AMERICAN JOURNAL OF HUMAN BIOLOGY 6:77-88 (1994)

Evolutionary Perspectives on Human Nutrition: The Influence of Brain and Body Size on Diet and Metabolism WILLIAM R LEONARD RNU MARCIA L. ROBERTSON School of Human Biology, Unrversrty of Guelph, Guelph, Ontano NIG 2WI, Canada

ABSTRACT Human dietary patterns and metabolic requirements are compared to those of nonhuman primate species in order to gain insights into the evolution of our nutritional needs. In general, primate diet quality (i.e., caloric and nutrient density) is inversely related to body size and total resting metabolic requirements (RMR). Humans, however, consume a diet of much higher quality than is expected for our size and metabolic needs. This energy-rich diet appears t o reflect an adaptation to the high metabolic cost of our large brain. Among primates, the relative proportion of resting metabolic energy used for brain metabolism is positively correlated with relative diet quality. Humans represent the positive extreme, having both a very high quality diet and a large brain that accounts for 20-2576 of resting metabolism. Evidence from the hominid fossil record implies that major changes in diet and relative brain metabolism occurred with the emergence of the genus Homo. 0 1994 Wiley-Liss, Inc In recent years, evolutionary perspectives on human nutrition have become a major focus of research within physical anthropology. Although interest in the evolution of the human diet has a long tradition, current research differs from earlier approaches in that it encompasses work done within all areas of the discipline from contemporary human biology to primate ecology and paleoanthropology (cf. Gordon, 1987; Leonard and Robertson, 1992; Milton, 1987). What is particularly interesting about this trend is that it was anticipated over 25 years ago by Stanley Garn in his 1966 presidential address to the American Association of Physical Anthropologists, “Nutrition in Physical Anthropology.” Garn closed the lecture with the following:

So it is a reasonable guess that a concern with nutrition in its broadest sense, including the physical form of the dietary and the energy balance will occupy u s more rather than less in the years to come, as we channel our interests to ascertain the directions of evolutionary and physiological change rather than limiting our endeavors to their simple description (Garn, 1966, p. 292).

0 1994 Wiley-Liss, Inc.

In the time since this address was given, perceptions on what the “natural” human diet is have changed substantially. Early theories of human nutritional evolution viewed hominids a s “carnivores’’ and speculated that it was our ancestors’ prowess a s upright-walking, tool-using hunters that separated us from the great apes (Dart, 1957; Ardrey, 1961; Washburn and Lancaster, 1968). By the late 1960s, however, the “Man the hunter” interpretation was strongly challenged by Richard Lee and others who noted that most contemporary hunting and gathering populations consume more plant than animal material (Lee, 1968). As a consequence, the importance of “meat eating” has been de-emphasized in many recent models of human origins (cf. Jolly, 1970; Tanner, 1981). At present, there is still considerable debate over the type of diet to which humans are adapted (cf. Eaton and Konner, 1985; Garn and Leonard, 1989). One way of gaining insights into the evolutionary significance of the human diet is by comparing hu-

Received January 14,1993; accepted May 20,1993. Address reprint requests to W.R. Leonard, School of Human Biology, University of Guelph, Guelph, Ontario N1G ZWI, Canada.

W.R. LEONARD AND M.L. ROBERTSON

78 Diet Quality

400

I

8

Anihropolde

I

Prorimlane

I 1 l L - U

0

0.01

C

+

0.1

1

1

*

-

+

I

,

,

I I Ill1

I

IIIII

I

100

10

,

#

,

I

1000

Body Weight (kg) Fig. 1. Plot of diet quality (DQ)relative to log-body weight (kg) for 72 nonhuman primate species (data from Sailer et al., 1985)and 5 human foraging populations (see Table 1 for sources). DQ was calculated using the index of Sailer et al. (1985). Among nonhuman primates, DQ is strongly negatively correlated with body weight (r = -0.661; P < 0.001). -41 five human groups have DQs that are substantially higher than predicted for their body weight.

man dietary patterns to those of other primate species. Such a n approach is useful for determining whether humans conform to important ecological patterns of dietary variation observed among primate species, and isolating key metabolic and nutritional differences between humans and other primates. I n this study, we will first examine the relationship between diet and body size among primates and how well humans fit that relationship. Potential metabolic correlates of dietary differences between humans and nonhuman primates will then be evaluated. Finally, we will examine the human fossil record to gain insights into when and how the dietary and metabolic differences between humans and other primates might have evolved. DIET AND BODY SIZE One of the most important correlates of dietary variation among primate species is body size. In general, there is a negative relationship between diet quality (the energy and nutrient density of food items) and body size (Clutton-Brock and Harvey, 1977; Sailer e t al., 1985). Large-bodied primates, such as gorillas and orangutans, have diets composed mostly of difficult to digest and

nutrient-poor foods such a s leaves and bark (Rodman, 1977, 1984; Goodall, 1977; Temerin e t al., 1984). In contrast, very small primates, such as the pygmy marmoset (Cebuellapygmaea, 110 g), tend to specialize on insects, gums, and sap (Richard, 1985). Figure 1depicts diet quality as a function of log-body weight for a sample of 72 nonhuman primate species (13 prosimians; 59 anthropoids). Diet quality (DQ) was measured using a n index developed by Sailer et al. (1985). This index ranks the energy and nutrient density of a diet based on the relative proportions of structural plant parts (e.g., leaves, stems), reproductive plant parts ( e g , fruit, bulbs), and animal material (including insects) by weight or feeding time. The equation for calculating DQ is:

DQ

=

s

+ 2(r) + 3.5(a)

where s = percent of diet from structural plant parts, r = percent of diet from reproductive plant parts, a = percent of diet from animal foods. DQ values range from a minimum of 100 (a diet of 100% foliage) to a maximum of 350 (100% animal material). It is apparent in Figure 1 that there is a strong negative correlation (r = -.661; P 0.001) between DQ and log-body

79

HUMAN NUTRITIONAL EVOLUTION

TABLE 1 Adult body weight and dietary data for fiur human foraging populations Population !Kong

Sex

M

Weight (kg)

Percent animal’

Diet quality index

46.0 41.0 59.6 51.0

33

235.5

Lee (1968, 19791

56

263.0

Hill et al. (1984)

68

287.0

Hurtado and Hill (1989)

96

343.4

44

252.5

Inuit

F M F M F M

Pygmies

F M

55.0 42.5

F

38.8

Ache Hiwi

57.8 48.5 61.0

References

Gaulin and Konner (1977) Rode and Shephard (1973) Dietz et al. (1989) Hill (1982)

‘Percent of daily energy intake derived from animal material

weight, that is, as size increases, DQ decreases. This association between size and DQ, sometimes referred to as the “JarmanBell” relationship, is seen among mammals in general and is thought to result from the relationship between body size and metabolism (Bell, 1971; Jarman, 1974; Gaulin, 1979). The Jarman-Bell model postulates that since resting metabolism scales to the 3/4th power of body weight (Kleiber, 19611, smaller animals have much higher metabolic costs per unit of body weight and thus need to consume foods of high caloric density. In contrast, large animals have higher total energy needs but relatively low requirements per unit weight. Consequently, these animals tend to fulfill their nutritional needs by consuming large amounts of widely available low quality foods such as leaves and bark. Humans, however, appear to depart substantially from the primate DQ-body weight relationship (Fig. 1). Dietary and body weight data for five human hunting and gathering populations are given in Table 1. Modern hunter-gatherers were used in this comparison since the foraging subsistence strategy represents the “primitive condition” for our species (i.e., the strategy utilized by early Homo sapiens). The DQ index was calculated from the contribution, by weight, of different food items from the references listed in Table 1. Food items included in the “animal” category were meat, insects, and larvae, while those categorized as “reproductive plant parts” included seeds, nuts, fruits, bulbs, and tubers. All foliage and leafy material were categorized as “structural plant parts.” Animal foods contribute anywhere from one third to more than 95% of energy intake in these contemporary foragers (mean =

59 +- 24%;median = 56%).While this would hardly qualify humans as “carnivores,” it is substantially more animal material than any other primate of our size. By comparison, chimpanzees, the most predatory of the large-bodied apes, derive only about 5-7% of their daily energy intake from animal material (2-396 from meat; 3 4 % from insects, estimated from data in Teleki, 1981; McGrew, 1974; Wu Leung, 1968). Hence, as shown in Figure 1, the diets of the five hunter-gatherer groups are of much higher quality than expected for primates of their size. This generalization holds for most all subsistence-level human populations, as even agricultural groups consuming cerealbased diets have estimated DQs higher than other primates of comparable size. For example, agricultural groups from highland and coastal South America which derive only 6 4 % of their calories from meat (Leonard and Thomas, 1989; Leonard et al., 1993) have DQs of 200-210. Thus, even in human populations where meat consumption is low, DQ is still much higher than in other largebodied primates because grains are much more calorically dense than foliage. If it is assumed that the last common ancestor of apes and humans had a diet similar to that of a modern ape, and that the earliest members of our species (Homo supiens supiens j consumed a diet whose composition fell within the range encompassed by modern hunter-gatherers, then it appears that hominid evolution was characterized by increasing diet quality. The distinctiveness of the human diet relative to other primates is likely associated with important metabolic differences. One possibility is that the high caloric density of our diet is reflective of elevated resting metabolic requirements. Indeed, the work of McNab (1986) and Nagy

80

W.R. LEONARD AND M.L. ROBERTSON

TABLE 2. Body weight and RMKs o f 31 primate species Species Lemuroidea and Lorisoidea Arctocebus calabarensis Cheirogales medius Galugo crassicaudatus G . demidouii G . elegantulus G. senegalensis Loris tardigradus Micmcebus murinus Nycticebus coucang Perodicticus potto Pmpithecus uerreaui Ceboidea Alouatta palliata Aotus triuirgatus Callathriz geofroyi Cehuella pygmaea Saimiri sriureus Cercopithecoidea Cercocebus torquatus Cercopithecus mitis Colobus guereza Erythmcebus patas Maraca fascicularis M. fusrata M. mulatta Papio anubis P. eynoeephnlus P. papio P. ursinus Hominoidea Hvlobates lar Pan troglodytes Pongo pygmaeus Homo sapiens

Weight (kg)

RMR (kcalldl

0.206 0.30 0.950 0.058 0.26 0.215 0.284 0.054 1.377 0.998 3.08

15.2 22.7 47.6 6.3 25.1 18.1 14.8 4.9 32.4 41.3 86.8

4.67 1.02 0.225 0.105 0.85

231.9 52.4 27.0 10.1 68.8

Milton et a1 (1979, Goffart (1977) Scholander et a1 (1950) Morrison and Middleton (1967) LeMaho et a1 (1981)

4.00 8.50 10.45 3.00 7.10 9.58 5.38 9.50 14.30 6.23 16.62

196.2 407.7 357.9 186.9 400.9 485.4 231.9 342.9 668.9 297.3 589.3

Bruhn (1934) Muller et a1 (1983) Muller et a1 (1983) Mahoney (1980) Tokura et a1 (1981) Nakavdma et a1 (1971) Bruhn (1934) Proppe and Gale (1970) Hohimer and Smith (1979) Bruhn (1934) Goldstone et a1 (1967)

1.90 18.30 16.20 54.00

123.4 581.9 569.1 1438.5

(1987) indicates that greater diet quality is associated with higher resting and total metabolic rates among mammalian species. An alternative hypothesis is that the high metabolic costs of our large brain necessitate an energy-rich diet (Martin, 1989; Leonard and Robertson, 1992). Each of these possibilities is examined subsequently. PRIMATE METABOLIC VARIATION Resting metabolic requirements Dietary patterns strongly influence metabolic requirements in mammalian species (McNab, 1978, 1986; Kurland and Pearson, 1986; Nagy, 1987).McNab (1986)found that mammals consuming high quality food items, such as, vertebrates, seeds, or nuts, tended to have high resting metabolic rates (RMR),while those consuming poor quality foods, e.g., leaves and woody plants, tended to be hypometabolic. If such a pattern holds for primates, it should be expected that metabolic rates will vary according to relative

References Hddwein 11972) McCormick (1981) Kurland and Pearson (1986) Hildwein (1972) Hildwein (1972) Dobler (1982) Muller et a1 (1985) Armstrong (1985) Muller 11975) Hildwein and Goffart (1975) Richard and Nicoll (1987)

Bruhn (1934) Riuhn (1934) Bruhn (1934) Bcnedict (1938)

diet quality. That is, deviations in DQ relative to body size should be correlated with deviations from predicted metabolic rate. Since humans have a higher DQ than their size would dictate, they should be expected t o have an elevated RMR (i.e., be hypermetabolic). Table 2 and Figure 2 present data on RMR (kcal/d) and body weight (kg) for humans and 30 nonhuman primate species (11 prosimians and 19 anthropoids). The solid line in Figure 2 represents the predicted RMR to body weight relationship based on the Kleiber (1961) equation: RMR

=

70.Wt0.75

where RMR = resting metabolic rate in kcal/d and Wt = weight in kilograms. As a group, primates conform closely to the Kleiber relationship; however, at an individual level, several species significantly depart from the predicted metabolic require-

HUMAN NUTRITIONAL EVOLUTION

81

Resting Metabolic Rate (kcal/d)

0 Anthropoids

10

1

1 0.01

1 0.1

10

1

100

Body Weight (kg) Fig. 2. Log-log plot of resting metabolic requirements (RMR in kcal/d) vs. body weight (kg) for 31 primate species. The solid line represents the predicted RMR-weight relationship based on the Kleiber (1961) equation, while the broken line represents the linear regression that best fits these data. A s a group, primates conform to the Kleiber relationship. However, several species (especially prosimians) depart substantially from their predicted RMRs.

ments. For prosimians, 9 of 11 species are hypometabolic, having metabolic rates that are more than 20% below their predicted levels. Among the anthropoids, two species are hypometabolic and three others are hypermetabolic (RMR more than 20% above predicted levels). Consequently, the allometric relationship between RMR and weight for anthropoids is almost identical to the Kleiber equation (RMR = 69[Wto.781; r2 = 0.98), while that of the prosimians departs substantially (RMR = 44[Wt0."]; r2 = 0.94). Contrary to expectations, the mean RMR for humans deviates by only +1%from what is predicted by the Kleiber equation. Moreover, when deviations from predicted DQ are regressed against deviations from predicted RMR for the entire sample, the correlation is low and not significant (r = .07; N.S.). Therefore, in contrast to the results of Kurland and Pearson (1986), there is no clear association between deviations in metabolic rate and diet quality. In other words, poor DQ relative to body size is not associated with hypometabolism and high DQ relative to body size is not associated with hypermetabolism. These results do not imply that body metabolism is unrelated to diet. To the contrary, Figure 3 shows that DQ is signifi-

cantly correlated with log-RMR (r = -0.557, P G 0.002 for entire sample; r = -0.654, P 6 0.001 when humans are excluded) as predicted by the Jarman-Bell model.'However, the amount of variance in DQ explained by RMR is no greater than that explained by body weight alone, a proxy measure of RMR. Hence, while RMR is an important correlate of diet (as predicted by the Jarman-Bell model), departures from predicted metabolic requirements are not. Humans, therefore, remain significantly above the regression line in Figure 3 (standardized residual [z] = 1.76), indicating that their diet quality is higher than expected for their metabolic needs. Brain size and metabolism

An additional explanation for the energyrich diet of our species is that it reflects an adaptation to the high metabolic costs of our large brains (Martin, 1983, 1989; Armstrong, 1985). If this hypothesis is correct, a correlation between relative brain size and relative diet quality among our primate sample should be expected.

'Because of differences in body weight between laboratory and free-living animals, RMlts in Figure 3 have been adjusted for the weight of animals in the wild (listed in Table 3).

W.R. LEONARD AND M.L. ROBERTSON

82

Diet Quality 400 7

I

++

1 I*

+

Prosimians Humans 4

0 3

I

I

I

0

I i l l 1 1

I

30

I

I

I

I

1 1 1 1

I

I

300

I

I

I

I

I

I

3000

Resting Metabolic Rate (kcal/d) Fig. 3. Plot of diet quality CDQ) relative to log-RMR (kcalldl for 25 primate species. As predicted by the Jarman-Bell model, DQ declines as total metabolic needs increase (r = 0.567 for total sample; r = -0.654 for nonhuman primates). Solid line represents the regression for the entire sample. Humans have a much higher quality diet than expected based on metabolic requirements. ~

Data on brain size and body size for the 31 primate species for which metabolic data are available, are presented in Table 3 and Figure 4. Three important points are evident from the log-logplot of brain size to body size in Figure 4: 1) brain size scales approximately t o the 0.75 power of body weight (slope = 0.81 for entire sample, 0.78 for nonhuman primates)'; 2) prosimians have small brains relative to body weight (z = -0.39); and 3) humans, of course, have relatively large brains (z = +3.21). The 3/4th scaling coefficient of brain-body weight is notable because it is the same coefficient observed for the RMR-body weight relationship. Therefore, when brain size is plotted relative to RMR, as shown in Figure 5, there is an isometric relationship (i.e., an exponent of 1)as the slope of the log-log regression is 0.98 (0.95 if humans are e x c l ~ d e d )This .~ means that there is a constant linear association between the two variables and implies that brain size is linked to metabolic output. Consequently, prosimians, which tend to be hypometabolic, do not depart from the brain-RMR regression line in Figure 5.

'For both regressions, the slopes are not significantly different from 0.75.

'FUvIRs have been adjusted for the body sizes presented in Table 3.

Thus, although prosimians have smaller brain:body weight ratios than anthropoids, they have comparable brain:RMR ratios (see also Armstrong, 1985, for a similar comparison between prosimians and monkeys). Departures from the brain-RMR regression indicate that a species expends a relatively larger or smaller proportion of its basal energy expenditure on brain metabolism. Humans, for example, significantly depart from the regression line spending about three to four times more on brain metabolism than other primates (20-25% of RMR vs. 8-9%; Holliday, 1986; Mink et al., 1981). If diet and brain metabolism are associated, deviations from the brain-RMR regression should be positively correlated with relative diet quality. Figure 6 plots the standardized residuals from the brain-RMR regression (relative brain size) vs. the residuals of DQ-RMR regression (relative diet quality) for the 25 species for which metabolic and dietary data are available. As predicted, there is a significant positive correlation between the two measures (r = 0.630; P s 0.001). The correlation remains significant when humans are excluded from the analysis (r = 0.571; P = 0.002). Thus, it appears that the amount of metabolic energy allocated to the brain is an important correlate of dietary patterns among extant primates.

HUMAN NUTRITIONAL EVOLUTION

TABLE 3. Body weight, brain weight and diet quality of31 primate species Weight’ Suecies

(ke)

Lemuroidea and Lorisoidea Arctocebus calabarensis Cheirogales medius Galago crassicaudatus G. demidovii G. elegantulus G. senegalensi.7 Loris tardigradus Microcebos murinus Nycticebus coucang Perodicticus potto Propithecus oerreauxi Ceboidea Alouattu palliata Aotus triuirgatus Callathrix geoffryi Cebuella pygmaea Saimiri sciureus Cercopithecoidea Cercocebus toryuatus Ceruopithecus mitis Culobus guereza Erythrocebus patas Macaca fascicularis M. fuscata M . mulatta Papio anubis P. cynocephalus P. papio P . ursinus Hominoidea Hylobates lar Pan troglodytes Pongo pygmueus Homo sapiens

Brain weight’ (8)

Diet quality index”

0.323 0.177 0.850 0.081 0.274 0.186 0.322 0.054 0.800 1.15 3.48

7.2 3.1 10.3 3.4 7.2 4.8 6.6 1.8 12.5 14.0 26.7

327.5

6.40 0.85 0.28 0.14 0.68

51.0 16.0 7.6 4.5 22.0

136.0 177.5 235.0 249.5 323.0

7.90 6.50 7.00 8.00 5.50 5.90 8.00 26.00 19.00 18.00 18.00

104.0 76.0 73.0 118.0 74.0 84.0 110.0 205.0 195.0 190.0 190.0

234.0 201.5 126.0

6.00 46.00 55.00 53.50

102.0 420.0 370.0 1.295.0

181.0 178.0 172.5 263.0

-

195.0 305.0 230.0

__

327.5

-

190.0 200.0

-

200.0 223.0 159.0 207.0 184.0 -

189.5

‘Body weight and brain weight data were derived from Strphan et at. (1981) and Jerison (1973). ‘DQ indices for nonhuman primates were calculated from dietary data presented in Hichard (1985) and Sailer et al. (1985).DQ index for humans 1s the median for t h e five foraging populations presented in Table 1.

These results imply that changes in diet quality during hominid evolution were linked with the evolution of brain size. The shift to a more calorically dense diet was probably needed in order to substantially increase the amount of metabolic energy being used by the hominid brain. Thus,while nutritional factors alone are not sufficient to explain the evolution of our large brains, it seems clear that certain dietary changes were necessary for substantial brain evolution to take place. DIET AND BRAIN SIZE IN HOMlNlD EVOLUTION

From the results presented thus far, it appears that human dietary needs were shaped by the metabolic demands of our

83

large brains. The question that remains is when in our evolutionary past these dietary and metabolic changes might have occurred. Although we can only speculate on the answer, the hominid fossil record provides some useful clues. Specifically, data on estimated brain size and body size can be used to determine how far these species deviate from the brain-body metabolism relationship in extant primates. Significant deviations in brain size relative to metabolic rate should be expected to signal the evolution of a more “human-like’’ (i.e., energy-rich) dietary strategy. Table 4 presents data on mean cranial capacity, estimated body weight (kg), and estimated RMR for six fossil hominid species. Cranial capacity data were derived from a number of different sources (Falk, 1987; Tobias, 1987; Wolpoff, 1984) so as to maximize sample sizes within each taxonomic group. Body weight data were derived from the most recent estimates of McHenry (1988, 1992a,b). RMRs (kcal/d) were estimated from the RMR-weight regression for the 20 anthropoid species in Table 2. The relationship between brain size and RMR for the six fossil groups and 16 extant anthropoid species is shown in Figure 7. The solid line is the regression derived from the extant primate sample and the dashed lines are the 95% confidence limits of that regression. All three of the australopithecine groups (Australopithecus afarensis, A. africanus, A. robustuslboisei) are accommodated within the 95% confidence limits for individual data points around the brainRMR regression. Homo habilis falls at about the 95% limit, while both H. erectus and H. sapiens fall beyond the 99% limits. It is intriguing that the clear departure from the general primate brain-metabolism regression occurs with the emergence of species of our own genus, H. habilis and H. erectus, since this is a time when other important anatomical and behavioral changes appear. Specifically, both the archeological and morphological evidence indicate that these early members of the genus Homo incorporated greater amounts of animal material in their diet than the australopithecines (Bunn, 1981; Wolpoff, 1980). With early Homo there is the first clear evidence of home bases, implying that resources were collected and brought back t o a central location where they were shared (Potts, 1988). Hence, it is likely that what supported the

W.R. LEONARD AND M.L. ROBERTSON

84 Brain Size (g)

-~

100

10

1

0.01

0.1

1

100

10

Body Weight (kg) Fig. 4. Log-log plot of brain size (g) vs. body size (kg)for 31 primate species. The solid line represents the regression for the entire sample (slope = 0.81), whereas the broken line is the regression for nonhuman primates (slope = 0.781. Brain size scales approximately to the 0.75 power of body weight. Thus, the allometric relationship between brain size and body size is the same as that between metabolic requirements and body size.

Brain Size (9)

I

* /

100 -

E

lo

1

t 1

I

3

p - i - L 1

1 , l I I l

30

I

I , I 1 , 1

I

I

I

300

l

l

1 1 ,

3000

Resting Metabolic Rate (kcal/day) Fig. 5. Log-log plot of brain size ( g ) vs. RMR (kcal/d) for 31 primate species. Best fit regression (solid line) shows that brain size scales isometrically with metabolic requirements (slope = 0.98).This relationship implies that the proportion of RMR spent on brain metabolism is constant (%9%)across primates of all size. Humans depart substantially from the regression, using 2@25% of RMR for brain metabolism.

rapid expansion of brain size in Homo habi-

lis and Homo erectus were both the higher quality and greater stability of the diet. CONCLUSIONS We have demonstrated that humans have a much higher quality diet than expected for

their size or their resting metabolic needs. Although we are hardly carnivores, humans do consume substantially more meat than any primate of our size. Contemporary human foraging groups obtain a t least 30% of their dietary energy from animals foods, compared to 557% in chimpanzees. Adapta-

HUMAN NUTRITIONAL EVOLUTION

85

Diet Quality Residuals

3--

-1

~

~~

~~

~

-

I

2l 't

t

+

,

I

t

+

ij

I

Prosimians

3

Anthropoids

f

Humans

I

,

,~~ , ' :"

,

-2

- 3 - 2 - 1

0

1

2

3

4

Brain Size Residuals Fig. 6. Plot of relative diet quality (DQ) vs. relative brain size for 25 primate species. Relative DQ values are standardized residuals from the DQ-log RMR regression; relative hrain sizes reflect standardized residuals from the log-hrain vs. log-RMR regression. There is a significant positive correlation between these two mea-

TABLE 4. Brain size. body weight, and estimated resting metabolic requirements for s i r fossil horninid species

Brain

Body

size'

RMR3 (kcaVd) 1,145 1,109 1,186 1,255 1,638 1.657

Species

(cc)

weight' (ka)

Australopithecus afarensis A. africanus A. mbustuslboisei Homo habilis H. erectus H. sapiens (early)

404 441 516 640 945 1155

37.0 35.5 38.7 41.6 58.6 59.5

'Cranial capacity estimates for Australopithecus ufarensis taken from Falk (1987); estimates for A. afrzcanu, A. robustus, Homo habilrs taken from Tobias (1987);estimates forrH.erectus and early K supiens taken from Wolpoff (1984). 'Body weight estimates taken from McHenry (lYB8,1992a,b). 3RMR estimated from the anthropoid regression of RMR vs. weight presented in this study (RMR = 69 x Wt0-77*).

tion to this calorically dense, easy to digest diet is evident in our gut morphology, as humans have a relatively reduced digestive tract in comparison to most other primates (Sussman, 1987; Chivers and Hladik, 1980; Milton, 1987). This distinct diet appears to be linked to the high metabolic costs of the human brain. In general, primate brain size varies as a direct (linear) hnction of body metabolism. This means that the proportion of metabolic energy spent on the brain is relatively constant across primates of all size (about 8-9%

sures ir = 0.630, P < 0.001 for entire sample;r = 0.571, = 0.002 for nonhuman primates only). This association implies that the amount of energy spent on brain metabolism is an important correlate of dietary variation in primates.

P

of RMR). Species spending a larger proportion of RMR on their brain have a higher quality diet than expected for their body size. Conversely, small brains relative t o metabolic turnover are associated with poor quality diets. Humans represent the positive extreme, having both a very high quality diet and a brain that accounts for 2025%of resting metabolic energy. Other researchers have previously noted the apparent link between metabolic rate and brain size (Armstrong, 1985; Mink et al., 1981; Martin, 1989,1990). In particular, Martin (1989) has argued that this relationship reflects the association between brain growth and maternal metabolism. This hypothesis posits that since the majority of brain growth in humans and other primates occurs prenatally and early in the postnatal period, it is maternal metabolic output (through pregnancy and lactation) that largely determines achieved adult brain size. If this hypothesis is correct, the results of the present study would imply that improvement in the stability and quality of maternal nutrition (to support the high metabolic demands of pregnancy and lactation) was a consequence of the selection for larger brain size in hominid evolution. Evidence from the fossil record indicates that such a dietary shift may have appeared

W.R. LEONARD AND M.L. ROBERTSON

86

I

I

I

/

I

/

2000

200

Resting Metabolic Rate (kcal/d) Fig. 7. Log-log plot of brain size (g) vs. R M R (kcaVd) for 6 fossil hominid species and 16 extant anthropoids. The solid line represents the regression for the extant primate data. The broken lines delimit the 95% confidence intervals of that regression. All of the australopithecine species (Australopithecus afarensis, A. africa-

nus, A. rohustudhoisei) are accommodated within the 95% limits. Homo hubilis falls on the 95% boundary while H.erectus and early H. supiens fall beyond the 99% limits. Thus significant deviations from the primate brain-metabolism relationship occurred with the emergence of the genus Homo.

with the emergence of the genus Homo. The evolution of early Homo was associated with both rapid rates of brain evolution and important changes in material culture (Falk, 1987; Gowlett, 1984; Potts, 1988).Moreover, this period appears t o have coincided with a marked drying trend in eastern and southern Africa (Behrensmeyer and Cooke, 1985; Vrba, 1988). Vrba (1988, 1993) has demonstrated that at about 2.5 million years ago a major period of global cooling had begun which led to a dramatic increase in the amount of open, grassland environments in Africa. Such ecological change would have directly influenced both the density and distribution of high quality plant foods and likely would have made animals an increasingly attractive food resource (Vrba, 1988). Different hominid species of that time appear to have adapted to the environmental drying in different ways. Early members of the genus Homo (H. habilis and H . erectus) appear to have included larger amounts of meat in their diet. while those of the robust australopithecine 'lineage (e.g., A. robustus, A. boisei) continued to subsist largely on fibrous lower quality plant foods. Thus, these

ecologically initiated changes in foraging behavior and diet likely provided the basis for sustaining selection for rapid brain evolution in early members of the genus Homo. Clearly, studies of comparative human and primate nutrition and metabolism can offer important insights into the evolution of our species. Indeed, as Stanley Garn correctly predicted back in 1966, the study of diet and nutrition has become a central part of our discipline. The challenge for us as human biologists is to continue to follow his example in framing the study of our species in an explicitly evolutionary perspective rather than being content to simply describe its variability. ACKNOWLEDGMENTS

We are grateful to Drs. A. Roberto Frisancho and Robert M. Malina for the invitation to participate in the symposium honoring Stanley Garn. Stanley's insights, support, and encouragement over the years are deeDlv amreciated. This research was SUDported by a grant from the Natural Sciencls and Engineering Research Council of Canada (OGP0116785). I

I

I I

HUMAN NUTRITIONAL EVOLUTION

LITERATURE CITED Ardrey R (1961) African Genesis. New York: Dell. Armstrong E (1985) Relative brain size in monkeys and prosimians. Am. J. Phys. Anthropol. 66t263-273. Behrensmeyer AK,and Cooke HBS (1985) Paleoenvironments, stratigraphy, and taphonomy in the African Pliocene and early Pleistocene. In E Delson (ed.): Ancestors: The Hard Evidence. New York: Alan R. Liss, pp. 60-62. Bell RII (1971) A grazing ecosystem in the Serengeti. Sci. Am. 225t86-93 (Jan 1). Benedict FG (1938) Vital energetics. Washington, DC: Carnegie Institution, Publication no. 503. Bruhn JM (1934) The respiratory metabolism of infrahuman primates. Am. J. Physiol. 110t477-484. Bunn HT (1981) Archeological evidence for meat eating by Plio-Pleistocene hominids from Koobi-Fora and Olduvai Gorge. Nature 291574-577. Chivers DJ, Hladik CH (1980) Morphology of the gastrointestinal tract in primates: Comparisons with other mammals in relation to diet. J. Morphol. 166; 337-386. Clutton-Brock TH, and Harvey PH (1977) Species differences in feeding and ranging behavior in primates. In TH Clutton-Brock (ed.): Primate Ecology: Studies of Feeding and Ranging Behavior in Lemurs, Monkeys and Apes. New York Academic Press, pp. 557-579. Dart FL4 (1957) The osteodontokeratic culture of Australopithecus prometheus. Memoirs of the Transvaal Museum. 1O:l-105. Dietz WH, Marino B, Peacock NR, and Bailey RC (1989) Nutritional status of Efe pygmies and Lese horticulturalists. Am. J. Phys. Anthropol. 78509-518. Dobler H J (1982) Temperaturregulation und Sauerstoffvebrauch beim Senegalund Zwergalago (Galago senegalensis, Galago (galagoides} demidouii). Bonn Zool. Beitr. lr3349. Eaton SB, and Konner M (1985) Paleolithic nutrition: A consideration of its nature and current implications. N. Eng. J. Med. 322t283-289. Falk D (1987) Hominid paleoneurology. Ann. Rev. Anthropol. 16tl3-30. Garn SM (1966) Nutrition in physical anthropology. Am. J. Phys. Anthropol. 24.289-292. Garn SM, and Leonard WR (1989) What did our ancestors eat? Nutr. Rev. 47:337-345. Gaulin SJC (1979)A JarmadBell model of primate feeding niches. Hum. Ecol. 7:l-20. Gaulin SJC, and Konner M (1977) On the natural diets of primates, including humans. In R Wurtman and J Wurtman (eds.):Nutrition and the Brain, Vol. 1. New York: Raven Press, pp. 1-86. Goffart M (1977) Hypometabolisme ches Aotus triuirgatus. C.R. SOC.Biol. (Paris) 171:114%1152. Goldstone BW, Savage N, and Steffens FE (1967) Relation of basal oxygen consumption to some aspect of body size in baboons (Papio ursinus). J. Appl. Physiol. 22:8&90. Goodall AG (1977) Feeding behavior of the mountain gorilla group (Gorilla gorilla berengei) in the Tshibinda-Kahuzi region (Zaire). In TH Cluttan-Brach (ed.): Primate Ecology: Studies of Feeding and Ranging Behavior in I,emurs, Monkeys, and Apes. New York: Academic Press, pp. 450479. Gordon KD (1987) Evolutionary perspectives on the human diet. In FE Johnston (ed.): Nutritional Anthropology. New York: Alan R. Liss, pp. 3-39. Gowlett JAJ (1984) Mental abilities of early man: A look

87

at some hard evidence. In R Foley (ed.): Hominid Evolution and Community Ecology. New York: Academic Press, pp. 167-192. Hildwein G (1972) Metabolisme energetique de quelques mamiferes et oiseaux de la foret equatoriale. Arch. Sci. Physiol. 26t387-400. Hildwein G, and Goffart M (1975) Standard metabolism and thermoregulation in a prosimian, Perodicticus potto. Comp. Biochem. Physiol. iiOAt201-213. Hill K (1982) Hunting and human evolution. J. Hum. Evol. IIt521-544. Hill K, Hawkes K, Hurtado M, and Kaplan H (1984) Seasonal variance in the diet of Ache hunter-gatherers in Eastern Paraguay. Hum. Ecol. 12t101-135. Hohimer AR, and Smith OA (1979) Decreased renal blood flow in the baboon during mild dynamic leg exercise. Am. J. Physiol. 236tH141-150. Holliday MA (1986) Body composition and energy needs during growth. In F. Falkner and J M Tanner (eds.): Human Growth: A Comprehensive Treatise, Vol. 2 (2nd Ed.). New York Plenum, pp. 101-117. Hurtado AM, and Hill K(1989) Seasonality in a foraging society: Variation in diet, work effort, fertility, and sexual division of labor among the Hiwi of Venezuela. J. Anthropol. Res. 34:293-346. Jarman PJ (1974) The social organization of antelope in relation to their ecology. Behaviour 58.215-267. Jerison H J (1973) Evolution of the Brain and Intelligence. New York: Academic Press. Jolly C (1970) The seed eaters: A new model of hominid differentiation based on a baboon analogy. Man 5 5 26. Kleiber M (1961) The Fire of Life: An Introduction to Animal Energetics. Huntington, NY:Krieger. Kurland JA, and Pearson J D (1986) Ecological significance of hypometabolism in nonhuman primates: Allometry, adaptation, and deviant diets. Am. J. Phys. Anthropol. 71:445-457. Lee RB (1968)What hunters do for a living, or how to make out on scarce resources. In RB Lee and I Devore (eds.): Man the Hunter. Hawthorne, Ny: Aldine, pp. 3M8. Lee RB (1979) The !Kung San: Men, Women and Work in a Foraging Society. Cambridge: Cambridge University Press. Le Maho Y, Goffart M, Rochas A, Felbabel H, and Chatonnet J (1981) Thermoregulation in the only nocturnal simian: The night monkey Aotus triuirgatus. Am. J.Physiol. 240tR156-165. Leonard WR, and Robertson ML (1992) Nutritional requirements and human evolution: A bioenergetics model. Am. J.Hum. Biol. 4t179-195. Leonard WR, and Thomas RB (1989) Biosocial responses to seasonal food stress in highland Peru. Hum. Biol. 61t6545. Leonard WR, DeWalt KM, Uquillas J E , and DeWalt BR (1993) Ecological correlates of dietary consumption and nutritional status in highland and coastal Ecuador. Ecol. Food Nutr. 31t67-85. Mahoney SA (1980)Cost of locomotion and heat balance during rest and running from 0" to 55°C in a patas monkey. J . Appl. Physiol. 49r789-800. Martin RD (1983) Human Brain Evolution in a n Ecological Context. Fifty Second James Arthur Lecture on the Evolution of the Human Brain. New York: American Museum of Natural History. Martin RD (1989)Evolution of the brain in early hominids. Ossa 14:4962. Martin RD (1990) Primate Origins and Evolution: A

88

W.R. LEONARD AND M.L. ROBERTSON

Phylogenetic Reconstruction. Princeton, NJ: Princeton University Press. McCormick SA (1981) Oxygen consumption and torpor in the fat-tailed dwarf lemur (Cheirogaleus medius): Rethinking prosimian metabolism. Comp. Biochem. Physiol. 68A:605-610. McGrew WC (1974) Tool use by chimpanzees in feeding upon driver ants. J . Hum. Evol. 3501-508. McHenry HM (1988) New estimates of body size in australopithecines. In FE Grine (ed.): Evolutionary History of the “Robust” Australopithecines. Hawthorne, NY:Aldine, pp. 133-148. McHenry HM (1992a) How big were early hominids? Evol. Anthropol. 1:15-20. McHenry HM (199213)Body size and proportions in early hominids. Am. J . Phys. Anthropol. 87:407-431. McNab BK (1978)Energetics of arboreal folivores: Physiological problems and ecological consequences of feeding on a ubiquitous food supply. In GG Montgomery (ed.): The Ecology of Arboreal Folivores. Washington, DC: Smithsonian Institution Press, pp. 153-162. McNab BK (1986) The influence of food habits on the energetics of eutherian mammals. Ecol. Monog. 56: 1-19. Milton K (1987) Primate diets and gut morphology: Implications for hominid evolution. In M Harris and EB Ross (eds.): Food and Evolution: Toward a Theory of Human Food Habits. Philadelphia: Temple University Press, pp. 93-115. Milton K, Casey TM, and Casey KK (1979) The basal metabolism of mantled howler monkeys (Alouatta palliata). J. Mammol. 60t373-376. Blumenschine RJ, and Adams DB (1981) RaMink JW, tio of central nervous system to body metabolism in vertebrates: Its constancy and functional basis. Am. J . Physiol. 241rR203-R212. Morrison PR, and Middleton E (1967)Body temperature and metabolism in the pygmy marmoset. Folia Primatol. 6t70-82. Muller E F (1975) Temperature regulation in the slow loris. Naturwissenchaften 62140-141. Muller EF, Kamau JMZ,and Maloiy GMO (1983)A comparative study of basal metabolism and thermoregulation in a folivorous (Colobusguereza) and an omnivorous (Cercopithecus rnitis) primate species. Comp. Biochem. Physiol. 31t332-337. Muller EF, Nieschalk U, and Meier B (1985)Thermoregulation in the slender loris (Loris turdigradus). Folia Primatol. 44216-226. Nagy KA (1987) Field metabolic rate and food requirement scaling in mammals and birds. Ecol. Monog. 57: 111-128. Nakayama T, Hori T, Nagasaki T, Tokura H, and Tadaki E (1971) Thermal and metabolic responses in the Japanese Monkey at temperatures of 5-38”. J. Appl. Physiol. 31:332-337. Potts R (1988) Early Horninid Activities at Olduvai. Hawthorne, Ny: Aldine. Proppe DW, and Gale CC (1970) Endocrine thermoregulatory responses to local hypothalamic warming in unanesthetized baboons. Am. J . Physiol. 219:202207. Richard AF (1985) Primates in Nature. New York: WH Freeman. Richard AF, and Nicoll ME (1987) Female social domi-

nance and basal metabolism in a Malagasy primate, Propithecus uerreaui. Am. J. Primatol. 12:309314. Rode A, and Shephard R J (1973) Growth, development and fitness of the Canadian Eskimo. Med. Sci. Sport 5:161-1 69. Rodman PS (1977) Feeding behaviour of orang-utans of the Kutai nature reserve, East Kalimantan. In TH Clutton-Brock (ed.):Primate Ecology: Studies of Feeding and Ranging Behaviour in Lemurs, Monkeys and Apes. New York: Academic Press, pp. 383413. Rodman PS (1984)Foraging and social systems in Orangutans and chimpanzees. In PS Rodman and JGH Cant (eds.): Adaptations for Foraging in Nonhuman Primates. New York Columbia University Press, pp. 13P 160. Sailer LD, Gaulin SJC, Boster JS, and Kurland JA (1985) Measuring the relationship between dietary quality and body size in primates. Primates26r14-27. Scholander PF, Hock R, Walters V, Johnson F, and Irving L (1950) Heat regulation in some arctic and tropical mammals and birds. Biol. Bull. 99:237-258. Stephan H, Frahm H, and Baron G (1981) New and revised data on volumes of brain structures in insectivores and primates. Folia Primatol. 35:l-29. Sussman RW (1987) Species-specific dietary patterns in primates and human dietary adaptations. In WG Kinzey (ed.): The Evolution of Human Behavior: Primate Models. Albany, Ny: SUNY Press, pp. 131-179. Tanner NM (1981) On Becoming Human. Cambridge: Cambridge University Press. Teleki G (1981)The omnivorous diet and eclectic feeding habits of chimpanzees in Gombe National Park, Tanzania. In RSO Harding and G Teleki (eds.): Omnivorous Primates: Gathering and Hunting in Human Evolution. New York: Columbia University Press, pp. 303-343. Temerin LA, Wheatley BP, and Rodman PS (1984) Body size and foraging in primates. In PS Rodman and JGH Cant (eds.): Adaptation for Foraging in Nonhuman Primates. New York: Columbia University Press, pp. 215-248. TobiasPV(1987)Thebrain ofHorno habilis: Anew level of organization of cerebral evolution. J. Hum. Evol. 16t741-761. Tokura H, Tanaka N, Nakagawa S, and Ohsawa W (1981) Thermal and metabolic responses in the Japanese Macaque (Macnca fuscata) acclimated to an ambient temperature of 5°C. Comp. Biochem. Physiol. 69A59 1-594. Vrba ES (1988) Late Pliocene climatic events and hominid evolution. In FE Grine (ed.): Evolutionary History of the “Robust” Australopithecines. Hawthorne, Ny: Aldine, pp. 405-426. Vrba ES (1993) The pulse that produced us. Nat. Hist. 102r47-51 (May). Washburn SL, and Lancaster CS (1968) The evolution of hunting. In RB Lee and I DeVore (eds.): Man the Hunter. Hawthorne, NY:Aldine, pp. 293-303. Wolpoff MH (1980) PaLeoanthropology. New York: Knopf. WolpoffMH (1984)Evolution ofHomo erectus: The question of stasis. Paleobiology IOr389406. Wu Leung W-T (1968) Food Composition Table for Use in Africa. Bethesda, MD: US Department of Health, Education and Welfare.