 23. Smith, N. J.: Iron as a Therapeutic Agent in Pediatrics, g. PEDIAT. 53: 37, 1958. 24. Schulz, g., and Smith, N. J.: Quantit a t i v e Study of the Absorption of Iron Salts in I n f a n t s and Children, A. M. A. J. Dis. Child. 95: 120, 1958. 25. Lahey, M. E., and Shubert, W. K.: A New Deficiency Syndrome Occurring in I n f a n c y , A. IV[. A. J. Dis. Child. 93: 31, 1957. 26. Sturgeon, P., and Brubaker, C.: Copper Deficiency in I n f a n t s , A. M. A. J. Dis. Child. 92: 254, 1956. 27. Zipursky, A., Dempsey, H., Markowitz, H., Cartwright, G., Wintrobe, M. M.: Studies on Copper Metabolism. XXIV. Hypocupremia in I n f a n c y , A. M. A. J. Dis. Child. 96: 148, 1958.

" . . . we are able to show, by some remarkable experiments, . . . that molecular and tailed forms of some material substance may be seen issuing out of (the corpuscles) and passing into the plasma, without much alteration of their form or colour; . . . there can be no diffiulty in concluding that the corpuscles, in their natural state, discharge matter into the plasma. I t must be confessed that but little can be demonstrated of the vital and depurating processes constantly going forward in the blood of the living person. Nevertheless, in the absence of such demonstration, we are warranted, by the experiment just referred to, and by the undisputed teaching of physiology, in concluding that the corpuscles of blood are sustained in their vital and chemical qualities by the plasma and the air; also, that their excretions are passed, p a r t l y into the plasma and partly into the air . . . . " Galstonian Lectures on " F e v e r and Inflammation," delivered before the Royal College of Physicians, London, by William Addison, M.D. From the British Medical Journal, April 30, ]859.