Please note that you are viewing the old Science in Africa website. Please see for our new site!

Science in AfricaLogo Merck: Distributors of fine chemicals and apparatus. Enter here for more information.



How to extract information about diet from those ancient teeth and bones 

 Dr Julia Lee-Thorp
Archaeology Department, 
University of Cape Town

In the last issue Dr Becky Rogers Ackermann described how our views on the shape of our family tree, or rather, family bush, are being changed by exciting new early hominid discoveries. But discovery is really just the start of the scientific 'detective' trail. After the immediate questions about relationships of the new finds to other hominids, judged from their morphology (size and shape), contexts and ages, follows another important set of questions about how they lived and behaved. 

       This is an area where recent advances in analysis of the fossil tissues themselves are enhancing our understanding of the lifeways of early hominids. I will focus mainly on learning more about the foods that were eaten, using a particular kind of chemical tool called stable light isotope analysis. South African scientists have pioneered its application to hominid diets.

Why is it important to know about the foods early hominids ate? 

       Most primates (not forgetting us humans) spend a large proportion of waking hours searching for, and eating food, so diet is a major factor underlying behaviour and ecology. Finding out what they ate is a very important step in establishing what "a day in the life" was like for an early hominid. And for the same reason, dietary differences and changes are inextricably bound with evolutionary pathways. 

        Enter here for full picture. Raymond Dart contemplating the Taung cranium shortly after the announcement. Illustration from frontispiece in P.V. Tobias (Ed) Hominid Evolution, Past, Present and Future, Alan Liss, New York.The primary importance of diet was first recognised long ago by Raymond Dart who discovered the first hominid to be found in Africa - the "Taung child" (Australopithecus africanus) (enter figure to the right for larger image). Dart puzzled over how these "man-like apes" had survived in an apparently open, arid environment (the Northern Cape) so hostile to their forest-loving great ape cousins. He suggested an expansion of the hominoid menu to include "insects and scorpions, lizards and bird's eggs, berries and grubs" (Dart, 1959, p. 7). 

        Debates about hominid diets have raged back and forth since then. As we shall see below, Dart's original idea has turned out to be pretty good, even though we now know that the ancient Taung environment did not look much like the arid place it is today! 

How can we investigate diets of species that have been extinct for millions of years? 

        Enter here for full picture.  Some very large teeth! Fragment of the mandible of an Australopithecus robustus specimen from Swartkrans. Photograph by David Brill, Transvaal Museum.It's not easy! The shape and structure of teeth (dental morphology) provides some guidance because teeth are adapted for processing food. For instance, the giant molars of the Australopithecines suggest that they needed to process very tough food (see figure to the right) (Ungar, 1998). But phylogenetic history also plays a role in tooth morphology, and adaptations are not necessarily the same as actual behaviour. For example, Papio baboons have tooth shapes indicative of fruit diets, but many modern baboons eat as much as 50% grass for which they are poorly equipped. The problem is worse in animals that are 'generalists' (ie. can eat a bit of everything) like hominids. 

        Some foods leave microscopic traces on teeth. Certain diets such as those rich in hard fruits or grasses leave tiny distinctive damage patterns on enamel surfaces. Based on different microwear patterns, Fred Grine suggested that Australopithecus africanus ate a diet with fleshy fruits and leaves, while A. robustus ate harder, more fibrous foods (Grine, 1981). Unfortunately microwear only reflects the consistency of foods eaten in the last few days or weeks, and many foods, such as animal flesh, are "invisible". Scatters of stone tools and animal bones in former living sites can provide some clues about how food was acquired although the stone tools do not tell us much about diet. It's also hard to tell which hominid was using them. In addition, stone tools appeared about 2.5 million years ago, so they cannot help us in the case of earlier hominids. 

Other options for investigating hominid menus: introducing stable carbon isotopes.

Valuable information is locked up in the crystal structure of fossil tooth enamel. The tooth enamel of fossils is usually remarkably well-preserved (see figure above and follow link for full image), and this is certainly the case for the fossils found in the South African sites. Tooth enamel is a crystalline mineral made up mostly of calcium and phosphate, but small amounts of other ions are also present, including carbonate ions. This tiny carbonate inclusion preserves, in the carbon isotope signatures, a record of certain classes of foods that were eaten when the tooth was forming!

The way it works goes something like this. The ratios of the two stable forms of carbon (distinguished only by their atomic mass, 13C and 12C), provide a natural tracer tool for the chemical and biochemical reactions of the carbon cycle. In African savanna or summer rainfall environments, trees, shrubs and herbs follow a photosynthesis pathway (called the C3 pathway) which discriminates strongly against 13C, whereas another pathway used by tropical grasses (the C4 pathway) does not. The result is that the two Enter here for the full image groups of plants have very distinct 13C/12C ratios (see figure to the right and follow link for full image). Animals incorporate the plant carbon they eat into their tissues, which then directly reflect proportions of C4 grasses and C3 plants eaten. For instance, grazers (such as wildebeests) are enriched in 13C compared to animals that eat C3 foods such as the leaves or fruits of trees and bushes (like giraffes and chimpanzees). In the case of fossil animals, analysis is limited to bones and teeth, which alone are preserved in the fossil record. Original isotopic signatures are best preserved in enamel in samples that are millions of years old, whereas bone tends to become altered. 

What does stable carbon isotope analysis tell us about hominid diets?

Analysis of A. africanus tooth enamel from Makapansgat Limeworks (about 3 million years old), and A. robustus and early Homo from Swartkrans (about 1.5- 1.7 million years old) produced some surprising results. If A. africanus was a fruit- and leaf- eater as suggested by the microwear analysis, they should show corresponding C3 carbon isotope signatures. But they do not! Instead, the results show that, on average, 25% of their dietary carbon came from grasses. For one individual at Makapansgat it was more like 50%. It was also surprising that the same pattern held for all the hominids at Makapansgat and Swartkrans, over a period of some 1 to 2 million years (see figure above). 

None of the hominids analysed so far ate a diet like that of the modern chimpanzee, gorilla, or even orangutan, all of which eat nearly 100% C3 foods. This is not to say that they did not eat fruits and leaves - they most probably did. But they also ate quantities of actual grasses, or animals that ate the grasses, or both. Grass itself is difficult to process and to extract the nutrients (unless one is well-equipped to do so, like a cow), so it's difficult to visualise how such a large ''grass" signature could occur unless the hominids ate some animal foods. C4 -consuming invertebrate and vertebrate animals were abundant and easily collected by hominids. Raymond Dart was on the right track all those years ago, even if his environmental scenario was not quite right! 

The important point is that we now know that all of these hominids were willing to eat C4 resources that are generally ignored by our primate cousins, the chimpanzees, gorillas, and orangutans. Chimpanzees, for instance, stick to C3 'forest' foods even when grasses or grass-eating animals are abundant. It seems that hominids early on became dietary generalists who broadened their diets and thus their resource base. This may have been the seminal step in the development of the hominid lineage. It makes sense when one considers that global climates changed between about 4 - 1.8 millions years ago, causing African forests to be replaced by woodlands and grasslands. 

More Information:

Dr Julia Lee-Thorp
Archaeology Department, 
University of Cape Town, 
Rondebosch 7701

References and further reading:

Brain, C.K. (1981). The Hunters or the Hunted? Chicago: University of Chicago Press.

Chang, K. (1999) What the hominid ate. ABC News Science

Dart, R.A. (1959). Adventures with the Missing Link. Hamish Hamilton, London, pp.1-251.

Grine, F.E. (1981). Trophic differences between gracile and robust australopithecines. South African Journal of Science 77, 203-230.

Lee-Thorp, J.A., Thackeray, J. F. and van der Merwe, N.J. (2001). The hunters and the hunted revisited. Journal of Human Evolution, 39, 565-576.

Mayell, H. (1999) Pass the meat: precursor to humans had a varied diet. National Geographic News.

Sponheimer, M. and Lee-Thorp, J.A. (1999a). Isotopic evidence for the diet of an early hominid, Australopithecus africanus. Science 283, 368-370.

Ungar, P. (1998). Dental allometry, morphology, and wear as evidence for diet in fossil primates. Evolutionary Anthropology 6, 205-217.

Vogel, G. (1999) Give me what the hairy guy's having. Science Daily InSCIght.

Science in Africa - Africa's First On-Line Science Magazine

Return to Home PageReturn to the TopYour FeedbackRegister with "Science in Africa"

Copyright  2001, Janice Limson. All Rights Reserved