Brain-size, Grey Matter and Race Fact or Fiction?’ PHILLIP V. TOBIAS Department of Anatomy, University of the Witwatersrand, Johannesburg ABSTRACT A critical review is given of those factors which may be accompanied by variations in brain weight, viz. sex, body size, age of death, nutritional state in early life, source of the sample, occupational group, cause of death, lapse of time after death, temperature after death, anatomical level of severance, presence or absence of cerebrospinal fluid, of meninges, and of blood-vessels. Valid comparisons between the brain-weight of human populations should take all, or several, of these variables into account; however, published studies have not done so, despite claims to the contrary. The ideal sample is from subjects who have died suddenly without prior disease: while three such samples are o n record for Europeans, none has been recorded for Negroes. The brain-weight of healthy Negroes is not known. Most published interracial comparisons are invalid. The histological, chemical and functional counterparts of big and small brains in modern man are not known. Published interracial comparisons of thickness of the cerebral cortex and, particularly, of its supragranular layer, are technically invalid: there is no acceptable proof that the cortex of Negroes is thinner i n whole, or in any layer, than that of Europeans. It is concluded that vast claims have been based o n insubstantial evidence. In this age of cellular and molecular biology, ever more refined techniques are being focussed on the central nervous system. Brains are being studied by electron microscopy, autoradiography, ultraviolet absorption spectromicrophotometry, biochemical analyses, cybernetic models, tissue culture and a host of other methods. In the light of these developments, it is almost inconceivable that anything remains to be said at all about so gross and crude a measure as the overall size of the brain. Yet, it is an astonishing fact that a great deal remains to be said and, I am afraid, much of it negative-for most that has already been asserted does not stand up to the cold light of scientific scrutiny. The theme of this paper - “Brain-size, Grey Matter and Rack - Fact or Fiction?” - is based upon the oft-repeated claims that the races of man have been shown to differ in quantity of brain-substance and, especially, of grey matter in the cerebral cortex. More specifically, it is based upon the claims that Negroes have smaller brains than White or Caucasoid peoples and upon the frequently drawn inference that an apparently lesser intellectual performance by Negroes is based upon such supposed differences of brain-size. AM. J. PHYS. ANTHBOP.,32: 3-26. It will be my aim to examine the evidence for such claims; to try to dissect apart the facts from supposed facts. We shall find that the evidence is inadequate or, in some respects, totally lacking. Further, I shall try to examine the meaning of brain-size differences in the light of recent neurohistological and neurophysiological researches, as well as from an evolutionary standpoint. We shall find that, even were differences in brain-size validly demonstrated among different populations, they would be unable to explain adequately variations in cerebral function and achievement among living human beings. Furthermore, we shall find that such variations, if they do exist, are apparently not of much importance in modern man. We shall see that in this expanding universe, ours is a world of contracting brains. Finally, we shall be led to regard somewhat more critically the claims made about grey matter in the men of today. 1Lecture delivered on 25 March 1969, on the first anniversary of the departure from South Africa of Dr. Raymond Hmenberg, formerly of the Department of Medicine, University of Cape Town. This was the 1st Raymond Hoffenberg Lecture, delivered under the auspices of the Department of Medicine, University of Cape Town. 3 4 PHILLIP V. TOBIAS Some claims about race and brain power It has often been claimed that races differ in intellectual capacity. For example, in 1961 and again in 1962, in a report on ‘The Biology of the Race Problem,” prepared by commission of the Governor of Alabama, George quotes I.&. test performances, crime statistics, behaviour and temperament traits, for all of which differences are claimed to exist between Whites and Negroes. These statements by George have been quoted by Hofmeyr (’61), Swan (’64), Putnam (’63, ‘67) and others, Not being a psychologist, a criminologist or a sociologist, I do not feel competent to offer any authoritative comment on the statistics, the objectiviy with which they were obtained, the degree to which such factors as economic and educational background were considered in the study of the crime statistics, nor the validity of claims that the alleged differences are based upon racial factors, i.e., inherent or genetic differences. However, 1 do feel better able to express an opinion when George (’62), Swan (’64), Putnam (’63, ’67) and others attribute these differences to inherent anatomical causes. The most important features cited by these writers are the size of the brain, the fissuration of the cortex and the thickness of the supragranular layer of the cortical grey matter. For example, we find Putnam (‘67) stating. tion of the degree to which these differences may be attributable to inherent morphological rather than to environmental causes.” He goes on to add: “Are there hereditary structural and other biological differences between individuals and races that might serve to explain the observed differences in intelligence and in behaviour in those areas of activity that make western civilization? The presence of such differences is not only a reasonable expectation but is supported by evidence.” (Op. cit., p. 25.) The validity of the argument hinges on two sets of premises: (i) the correctness of the statements about the differences in the anatomy of the brains; and (5)the validity of the inference that these anatomical variations, if present, connote functional differences relevant to social behaviour, learning ability, ‘behaviour in those areas of activity that make western civilization,” and so on. Perhaps the most important claimed anatomical difference relates to brain size. Thus, Swan (’64, p. 33) states: “Despite sampling difficulties and variations in methods, enough comparative racial studies of brain weight have been completed to warrant a quantitative description of the racid differences. The average cranial capacity of European Whites is from 100 to 175 cm3 greater than that of African Negroes, and their average brain weight is some 8 to 12% greater.” (op. cit., p. 33). ‘We have enough studies of the Negro What facts do we have about brain-size? brain, under varied enough circumstances, to speak with assurance of its First, what measure of brain-size should relative weight. When this factor is we use? If the actual brain is available, combined with studies of its other feawe may weigh it or, less satisfactorily, detures, such as sulcification of the cortermine its volume. If the brain itself is tex and thickness of the supragranular layer, we can also speak with assurance not available, we can still derive an apof its relative evolutionary status.” (Op. proxima tion to brain-size by determining cit., p. 103.) the capacity of the cranium or brain-case. These writers claim that this group of However, the brain-case accommodates a supposed facts points to the Negro’s brain great deal more than simply brain. Thus, being of lower evolutionary status, which, when we say that a skull has an endocranin turn, implies a lesser capacity to learn ial capacity of 1400 cm3, this includes not (e.g., Putnam, ’67, p. 106). only brain-tissue, but the roots and intraSimilarly, after detailing crime statistics cranial trunks of no fewer than 24 cranial among Negro and White communities, nerves, the thick outer brain covering or George (’62, p. 25) turns to consider the dura mater, the two thinner, inner coverbrains of Negroes and Whites with the ings, the arachnoid and pia mater, the words, “We must now turn to a considera- subarachnoid space and its enlargements BRAIN-SIZE, GREY MATTER AND RACE called cisterns containing fluid, numerous blood-vessels including special enlarged venous channels called cranial venous sinuses, blood, cerebrospinal fluid. Thus, only a proportion of the cranial capacity is made up of brain-tissue: estimates of this proportion vary from 10% (Brandes, ’27) to 33.33% (Mettler, ’55). Furthermore, the ratio is not a constant figure within the adult lifetime of any one individual, for the brain shrinks with age, in certain illnesses and under some other conditions, though the brain-case itself has not been shown to diminish. Hence, as the quality of life changes, the amount of nonneural cranial capacity increases and one’s head may become progressively N e d more and more with emptiness. For these reasons, cranial capacity is a less reliable indicator of brain-size than brain-weight itself. When dealing with fossil man, we are perforce obliged to consider cranial capacity, as our sole guide to brain-size. But when speaking of living man, we are able to determine brain-size itself and thus eliminate at least some unnecessary sources of error. In this discussion on the brain-size of living man, attention will be confined largely to the actual brain-weight. We shall see that even this measure of brain-size is not as easy to determine as it may sound and that, even though we avoid certain errors by not relying on cranial capacity, we nevertheless remain with a large number of possible sources of variance. Some facts about brain-weight For over 130 years, data have been accumulating on the weight and/or volume of the human brain, or the cranial capacity, in samples drawn from different racial groups. As long ago as 1849, S. G. Morton published observations on the cranial capacity of skulls drawn from different populations: the mean capacity of a small sample of Negro skulls was 91.3% of the mean for a very small sample of Caucasoid skulls, the values being 1360.1 cm3 and 1489.6 crn3 (converted to metric units by Pearl, ’34). Other early studies on cranial capacity gave similar results (Peacock, 1865 -88.4% on 9 Negro skulls and 16 “European” skulls: Duckworth, ’04 92.5% ). 5 As far as can be traced, the earliest references to the brain-weight of the Negro were those of Soemmering (1788) who weighed two brains, Tiedemann (1836) who weighed one and Sir Astley Cooper one (cited by Peacock, 1865). In 1865, Peacock weighed another five brains of Negroes; according to Pearl (’34), Peacock copied an additional two brain-weights from tables compiled by John Reid. Five of the seven individuals represented were males, and three of them are reported to have died of phthisis in a greatly emaciated state. The mean weight for these five was computed by Pearl (’34) as 1256.8 gm. By 1885, Topinard could cull data for 29 male Negro brains from diverse sources in the literature: the mean was 1234 gm. In 1894, Waldeyer recorded a mean brain-weight of as low as 1148 gm for 12 male African Negro brains, 11 of which were weighed fresh. But, as Pearl pointed out (’34, p. 432), two of these belonged to 15-year-olds, while several had died of wasting diseases and were severely emaciated. In 1909, Mall made the forthright statement that the average brain-weight of the Negro is about 100 gm less than that of Whites. Bean, too, found a lower brainweight in Negro males than in White males (’17). Perhaps the most frequently quoted study is that of Vint (’34) who found an average brain-weight of 1276 gm in 389 adult male Kenya Africans. This is the longest Negro series on record. Pearl (’34) analysed brain-weights collected by Surgeon Ira Russell of the Eleventh Massachusetts Volunteers during the American Civil War and published by S . B. Hunt in London in 1869. The sample of brains of enlisted soldiers had been broken down a century ago into White, varying grades of “mixtures”, and Black. It seems that the various mixtures were determined by the simple, though grossly misleading expedient, of examining skin colour. Pearl found that the mean for 139 brains of Negroes was 1354.8 gm, that is, 92.1% of the mean for 24 brains of Whites (1470.6 gm). Strangely enough, the mean for 240 brains of people believed to be mixtures was 1333.6 gm i.e., lower than both the White and Negro means! (table 1). 6 PHILLIP V. TOBIAS TABLE 1 Pearl’s (‘34) analysis of brain weights of American Civil W a r dead. Measurements made originally by uncertain technique by surgeon Ira Russell of the Eleventh Massachusetts Volunteers Classification Number Average brain weight White Mixtures Black 24 240 139 1470.6 1333.6 1354.8 gm As I have recited them, these figures look consistent and convincing. Indeed, they have been immediately and uncritically accepted as indicating a lower average level of intelligence of the Negro. As Putnam has put it: “The evidence is simply that, as a racial average, the Negro brain is lighter t h a n the White and that this, in turn, indicates a lower average level of intelligence.” (’63, p. 10) We shall return later to the claim that less brain-weight means less intelligence. For the moment, let us examine the figures on brain size more closely. It is true that the studies just cited have demonstrated a lower average brain weight, or smaller average capacity of the braincase, in Negroes than in Whites. The biggest series of White or Caucasoid brains to have been weighed was 2752 Americans: their average weight was 1301 gm, or, if 672 diseased brains are excluded, the average for the remaining 2080 brains was 1305 gm. (Appel and Appel, ’42). This is only 29 gm heavier than the average for the sample of Kenya African brains. On the other hand, Davis (1868) recorded capacitites from which Pakkenberg and Voigt (’64) computed a mean brain-weight in “Africans” of 1293 gm,whilst the average for 139 American Negro brains was 1355 gm (Pearl, ’34) - just 50 gm heavier than the mean of the most comprehensive White series. Other Caucasoid series range up to means of 1440 gm for 724 adult male Danes (Pakkenberg and Voigt, ’64), 1450 gm for 581 brains weighed at the Institute for Forensic Medicine in Prague (Matiegka, ’02) and 1455 for 372 brains of Bohemians (Pearl, ’05). That is, the means for three fairly long Negro series range from 1276-1355 gm, while the means for some 11 White or Caucasoid series range from 1301-1455 gm. It must be stressed that these are ranges of averages; ranges of individual values are much wider and overlap to a greater extent. For comparison, means for three Mongoloid series range from 1360-1375 gm (Kusumoto, ’34; Shibata, ’36 - cited by Bailey and von Bonin, ’51 and by Pakkenberg and Voigt, ’64). Bailey and von Bonin (’51) have given an overall average (weighted mean) of 1344 gm for many modern human series irrespective of race, i.e., for the living species Homo sapiens at large (table 2). Negro populations sampled have averages above and below the overall modern human average. None the less, the overall Negro average would seem to be somewhat lower than the overall European average in absolute terms, that is, when uncorrected for general bodily size and other related variables. However, a mere comparison of average figures may be misleading unless allowance is made for these other TABLE 2 Average brain weight gm Kenya Africans (389) Africans American Negroes (139) 1276 1293 1355 Vint (’34) Davis (1868) Pearl (’34) American Whites (2752) 11 White (Caucasoid) Series 1301 1301-1455 Appel and Appel (’42) 3 Mongoloid Series 1360- 1375 Kusumoto (’34) Shibata (’36) 1344 Bailey and Von Bonin (’51) Modern Homo sapiens (weighted mean) 7 BRAIN-SIZE, GREY MATTER A N D RACE variables : we shall consider first the simple notion that one might expect people with bigger body sizes to have bigger brain sizes. Brain size and body size The size of the brain in relation to body size has been employed chiefly in comparisons between species, in an effort to show how predominant the brain has become in man. It was Cuvier who first introduced the concept of relative brainweight, that is, the weight of the brain expressed as a fraction of the weight of the body (Krompecher and Lipak, '66). A picture given by Cobb of a small Asian woman standing alongside a rhinoceros shows graphically the difference in relative sizes of brains. The woman may have a brainweight of 1200 gm in a body weighing, say, 45 kg, whereas the rhinoceros has a brain which may weigh 600 gm in a body of about 2000 kg (Cobb, '65). The brain: body ratio in the rhinoceros is 1: 3,000 in such a case, that of the woman 1:38. Even more revealing is a glance at the Brontosaurus, one of the great dinosaurian ruling reptiles, perhaps 65 feet long, that lived in the Jurassic period: its brain constituted a mere 1/100,000 of its 35-ton bulk. A whale has a brain/body weight ratio of 1:10,000; an elephant 1:600; a gorilla 1:200 and a man about 1:45. But it is sobering to see that while Man's exalted brain constitutes just over 2% of his body weight, this percentage is surpassed by the lowly house mouse (2.50% or 1:40), the porpoise with 1:38, the marmoset with 1:19, and the attractive little squirrel monkey of tropical America whose brain occupies 1/12 or 8.50% of its body weight (Cobb, '56)! Because Man, the sapient, did not come out at the top, it is not surprising that Man, the vainglorious and the arrogant, has been searching ever since for an index which would place him unequivocally and unassailably on the highest branch of the tree of life. For example, when the length of the hypothalamus is expressed as a fraction of the cerebrum, Man has the lowest fraction and so comes out on top (Kummer, '61); when the weight of the spinal cord is expressed as a fraction of the brainweight, Man has the lowest fraction and so conies out on top (Latimer, '50; Krompecher and Lipak, '66); when the cranial capacity is related to the area of the foramen magnum, Man has the highest value and still ends up on top (Radinsky, '67) I All the indices I have cited have a certain though limited usefulness when comparisons are made between one species and another. However, few studies have seriously addressed themselves to the problem within the species of man. Matiegka ('02) and Pearl ('05) claimed that, in Man, brain weight varies with body weight and with body height. That is to say, they claimed to find that taller people and heavier people have larger brains. However, more recent workers have questioned the correlation claimed to exist between brain weight and body weight within the human species. Pakkenberg and Voigt ('64) have made a more refined statistical analysis on the brains of European subjects. They showed that the increasing brain weight with increasing body weight found by the earlier workers is really owing to the fact that people with higher body weight are usually taller than average (chart 1). The earlier workers had failed to correct for body height when evaluating the relation between brain weight and body weight. When this correction was made (chart 2), it was Pound that brain weight depends significantly on body height but not on body weight (Pakkenberg and Voigt, '64, p. 303). CHART 1 Taller people tend to have bigger brains Heavier people tend to be taller :. Heavier people APPEAR to have bigger brains CHART 2 Heavier people, adjusted for height, do NOT have bigger brains Taller people have bigger brains irrespective of weight :. Brain-weight is positively correlated with body height, but not with body-weight 8 PHILLIP V. TOBIAS ence to his “initial hypothesis” in this context is both seriously misleading and an error of logic. The wording may eliminate such sources of variance as different mean body-weights from the hypothesis, but this by no means eliminates these sources of variance from the facts to which Putnam tries to apply the hypothesis. By citing the “initial hypothesis” as eliminating bodyweight and other variables, Putnam creates the impression that the value he quotes of 8-12% lower brain-weight in the Negro has been corrected for body size. He states explicitly in another place (’67,p. 43), “Brain weight or size comparisons were never attempted without first allowing for sex and body size, and when these adjustments were made the averages were both consistent and clear.” But no height measurements were recorded for Vint’s Kenya Africans nor for Pearl’s Civil War Negroes, the two longest series of Negro data on record, nor for any of the earlier cited studies. Where and by whom were the claimed adjustments made? Putnam’s statement is totally erroneous: in none of the studies cited, nor in Putnam’s own writings, has a correction been made for the varying heights of the subjects concerned, nor are corrected or adjusted figures recorded. The figures of 8-12% shortfall cited by him are absolute figures as measured, not as corrected for body height differences. In comparisons between different animals, Lashley (’49) suggested that the “There is, I suppose, no dispute about the fact that, other things being equal total amount of brain material, expressed (such as sex, body size, proportion of as a fraction of total body size, “seems to parts and sulcification), the weight of represent the amount of brain tissue in exthe brain correlates with intelligence.” cess of that required for transmitting im(Putnam, ’63, p. 10.) pulses to and from the integrative centres” Elsewhere, too, Putnam (’67,p. 53) states (op. cit., p. 33). Following up this notion, “, . constant efforts like Simmons’ oc- Jerison (’63) has demonstrated that brain curred to confuse the issue by injecting size may be considered as two separate variables which properly were eliminated components: one is directly related to the in the initial hypothesis.” Putnam’s refer- size of the body, and it is bigger in Primates with bigger body size, and vice versa. The CHART 3 other component seems to vary independently of body size; it comprises the Adult men have on the average 8.3gm of brain “surplus” nerve-cells which are present for every centimetre of body height over and above those required for the satisAdult women have on the average 8.0 gm of brain faction of immediate bodily needs. These for every centimetre of body height “surplus” nerve-cells, Jerison suggests, axe available for response to the challenge of SPA” AND DUSTMANN (’65) the environment through a wider range Exactly similar results were obtained by Spann and Dustmann (’65) in a study of 1229 male and 632 female brains in the medico-legal institute at the University of Munich: brain weight rose with increasing stature (chart 3), but no association could be determined between brain-weight and body-weight. In 1966,Schreider reanalysed some old brain and body measurements made in Paris in 1865-1870 by Paul Broca, one of the great anatomists of the last century. Schreider found that between brainweight and body-height there was a positive correlation coefficient of 0.26 for 224 males and 0.31 for 111 females. Clearly, if one does not allow for the varying average body-sizes of different populations, one may be misled into making false statements about the existence of real differences in average brain size among the populations compared. With this in mind, one has reexamined the main original sources on Negro brainweight in comparison with White brainweight. Neither Davis (1868) nor Vint (’34),nor Pearl (’34),the three main students of Negro brain-weight, have taken body height into account, nor made any allowance for it. Yet, when it is pointed out that many Negro groups have a lower average body size than many White groups, Putnam’s reply is that such variables are eliminated in the initial hypothesis, which reads as follows: . 9 BRAIN-SIZE, GREY MATTER AND RACE of brain-behaviour mechanisms; that is, for intelligent adjustment. On the basis of cell-counts in a variety of Primates and, given certain assumptions, Jerison claims that it is possible to estimate the number of cortical nerve-cells, not only in the brain as a whole, but in each of the two components. He has developed a series of equations for the calculation of these neuronal values, given the size of the brain and the size of the body. By applying these formulae, he has been able to compute the number of ‘extra’ neurones regarded as being available for brain:behaviour adaptive mechanisms. With this second parameter, the number of “excess neurones,” he has found it possible to distinguish the Primates and, especially, the family of Man, on the basis of the relative development of the extra neurones. Modern man has f a r more excess neurones than, say, the chimpanzee and gorilla. While the African great apes can be shown by Jerison’s equations to possess 2.4-3.6 billion excess neurones, modern man has over eight billion. The method involves many assumptions. It does not adequately take into account regional variations in the density of neurones, in the ratio between neuroglia and neurones, in the size of nervecell bodies, in the length and complexity of the dendritic processes of the neurones. None the less, itprovides a novel, approximate gauge of cortical development. Above all, the method has been developed for comparisons between species, However, as an exercise, I have tentatively assumed that similar patterns of variation may occur between populations within a species, like Homo sapiens, to those which occur between species. Purely to emphasize the need for body-size to be taken into account in discussions on human brain-sizes, I have made some calculations, using Jerison’s formulae, of the number of such “surplus” nerve-cells in various races, for which large samples of data on male brain size and body size are available. The results are given in table 3. These results suggest that the differences among various racial or population groups are negligible, once allowance has been made for body size. Further, it should be noted that the figures quoted are averages for each group: but individuals within each group vary above and below the average, so that considerable overlap occurs. It is stressed that this is only an exercise: it has yet to be shown whether Jerison’s method of analysis is valid in general, and, more particularly, whether it can validly be applied in this manner within a species. We may conclude this section by stating that no comparisons between the mean brain-size of different populations or races permit valid statements to be made on interracial differences, unless corrections have been made for differences in body height. On this basis alone, all comparisons between Negro and White brain-sizes to date are invalid. BRAIN-SIZE AND AGE It has long been known that the average weight of the human brain decreases with advancing years (e.g. Pearl, ’05; Appel and Appel, ’42; Tiakahasi and Suzuki, ’61; Pakkenberg and Voigt, ’64; Spann and TABLE 3 Excess 7tezLrms in populatiolts of modem man (Estimated by the method of Jerlson, ’63) Population n Average brain weight Kenya Negro American White French English American Negro Japanese Korean Swedish 389 2080 292 457 139 342 136 416 1276 1305 1325 1333 1355 1360 1370 1400 Millions of excess nervecells gm 8,400 8,500 8,600 8,600 8,700 8,900 8,900 8,900 10 PHILLIP V. TOBIAS mental animals may lead to permanent impairment of the brain, including diminution of brain-size. This effect occurs provided the period of undernutrition begins before the cerebral cortex has reached its adult state, or, at least, during the period of maximum vulnerability of the brain to stress. If the nutritional insults are administered within this critical post-natal period, or even if it is the pregnant mother which is undernourished, the impairment to the size, stature and chemistry of the brain is not reversed by subsequent restoration of normal nutrition. These results have been obtained on pigs, rats and mice (Jackson and Stewart, '20; Dobbing and Widdowson, '65; Davidson and Dobbing, '65; Dickerson and Dobbing, '67; Dickerson and McCance, '67; Dickerson and Walmsley, '67; Guthrie and Brown, '68; Zamenhof, van Marthens and Margolis, '68; Chase, Lindsley and O'Brien, '69). Is there any evidence for impaired brain growth in man under conditions of malnutrition? The studies of Engel ('56), Nelson ('59, '63), Stoche and Smythe ('63, '67), Cravioto and Robles ('65), Brown ('66), Eichenwald and Fry ('69) and Baraitser and Evans ('69) have all noted functional impairments of the human brain following early undernutrition. These workers have studied electroencephalographic readings, psychometric testing, rate of psychomotor development, and head circumference, to demonstrate abnormalities. It is still too early to say whether the changes demonstrated are permanent; the analogy of animal experiments would suggest a permanent effect, while the recent study of Baraitser and Evans ('69) in the Department of Neurology at the Groote Schuur BRAIN-SIZE AND NUTRITIONAL STATE Hospital, Cape Town, would tend to s u p Evidence has accumulated, especially port the notion that the E.E.G. changes in recent years, that undernutrition at outlast the acute stage of the nutritional critical stages during ontogeny of experi- insult. In their recent review, Nutrition and CHART 4 Learning, Eichenwald and Fry ('69) summarise as follows: Average brain weight : Decline with years Dustman, '65; Schreider, '66), and that the decline may begin as early as the twentieth year. The drop in average brain weight from young adulthood to about 80 years of age varies from 95 gm in Swedes (Pearl, '05, after Retzius, 1900) to 170 gm in Danes (Pakkenberg and Voigt, '64). Among the Danes, the drop from the age of 25 years to the age of 70 years is approximately 100 gm. These figures suggest a drop in the average of 7 to 11% of the maximum brain size (chart 4). Schreider ('66) showed negative correlation coefficients between brain weight and age as high as -0.39 in males and -0.46 in females. The cause of the fall with age need not concern us here, but the relevance of this finding is that if one series of brains includes a large proportion from old people, the overall average would tend to be lower than that of another series which includes more young adult brains, even though both series may stem from the same population. For many European series the ages at death are known (e.g. Schreider, '66, on Broca's material), but this is not true for Negroid series, such as that of Vint ('34) on Kenya African brains. We simply do not know the ages of his Negroid subjects and cannot therefore decide to what extent his relatively small average figure of 1276 gm is attributable to age, and how much to stature and to other variables in that population. A valid comparison between two series of brains drawn from different racial groups should take the age composition of the sample into account. It cannot be claimed that this has been done for comparisons between the brainsizes of Caucasoids and Negroids. Young adulthood - Drop of 95 gm Drop of 170 gm Drop in average f 80 years Swedes Danes 7%-11% "Observations on animals and human infants suggest that malnutrition during a critical period of early life results in short stature and may, in addition, permanently and profoundly affect the future intellectual and emotional development of the individual. In humans, it BRAIN-SIZE, GREY MATTER AND RACE is not known whether these results may be caused by malnutrition alone or whether such intimately related factors as infection and an inadequate social and emotional environment contribute significantly to the problem. Field studies to test these hypotheses are, at best, difEcult to design and to carry out; it seems likely that it wiU prove impossible to separate clearly the individual effects of malnutrition, infection and social environment.” In a subsequent editorial in Science, Abelson (’69) summarised the results of their study and of an International Conference on Malnutrition, Learning, and Behaviour (Scrimshaw and Gordon, eds., ’68) as follows: “Children reared in poverty tend to do poorly on tests of intelligence. In part this is due to psychological and cultural factors. To an important extent it is a result of malnutrition early in childIt seems likely that millions hood of young children in developing countries are experiencing some degree of retardation in learning because of inadequate nutrition, and that this phenomenon may also occur in the United States animal experiments suggest that (in the human infant) good nutrition during the first three years of life is particu,” larly important ... ... .. Further follow-ups are obviously necessary to confirm the hypothesis that malnutrition at a critical age results in permanent braindamage, with permanent impairment of intellect and emotions, and of brain rhythms. The substantial body of evidence for such impairment leaves undecided the question of whether changes have occurred in the brain-size and structure. What direct evidence do we have for such changes following early undernutrition? Several students have demonstrated that a smaller brain-weight in man may result from early undernutrition. Kerpel-Fronius (’61) found a reduction of 10-20% in brain-weight of the malnourished human infant. Brown (’66),working on Uganda African children at Mulago Hospital, Kampala, found a significant reduction in the brain weight of malnourished children compared with other Ugandan children. Even among his control “non-malnourished children,” Brown stressed, many, if not most, showed suboptimal nutrition, though not rating the 11 autopsy diagnosis of kwashiorkor, marasmus or general undernourishment. Recently, belated after-effects of starvation and other forms of maltreatment of Second World War prisoners held in concentration camps have been investigated in Norway (Strprm, ’68). Nearly 20 years later, a number of survivors showed a reduction in the size of the brain, accompanied by signs of intellectual deterioration. Further, an article by W. Osler entitled “Belated scars of war prison,” which appeared originally in the London “Sunday Times” and was quoted by the Johannesburg “Star” on November 11, 1968, stated, “Research in Japanese camps had shown that starvation causes the brain to swell, and it is thought that on return to normal conditions the brain then shrinks to less than its original size.” On the structural side, Eayrs and Horn (’55) showed histologically that undernutrition impairs the elaboration of nervecell processes: this may be one of the mechanisms behind both the reduced brain-size and impaired brain-function. Fishman, Prensky and Dodge (’69) have recently shown changes in the brain lipids of starved (or chronically malnourished) human beings, resembling those found in experimental animals. Not only is the total lipid reduced, but most severely affected are those classes of lipids of which myelin is composed. Here, too we may have a clue as to the mechanism, or one of the mechanisms, responsible for the lower brain-weight and the impaired function in undernourished subjects. One interesting point that emerges from the studies on both men and experimental animals is that the brain-weight reduction, though definite and significant, is not as marked in degree as that of the body as a whole, or as those of other organs, such as muscles, spleen and liver (Platt et al., ’65; Brown, ’66; Guthrie and Brown, ’68; Chase et al., ’69; etc.). In the study of Kerpel-Fronius (’61), while the infant’s body weight was 50% below the expected value, the brain-weight was only 10-20% reduced. Thus, although the absolute brainweight was depressed, the brain,/body weight ratio is elevated in such caseswhich perhaps provides one further indica- 12 PHILLIP V. TOBIAS tion of the relative meaninglessness of the brain/body weight ratio! It seems that some protective mechanism may be operating which gives the brain priority in access to nutrients during periods of undernutrition (Hammond, ’44; Kerpel-Fronius, ’61 Nevertheless, enough has been said to show that in any population widely exposed to varying degrees of undernutrition, it would be reasonable to expect that heads and brains would be signiikantly smaller, on the average, than those of a population on a good average level of nutrition. The widespread incidence of varying degrees of suboptimal nutrition in the African population is well-known (cf. Gillman and Gillman, ’51; Tobias, ’58; Brown, ’66; Smit, ’69). I am referring here not merely to an extreme form, such as kwashiorkor, but to lesser and subtler grades of undernutrition, such as have been uncovered by the multidisciplinary study of Pretoria schoolchildren of African, Asian, Coloured and European origin, by the National Nutrition Research Institute of South Africa (Smit, ’69). The relevance of these malnutritional effects for the interpretation of the mean brain-weight of the Negro, and of other poorly-nourished and emergent peoples, is obvious: if lower mean brain-weights can be validly demonstrated (and they have not yet been validated), the degree to which undernourishment is responsible for the shortfall would have to be considered, before such differences are lightly ascribed to racial or genetic differences. BRAIN-SIZE AND THE INFLUENCE OF THE (NON-NUTRITIONAL) ENVIRONMENT Is there a relationship between brainsize and structure on one hand, and the degree of challenge, of enrichment, or of impoverishment of the environment on the other? Although there is little direct evidence for man, indirect evidence of several kinds is available. Marian Diamond and her co-workers at Berkeley (’64, ’66) have studied the brains of rats reared in an enriched, stimulating environment and of their littermates reared in an impoverished and isolated environment. Definite structural differences have been demonstrated between the brains of the two groups. Those from the enriched milieu showed a significantly deeper visual cortex and a higher glia-neuron ratio than those from the isolated environment. Unforunately, I have not been able to trace any reference to brain-size measurements made in such experiments. At any rate, one can say clearly that the “educational environment” of a growing rat influences the fine internal structure of the cerebral cortex. A second line of evidence is the demonstration that closely related animals may differ in brain-size, according to the amount of activity and the challenges, or protective influences, emanating from their environments. Thus, Poliakova (’60) has shown that the agile, climbing, red fieldmouse possesses a larger brain-weight than the relatively sluggish, grey field mouse, although both are classified as belonging to the same species. Another series of studies on dogs, pigs, rabbits, rats, guinea-pigs, cats and ducks showed that domestication causes a decrease of about 20%, in some cases as high as 30%, in brain-weight (Herre, ’58). Klatt (quoted by Herre, ’58) showed that when wild animals are kept in captivity from an early age, their brain weight decreases markedly. Conversely, asses which have been allowed to run wild on the pampas of Argentina and Peru show an increase in brain size of about 15%, A similar figure obtains for cats which have been allowed to run wild (Klatt). It would seem that the docile life, in which animals are fully protected from their enemies, provides a similar absence of environmental stimulation to the developing brain of young animals, as obtains with Marian Diamonds isolated rats reared in an impoverished environment. One cannot help thinking of the enriched environment in which the ‘?laves” are reared and the deprived and neglected upbringing of the ‘3ave-nots.” There is a distinct possibility that such differences in the early home-upbringing and nursery school facilities between members of different populations may contribute towards structural and even size differences in their brains. Here is one factor, which, as far as I can determine, has not been examined in studies on human brain-size structure. BRAIN-SIZE, GREY MATTER AND RACE BRAIN-SIZE AND THE SOURCE OF THE SAMPLE Yet another variable which may influence the average brain weight is the source from which the brains were obtained. As Pakkenberg and Voigt (’64) have recently pointed out, most of the series published have been from hospitals, including mental hospitals (as is the case with the largest series, that of Appel and Appel (‘42)). This implies that, with relatively few exceptions, the brain weight of the normal or healthy human being is not known! The only real exceptions are brains which have been obtained through institutes of forensic medicine, i.e., following sudden death without prior disease. Only three such series have been traced in the literature, namely those of Matiegka (’02) from the Forensic Institute in Prague, of Pakkenberg and Voigt (’64) from the Neuropathological Laboratory of the Kommunehospitalet and the University Institute of Forensic Medicine in Copenhagen, and of Spann and Dustmann (’65) from the Institut fur Gerichtliche Medizin and Versicherungsmedizin at the University in Munich. No Negro series from such a source has been published: we do not know the brain weight of the normal, healthy Negro! Vint (’34), whose low average for the Negro brain (1276 gm) has been most often quoted, obtained his material from hospitals in Nairobi, excluding the Mental Hospital. His series therefore did not fulfill the ideal requirement of sudden death without prior disease. Although many hospital patients may suffer and die from diseases which probably do not affect brain weight (and even on this point we lack precise information), hospital populations as such are often not a good cross-section of all levels of society. They tend to reflect the lower social strata, while the upper strata tend to go to private nursing homes or to receive medical care at home. Vint even says of his Kenya African series, “These 100 brains represent the average native population, but do not include any of the so-called educated class” (’34, p. 216). Even bodies which are studied and dissected in Medical Schools are a far from random sample. The law caters for two main groups of subjects: those who die as 13 paupers in State Institutions - in South Africa these tend to be largely Africans; and those who bequeath their bodies to the Medical Schools - and these people are commonly better educated and often of the higher social strata. As Burial Societies come more and more within the reach of the poorest sections of the community, fewer and fewer people die as unclaimed paupers: and so gradually the face of the dissecting room population itself changes. This phenomenon is particularly vividly to be seen at present in the Anatomy School of the Witwatersrand University. T. Wingate Todd (’27) has recounted how dramatic an effect an economic depression had on the dissection room population; all sorts of people took to dying as unclaimed paupers who had previously been numbered among the university’s financial benefactors. The average brainsize of the cadaver population increased. Todd entitled his address, “A liter and a half of brains!” We may conclude that no series of brainweights is available for healthy, normal Negroes who have died a sudden death; no Negro series can therefore be compared validly with the best available European series from Forensic Medical Institutes. Brain-size and occupational groups Closely related to the source from which the brain was obtained is the question of the occupational groups represented in the sample (chart 5). Matiegka (’02) showed an association between brain-weight and occupation in his studies at the Forensic Medical Institute in Prague. Results obtained in the recent study of Spann and Dustmann (’65) tally remarkably closely with those of Matiegka. Thus, in both studies, unskilled workmen had a mean brain-size of just over 1400 p, and the academic, the professional and the senior official group of just on 1500 gm. Spann and Dustmann found some correlation among women: three groups of skilled and unskilled workers, handworkers, and lesser officials all had mean brain-weights between 1300 and 1315 gm, while six female academics had the absurdly high mean value of 1422 gm (just within the 14 PHILLIP V. TOBIAS CHART 5 Brain-size and occupation Matiegka-Prague, '02; Spann and Dustmann, Munich, '65 Average brain size Occupation gm Unskilled workmen 1400 Academics, professional men and senior officials 1500 Paul Broca, Paris, 1865-70 Unskilled labourers 1365 Semi-skilled and skilled workers 1420 But the latter were taller than the former1 Question: Has there been a secular trend towards increased brain size, in parallel with the secular trend towards increased stature? lower end of the range of occupational means among males!). The tendency is apparently not a new one: Broca's results from the late 1860's yielded a mean brain-weight of 1365 gm for 51 unskilled labourers and of 1420 gm for 24 semi-skilled and skilled workers: the difference is almost significant at the 5% level (Schreider, '66). Schreider comments, however, that the differences in brain weight between the occupational groups are probably related to differences in body height, for the skilled and semiskilled group were on the average 1.34 cm taller than the unskilled group. The degree to which any sample of brains represents, or departs from, a cross-section of occupations may thus affect the validity of any comparisons between such a sample and any other. Brain-size and cause of death In a large hospital population, Appel and Appel ('42) found that the mean weight of the brain was highest (1374 gm) in cases of violent and accidental deaths. On the other hand, patients who died of arteriosclerosis and cancer had average brain weights of 100 gm less (1273 and 1275 gm respectively). For other categories of causes of death, the average weights lay between these two extremes. Of the various disease groups, tuberculotics showed the highest mean weight (1328), that is, 55 gms above the mean for arteriosclerotics and 46 gms below the mean for victims of accidental and violent deaths (chart 6). In the large Copenhagen medico-legal series of Pakkenberg and Voigt ('64), the highest mean brain weights were found among those who had died from hanging, poisoning (mainly by carbon monoxide or barbiturates) and shock: similar results were observed in the Munich series of Spann and Dustmann. Low values, on the other hand, were encountered in those who had died following brain diseases, general pathology, head trauma, drowning, asphyxiation, electrocution and protracted haemorrhage. CHART 6 Brain-size and cause of death S t . Elizabeth's Hospital, Washington, D. C . (Male white data - after Appel and Appel '42) Cause of death Number Average brain weight 54 533 200 284 1023 955 1374 I328 1315 1307 1300 1295 430 112 358 1285 1275 1273 gm Violence/accidents Tuberculosis Urogenital diseases Digestive diseases Respiratory diseases Circulatory disease (incl. Arteriosclerosis1 Nervous diseases Cancer Arteriosclerosis ZIt is doubtful however whether this relatively small difference in stature is adequate to explain all the difference in brain-weight between the two groups. On Spann and Dustmann's figure of 8.3 gm of brain for e v e n centimeter of bodv heizht in men. the difference of'1.34 cm might account-for some i l . 0 g m 04 brain-weight difference: this would reduce the discrepancy to non-significant levels (with these small samples), but there remains a shortfall of 44 gm be7 tween the two groups. BRAIN-SIZE, GREY MATTER AND RACE Again, in his study of Broca’s centuryold data, Schreider (’66) showed that those who died from accidents had a mean brain weight 60 gm greater than those who had died of infectious diseases. It seems that even 100 years ago, accidents were common in the streets of Paristhough they were seemingly often owing to horse kicks. Thus, whether death is lingering or sudden, disease processes themselves may influence brain weight at death and, indirectly, the rate of increase in brain weight with each passing hour after death. It is well known that the patterns of fatal diseases differ appreciably among M e r e n t populations, especially where one group lacks adequate medical and hospital facilities (as is the case with many African Negro populations). Thus, two groups may show differences in average brain-weight simply because of a different pattern and incidence of fatal diseases in the two groups. Such a factor may account for some of the discrepancy between, for instance, two Negro series, one in Africa with an average brain-weight of 1276 gm and one in America with an average of 1355 gm. Similarly, some of the differences found in interracial comparisons may be accounted for by such differential diseasepatterns. Brain-size and the lapse of time after death Another source of variation in the average brain-size is the time that elapses between death and the removal of the brain (chart 7 ) . Chemical changes go on in the brain for several days after death, tending towards an equilibrium that is appreciably heavier than the original weight (Appel and Appel, ’42). Thus, if left for 24 hours after death, the brain weight is no less than 40 gm heavier than if removed within CHART 7 Time after death Brain weight g m 24 hours 48 hours Maximum change +++4090 75 (+9%) 15 a few hours of death; 48 hours after death, it weighs on the average 75 gm more. The average maximum increase in weight is 90 gm, whilst the increase from the lowest mean weight to the highest is no less than 117 gm,or 9% of the average weight for the total array. Statistically and biologically, this is a significant and appreciable increment. It is almost as great as the decrease with age ( 1 1% ) and it is greater than most of the recorded differences between the averages for various racial groups, including Negroes and Whites. Then, too, the temperature at which the cadavers are kept between death and removal of the brains may affect the rate of chemical change and, so, the increase in weight after death. Brain-size and the treatment of the brain after death “To weigh the human brain is not as easy as it sounds” (Bailey and von Bonin, ’51). Thus, the measured weight will depend upon where exactly one severs the brain from the spinal cord. It is an agreed definition that everything cranial to the pyramidal decussation is “brain”; but the decussation is not a point: it extends for some little distance down the central nervous system. Different workers may sever the brain from the cord at a systematically higher or lower level (Bailey and von Bonin, ’51). Other technical points on which no standardization has been achieved include whether and how one drains the cerebrospinal fluid from the brain; Brandes (’27) stated that the complete drainage of the cerebrospinal fluid from the fresh brain may change its weight by as much as 50 gm. Again, there is the question: which of the covering membranes (or meninges ) one removes from the surface of the brain and which of them one includes; Brandes (’27) determined the weight of the outer, thick covering, the dura mater, as no less than 50-60 gm. Should one strip the bloodvessels from the brain surface? Inconsistency between workers in every one of these points may import additional variance in the results obtained by different workers on different races or populations. Some studies in the literature (e.g., Pakkenberg and Voigt, ’64; Spann and Dust- 16 PHILLIP V. TOBIAS mann, ’65) carefully define their methods; in others, it is stated simply that the brains were weighed. As in other branches of science, standardization of technique and precise description of the methodology followed in each study must obviously precede any meaningful comparisons between any two sets of data, On the meaninglessness o f most studies on brain-weight For all these reasons, it is not surprising that contradictory results appear in the scientific literature for the brain-sizes of Negroes and Whites. We have seen that brain-size may vary according to body size, the age at death, the cause of death, the selection of the sample, the sex, the early environment, especially nutritional, the lapse of time after death, the level at which the brain is severed from the spinal cord, the mode of drainage of the cerebrospinal fluid and the treatment of the covering membranes and blood-vessels; other factors may intrude as well, such as the temperature at which the body is kept between the time of death and the removal of the brain (chart 8). These factors may cause fluctuations in the measured brain-size up to 9% too high if the brains are left for several days after death, or 11% too low if the sample of brains comes from elderly subjects. This order of variability, stemming from only one or two of these possible variables, is greater than most of the supposed interracial differences which have appeared in the literature. The ideal sample is from subjects who have died CHART 8 Measured brain size varies with: - 1. Sex 2. Body size 3. Age at death 4. Nutritional state in early life 5. Non-nutritional environment in early Iife ( ? ) 6. Source of sample 7. Occupational group a. Cause of death 9. Lapse of time after death 10. Temperature after death 11. Anatomical level of severance 12. Presence or absence of C.S.F. 13. Presence or absence of meninges 14. Presence or absence of blood-vessels suddenly without prior disease, as from vie lence and accidents. No such study has been recorded for Negro subjects. It must be concluded that among the widely varying estimates for Negro brainsize, none reflects the normal brain-size of a healthy Negro group. W e do not know the brain-size of healthy Negroes who have died suddenly without prior disease. Few of the published studies on any racial group have been corrected for body size, age at death and the lapse of time after death; it is clearly difficult, if not impossible, to correct for all the other factors enumerated. These considerations largely invalidate most interracial comparisons so far published. Certainly, it is invalid at this stage in our knowledge to claim, as Swan (’64) and Putnam (’63) have done, that the average brain weight of European Whites is some 8-12% greater than that of African Negroes. Equally misleading is the bald statement of Gates (’46) that “Comparative weights are given as 1380 gm for Caucasoids, 1300 gm for Mongoloids, 1280 gm for East Africans, 1240 gm for Negroes, and 1180 gm for Australian Aborigines.” These figures are meaningless without any attempt to consider the body size of the various groups, let alone all the other variables. A fortiori, it is totally invalid at this stage of our knowledge to cite the alleged smaller brain-weight of Negroes as the physical basis for differences in intelligence and behaviour tests. W h a t do differences in brain-size mean? It has been known for long that in living men, enormous variations in brain-size may occur within the limits of normal healthy functioning. Even well-known men of high achievement may vary considerably (chart 9). Thus, the great French writer, Anatole France, is reported to have possessed a brain weight of as little as 1017 gm (or an endocranial capactiy of about 1100 cm3); similar in size were the brains of Franz Joseph Gall, the German anatomist, physiologist and founder of the pseudo-science of phrenology; and of the French statesman, Ikon Gambetta (Dart, BRAIN-SIZE, GREY MATTER AND RACE CHART 9 Brain size and achievement. A pot-pouri-i of big and small bTains Brain weirrht gm 1000-1 100 Anatole France Franz Joseph Gall Leon Gambetta 1282 Walt Whitman ‘-c 1900 2230 Daniel Webster Ivan Turgenev Dean Swift Otto Von Bismarck Oliver Cromwell George Gordon, Lord Byron ’56); while only slightly larger was the brain of the great American poet, Walt Whitman (1282 gm) (Cobb, ’65). At the other extreme, George Gordon, Lord Byron, is said to have had a brainweight of 2,238 gm and Oliver Cromwell of 2,231 gm (Keith, ’12, cited by Dart, ’56). Brain-weights of about 1900 gm were possessed by Daniel Webster, the American lawyer, statesman and orator; by Ivan Turgenev, the Russian novelist; by Dean Swift, the English satirist, and by Bismarck, the Prussian statesman and creator of the German Reich (Dart, ’56). Thus, considerable human achievement is possible at very nearly both ends of the scale of modern human brain-size. Further, as Dart (’56) has pointed out, apparently normal human beings have existed with brain-sizes in the 700’s and 800’s.Schlaginhaufen (’50-’51) reported the skull of a Melanesian woman from one of the Feni Islands in the Bismarck Archipelago with an endocranial capacity of only 790 cm3. As far as Dart (’56) could trace in the literature, this is the smallest capacity yet attested for a normal human skull - for it was not the skull of a pathological microcephalic! This capacity was found not in an ape-man or a pre-man, but in an individual classifiable as a member of our living species, Homo sapiens (Dart, ’56); yet her capacity was a mere 38 cm3 greater than the maximum capacity recorded for an anthropoid ape, namely a gorilla with a cranial capacity of 752 cm’ (Schultz, ’62). 17 Thus, normal human beings exist with brain-sizes three times that of other human beings. So fantastic a range would single out brain-size as one of the most highly variable parameters in modern man. Perhaps only body-weight exceeds brain-size in the discrepancy between the top and bottom of the range in normal healthy human beings. What then do the W e r e n t brain-sizes in normal men signify? Do people with bigger brains have more nerve-cells, or bigger nerve-cells, or more neurogkia (the non-neural elements which lie between the neurons), or more nerve-processes, or longer nerve-processes, or tnicker nerveprocesses, or more highly-branched nerveprocesses? Or are larger brains larger through a combination of any two or more of these seven possible variables? We simply do not know the answers to these questions for modern human brains within the range of normality. We shall see in a moment that we have a little more information when it comes to comparisons between different animals of different average brain-sizes. But within our species, we do not know what the microscopic or cellular basis is of varying brain-sizes. If we do not know that simple physicophysical correlation, how can we hope to make meaningful statements about the correlation between gross brain-size and cellular structure on one hand, and about psychical and behavioural attributes on the other? For the physico-physical correlation is basic to the physico-psychical association. We do not have the requisite information at either level. No wonder that a leading neurologist, Gerhardt von Bonin (’50), was led to remark, “The correlation between brain-size and mental capacity is insignificant” and, again, “Brain size as such is none too meaningful.” Similarly, Holloway (’68), moving outside the human species, has concluded that “Gross size of the brain alone does not explain differences of behaviour within the primate order.” He is at pains to point out that “such correlations (as between brain-size and specific behavioural traits like memory, insight, forethought, symbolization) are not causal analyses, and that a parameter such as 18 PHILLIP V. TOBIAS We are back at square one. brain weight in grams, or volume in milliliters, or area in square millimeters, canWe must confess our ignorance of the not explain the differences in behaviour functional meaning and value of different which are observed.” (op. cit., p. 125) sized brains in modern human individuals. (italics mine). A more encompassing Will a long-term evolutionary look at the theory, Holloway states, should entail not problem throw any light on it? merely the changes in brain weight which Brain-size, evolution and survival have occurred in evolution, but the internal reorganization of the cellular maWe showed above that, when allowance terial of the brain. It is precisely at this is made for body size, the differences in the level that we are most ignorant. computed numbers of “excess” nerve-cells If we compare different species of mam- in the cerebral cortex of various racial or mals, many studies have demonstrated population groups are negligible - despite that bigger brains are correlated with variations in the source, composition and clearly defined cellular and chemical fea- treatment of the brains. Even if further tures. For instance, the bigger the brain, studies did demonstrate that one group had the lower is the density of nerve-cells in fewer surplus nerve-cells on the average that brain (Nissl, 1898; von Bonin, ’48; than another, how valid would it be to Tower and Elliott, ’52; Shad€, ’53; Tower, jump to the conclusion that fewer surplus ’54). Further, it has been claimed that nerve-cells automatically connote lesser neurones are bigger and nerve-cell proc- cerebral development (cf. Vint) or lesser esses longer and more complex in bigger evolutionary advancement (cf. Gates)? brains. The glia-neurone ratio is likewise Putnam (’63) assumes that there is no dishigher. The claims and their validity have been well summarised by Holloway (’68:). pute on this question, but his assumption As he points out, an increase in dendritic is far from justified. When one takes a long-term evolutionbranching means more synapses, more connectivity, and with this goes more com- ary look at this problem, it seems clear that increasing brain-size (and the organizaplex beh aviour. On this kind of analysis of W e r e n t tion, fine structure and chemistry that went species, increase in brain-size is coming with it) may once have been vitally imto be meaningfully analysed in terms of its portant in aiding survival - for instance structural units and these, in turn, may in a world of wild animals and ape-men, provide a more rational basis for under- bereft of fire. It seems, too, that the furstanding increasingly complex behaviour. ther development of man has placed less But all this has been shown to apply in an and less of a premium on the size of brain, assumed evolutionary progression from the numbers of nerve-cells and other inone form to another as one passes up the ternal organizational details. For during scale. To a lesser extent, similar changes his evolution, culture and the benevolence have been shown to apply to the ontoge- of social life have taken the place of nimnetic development of individuals within ble wits as an insurance policy against a species. At the adult level, however, I extinction. Beyond a certain stage in the reiterate the view expressed earlier: we do increase of brain-size, we have no evinot have any clear picture of the histologi- dence that further increase in any way imcal and chemical differences between big proved man’s adaptive abilities. For aught and small brains among members of the we know, the slight preponderance of same species. Therefore, we cannot pin- brain-size which is claimed for Caucasoid point any cellular and chemical differences or White men of today may, if it exists as between big and small brains which would a genetic feature at all, be a somewhat indicate a basis for different behaviour. Are superfluous and gratuitous heritage from there differences in behaviour between in- Stone Age ancestors, like Cro-Magnon and dividuals with big brains and with small Neandertal men. For these ancestral men brains? Evidence for such differences does had bigger brains on the average than their present-day European descendants. not seem to exist. BRAIN-SIZE, GREY MATTEK AND RACE In fact, there is some evidence to suggest that the trend towards increased brainsize, which marked the first two million years of human evolution, has spent itself; the wave of brain expansion has passed its peak. This is true, as I have just said, of early and modern Europeans; to a certain extent, it may be true of Africa as well. Some of our Stone Age men had carried the twin processes o€ reduction of teeth and jaws, on the one hand, and expansion of brain, on the other, so far that they seem to represent an ancient foreshadowing of the popular idea of the man of the future. At any rate, that is the view of them adopted by Professor Loren Eiseley of Philadelphia, in his delightful and penetrating book of essays called “The Immense Journey” (’58). Listen to the way in which he describes the man of the future: “I have stared so much at death that I can recognize the lingering personalities in the faces of skulls and feel accompanying affinities and repulsions “One such skull lies in the lockers of a great metopolitan museum. It is labelled simply: Strandloper, South Africa. I have never looked longer into any human face than I have upon the features of that skull. I come there often, drawn in spite of myself. It is a face that would lend reality to the fantastic tales of our childhood. There is a hint of Wells’s Time Machine folk in i t those pathetic, childlike people whom Wells pictures as haunting earth’s autumnal cities in the far future of the dying planet. “Yet this skull has not been spirited back to us through future eras by a Time Machine. It is a thing, instead, of the millennia1 past. It is a caricature of modern man, not by reason of its primitiveness but, startlingly, because of a modernity outreaching his own. It constitutes, in fact, a mysterious prophecy and warning. For at the very moment in which students of humanity have been sketching their concept of the man of the future, that being has already come, and lived, and passed away.” (Op. cit., pp. 127-128.) ... 19 ‘Their voices,” says Eiseley, “‘ring with youthful confidence, the c o d dence engendered by my persuasive colleagues and myself. At times I glow a little with their reflected enthusiasm. “I should like to regain that confidence, that warmth. I should like to but “There’s just one thing we haven’t quite dared to mention. It’s this, and you won’t believe it. It’s all happened already. Back there in the past, ten thousand years ago. The man of the future, with the big brain, the small teeth. “Where did it get him? Nowhere. Maybe there isn’t any future.Or, if there is, maybe it’s only what you find in a little heap of bones on a certain South African beach. “Many of you who read this belong to the White race. We like to think about this man o f the future as being White. It flatters our ego. But the man of the future in the past I’m talking about was not White. He lived in Africa. His brain was bigger than your brain. His face was straight and small, almost a child’s face. He was the end evolutionary product in a direction quite similar to the one anthropologists tell us is the road down which we are travelling.” (Op. cit., p. 129-130.) . .. In Africa, no less than in Europe, it seems a corner has been turned, a peak has been surmounted and something of a downslope presents itself. Perhaps, we can put it this way: the selective pressures which once placed a considerable premium on big brains have been somewhat relaxed. Perhaps, we have reached a stage in human evolution, and I am convinced we reached it thousands of years ago, where 100 people with smaller brains stand just as good a chance of surviving to childbearing age as 100 with bigger brains, and are likely to leave no fewer children than the others. Brain-size is no longer a yardstick to survival as it may once have been. We have used these very brains to develop new mechanisms of adaptation, tools, shelters, clothing, fire, social institutions and central heating, air-conditioning, refrigeration, disinfection, mink coats and When Eiseley shows his students in sun shades. You can have these things with Philadelphia a picture of what the man of a size 5.50 hat or with a size 8.50, just as the future may be expected to look like, it makes no difference what size shoe you they say, “It’s O.K. Somebody’s keeping an take. eye on things. Our heads are getting bigBrain-size seems to make no difference ger and our teeth are getting smaller. to your ability to avail yourself of the joys Look!” of modern living. So, too, it seems that - 20 PHILLIP V. TOBIAS publications mentioned, although the shortfall claimed by Vint for one other layer of the cortex, the internal granular layer, is greater. Yet, when one goes back to the original sources, one is left with serious doubts about the validity of the data on which the comparison was based. First, let us look “Certainly, the weight of the brain is at what Vint actually reported: he claimed a very poor indicator of its functional value” and, again, ‘Brain size as such that the average total reduction of the i s a very poor indicator of mental abilwhole cortex, as compared with that of the ity.” European, is 14.8% in the cortical areas A century earlier, James Hunt (1863) had examined. His figures for the Negro were stood before the Anthropological Society of based on one single study of Negro brains, London and declared, which, he states, did not include any of . . we know that it is necessary to be the “so-called educated class.” He did not most cautious in accepting the capacity study any European brains himself, alof the cranium as any absolute test of the intellectual power of any race” (p. though at least two of the references 13). quoted give the impression that Vint studIn the face of these individual variations ied European brains directly. Thus, Swan of brain-size and achievement within a (’64) speaks of “Dr. F. W. Vint’s careful single race, in the face of a decline in aver- comparative studies of the cortical histolage brain-size over the last tens of thou- ogy of the brains of European Whites and sands of years, in the face of our knowl- African Negroes” (p. 27). Similarly, and edge that a large part of man’s adaptabil- equally erroneously, Putnam (’67) states, ity to new environments and to new ways “In 1934 F. W. Vint of the Medical Reof life is culturally determined,- who search Laboratory, Kenya, Africa, pubwould confidently assert that slightly lished the results of a Comparative study smaller brains could be any deterrent what- of Negro and European brains in which he ever to achievement, or set any brake upon found that the supragranular layer of the mental capacity? Negro cortex was about 15 per cent thinner than the Whites” (p. 51) (italics mine). T h e amount of grey matter Thirty five years ago, Vint (’34) exam- In both quotations, the wording conveys ined the grey matter of the cerebral cortex the impression that Vint had himself made of Kenya Africans under the microscope a careful study of the cortex of European and claimed that it was narrower on the brains, as well as of Negro brains, whereas average than that of the European. This he had in fact studied only Negro cortices work has often been quoted in publica- and compared his results with those obtions distributed by the Putnam Letters tained in one other study, by a different Committee: e.g., by George (’62), by his worker, namely von Economo (’29) in reviewer, Sanborn (n.d.), by Swan (’64) Europe. In eight different regions of the cortex, and by Putnam (’63, ’67), as well as by Hofmeyr (’61). More particularly, this he measured the total cortical thickness oup of writers has drawn attention to and the thickness of each of four layers e supragranular layer of nerve-cells in comprising the total cortex, namely the the cortex, which they claim Vint found to lamina zonalis, the supragranular zone, be 14% (Sanborn, n.d.), more than 14% the internal granular zone and the infra(Putnam, ’63, p. 9), about 15% (Putnam, granular zone. Each of the five measure’67,p. 51), about 14% (George, ’62, p. ments in seven out of eight regions was 33) and 16% (Swan, ’64, p. 28) smaller then compared with a corresponding measthan that of Europeans. This layer is held urement as given in von Economo’s (’29) by some to be the “most advanced” layer in book, and a percentage increase or dethe cortex; and that is presumably why crease recorded, as compared with “the its supposed shortfall is stressed in all the European brain.” brain-size does not limit your ability to contribute to society, culture, science. Some gifted people have had very small brains. Others, also gifted, have had large brains. And some very ordinary persons had equally large brains. Small wonder that Gerhardt von Bonin could say in 1963. “. 21 BRAIN-SIZE, GREY MATTER AND RACE There are many possible sources of error in these comparisons. First, Vint himself points out that there is even some doubt among different workers as to where exactly to cut the little section of tissue to measure the thickness of the layers whether from the tops of the convolutions of grey matter, or from their sides, or from the depths of the fissures between them. Vint states that Bolton (’03, ’14) had recorded measurements which were the average in each case between thickness readings (i) on the flat external surface of the convolution; (ii) at the lip of the adjacent sulcus or fissure; (iii) OR the side of the fissure; and (iv) at the bottom of the fissure. Von Economo (’29) pointed out that the thickness of the cortex on the summits of the convolutions was twice as great as in the floor of the fissures, but unfortunately he did not make clear whether he had followed Bolton’s averaging method or some other approach, Thus, Vint makes clear that he is not exactly certain whether von Economo’s measurements on European brains were made in the same area as were his (Vint’s) measurements on Negro brains, namely only on the summit or crown of the convolutions. Secondly, the actual measuring of the layers of the cortex is not as easy as it may sound. Although the various layers are fairly readily identified, the boundaries between them are not. The boundaries are frequently wavy and exactly where to set the termini for a particular measurement may vary from worker to worker, even in the same laboratory. It would be easy for one worker consistently to measure from a different part of such a wavy boundary, compared with another, and so to introduce a systematic error into his results. Vint had received no specific training under von Economo, as far as we know, and there is no reason to believe that his technique of making measurements would have been precisely the same as that of von Economo. Furthermore, Vint nowhere states how his measurements were made, nor are the boundaries of the layers marked in on his photographs, in contrast, say, with the illustrations on the human cortex by Bailey and von Bonin (’51) . A third most serious difficulty is the technical procedures employed. To prepare a brain for microscopical study requires a number of chemical and staining procedures, a variety of which is available to workers in most laboratories. Different chemical treatments are known to produce different degrees of shrinkage or swelling. These influences readily affect measurements, especially when one is dealing with a total cortical thickness as small as 1.886 to 3.006 mm (the range of average cortical thicknesses found by Vint in 8 different areas of the brain). As an illustration, Diamond et al. (’64, ’66) measured the thickness of the rat’s visual cortex by two different techniques: frozen sections stained by thionin and elloidin sections stained by Windle’s modified Nissl stain. The thickness measurements by the second method were only 60.5% of those by the former. Even within one Laboratory, with the use of only a single technique over and over again, most variable results may be obtained in the state of preservation (fucation) and staining of brains. As Carothers (‘53,p. 82) has pointed out, “Comparative assessments of cerebral histology are notoriously difficult and require special knowledge of this type of work. .” Here, then, we have the unsatisfactory position of a series of h e measurements made on brains treated in one laboratory by one worker and derived from Negroes, being compared with the results of another series of fine measurements made by another worker on European brains prepared in a different laboratory, possibly under very different conditions. And Vint’s are the only data on record for Negro cortical thickness1 If we left the matter there, anyone with a logical mind would realise that no conclusions could validly be drawn about alleged differences between races culled from such unsatisfactory material. To quote Carothers again, “Comparative assessments of cerebral histology . . are of little value d e s s the preparation of material and the techniques that are used are virtually identical for the two groups. These requirements were not convincingly fulfilled (in Vint’s comparisons)” (op. cit., p. 82). . . 22 PHILLIP V. TOBIAS Vint himself points out a further serious shortcoming. He states specifically, “Owing to climatic and other conditions, it was impossible to fix the brain by injection of the carotid arteries.” If all these sources of variation apply to the thickness of the entire cortex, how much more strongly do they apply to the supragranular layer alone, which Vint found to vary in average thickness from 0.338 to 0.922 of a millimetre in six different parts of the cortex! (chart 10). It is clear that Vint’s study, which has been so widely quoted, permits only one definite coriclusion about cortical thicknesses, namely that there is a need for more carefully controlled studies. Yet, we b d the alleged 15% difference in total cortical thickness and the supposed 16% shortfall in the supragranular layer being quoted and re-quoted as established facts, as further evidence of the supposed inferiority of the Negro’s brain. What is more, the inferences drawn from these so-called facts are used to Create the impression that the Negro is a less highlyevolved being as for instance when Putnam (’63, p. 9) states, ‘The thickness of the supragranular layer of the cortex, which is found to increase as we move up the scale from animal to man, may thus be said to be another measure of evolutionary development.” We are led to conclude that there is no scientifically acceptable proof that the cerebral cortex of Negroes is thinner in whole, or in any layer, than that of Europeans. - CONCLUDING THOUGHTS ON BRAINS AND RACES I have singled out only two of the socalled well-established differences between the brains of Negroes and of Whites, namely brain-size and the thickness of the CHART 10 Range of thickness measurements (Vint, ’34) ACKNOWLEDGMENTS Cortex 1.886 - 0.338 - 3.006 cortex, and shown that they provide highly unsatisfactory evidence of structural differences in the Negro’s brain, and of differences in intellectual capacity. Other claims have been made, such as one about the size of the frontal lobe and the degree to which it is broken into convolutions, claims which have been denied equally strenuously by many other workers. From my little venture into the study of the brain, I have emerged with the conviction that vast claims have been based on insubstantial evidence. I conclude that there is no acceptable evidence for such structural differencesin the brains of these two racial groups; and certainly nothing which provides a satisfactory anatomical basis for explaining any difference in I.&.or in other mental and performance tests, in temperament or in behaviour. The exploding of the myth about brainsize and grey matter differences leads to the realisation that, in science, the truth does not gain acceptance by mere repetition of a set of facts. Hypotheses are not confirmed by statement and re-statement of the hypotheses. Science requires rather the patient testing of the facts, the repeating of early studies by more modern, better controlled and better standardized methods, the constant re-examination and critical re-appraisal of premises and assumptions, the elasticity of mind which permits, nay demands, old hypotheses to be modaed when new facts emerge which cannot be adequately explained by them, resistance to the tendency to develop a vested interest in a particular viewpoint, avoidance of ascribing motives to scientists of opposite viewpoint, in favour of the unbiassed examination of the evidence they may advance, the eschewing of premature hypotheses based on over-tenuous evidence. It is to these stern and rigorous demands of scientific method and to one who proved himself a trenchant devotee of this philosophy that my address is dedicated. millimetres Supragranular k y e r 0.922 millimetres I should like to express my gratitude to Mrs. E. Hibbett, Mrs. E. Judd, Miss C. Orkin and Mrs. S. Boles; and to Brian Greis and Brian Hume. 23 BRAIN-SIZE, GREY MATTER AND RACE LITERATURE CITED Abelson, P. H. 1969 Malnutrition, learning and behavior. Science, 164: 17. Appel, F. W., and E. M. Appel 1942 IntracraniaI variation in the weight of the human brain. Human Biology, 14: 48-68 and 235-250. Bailey, P., and G. von Bonin 1951 The Isocortex of Man. Urbana: University of Illinois Press. pp. 1-301. Baraitser, M., and D. E. Evans 1969 The effect of undernutrition o n brain-rhythm development. s. Afr. Med. J., #: 56-58. Bean, R. B. 1917 The weights of the organs i n relation to type, race, sex, stature and age. Anat. Rec., 11: 326-328. Bolton, J. S. 1903 The histological basis of amentia and dementia. Archiv. Neurol., 2: 424-620. 1914 The Brain in Health and Disease. London: Arnold, Brandes, K. 1927 Liquorverhatnisse a n der Leiche und Hirnschwelling. Frankfurt. Ztschr. f. Path., 35: 274-301. Brown, R. E. 1966 Organ weight in malnutrition with special reference to brain weight. Develop. Med. Child. Neurol., 8: 512-522. Carothers, J. C. 1953 The African Mind in Health - a n d Disease. World Health Organization Monograph Series, No. 17. Chase, H. P., W. F. B. Lindsley and D. O'Brien 1969 Undernutrition and cerebellar development. Nature, 221: 554-555. Cravioto, J., and B. Robles 1965 Evolution of adaptive and motor behaviour during rehabilitation from kwashiorkor. Amer. J. Orthopsychiat., 35: 449-464. Cobb, S. 1965 Brain size. Arch. Neurol. (Chicago), 12: 555-561. Dart, R. A. 1956 The relationships of brain size and brain pattern to human status, S. Afr. J. Med. Sci., 21 (1 and 2 ) : 23-45. Davis, J. B. 1868 Contributions towards determining the weight of the brain in different races of man. Philos. Trans., 158: 505-527. (Cited by Pakkenberg and Voigt, '64.) Davison, A. N., and J. Dobbing 1965 Myelination as a vulnerable period in brain development. Brit. Med. Bull., 22: 4044. Diamond, M. C., D. Krech and M. R. Rosenzweig 1964 The effects of a n enriched environment o n the histology of the rat cerebral cortex. J. Comp. Neur., 123: 111-119. Diamond, M. C., F. Law, H. Rhodes, B. Lindner, M. R. Rosenzweig, D. Krech and L. Bennett 1966 Increases in cortical depth and glia numbers in rats subjected to enriched environment. J. Comp. Neur., 128: 117-125. Dickerson, J. W. T., and J. Dobbing 1967 Prenatal and postnatal growth and development of the central nervous system of the pig. Proc. roy. SOC.,B, 166: 384-395. Dickerson, J. W. T., and R. A. McCance 1967 Effect of undernutrition on the postnatal development of the brain and cord in pigs. Proc. roy. SOC.,B, 166: 396.407. Dickerson, J. W. T., and A. L. Walmsley 1967 The effect of undernutrition and subsequent re- habilitation on the growth and composition Of the central nervous system of the rat. Brain, 90: 897-906. Dobbing, J,, and E. M. Widdowson 1965 The effect of undernutrition and subsequent rehabilitation on myelination of rat brain as measured by its composition. Brain, 88: 357-366. Duckworth, W. L. H. 1904 Morphology and Anthropology. Cambridge University Press, xxviii 564 pp. Eavrs. 1. T., and G.Horn 1955 Development of cerebral cortex in hypothyroid and starved rats. Anat. Rec., 121: 53-61. Eichenwald, H. F.. and P. C. Fry 1969 Nutrition and learning. Science, 163: 644-648. Eiseley, L. 1958 The Immense Journey. London: Victor Gollancz. Engel, R. 1956 Abnormal brain-wave patterns in kwashiorkor, Electroenceph. and Clin. Neurophysiol., 8: 489-500. Fishman, M. A., A. L. Prensky and P. R. Dodge 1969 Low content of cerebral lipids in infants suffering from malnutrition. Nature, 221: 552-553. Gates, R. R. 1946 Human Genetics, New York: Macmillan. George, W. C. 1961 Race, Heredity and Civilization. (Cited by Hofmeyr, '61.) 1962 The Biology of the Race Problem. Report commissioned by the Governor of Alabama, 1-87. Gillman, J., and T. Gillman 1951 Perspectives in Malnutrition. New York: Grune and Stratton. pp. 1-584. Guthrie, H, A,, and M. L. Brown 1968 Effect of severe undernutrition in early life on growth, brain-size and composition in adult rats. J. Nutrition, 94: 419-426. Hammond, J. 1944 Physiological factors affecting birth weight. Proc. nutr. Soc., 2: 8-12. Herre, W. 1958 The i d u e n c e of the environment on the brain of the mammds. Dtsch. Med. Wschr., 83: 86. Hofmeyr, J. D. J. 1961 Gedifferensieerde ontwikkeling : h genetiese standpunt. Pretoria, University of Pretoria Branch of Afrikaanse Studentebond, pp. a-15. Holloway, R. L. 1968 The evolution of the primate brain: some aspects of quantitative relations. Brain Research, 7: 121-172. Hunt, 3. 1863 On the Negro's place in Nature. Anthrop. SOC.Lond., viii 60 pp. Hunt, S. B. 1869 Journ. Anthrop. SOC. Lond., 7: 40-54. (Cited by Pearl, '34.) Jackson, C. M.. and C. A. Stewart 1920 The effects of inanition in the young upon the ultimate size of the body and the various organs in the albino rat. J. Exp. Zml., 30: 8. Jerison, H. J. 1963 Interpreting the evolution of the brain. Human Biol,, 35 (3): 263-291. Kerpel-Fronius, E. 1961 Somatic stability in the newly born. Ciba Foundation Symposium. Krompecher, St., and J. Lip6k 1966 A simple method for determining cerebralization. Brain weight and intelligence. J. Comp. Neurol., 227: 113-120. Kummer, H. 1961 Beitrag zur quantitativen Bestimmung der Entwicklungshohe des Sauge- + + 24 PHILLIP V. TOBIAS tiergehirnes. Psychiat. Neurol., Basel, 142:352375. Kusumoto, M. 1936 On the biometrical constants of the Japanese brainweight. Jap. J. Med. Sci., 6(91) (abstract) (cited by Bailey and von Bonin, ’51). Lashley, K. S. 1949 Persistent problems in the evolution of mind. Quart. Rev. Biol., 24:2& 42. Latimer, H. B. 1950 The weights of the brain and its parts and the weight and length of the spinal cord in the adult male guinea pig. J. Comp. Neur., 93: 37-51. Mall,F. P. 1909 On several anatomical characters of the human brain said to vary according to race and sex, with special reference to the frontal lobes. Am. J. Anat., 9: 1-32. Matiegka, H. 1902 Uber das Hirngewicht, die Schadelkapazitat und die Kopfform, Sowie deren Beziehungen zur psychischen Thatigkeit des Menschen. S.B. kgl. bohmischen Ges. Wiss. Prag. math-nat. Cl., No. 20, 1-75. Mettles, F. A. 1955 Culture and the structural evolution of the neural system. James Arthur Lecture on the Evolution of the Human Brain. New York: American Museum of Natural History. Morton, S. G. 1849 Proc. Nat. Acad. Sci. Philadelphia, 4: 221-224. (Cited by Pearl, ’34.) Nelson, G. K. 1959 The electroencephalogram in kwashiorkor. Electroenceph. Clin. Neurophysiol., 11: 73-84. 1963 Electroencephalographic studies in sequelae of kwashiorkor and other diseases in Africans. 777-787. Proceedings of the Central African Scientific and Medical Congress, August 1963. In: Science and Medicine in Central Africa. George J. Snowball, ed. Pergamon Press. Nissl, F. 1898 Nervenzellen und graue Substanz. Munch. med. Wschr., 45: 988. Osler, W. 1968 Belated scars of war prisons. The Star, Johannesburg, 11 November 1968. Pakkenberg, H., and J. Voigt 1964 Brain weight of the Danes. Acta Anat., 56: 297-307. Peacock, T. B. 1865 Mem. Anthrop. Soc. Lond., 1 , 65-71 and 520-524. (Cited by Pearl, ’34.) Pearl, R. 1905 Biometrical studies in man. I. Variation and correlation in brain weight. Biometrika, 4: 13-104. 1934 The weight of the Negro brain. Science, 80: 431-434. Platt, B. S., G.Pampliglione and R. J. C. Stewart 1965 Experimental pmteincalorie deficiency. Develop. Med. Child Neurol., 7: 9. Poliakova, R. S. 1960 Interspecial differences in brain size. Arkh. Anat. Gistol. i. Embriol., 39: 58-64. (Read in abstract.) Putnam, C. 1963 Three new letters on science and race. New York: National Putnam Letters Committee, 3-23. 1967 Race and Reality: A Search for Solutions. Washington, Public Affairs Press, 1-192. Radinsky, L. 1967 Relative brain size: a new measure. Science, 155: 836-838. - Sanbom, H. C. n.d. Dr. W. C. George’s “The Biology of the Race Problem”: a review. New York, National Putnam Letters Committee, 1-8. Schlaginhaufen, 0. 1950-51 Ein Melanesierschsdel von ungew6hnlich kleiner Kapazitiit. Bull. Schweiz. Ges. Anthrop. u. Ethnol., 27: 2637. Schreider, E. 1966 Brain weight correlations calculated from original results of Paul Broca. Am. J. Phys. Anthrop., 25: 153-158. Schultz, A. H. 1962 Die Schadelkapazitat mannlicher Gorillas und ihr Hiichstwert. Anthrop. Anz., 25: 197-203. 1968 Scrimshaw, N. S., and E. Gordon ( 4 s . ) Malnutrition, Learning and Behaviour. Cambridge, MIT Press. S h e , G. A. 1953 Cell counts in the primate cerebral cortex. J. Comp. Neur., 98: 3d1-400. Shibata, I. 1936 Brain weight o f the Korean. Am. J. Phys. Anthmp., 22: 27-35. Smit, P. J. 1969 Anthmpometric, motor performance and physiological studies o n South African children involved in a nutritional status survey. Thesis accepted for the degree of Ph.D., Anatomy Dept., University of the Witwatersr and. Soemmering, S. T. 1788 Vom H i m und Ruckenmark. Mainz: P. A. Winkpp. 115 pp. (Cited by Bailey and von Bonin, ’51.) Spann, W., and H. 0. Dustmann 1965 Das menschliche Hirngewichte und seine AbhHn zigkeit von Lebensalter, Karperl;inge, Todesursache und Beruf. Deutsche Zeitschrift fur gerichtliche Medizin, 58: 299-317. Stoch, M.B., and P. M. Smythe 1963 Does undernutrition during infancy inhibit brain growth and subsequent intellectual development? Arch. Dis. Childh., 38: 546-552. 1967 The effect of undernutrition during infancy on subsequent brain growth and intellectual development. S. Afr. Med. J., 41: 1027-1031. StrBm, A. (ed.) 1968 Norwegian Concentration Camp Survivors. Oslo: Universitetsforlaget. pp. 9-186. Swan, D. A. 1964 Juan Comas on “ ‘Scientific’ racism again?”: A scientific analysis, Mankind Quarterly, 2 (4): Reprinted in Mankind Monographs VI (3), 24-36. Tiakahasi, K., and I. SuzuM 1961 The brain weight of Japanese. Sapporo Med. J., 20: 179184. Tiedemann, F. 1836 On the brain of the Negro, compared with that of the European and the orang-utang. Phil. Trans. Roy. Soc. Lond., 126: 497-527. Tobias, P. V. 1958 Some aspects of the biology of the Bantu-speaking African. Leech, 28: 312. Todd, T. W. 1927 A liter and a half of brains. Science, 66: 122-125. Topinard, P. 1885 Elements d‘Anthropologie Gh6rale. xv, 1157 pp. Paris, A. Delahaye and E. Lecrosnier. Tower, D. B. 1954 Structural and functional organization of mammalian cerebral cortex: the correlation of neurone density with brain size. J. Comp. Neur., 101: 19-53. 25 BRAIN-SIZE, GREY MATTER AND RACE Tower,I). B., and K. A. C. Elliott 1952 Activity of the acetylcholine system in the cerebral cortex of various unanaesthetized mammals. Am. J. Physiol., 168: 747-759. Vint, F. W. 1934 The brain of the Kenya native. J. Anat. Lond., 68: 216-223. von Bonin, G. 1948 The frontal lobe of the primates, cytoarchitectural studies. Res. Publ. Assoc. nerv. ment. Dis., 27: 67-83. 1950 Essay on the Cerebral Cortex. Charles C Thomas, Springkld, Illinois. 1963 The Evolution of the Human + 92 pp. Brain. Chicago: University Press. xiv von Economo, C. 1929 The Cyto-architectonics of the Human Cerebral Cortex. Oxford University Press. Waldeyer, W. 1894 Sitzungber. d.k. Preuss. Akad. d. Wiss. Berlin, pp. 1213-1221 (cited by Pearl, '34). Zamenhof, S., E. van Marthens and F. L. Margolis 1968 DNA (cell number) and protein in neonatal brain: alteration by maternal dietary protein restriction. Science, 160: 322-323.