Newborn:Adult Brain Ratios in Hominid Evolution HAROLD V. F. JORDAAN Department of Anatomy, University of Cape Town Medical School, Cape Town, South U r i c a
KEY WORDS relations.
Allometry
. Fetal cranial
capacity
. Cephalo-pelvic
ABSTRACT The ratio of newborn to adult brain size varies widely in primates. These variations provide a n index of the different degrees of postnatal brain growth in the different members of the primate order. The uniquely low figure for Homo sapiens indicates a greater degree of postnatal brain growth and therefore postnatal dependence and also a greater need and opportunity for social organisation. An attempt is made to determine the newborn:adult brain ratio in a proto-human population, Australopithecus africanus. Two possible causes of the reduction of the ratio in hominid evolution are discussed. The first is the limiting confines of the maternal pelvis adapted primarily for orthograde progression rather than parturition. The second concerns the resultant of a set of three paired variables between the members of each pair of which there exists a n allometric relationship. These are the relation between brain and body size in the adult, feto-maternal weight allometry and the relation between newborn brain-size and birth weight.
Primates show variation in the relation between maternal pelvic capacity and fetal cranial dimensions. In Homo supiens the approximation between fetal cranial size and the principal diameters of the true pelvis is so close that Schultz (’68) described the human pelvis as “shockingly crowded’ in parturition. Two factors have been responsible for the greater relative size of the fetal neurocranium in man: the greater size of the newborn; and the differential enlargement of the hominid pelvis in response to the requirements for efficient bipedalism. The size of the newborn. The allometric relationship between newborn and maternal size has been studied in many Eutheria. The regression line (Leutenegger, ’72) defining this relationship in subhuman primates is F = 1.0 M.W.O.70
where F = log,, fetal weight in gm and M.W. = log,, maternal weight in gm. If it is assumed that the exponential of maternal weight to which birth weight is related is constant for all primates, then it is inferred (Leutenegger, ’72) that the reAM. J.
PHYS.ANTHROP.. 44: 271-278.
gression formula for modern man and the extinct hominid genus, Australopithecus africunus, is F = 1.47M.W.o.70
Man’s much vaunted brain size in relation to total body weight has been held by many to be a distinctive feature, though higher brain:body ratios have been recorded in, for example, squirrel and marmoset monkeys (Tobias, ’70). Nonetheless hominid values average higher than in subhuman primates and rise throughout the hominid progression. This increasing cephalisation is not apparent when comparing the relative endocranial capacities in the newborn of human and sub-human primates. Thus Schultz (‘41) found the respective ratios to be 9.9 and 9.7%, the difference of 0.2 being non-significant. It should be noted however, that the relative endocranial capacity is not the same as the relative brain size and that the relation between the two indices varies in different members of the Primate order. The pelvis. While it is probable there has been a significant increase in total body size in human evolution, accompanied 271
2 72
HAROLD V. F. JORDAAN
by a corresponding increase in many allometrically related organs, the enlargement of the pelvis did not conform to these rules. The dominant forces which controlled the size and differential expansion of the pelvis in the emergence of the Hominidae were the requirements for orthograde progression and weight transmission. Thus, in the evolution of the hominid pelvis from the inferred ancestral primate form, the emphasis has been on greater transverse enlargement to provide a wider stance in the erect position, while the outward movement of the ischia ensured more advantageous action of the hamstrings now acting in the coronal plane of each hip joint. The reduction of the iliac blade, and therefore of the conjugate length of the brim and the brim index, together with the enlarged sacro-iliac articular surface, provided greater stability of the lumbosacral-iliac complex. A more direct line of weight transmission from the vertebral column to the femora was achieved by an altered inclination of the pelvic brim. Some of these changes admittedly aided the function of the girdle as a birth canal. The greater transverse pelvic inlet diameter, for example, permitted the alignment of the fetal occipito-frontal dimension in the form of engagement unique for man and, it is inferred (Jordaan, '73), the extinct Hominidae. However, the net result of these changes was to cause a relative reduction in the capacity of the structure as a birth canal. The obstetric function of the pelvis was subordinated to the demands of efficient bipedalism. Changes necessary to maintain patency commensurate with successful parturition took place secondarily to the structural alterations for efficient bipedalism. The more marked occurrence of these secondary changes in the human female pelvis produces sexual dimorphism in the pelvic girdle (Jordaan, '73). Sexual dimorphism in the pelvis of lower primates, however, does not appear to be directly related to the requirements of parturition. Thus in the gibbon, in which fetal cranial diameters closely approximate the maternal pelvic dimensions, pelvic sexual dimorphism is hardly detectable; in Gorilla gorilla dimorphism is marked although fetal cranial size is small in relation to adult pelvic capacity (Schultz, '49).
The precise chronological relationships between the responses of the evolving hominid pelvis to the requirements of bipedalism and parturition are important. The adjustments did not take place contemporaneously. The requirement of the older reproductive process was not important initially. Thus, Leutenegger ('72) points out that the australopithecine pelvis, which is hominid in its total morphological pattern, would have permitted the easy passage of the fetal head in parturition. He used the hominid feto-maternal regression formula to estimate birth size in the gracile australopithecines and applied the ratio of 9.8% to derive the relative endocranial capacity. Newborn neurocranial dimensions homomorphic with the neurocranium of a newborn chimpanzee were matched against the pelvic inlet diameters of A. africanus (STS 14), viz., a transverse diameter of 99 mm and sagittal axis of 85 mm. This australopithecine fossil is considered to be a mature, possibly female individual, according to Robinson ('71) and definitely female, according to Wells ('73). This then is the background against which the significant differences in cephalopelvic relations in parturition as well as differences in newborn: adult endocranial and brain ratios between human and subhuman primates must be viewed. 1. The human fetus at term is absolutely larger because it is allometrically related to the rising adult size in the hominid progression. 2. The human fetus is relatively larger because of the allometric shift in the regression of fetal on maternal weight, as indicated by the constant of 1.47 (as o p posed to 1.0) when contrasting the hominid with the simian formula defining this relation. 3. The hominid pelvis has been remodelled for bipedal efficiency rather than parturition. MATERIALS AND METHODS
A. Materials The purpose of the investigation was to emphasize the importance of obstetrical considerations in hominid evolution by comparing cephalo-pelvic relations in parturition in modern man with those of A. africanus. It was, therefore, necessary to have data on the neurocranial dimen-
N E W B 0 R N : A D U L T BRAIN RATIOS
sions of the fullterm fetus and the diameters of the maternal true pelvis in these two ho minid species. I. Human data These were collected in the obstetric unit of the University of Cape Town/Groote Schuur Hospital complex. (a) F u ~ ~ t e rfetal m neurocranial dimensions, The series consist of 68 newborns. Of these 18 were unsuitable for this study. The basis of selecting the newborns in the study group was the elimination of infants suspected of being growth retarded in the period of antenatal care, by a study of maternal weight gain, fetal ultrasonic biparietal trajectories, and maternal urinary estriol assays. There were 50 cases in the study group. (b) Newborn brain and body size. The brain weight and birth weight of newborns delivered at term and dying within 24 hours of delivery were obtained from the autopsy records of the Department of Pathology. The criteria for inclusion in the study group were the same as those applied in selecting newborns for cephalometric studies. Five hundred and seventy-six autopsy reports were scrutinized; of these 393 were rejected as unsuitable, i.e., the series consists of 183 case reports. Cases which fulfilled the following criteria were included in the study group. 1. The gestational age (calculated on the menstrual history) showed agreement with the clinical estimation of maturity by palpation. 2. No evidence of intrauterine growth retardation was present. Cases of prolonged pre-eclamptic toxemia, hypertension and chronic nephritis were excluded. 3. In multiparous subjects the weight of the newborn was commensurate with the birth weights of previous full-term infants. 4. Cases in which genetic and congenital anomalies were present were excluded. 5. Cases of intracranial haemorrhage and those dying of sepsis were excluded. The data were stratifled according to birth weight using a class interval of 400 gm. By use of double logarithmic plots the
273
regression of brain weight on birth weight was determined and the least squares equation calculated. (c) T h e regression of birth weight o n maternal height. Maternal stature is an important determinant of birth weight in human subjects. Data on maternal height and birth weight (for birth orders one to five) were extracted from the computer records of the University of Cape Town Obstetric unit. Cases of prematurity, stillbirth, congenital anomaly, and plural pregnancy were excluded. The series consisted of 1999 cases. The regression of birth weight on maternal height for the range 55 to 69 inches was determined. (d) Pelvic inlet dimensions. One hundred and nine unselected adult females were X-rayed. The pelvic transverse inlet dimension was measured on an anteroposterior view by the method of parallax shift; the inlet sagittal diameter was determined in a lateral view by the isometric method. Ninety-two X-rays showed the landmarks with sufficient clarity to be used in this study. 11. Data o n A. africanus (a) Birth weight. Two estimates of birth weight were made by applying the simian and hominid feto-maternal regression formulae to Robinson's ('71) estimate of body weight of adult A. africanus (22,500 gm). Estimated birth weight values are 1,220 and 1,660 gms respectively. (b) Pelvic dimensions. The pelvic inlet dimensions used in this study are those quoted by Leutenegger ('72) viz, a transverse diameter of 99 m m and a sagittal axis of 85 mm. (c) Adult cranial capacity. The mean endocranial capacity used in this study is that determined by Tobias ('71) viz, 494 cc. 111. Data on the chimpanzee The mean cranial dimensions for newborn chimpanzees are those quoted by Schultz ('49) viz, 83 mm (head length) and 71 mm (head breadth). Summary of h u m a n data used in this study (i) Fullterm fetal neurocranial dimensions (BPD, biparietal diameter; OFD, occipito-frontal diameter)
274
HAROLD V. F. JORDAAN Gestational (Weeks)
Weight (gm)
BPD (mm)
OFD (mm)
39.97
3384.22
96.53
119.70
0.47
494.95
4.86
3.28
0.07
30.00
0.59
0.46
age
Mean Standard deviation Standard error
(ii) Newborn brain and body weight (B.W.) in gm (a) Brain weight = 2.21 B.W.O.64; (b) Correlation coefficient: r = 0.9990; (c) Regression line: Y = 0.64X + 0.35 (where Y = log,, brain weight; X = log,, birth weight). (iii) Correlation between maternal height ( M . H . ) in c m and birth weight (B.W.) in gm B.W. = 15.118 M.H. 791.99 (r = 0.59).
+
(iv) Maternal pelvic inlet dimensions (a) Obstetric conjugate (mm) Mean 119.94 Standard deviation 9.79 Standard error 1.01
(b) Inlet transverse diameter ( m m ) Mean 121.10 Standard deviation 8.01 Standard error 0.83
B. Method Among the Ponginae the orangutan has the low newborn:adult endocranial ratio of 40.4%. This figure is substantially higher than the 23.3% for Homo (Todd, ’33). If the newborn of man had to retain the ratio applicable to the great apes the fetal endocranial capacity would be at least 700 cm. If this fullterm fetal head were homomorphic with the proportions of the presentday fetus, viz, a biparietal diameter of 96 mm, an occipito-frontal dimension of 119 mm, and a cephalic index of 80, the corresponding neurocranial dimensions would be 117 and 145 mm, respectively. Clearly such a fetal head would not pass through the average or even the larger maternal pelves. The ratio of fetal endocranial size to final adult size had to be reduced. Even at the value of 23.3% problems of cephalopelvic disproportion in parturition not infrequently still occur.
In attempting to determine whether A. africanus had reached, solved, and passed the point at which cephalo-pelvic disproportion on the basis of relative and absolute brain size and endocranial volumes was a problem, two hypothetical models of newborn australopithecine skull are defined. One model, which may be described as hominid, is based on human data: newborn weight is determined by the hominid feto-maternal regression formula, the relative endocranial size is taken as 9.9%, the skull is homomorphic with the unmoulded proportions of human newborn. Another model has all the attributes of a newborn chimpanzee: newborn size is determined by the simian feto-maternal regression formula, the relative endocranial capacity is 9.7% and the neurocranium is homomorphic with that of a newborn chimpanzee. An intermediate model is also presented. It is designated pongid-hominid and is based on Leutenegger’s (‘72) estimates: birth size is calculated according to hominid feto-maternal allometry, the relative endocranial capacity is the mean for modern man and the chimpanzee, viz, 9.8 % , the newborn australopithecine skull is homomorphic with that of a newborn chimpanzee. The three sets of data (table 1) have been matched against the pelvic inlet dimensions of australopithecine specimen STS 14 (figs. 1, 2, 3). Two forms of engagement of the fetal head in parturition are presented for each set of data. In each illustration “a” defines engagement with the occipito-frontal dimension aligned in the sagittal axis of the maternal pelvis; “b” depicts the alternative form of engagement in the transverse pelvic inlet diameter. RESULTS
When the hypothetical data on australopithecine fetal cranial dimensions are matched against the known pelvic dimensions of australopithecine specimen STS 14 the following are noted. (i) Whether hominid or pongid values or a combination of both are assigned to the fetus the head could easily traverse the pelvic inlet, provided it engaged with its occipito-frontal diameter aligned in the transverse axis of the maternal pelvis (figs. 1, 2 , 3).
275
NEWB0RN:ADULT BRAIN RATIOS
Figure 1
Hominid model.
(b)
(a) Figure 2
Horninid-pongid model.
Figure 3
Pongid model.
Figs. 1-3 Scaled diagrammatic representation of cephalo-pelvic relations in parturition in A. africanus based on hypothetical newborn neurocranial dimensions. In each illustration (a) represents engagement of the fetal head (shaded area) with the occipito-frontal diameter aligned in the sagittal pelvic axis; (b) defines the transverse form of engagement. Scale 1:2.
276
HAROLD V. F. JORDAAN TABLE 1
Three sets of hypothetical data on newborn australopithecines Cranial diameters Hypothetic a1 model
Hominid Pongid Pongid-hominid
Birth weight (gm)
Endocranial capacity
1660 1220 1660
164 118 162.7
(CC)
N.E.C. BPD (mm)
OFD (mm)
71 63 71
a8 74 a3
100
-X
A.C.C.
33.2 23.9 32.9
N.E.C., Newborn endocranial capacity; A.C.C., Adult endocranial capacity; BPD, Biparietal and OFD, Occipitc-frontal diameter.
the rates of growth of the brain and the non-neural contents of the cranial cavity, as well as the rate of growth of the braincase itself, may have been different. He adds that i t would be unwise to take for granted that the human percentage differences between brain and endocranial size applied age for age to A. afncanus or any other early hominid. It is not possible to determine, therefore, how closely the newborn: adult endocranial index reflected the newborn:adult brain ratio in A. africanus. However, bearing in mind the corresponding figures for the extant modern primates and their relation to newborn: DISCUSSION adult brain ratios, the value of 23.9% Of the three hypothetical models the may be regarded as broadly implying a pongid model is rejected because of the lower newborn:adult brain ratio than is low ratio (23.9%) of newborn to adult en- consistent with what is known and inferred docranial capacity. Even when Holloway's concerning the gracile australopithecines. ('70) estimate (442 cc) of adult endocraThe differences between the hominid nial capacity is used in this calculation, and hominid-pongid cranial models are the ratio derived is only 26.69%. Both slight. In regard to the latter, it does seem values are significantly lower than the cor- questionable to use the hominid feto-maresponding figures for the great apes which ternal regression formula to estimate birth range from 35 to 61% (Schultz, '68). The size, a hominid-pongid or intermediate figure of 23.9% is similar to that for man, relative endocranial capacity to determine viz, 23.3%, while the second figure of newborn endocranial volume, and to im26.69% does not differ significantly from pute to the newborn australopithecine it. The low ratio of newborn to adult endo- skull the pongid proportions of the chimcranial capacity in man reflects very closely panzee. However, when the two sets of the percentage relation between newborn cranial data are used for the theoretical and adult brain size of 25%. While it is reconstruction of cephalo-pelvic relations tempting to conclude that the ratios of in parturition the same conclusions are 23.9% or 26.69% for the pongid model reached, viz, that the approximation bedenote a correspondingly low ratio of new- tween fetal cranial dimensions and the born to adult brain size, such an inference diameters of the maternal true pelvis was cannot be made, for i t implies similar re- so close that the transverse form of enlations between brain size and endocranial gagement would have been obligatory, becapacity in newborn and adult australo- cause of the need to align the longest fetal pithecines to those of modern man. There cranial dimension in the largest available is no means of determining age changes pelvic axis. This form of engagement is in these relations in this extinct hominid unique for Homo among extant primates. A second observation is that the estipopulation. Tobias ('71) points out that
(ii) In the case of the hominid and hominid-pongid models engagement of the head with its long axis corresponding to the maternal sagittal pelvic inlet axis would have been associated with the mechanical dystocia (figs. 1, 2). The transverse form of engagement would have provided easier transit for even the significantly smaller pongid cranial model with less danger of damage to the maternal soft tissues (fig. 3). (iii) The ratio of newborn to adult endocranial capacity ranges from 23.9% (pongid model) to 33.2% (hominid model).
277
NEWB0RN:ADWLT BRAIN RATIOS TABLE 2
Human variations in newborn:adult brain ratio w i t h increasing maternal size Newborn Mot her
Newborn:Adult brain ratio
~
Brain size Height (cm)
Brain (B) (gm)
Weight (gm)
130 140 150 160 165 170
1040 1120 1200 1280 1320 1360
2757 2908 3060 3211 3286 3362
~
(a)
1
257.4 271.5 285.7 299.8 306.8 313.9
~
(b) 2
351.8 364.0 376.0 388.0 393.7 399.4
+ %
24.74 24.24 23.80 23.42 23.24 23.08
% Cb/ '
33.83 32.5 31.33 30.33 29.83 29.36
1 Newborn brain size is calculated as 94.3% of the relative endocranial capacity (9.9%). One cc of brain is taken to weigh 1 gm. 2 Newborn brain size is calculated according to the formula: Brain weight = (Birth weight)o 64.
mated newborn: adult endocranial ratio brain size. Calculations on human material (33.2%) of the hominid model is lower undertaken to determine the effect of inthan the range for the great apes, but is creasing maternal size on this ratio are setstill significantly higher than in modern out in table 2. Calculations are based on man. When newborn endocranial capacity height in preference to body weight, as is calculated at 40% of adult endocranial brain weight in adults depends signifivolume, the resultant newborn hominid cantly on height but not on body weight cranial dimensions would be 76 mm (head (Tobias, '71). The figure used is 8 gms of breadth) and 94 m m (head length). A fetal brain for every centimetre of stature. The head of these measurements would pass estimated birth weight in gms (B.W.) for through the pelvic inlet of STS 14 with a given maternal height in cms (M.H.) is difficulty, while the dangers to the ma- determined according to the regression 791.99. ternal soft tissues would be considerable. function: B.W. = 15.118 M.H. At lower reaches in the long bony birth Two calculated brain weights are recorded canal the progress of parturition would be for each birth weight. One is based o n the arrested because of disproportion. It ap- finding that the relative endocranial capears then that the reduction in the new- pacity in human newborn is 9.9% and born:adult endocranial ratio was an im- that the brain occupies 94.3% of the endoportant factor in permitting successful cranial volume at birth (Tobias, '71). By parturition. It is likely that the limiting this method the relative brain size is confines of the maternal pelvis were re- 9.34% for all birth weight strata. The second estimate is based o n the sponsible for the reduction in the ratio. A corresponding though not necessarily a regression of brain (Br.W) on body weight proportionate reduction in the ratio of (B.W.): Br.W. = 2.21 B.W.0.'j4. The estinewborn: adult brain size would then have mates derived by this method provide a taken place with greater emphasis on post- range of relative brain sizes from 11.88 to natal as opposed to prenatal brain growth. 12.76% for the highest and lowest birth The allometric relation between three weight categories respectively. These compaired variables determines the newborn: paratively high values for the relative brain adult brain ratio in man, viz, adult body size reflect the stringent application of the size and brain size; adult body size and the criteria of normal growth in selecting masize of the newborn; and newborn body terial for the study. By whatever method newborn brain size and brain size. The resultant of the interrelation of these paired items finally de- is estimated (table 2) the human newborn: termines the newborn:adult brain ratio. adult brain ratio falls with increasing maIn man greater brain size is generally found ternal size. Taking the newborn:adult enin association with larger body size. The docranial ratio as an approximation of latter, therefore, influences both the brain the corresponding brain ratio in the gracile weight of the adult female as well as the australopithecines, the index has the high weight of her newborn and, therefore, its value of 33.2%. When this figure is linked
+
278
HAROLD V. F. JORDAAN
to the highest calculated human newborn: adult brain ratio of 24.75% (based on a relative newborn brain size of 9.43% of endocranial volume) the resultant interspecific trend appears to be a continuity with the intraspecific variations in man. However, an analysis shows that the australopithecine index does not lie on or even near the least squares regression line of newborn:adult brain ratio ( N / A % ) on adult brain size in kg (A.B.S.), viz, N/A% = 30.09 - 5.19 A.B.S. Using the hominid feto-maternal allometric formula to determine birth size in the gracile australopithecines is consistent with their taxonomic status within the Hominidae. The newborn: adult endocranial volume index approaches the lower limits of the range for the modern subhuman primates, but is significantly higher than that of man. If this is taken as an approximation of the newborn: adult brain ratio, the value is consistent with the position of A. af7icanus in the scale of hominisation, viz, the “awakening dawn” between “the limitations of the world of the great apes,” on the one side and, on the other, the “settling in of Homo habitis, the maturation of Homo erectus, and the
flowering and prospering of Homo sapiens” (Tobias, ’71). LITERATURE CITED Holloway, R. L. 1970 New endocranial values for the australopithecines. Nature, 227: 199-200. Jordaan, H. V. F. 1973 A study in maternal pelvic capacity and fetal cranial dimensions in the South African Negro and its bearing on human evolution. Ph.D. thesis, University of Cape Town. Leutenegger, W. 1972 Newborn size and pelvic dimensions of Australopithecus. Nature, 240: 568569. Robinson, J. T. 1971 Topics in the Study of Life. Harper and Row, New York. 449 Schultz, A. H. 1941 The relative size of the cranial capacity in primates. Am. J . Phys. Anthrop., 28: 273-287. 1949 Sex differences in the pelvis of primates. Am. J. Phys. Anthrop., 17: 4 0 1 4 2 3 . 1968 In: Perspective in Human Evolution. S. L. Washburn and Phyllis C. Jay, eds. Holt, Rinehart and Winston, New York. Tobias, P. V. 1970 Brain size, grey matter and race -fact or fiction. Am. J. Phys. Anthrop., 32: 3-26. 1971 The Brain in Hominid Evolution. Columbia University Press, New York and London. Todd, T. W. 1933 In: Growth and Development of Child. Part 2, 26. Whitehouse Conference, New York. Wells, L. H. 1973 Personal communication.
KEY WORDS relations.
Allometry
. Fetal cranial
capacity
. Cephalo-pelvic
ABSTRACT The ratio of newborn to adult brain size varies widely in primates. These variations provide a n index of the different degrees of postnatal brain growth in the different members of the primate order. The uniquely low figure for Homo sapiens indicates a greater degree of postnatal brain growth and therefore postnatal dependence and also a greater need and opportunity for social organisation. An attempt is made to determine the newborn:adult brain ratio in a proto-human population, Australopithecus africanus. Two possible causes of the reduction of the ratio in hominid evolution are discussed. The first is the limiting confines of the maternal pelvis adapted primarily for orthograde progression rather than parturition. The second concerns the resultant of a set of three paired variables between the members of each pair of which there exists a n allometric relationship. These are the relation between brain and body size in the adult, feto-maternal weight allometry and the relation between newborn brain-size and birth weight.
Primates show variation in the relation between maternal pelvic capacity and fetal cranial dimensions. In Homo supiens the approximation between fetal cranial size and the principal diameters of the true pelvis is so close that Schultz (’68) described the human pelvis as “shockingly crowded’ in parturition. Two factors have been responsible for the greater relative size of the fetal neurocranium in man: the greater size of the newborn; and the differential enlargement of the hominid pelvis in response to the requirements for efficient bipedalism. The size of the newborn. The allometric relationship between newborn and maternal size has been studied in many Eutheria. The regression line (Leutenegger, ’72) defining this relationship in subhuman primates is F = 1.0 M.W.O.70
where F = log,, fetal weight in gm and M.W. = log,, maternal weight in gm. If it is assumed that the exponential of maternal weight to which birth weight is related is constant for all primates, then it is inferred (Leutenegger, ’72) that the reAM. J.
PHYS.ANTHROP.. 44: 271-278.
gression formula for modern man and the extinct hominid genus, Australopithecus africunus, is F = 1.47M.W.o.70
Man’s much vaunted brain size in relation to total body weight has been held by many to be a distinctive feature, though higher brain:body ratios have been recorded in, for example, squirrel and marmoset monkeys (Tobias, ’70). Nonetheless hominid values average higher than in subhuman primates and rise throughout the hominid progression. This increasing cephalisation is not apparent when comparing the relative endocranial capacities in the newborn of human and sub-human primates. Thus Schultz (‘41) found the respective ratios to be 9.9 and 9.7%, the difference of 0.2 being non-significant. It should be noted however, that the relative endocranial capacity is not the same as the relative brain size and that the relation between the two indices varies in different members of the Primate order. The pelvis. While it is probable there has been a significant increase in total body size in human evolution, accompanied 271
2 72
HAROLD V. F. JORDAAN
by a corresponding increase in many allometrically related organs, the enlargement of the pelvis did not conform to these rules. The dominant forces which controlled the size and differential expansion of the pelvis in the emergence of the Hominidae were the requirements for orthograde progression and weight transmission. Thus, in the evolution of the hominid pelvis from the inferred ancestral primate form, the emphasis has been on greater transverse enlargement to provide a wider stance in the erect position, while the outward movement of the ischia ensured more advantageous action of the hamstrings now acting in the coronal plane of each hip joint. The reduction of the iliac blade, and therefore of the conjugate length of the brim and the brim index, together with the enlarged sacro-iliac articular surface, provided greater stability of the lumbosacral-iliac complex. A more direct line of weight transmission from the vertebral column to the femora was achieved by an altered inclination of the pelvic brim. Some of these changes admittedly aided the function of the girdle as a birth canal. The greater transverse pelvic inlet diameter, for example, permitted the alignment of the fetal occipito-frontal dimension in the form of engagement unique for man and, it is inferred (Jordaan, '73), the extinct Hominidae. However, the net result of these changes was to cause a relative reduction in the capacity of the structure as a birth canal. The obstetric function of the pelvis was subordinated to the demands of efficient bipedalism. Changes necessary to maintain patency commensurate with successful parturition took place secondarily to the structural alterations for efficient bipedalism. The more marked occurrence of these secondary changes in the human female pelvis produces sexual dimorphism in the pelvic girdle (Jordaan, '73). Sexual dimorphism in the pelvis of lower primates, however, does not appear to be directly related to the requirements of parturition. Thus in the gibbon, in which fetal cranial diameters closely approximate the maternal pelvic dimensions, pelvic sexual dimorphism is hardly detectable; in Gorilla gorilla dimorphism is marked although fetal cranial size is small in relation to adult pelvic capacity (Schultz, '49).
The precise chronological relationships between the responses of the evolving hominid pelvis to the requirements of bipedalism and parturition are important. The adjustments did not take place contemporaneously. The requirement of the older reproductive process was not important initially. Thus, Leutenegger ('72) points out that the australopithecine pelvis, which is hominid in its total morphological pattern, would have permitted the easy passage of the fetal head in parturition. He used the hominid feto-maternal regression formula to estimate birth size in the gracile australopithecines and applied the ratio of 9.8% to derive the relative endocranial capacity. Newborn neurocranial dimensions homomorphic with the neurocranium of a newborn chimpanzee were matched against the pelvic inlet diameters of A. africanus (STS 14), viz., a transverse diameter of 99 mm and sagittal axis of 85 mm. This australopithecine fossil is considered to be a mature, possibly female individual, according to Robinson ('71) and definitely female, according to Wells ('73). This then is the background against which the significant differences in cephalopelvic relations in parturition as well as differences in newborn: adult endocranial and brain ratios between human and subhuman primates must be viewed. 1. The human fetus at term is absolutely larger because it is allometrically related to the rising adult size in the hominid progression. 2. The human fetus is relatively larger because of the allometric shift in the regression of fetal on maternal weight, as indicated by the constant of 1.47 (as o p posed to 1.0) when contrasting the hominid with the simian formula defining this relation. 3. The hominid pelvis has been remodelled for bipedal efficiency rather than parturition. MATERIALS AND METHODS
A. Materials The purpose of the investigation was to emphasize the importance of obstetrical considerations in hominid evolution by comparing cephalo-pelvic relations in parturition in modern man with those of A. africanus. It was, therefore, necessary to have data on the neurocranial dimen-
N E W B 0 R N : A D U L T BRAIN RATIOS
sions of the fullterm fetus and the diameters of the maternal true pelvis in these two ho minid species. I. Human data These were collected in the obstetric unit of the University of Cape Town/Groote Schuur Hospital complex. (a) F u ~ ~ t e rfetal m neurocranial dimensions, The series consist of 68 newborns. Of these 18 were unsuitable for this study. The basis of selecting the newborns in the study group was the elimination of infants suspected of being growth retarded in the period of antenatal care, by a study of maternal weight gain, fetal ultrasonic biparietal trajectories, and maternal urinary estriol assays. There were 50 cases in the study group. (b) Newborn brain and body size. The brain weight and birth weight of newborns delivered at term and dying within 24 hours of delivery were obtained from the autopsy records of the Department of Pathology. The criteria for inclusion in the study group were the same as those applied in selecting newborns for cephalometric studies. Five hundred and seventy-six autopsy reports were scrutinized; of these 393 were rejected as unsuitable, i.e., the series consists of 183 case reports. Cases which fulfilled the following criteria were included in the study group. 1. The gestational age (calculated on the menstrual history) showed agreement with the clinical estimation of maturity by palpation. 2. No evidence of intrauterine growth retardation was present. Cases of prolonged pre-eclamptic toxemia, hypertension and chronic nephritis were excluded. 3. In multiparous subjects the weight of the newborn was commensurate with the birth weights of previous full-term infants. 4. Cases in which genetic and congenital anomalies were present were excluded. 5. Cases of intracranial haemorrhage and those dying of sepsis were excluded. The data were stratifled according to birth weight using a class interval of 400 gm. By use of double logarithmic plots the
273
regression of brain weight on birth weight was determined and the least squares equation calculated. (c) T h e regression of birth weight o n maternal height. Maternal stature is an important determinant of birth weight in human subjects. Data on maternal height and birth weight (for birth orders one to five) were extracted from the computer records of the University of Cape Town Obstetric unit. Cases of prematurity, stillbirth, congenital anomaly, and plural pregnancy were excluded. The series consisted of 1999 cases. The regression of birth weight on maternal height for the range 55 to 69 inches was determined. (d) Pelvic inlet dimensions. One hundred and nine unselected adult females were X-rayed. The pelvic transverse inlet dimension was measured on an anteroposterior view by the method of parallax shift; the inlet sagittal diameter was determined in a lateral view by the isometric method. Ninety-two X-rays showed the landmarks with sufficient clarity to be used in this study. 11. Data o n A. africanus (a) Birth weight. Two estimates of birth weight were made by applying the simian and hominid feto-maternal regression formulae to Robinson's ('71) estimate of body weight of adult A. africanus (22,500 gm). Estimated birth weight values are 1,220 and 1,660 gms respectively. (b) Pelvic dimensions. The pelvic inlet dimensions used in this study are those quoted by Leutenegger ('72) viz, a transverse diameter of 99 m m and a sagittal axis of 85 mm. (c) Adult cranial capacity. The mean endocranial capacity used in this study is that determined by Tobias ('71) viz, 494 cc. 111. Data on the chimpanzee The mean cranial dimensions for newborn chimpanzees are those quoted by Schultz ('49) viz, 83 mm (head length) and 71 mm (head breadth). Summary of h u m a n data used in this study (i) Fullterm fetal neurocranial dimensions (BPD, biparietal diameter; OFD, occipito-frontal diameter)
274
HAROLD V. F. JORDAAN Gestational (Weeks)
Weight (gm)
BPD (mm)
OFD (mm)
39.97
3384.22
96.53
119.70
0.47
494.95
4.86
3.28
0.07
30.00
0.59
0.46
age
Mean Standard deviation Standard error
(ii) Newborn brain and body weight (B.W.) in gm (a) Brain weight = 2.21 B.W.O.64; (b) Correlation coefficient: r = 0.9990; (c) Regression line: Y = 0.64X + 0.35 (where Y = log,, brain weight; X = log,, birth weight). (iii) Correlation between maternal height ( M . H . ) in c m and birth weight (B.W.) in gm B.W. = 15.118 M.H. 791.99 (r = 0.59).
+
(iv) Maternal pelvic inlet dimensions (a) Obstetric conjugate (mm) Mean 119.94 Standard deviation 9.79 Standard error 1.01
(b) Inlet transverse diameter ( m m ) Mean 121.10 Standard deviation 8.01 Standard error 0.83
B. Method Among the Ponginae the orangutan has the low newborn:adult endocranial ratio of 40.4%. This figure is substantially higher than the 23.3% for Homo (Todd, ’33). If the newborn of man had to retain the ratio applicable to the great apes the fetal endocranial capacity would be at least 700 cm. If this fullterm fetal head were homomorphic with the proportions of the presentday fetus, viz, a biparietal diameter of 96 mm, an occipito-frontal dimension of 119 mm, and a cephalic index of 80, the corresponding neurocranial dimensions would be 117 and 145 mm, respectively. Clearly such a fetal head would not pass through the average or even the larger maternal pelves. The ratio of fetal endocranial size to final adult size had to be reduced. Even at the value of 23.3% problems of cephalopelvic disproportion in parturition not infrequently still occur.
In attempting to determine whether A. africanus had reached, solved, and passed the point at which cephalo-pelvic disproportion on the basis of relative and absolute brain size and endocranial volumes was a problem, two hypothetical models of newborn australopithecine skull are defined. One model, which may be described as hominid, is based on human data: newborn weight is determined by the hominid feto-maternal regression formula, the relative endocranial size is taken as 9.9%, the skull is homomorphic with the unmoulded proportions of human newborn. Another model has all the attributes of a newborn chimpanzee: newborn size is determined by the simian feto-maternal regression formula, the relative endocranial capacity is 9.7% and the neurocranium is homomorphic with that of a newborn chimpanzee. An intermediate model is also presented. It is designated pongid-hominid and is based on Leutenegger’s (‘72) estimates: birth size is calculated according to hominid feto-maternal allometry, the relative endocranial capacity is the mean for modern man and the chimpanzee, viz, 9.8 % , the newborn australopithecine skull is homomorphic with that of a newborn chimpanzee. The three sets of data (table 1) have been matched against the pelvic inlet dimensions of australopithecine specimen STS 14 (figs. 1, 2, 3). Two forms of engagement of the fetal head in parturition are presented for each set of data. In each illustration “a” defines engagement with the occipito-frontal dimension aligned in the sagittal axis of the maternal pelvis; “b” depicts the alternative form of engagement in the transverse pelvic inlet diameter. RESULTS
When the hypothetical data on australopithecine fetal cranial dimensions are matched against the known pelvic dimensions of australopithecine specimen STS 14 the following are noted. (i) Whether hominid or pongid values or a combination of both are assigned to the fetus the head could easily traverse the pelvic inlet, provided it engaged with its occipito-frontal diameter aligned in the transverse axis of the maternal pelvis (figs. 1, 2 , 3).
275
NEWB0RN:ADULT BRAIN RATIOS
Figure 1
Hominid model.
(b)
(a) Figure 2
Horninid-pongid model.
Figure 3
Pongid model.
Figs. 1-3 Scaled diagrammatic representation of cephalo-pelvic relations in parturition in A. africanus based on hypothetical newborn neurocranial dimensions. In each illustration (a) represents engagement of the fetal head (shaded area) with the occipito-frontal diameter aligned in the sagittal pelvic axis; (b) defines the transverse form of engagement. Scale 1:2.
276
HAROLD V. F. JORDAAN TABLE 1
Three sets of hypothetical data on newborn australopithecines Cranial diameters Hypothetic a1 model
Hominid Pongid Pongid-hominid
Birth weight (gm)
Endocranial capacity
1660 1220 1660
164 118 162.7
(CC)
N.E.C. BPD (mm)
OFD (mm)
71 63 71
a8 74 a3
100
-X
A.C.C.
33.2 23.9 32.9
N.E.C., Newborn endocranial capacity; A.C.C., Adult endocranial capacity; BPD, Biparietal and OFD, Occipitc-frontal diameter.
the rates of growth of the brain and the non-neural contents of the cranial cavity, as well as the rate of growth of the braincase itself, may have been different. He adds that i t would be unwise to take for granted that the human percentage differences between brain and endocranial size applied age for age to A. afncanus or any other early hominid. It is not possible to determine, therefore, how closely the newborn: adult endocranial index reflected the newborn:adult brain ratio in A. africanus. However, bearing in mind the corresponding figures for the extant modern primates and their relation to newborn: DISCUSSION adult brain ratios, the value of 23.9% Of the three hypothetical models the may be regarded as broadly implying a pongid model is rejected because of the lower newborn:adult brain ratio than is low ratio (23.9%) of newborn to adult en- consistent with what is known and inferred docranial capacity. Even when Holloway's concerning the gracile australopithecines. ('70) estimate (442 cc) of adult endocraThe differences between the hominid nial capacity is used in this calculation, and hominid-pongid cranial models are the ratio derived is only 26.69%. Both slight. In regard to the latter, it does seem values are significantly lower than the cor- questionable to use the hominid feto-maresponding figures for the great apes which ternal regression formula to estimate birth range from 35 to 61% (Schultz, '68). The size, a hominid-pongid or intermediate figure of 23.9% is similar to that for man, relative endocranial capacity to determine viz, 23.3%, while the second figure of newborn endocranial volume, and to im26.69% does not differ significantly from pute to the newborn australopithecine it. The low ratio of newborn to adult endo- skull the pongid proportions of the chimcranial capacity in man reflects very closely panzee. However, when the two sets of the percentage relation between newborn cranial data are used for the theoretical and adult brain size of 25%. While it is reconstruction of cephalo-pelvic relations tempting to conclude that the ratios of in parturition the same conclusions are 23.9% or 26.69% for the pongid model reached, viz, that the approximation bedenote a correspondingly low ratio of new- tween fetal cranial dimensions and the born to adult brain size, such an inference diameters of the maternal true pelvis was cannot be made, for i t implies similar re- so close that the transverse form of enlations between brain size and endocranial gagement would have been obligatory, becapacity in newborn and adult australo- cause of the need to align the longest fetal pithecines to those of modern man. There cranial dimension in the largest available is no means of determining age changes pelvic axis. This form of engagement is in these relations in this extinct hominid unique for Homo among extant primates. A second observation is that the estipopulation. Tobias ('71) points out that
(ii) In the case of the hominid and hominid-pongid models engagement of the head with its long axis corresponding to the maternal sagittal pelvic inlet axis would have been associated with the mechanical dystocia (figs. 1, 2). The transverse form of engagement would have provided easier transit for even the significantly smaller pongid cranial model with less danger of damage to the maternal soft tissues (fig. 3). (iii) The ratio of newborn to adult endocranial capacity ranges from 23.9% (pongid model) to 33.2% (hominid model).
277
NEWB0RN:ADWLT BRAIN RATIOS TABLE 2
Human variations in newborn:adult brain ratio w i t h increasing maternal size Newborn Mot her
Newborn:Adult brain ratio
~
Brain size Height (cm)
Brain (B) (gm)
Weight (gm)
130 140 150 160 165 170
1040 1120 1200 1280 1320 1360
2757 2908 3060 3211 3286 3362
~
(a)
1
257.4 271.5 285.7 299.8 306.8 313.9
~
(b) 2
351.8 364.0 376.0 388.0 393.7 399.4
+ %
24.74 24.24 23.80 23.42 23.24 23.08
% Cb/ '
33.83 32.5 31.33 30.33 29.83 29.36
1 Newborn brain size is calculated as 94.3% of the relative endocranial capacity (9.9%). One cc of brain is taken to weigh 1 gm. 2 Newborn brain size is calculated according to the formula: Brain weight = (Birth weight)o 64.
mated newborn: adult endocranial ratio brain size. Calculations on human material (33.2%) of the hominid model is lower undertaken to determine the effect of inthan the range for the great apes, but is creasing maternal size on this ratio are setstill significantly higher than in modern out in table 2. Calculations are based on man. When newborn endocranial capacity height in preference to body weight, as is calculated at 40% of adult endocranial brain weight in adults depends signifivolume, the resultant newborn hominid cantly on height but not on body weight cranial dimensions would be 76 mm (head (Tobias, '71). The figure used is 8 gms of breadth) and 94 m m (head length). A fetal brain for every centimetre of stature. The head of these measurements would pass estimated birth weight in gms (B.W.) for through the pelvic inlet of STS 14 with a given maternal height in cms (M.H.) is difficulty, while the dangers to the ma- determined according to the regression 791.99. ternal soft tissues would be considerable. function: B.W. = 15.118 M.H. At lower reaches in the long bony birth Two calculated brain weights are recorded canal the progress of parturition would be for each birth weight. One is based o n the arrested because of disproportion. It ap- finding that the relative endocranial capears then that the reduction in the new- pacity in human newborn is 9.9% and born:adult endocranial ratio was an im- that the brain occupies 94.3% of the endoportant factor in permitting successful cranial volume at birth (Tobias, '71). By parturition. It is likely that the limiting this method the relative brain size is confines of the maternal pelvis were re- 9.34% for all birth weight strata. The second estimate is based o n the sponsible for the reduction in the ratio. A corresponding though not necessarily a regression of brain (Br.W) on body weight proportionate reduction in the ratio of (B.W.): Br.W. = 2.21 B.W.0.'j4. The estinewborn: adult brain size would then have mates derived by this method provide a taken place with greater emphasis on post- range of relative brain sizes from 11.88 to natal as opposed to prenatal brain growth. 12.76% for the highest and lowest birth The allometric relation between three weight categories respectively. These compaired variables determines the newborn: paratively high values for the relative brain adult brain ratio in man, viz, adult body size reflect the stringent application of the size and brain size; adult body size and the criteria of normal growth in selecting masize of the newborn; and newborn body terial for the study. By whatever method newborn brain size and brain size. The resultant of the interrelation of these paired items finally de- is estimated (table 2) the human newborn: termines the newborn:adult brain ratio. adult brain ratio falls with increasing maIn man greater brain size is generally found ternal size. Taking the newborn:adult enin association with larger body size. The docranial ratio as an approximation of latter, therefore, influences both the brain the corresponding brain ratio in the gracile weight of the adult female as well as the australopithecines, the index has the high weight of her newborn and, therefore, its value of 33.2%. When this figure is linked
+
278
HAROLD V. F. JORDAAN
to the highest calculated human newborn: adult brain ratio of 24.75% (based on a relative newborn brain size of 9.43% of endocranial volume) the resultant interspecific trend appears to be a continuity with the intraspecific variations in man. However, an analysis shows that the australopithecine index does not lie on or even near the least squares regression line of newborn:adult brain ratio ( N / A % ) on adult brain size in kg (A.B.S.), viz, N/A% = 30.09 - 5.19 A.B.S. Using the hominid feto-maternal allometric formula to determine birth size in the gracile australopithecines is consistent with their taxonomic status within the Hominidae. The newborn: adult endocranial volume index approaches the lower limits of the range for the modern subhuman primates, but is significantly higher than that of man. If this is taken as an approximation of the newborn: adult brain ratio, the value is consistent with the position of A. af7icanus in the scale of hominisation, viz, the “awakening dawn” between “the limitations of the world of the great apes,” on the one side and, on the other, the “settling in of Homo habitis, the maturation of Homo erectus, and the
flowering and prospering of Homo sapiens” (Tobias, ’71). LITERATURE CITED Holloway, R. L. 1970 New endocranial values for the australopithecines. Nature, 227: 199-200. Jordaan, H. V. F. 1973 A study in maternal pelvic capacity and fetal cranial dimensions in the South African Negro and its bearing on human evolution. Ph.D. thesis, University of Cape Town. Leutenegger, W. 1972 Newborn size and pelvic dimensions of Australopithecus. Nature, 240: 568569. Robinson, J. T. 1971 Topics in the Study of Life. Harper and Row, New York. 449 Schultz, A. H. 1941 The relative size of the cranial capacity in primates. Am. J . Phys. Anthrop., 28: 273-287. 1949 Sex differences in the pelvis of primates. Am. J. Phys. Anthrop., 17: 4 0 1 4 2 3 . 1968 In: Perspective in Human Evolution. S. L. Washburn and Phyllis C. Jay, eds. Holt, Rinehart and Winston, New York. Tobias, P. V. 1970 Brain size, grey matter and race -fact or fiction. Am. J. Phys. Anthrop., 32: 3-26. 1971 The Brain in Hominid Evolution. Columbia University Press, New York and London. Todd, T. W. 1933 In: Growth and Development of Child. Part 2, 26. Whitehouse Conference, New York. Wells, L. H. 1973 Personal communication.
