How have metabolic constraints affected human evolution?

9 minute read

In the 2002 Annual Reviews in Anthropology, Leslie Aiello and Jonathan Wells provide a synopsis of the ways that morphological evolution in the human lineage have affected the energy utilization of our species and its ancestors. These considerations focus on body size, because it is tightly correlated with metabolic rates among mammals. Secondarily, they focus upon the relative sizes and proportions of different organs, especially the relative sizes of the brain and of the gut.

Aiello and Wells begin with the origin of early humans (which they assign to Homo ergaster). The most important change to happen at this time was an increase in body mass, likely to nearly twice the mass of female australopithecines. But there was also a complex of other changes that occurred at this time, including a change to more humanlike body proportions, with longer legs, and a change in diet toward higher energy food resources, probably including meat. One aspect of this change that the review summarizes is this our regulatory benefits of longer limbs, more linear physiques, and the effects of heat load upon a birthweight and energy expenditure during pregnancy, following the work of Wheeler (e.g. 1991), Ruff (1991), and Wells. From the increase in body mass, they calculate the necessary increase in resting metabolic rate following Kleiber's equations. They estimate that compared to Australopithecus afarensis, early humans would have required 39 percent more energy to meet resting metabolic requirements, with a much larger increase in females than males (resulting from the marked decrease in sexual dimorphism).

The authors spend a section considering what the dietary balance of these early humans must have been. They note that modern hunter gatherers with especially high energy requirements, such as Eskimos, meet these requirements by including a very high proportion of meat in their diets. Interestingly, they argue against this dietary model for the earliest humans, not on the basis of ecological reconstruction or arguments about scavenging vs foraging, but instead on the basis of the thermoregulatory requirements of meat digestion. They argue that digestion of meat produces more heat than the digestion of other kinds of foods, especially if the meat is protein-rich and fat-poor. They also note that the digestion of protein requires a great deal of water, which would be relatively scarce for savanna-based hominids. They do not use these arguments to suggest that early humans lacked meat in their diets, but instead emphasize that a balance between different dietary sources would be more advantageous, particularly if dietary fat were relatively unavailable.

Aiello and Wells then turn to the expensive tissue hypothesis. In brief, this hypothesis builds on the observation that some tissues require more energy for their resting metabolism than others. In particular nervous tissue is very expensive and digestive tissue is also quite expensive. The relative sizes of most body tissues are relatively constrained by functional requirements. But a reduction in dietary bulk might allow natural selection to pare away digestive tissues that were less necessary for food absorption, making energy available for the expansion of other tissues such as brain tissue (Aiello and Wheeler 1995). A novel element in this review is the inclusion of body fat and its potential complication in the estimation of resting metabolic rate. Aiello and Wells hypothesize that later members of the genus Homo were relatively fatter than earlier fossil hominids. It is well recognized that living humans in Westernized societies have a higher percentage of body fat than most mammals in wild populations. Fat tissue, called adipose tissue, has a relatively low contribution to overall metabolic rate. This means that an a fatter person will have a relatively lower metabolic rate than a leaner person of the same mass. If it is true that recent humans generally have been fatter than earlier humans, then even if their mass remained unchanged, the resting metabolic rates of human populations may have actually increased relative to their fat-free mass. This is a bit of a convoluted argument, dedicated to a single problem. Considering that recent humans have larger brains than their ancestors, the expensive tissue hypothesis predicts that either human metabolic rates must have increased over time, or that the relative mass of some other expensive tissues must have decreased. If humans became fatter at the same time that their brains increased in size, then the increase in fat mass might contribute to a relative reduction in the energetic requirements of the overall body mass. This reduction would have allowed an increase in the metabolic requirements of brain tissue. Aiello and Wells propose that these two opposing forces may have balanced each other, with the change in body composition underwriting the increase in size of the human brain.

Needless to say this vision is complicated by the fact that we have no evidence of soft tissue proportions in fossil humans. Likewise, there is no special reason to believe that recent humans--except for people in industrialized societies--actually are much fatter than their fossil ancestors. Aiello and Wells cite a study (Lawrence et al. 1987) that estimates body fat percentage in women in harsh environments at around 20 percent. This estimate would of course be higher for women than for men, considering the increased fat storage available in breast tissue and other sources in women. But it is far from clear that even human women have higher fat storage than females in other primate species. This is a subject that clearly needs further study.

There are several interrelated forces that do suggest that increased fat storage may have been an important human adaptation, possibly as early as the earliest fossil evidence of humans. One of these is the relatively high reproductive rate of humans compared to other hominoids. The short birth intervals of humans, combined with the rapid transition from weaning to new pregnancies among human mothers is significantly enhanced by the ability to store energy and smoothe out fluctuations in dietary resource availability. This hypothesis is supported by the observations that female reproductive fitness appears to depend on fat stores to some extent, and very thin women have a higher rate of miscarriage and a higher likelihood of low birthweight babies. Another is the fact that humans modify their group sizes with much less seasonal variation that is observed in chimpanzees.

Another novel element of this review is the consideration of energy costs as they change during early ontogeny. Aiello and Wells note that the metabolic costs of the brain are especially high very early in life, because of the early growth of the brain. They cite an estimate that the brain requires up to 70 percent of the total energy costs of the individual (Holliday 1986), although this sounds like an overestimate. They suggest that human children meet these energy requirements in part by compromising their rate of growth, especially during the time period between 2 and 12 years of age. The unique ontogeny of human growth in includes this time span during which chimpanzees actually have faster growth rates than humans, followed by the "adolescent growth spurt" in humans, when childhood growth rates continue and may even increase (Ulijaszek 1995). This alteration in rates may reflect the increased resting metabolic requirements of tissue proportions in human children. The authors also note that parent-offspring conflict provides another explanation for slowed growth rates in human children, considering that parents and children must both depend on the same resources, collected by parents. I might add to that the competition between children and new siblings, considering that at the weaning age it is likely that most human children would very shortly have seen their mothers investing their energetic resources in growing pregnancies and the care of subsequent offspring.

In the final section, the authors make suggestions about the relationship between energetic evolution and social organization. Citing Aiello and Key (2002), they provide estimates that indicate that lactating early humans would have had energy requirements 45 percent higher than lactating australopithecines, and "almost 100 percent higher than for a nonlactating and nongestating smaller-bodied hominin" (333). That paper argues that a reduction in the birth interval would have reduced the energy costs per offspring, at the same time that it increased overall reproductive output. The reduction is most noted in the length of lactation, which is the most expensive part of direct maternal investment. In that paper, they note that only a major shift in diet could allow this change to happen, because the resources lost to children by a reduction in lactation length would have to be replaced by other foods, presumably high-energy foods like meat. But they also note that other individuals besides the mother might be involved in providing these resources to children. They focus on the possible effects of grandmothers, following Kristin Hawkes and colleagues (1997). One might include on this list the possibility of paternal investment, or the contributions of other group members. In any event, the social changes necessary for effective hunter-gatherer foraging strategies, which necessitate the sharing of both risks and benefits of hunting, would help to support this strategy.

These energetic ideas leave several questions unanswered, mostly based around assumptions that cannot be verified. For example, what was the birth interval of australopithecines? If this birth interval were short, as in recent humans, then the transition to a more human-like body size might not been accompanied by such extensive social changes. And were the predecessors of early humans--such as Homo habilis--themselves fat? One might expect that the origins of tool use corresponded with the origins of some of the dietary changes that Aiello and Wells argued characterized early humans. If so, then energy storage must have been an integral element of this toolmaking adaptation. Indeed, the evidence the brain size first increase in Homo habilis or another small-bodied hominid directly detracts from the idea that energetic changes were the principal driving factors of changes in body size, sociality, or other early human characteristics. It goes without saying that we don't directly know what the proportion of different "expensive" tissues in australopithecines or any other early hominid might have been, beyond the suggestion that the reconstructed skeleton of Lucy had a broad gut and pelvic cavity.


Aiello LC, Key C. 2002. The energetic consequences of being a Homo erectus female. Am J Hum Biol 14:551-565.

Aiello LC, Wells JCK. 2002. Energetics and the evolution of the genus Homo. Annu Rev Anthropol 31:323-338.

Aiello LC, Wheeler P. 1995. The expensive-tissue hypothesis: the brain and the digestive system in human and primate evolution. Curr Anthropol 36:199-221.

Hawkes K, O'Connell JF, Blurton-Jones NG, Alvarez H, Charnov EL. 1998. Grandmothering, menopause, and the evolution of human life histories. Proc Natl Acad Sci U S A 95:1336-1339.

Holliday MA. 1986. Body composition and energy needs during growth. In: Falkner F, Tanner JM, editors, Human growth: A comprehensive treatise, 2 edition. New York: Plenum. p 101-107.

Ruff CB. 1991. Climate and body shape in hominid evolution. J Hum Evol 21:81-105.

Ulijaszek SJ. 1995. Energetics and human evolution. In: Ulijaszek SJ, editor, Human energetics in biological anthropology. Cambridge, UK: Cambridge University Press. p 166-175.

Wheeler PE. 1991. The thermoregulatory advantages of hominid bipedalism in open equatorial environments: the contribution of increased heat loss and cutaneous evaporative cooling. J Hum Evol 21:107-115.