The Obligate Fruigivore

An Evolutionary Hypothesis
J S Coleman


The purpose of this article is to explore the relationship between diet, bipedalism, mothering, and how these factors enabled our superior cognitive development. It has become something of an obsession for some anthropologists to want to seek out the nutritional 'causes' of certain human distinctions. Thus we see people (e.g. R. Leakey citing R. Martin, T. Perper, C. Schrire, J. Robinson and R. Ardrey in The Origin of Humankind) suggesting raw animal flesh eating to be the determining factor in humans developing larger brains and proto-culture, because is was a 'rich source' of some nutrient or other.

There is a popular notion that anthropology can offer useful insights for forming the basis of a dietary philosophy. Anthropology is a science which is only just starting to mature, previously having been little more that a systematic, but lose, body of "say-so" information which attempted to explain our species history and origins. With advances in dating methods, including DNA analysis and more fossil finds, the science is now embarking on its integration with biology. Previously, anthropology was a pseudo-scientific marriage of traditional views attempting to link the findings of robust sciences, such as geology, palaeontology and archaeology. However, even though anthropologists like Richard Leakey are aware that their 'science' is often "based on unspoken assumptions" (The Making of Mankind, p. 82, R. Leakey), they show that they will persist in making them.

Anthropologies 'Man The Hunter' concept is still used as a reason for justifying the consumption of animal flesh as food. This has even extended as far as suggesting that animal foods have enabled or caused human brain enlargement. Allegedly this is because of the greater availability of certain kinds of fats and the sharing behaviour associated with eating raw animal food. The reality is that through natural selection, the environmental factors our species have been exposed to selected for greater brain development, long before raw animal flesh became a significant part of our ancient ancestors diet. The elephant has also developed a larger brain than the human brain, on a diet primarily consisting of fermented foliage and fruits. It is my hypothesis that it is eating fruits and perhaps blossoms, that has, if anything, contributed the most in allowing humans to develop relatively larger brains than other species. The ability of humans to develop normal brains with a dietary absence of animal products is also noted.

Bipedalism is often thought to be one of our most unusual characteristics. However birds, with ancestors like the Emu, probably exceeded our humble two legged efforts far earlier than any human ancestor, as did many dinosaurs, even if aided by tails. It has also been suggested that our large brains are required to assist us with this supposedly great balancing act. I think it must be obvious that Emus and dinosaurs managed to balance somewhat more impressive structures with rather smaller brains. It is our highly developed brain which most dramatically seperates us from other species. Certainly bipedalism may have freed up our hands to become more inquisitive, but on its own it is not an 'advanced' feature. There is now good evidence that hominid bipedalism evolved well before 'super brains.'

I am not alone in thinking that bipedalism evolved in humans as a means to better obtain low growing fruits. As K. D. Hunt suggests below, chimps are observed to stand up when they feed from small fruit trees. Perhaps because our human ancestors ate so much fruit in this way, they evolved the unusual upright anatomy even further.

Hunt's chimpanzee feeding observations showed that chimps were bipedal when they ate small fruits from short trees, as the small-object feeding hypothesis suggested (Jolly & Plog, 1985). .... When reaching up from the ground, small fruits encourage bipedalism because it takes two hands to fill the mouth. .... The presence of chimp-like shoulders, torsos and hands in australopithecines, and a mechanically inefficient locomotor bipedalism suggests that bipedalism evolved, not for walking, but for postural harvesting, often in the trees with the body stabilized by arm-hanging.

The Evolution of Human Bipedality: Ecology and functional Morphology
Kevin D. Hunt; Journal of Human Evolution ('94) 23 pp. 183-202.

I believe it is quite reasonable to conclude that an animal as small as Ramapithecus would have walked upright very readily when it was on the ground .... And this habit would have been reinforced if it spent a good deal of time feeding while on the ground, taking fruit, berries and nuts from low bushes, for instance.

David Pilbeam quoted in The Making of Mankind, Richard E Leakey, pp. 51-52

The advantages of a smaller, and less sophisticated, brain is that the construction time is less, and so the simple brain may be used effectively soon after birth. Compared to other great apes, the large human brain takes much longer to complete construction of its neural network, so leaving our offspring unable to do anything much for themselves, other than feed, for many of the earlier years of their life. The evolutionary benefits of superior intellect seem to outway the loss of specialised adaptations, perhaps because it can be applied to advantage in any given scenario, and often to great effect. Once animals have exhausted their ability to improve on running, chasing and breeding, the development of superior intelligence* becomes the next obvious progression in interspecial 'arms races.' It is the selection of the more successful, and smarter, hunting carnivores which, in turn, leads to the selecting of the smarter prey species for survival. Herbivores have evolved greater intelligence due to natural selection, and not as a direct consequence of eating meat themselves. As Richard Dawkins says, in The Blind Watch Maker, '. . . there is a tendency for brains to become bigger as the millions of years go by. At any given time, the current herbivores tended to have smaller brains than the contemporary carnivores that preyed on them. But later herbivores tended to have larger brains than earlier herbivores, and later carnivores larger brains than earlier carnivores.'

* see: The Blind Watchmaker, R. Dawkins, 1991, p.178

The Role of Fruit in the Economy of Nurturing

This dryopithecine skull, found by Mary Leakey in 1948 was given the name Proconsul africanus, and Richard Leakey says '. . . dryopithecines were forest dwellers, feeding on fruit, soft leaves and shoots, flowers and probably insects.' Jerison (1973) defined the encephalisation quotient (EQ) as the ratio of actual brain size to expected brain size for living mammals. Put crudely, this formula tells us how 'big headed' a mammal is. Using modern humans as a reference point with a relative EQ of 100, Proconsul africanus scores 48.8 (Walker et al., 1983, in An Introduction to Human Evolutionary Anatomy, Leslie Aiello and Christopher Dean, 1996). This score is much higher than any of the great apes. Given that primates would not have competed for the same resources successfully, it is highly likely that P. africanus developed its own feeding strategy, with specialisations quite unlike the other primates. This is all the more obvious when one considers the unusual anatomy of this potential human ancestor, with its superior brain. It is more probable that P. africanus developed its superior intelligence in order to forrage and avoid predation rather than to engage in it.

It is highly probable that plant foods were indeed the major part of the early hominid economy, and unequivocal evidence for hunting as against scavenging carrion does not appear until relatively late in the fossil record, probably not earlier than half-a-million years ago.

Leakey, The Making of Mankind

Fruigivory is an intellectually demanding feeding behaviour demanding the development of strategic planning, whereas the folivores feeding behavior engages relatively simple tactics. According to Caroline E. G. Tutin et al. 'Allometric analyses suggest a relation between brain size (relative to body mass) and diet, with frugivores having relatively larger brains . . . Maintaining a frugivorous diet presents huge intellectual challenges of memory and spatial mapping compared with the relative ease of harvesting abundant foliage foods.' Tutin et al. also say that:

Studies of frugivorous communities elsewhere suggest that dietary divergence is highest when preferred food (succulent fruit) is scarce, and that niche separation is clear only at such times (Gautier-Hion & Gautier 1979: Terborgh 1983).

Foraging profiles of sympatric lowland gorillas and chimpanzees in the Lopé Reserve, Gabon, p.179, Philosophical Transactions: Biological Sciences vol 334, 159-295, No. 1270

During the earliest months of a great apes, and especially the human childs life, it is exclusively dependant upon its mother for food security and transport. This would certainly have been a far greater factor in the development of the earliest proto-human children than for other primates. In order to allow for greater brain development a greater period of dependency on the mothers nurturing is required. This factor creates limitations upon what the mother can do during the early phase of child growth, and it will certainly limit her feeding behaviour. Indeed, her whole life would change because, prior to contraception, continuous nursing would have been normal. This limitation could, in turn, place restrictions on the biochemistry of her offspring and herself.

There is a certain 'economics' of biology, and this means that a benefit such as more nurturing for greater brain growth must have its costs. In the case of primates, nurturing a helpless infant for many months, even years, would tend to make the most readily available foods the most desireable ones to be able to exploit. So which foods in our ancestral environment were the most economic ones to obtain? Some anthropologists have suggested that 'meat' has been at the heart of the uniquely human evolutionary developments. I have used the term raw animal flesh in this article to draw a distinction between what wild animals eat, and the modern cooked commodity which we call 'meat', that is quite unlike any food found in a natural environment.

The Hunting Hypothesis

The anthropologist Sally Slocum has successfully debunked the 'man the hunter' theories promoted most actively by another anthropolgist Sherwood Washburn, who said in Princeton in 1956:

'Hunting not only necessitated new activities and new kinds of co-operation but changed the role of the adult male in the group. Among the vegetarian primates, adult males do not share food. They take the best places for feeding and may even take food from less dominant animals. However, since sharing the kill is normal behaviour for many carnivores, economic responsibility of the adult males and the practice of sharing food in the group probably resulted from being carnivorous. The very same action that caused man to be feared by other animals led to food-sharing, more co-operation, and economic interdependence.'

The Making of Mankind, Richard E. Leakey, p.92

Wasburn has aided in the perpetual myth of raw animal flesh as the center of human intellectual and social development, Leakey has sided with Slocum in saying:

The argument seems persuasive when one considers the example of the co-operative, social carnivores such as hunting dogs but only a few carnivores share their kill in this way. Indeed, many carnivores are solitary except in the breeding season.

Leakey, The Making of Mankind

Wasburn together with C. S. Lancaster even suggested as late as 1968 that 'The biology, psychology and customs that separate us from the apes - all these we owe to the hunters of time past.' In 1976, Robert Ardrey* wrote a book about the hunting hypothesis, with the same title, in which he claims 'Man is man, and not a chimpanzee, because for millions upon millions of years we killed for a living.' (Leakey, The Making of Mankind) Clearly the student needs to be aware that numerous, and even eminant biologists and anthropologists, unwittingly seem to pass off their own maladaptive cultural behaviours as being part of the greater evolutionary scheme.

* see also: African Genesis, R. Ardrey

What Sally Slocum has pointed out is that among most contemporary hunter-gatherers, plant foods collected by the females provide the greater proportion of the diet. She also suggests that the trigger that set hominids in this new direction was the ever-lengthening period of infant dependency: saying 'The mothers would begin to increase the scope of their gathering to provide food for their still dependent infants.'

The hunting hypothesis has now been discreditted in favour of the gathering hypothesis developed by, amongst others, Sally Slocum, Adrienne Zihlman and Nancy Tanner. For Leakey, although better than the hunting hypothesis, this new hypothesis seemed too exclusive an explanation, so he favours the ideas of Glynn Isaac's food-sharing hypothesis. In this, gathering is the centre of the food economy with raw animal flesh, and later perhaps cooked meat, seen as a peripheral food. Isaac has refined his hunting hypothesis by analysing the solid scientific evidence available (mostly site 50). R. Leakey quotes Glynn Isaac in The Origin of Humankind, as saying that 'Although meat was an important component of the diet, it might have been acquired by hunting or by scavenging . . . . You would be hard pressed to say which, given the kind of evidence we have from most archeological sites.'

The historical role of meat in human diets has probably varied significantly, as Jared Diamond, who is not an authority on human evolution, points out:

.... while early humans ate some meat, we do not know how much meat they ate, nor whether they got the meat by hunting or scavenging. It is not until much later, around 100,000 years ago, that we have good evidence about human hunting skills, and it is clear that humans then were still very ineffective big-game hunters. Human hunters of 500,000 years ago and earlier must have been more ineffective.

Western male writers and anthropologists are not the only men with an exaggerated view of hunting. In New Guinea I have lived with real hunters, men who recently emerged from the stone age. .... To listen to my New Guinea friends, you would think that they eat fresh kangaroo for dinner every night and do little each day except hunt. In fact, when pressed for details, most New Guinea hunters admit that they have bagged only a few kangaroos in their whole life.

The Rise and Fall of the Third Chimpazee, Jared Diamond, 1991, pp.33-34

In The Origin of Humankind, Richard Leakey mentions Lewis Binford, who suggested that systematic hunting of any kind began to appear only when modern humans evolved, giving dates of 45,000 to 38,000 years ago.

What is obvious is that a mother and her infant cannot engage in hunting, or any other arduous food gathering activity. Of all the primate foods, fruits are the most easily gathered. They may be obtained without the use of digging or cracking tools, climbing, and digested with no need for hind-gut fermentation, such as is the case with foliage. Most significantly, it is the feeding limitations of the nursing mother which determine what foods she, and her offspring will have continually available.

A simple calculation based on comparisons with other primates reveals that gestation length in Homo sapiens . . . . should be twenty-one months, not the nine months it actually is. Human infants therefore have a year's growth to catch up on when they are born, hence their helplessness.

The prolongation of childhood in modern humans is achieved through a reduced rate of physical growth compared with that in apes. As a result, humans reach various growth milestones, such as tooth eruption, later than apes do.

Present-day apes are heavily built for their stature, being twice the bulk of a human of the same height.

Richard Leakey, The Origin of Humankind

The Fruigivore Hypothesis

Given a plentiful supply of fruits the mother does not have to risk expending much of her effort obtaining difficult to get foods like raw animal flesh, insects, nuts and roots. Furthermore, fruits contain abundant supplies of sugars which the brain solely uses for energy. The mother whos genes better dispose her for an easy life on fruits would have an advantage of those who do not, and similarly, the fruit species which is the best food for mother and child nutrition, would tend to be selected for. There is now little doubt amongst distinguished biologists that fruit has been the most significant dietary constituent in the evolution of humans. In order to gain an understanding of how plant foods may have contributed to our ancestral diet Peters and O'Brian have a paper in which they state:

Our interests lie in the edible flora of sub-Saharan Africa and the evolution of the hominid diet. .... In its extreme form, this interest can be expressed as the African Dietary Hypothesis: it is the African sub-tropical flora in particular that has nurtured human evolution through most of earth's recent pre-history; this is the nutritional environment to which our dietary physiology is most fundamentally adapted. An extreme position to be sure, perhaps only partially true.

Charles R Peters and Eileen M O'Brian in The Digestive System In Mammals, p.166

What did Peters and O'Brian discover? They analysed 152 edible plant species from Gaborone and Harare. Eliminating the plant matter known to be toxic, or that which was cooked (primarily leaves) and also gums which were reported as poorly digested or any other unpleasant parts, we get the following breakdown of species edible parts: 81% fruit, 5% flowers, 4% seeds, 4% wood, 3% sprouts, 2% nuts and 1% roots. This is shown in the pie chart. Roots and sometimes wood are primarily used as a source of water.

It seems inevitable that our ancestral human biochemistry would have evolved with a significant shift towards the processing of fruits by natural selection, and that once this shift occurred a selective positive feedback effect would drive our genes towards this extreme, until a limit is reached. But our behaviour may have been favourably selected if it was opportunistic, and included occassional use of meat and other harder to get edibles when we happened upon them. Richard Dawkins, an authority on Neo-Darwinism, says:

Genes for making teeth suitable for chewing meat tend to be favoured in a 'climate' dominated by genes making guts suitable for digesting meat. Conversely, genes for making plant-grinding teeth tend to be favoured in a climate dominated by genes that make guts suitable for digesting plants. And vice versa in both cases. Teams of 'meat-eating genes' tend to evolve together, and teams of 'plant-eating genes' tend to evolve together. Indeed, there is a sense in which most of the working genes in a body can be said to cooperate with each other as a team, because over evolutionary time they (i.e. ancestral copies of themselves) have each been part of the environment in which natural selection has worked on the others. If we ask why the ancestors of lions took to meat-eating, while the ancestors of antelopes took to grass-eating, the answer could be that originally it was an accident. An accident, in the sense that it could have been the ancestors of lions that took up grass-eating, and the ancestors of antelopes that took up meat-eating. But once one lineage had begun to build up a team of genes for dealing with meat rather than grass, the process was self-reinforcing. And once the other lineage had begun to build up a team of genes for dealing with grass rather than meat, that process was self-reinforcing in the other direction.

The Blind Watchmaker, Richard Dawkins, 1991, p.172

The above explains why, although humans have eaten raw animal flesh and even cooked 'meat' for 500,000 years or more, we have not yet seen an example of a human with unequivable carnivorous adaptations. Certainly if such evidence existed it might be used by the industrial interests to promote meat eating. If any adaptations have occurred in half a million years, they would be small changes, micro evolution at the molecular level, and would have happened against the teams of opposing genes specialised for the fruit processing metabolism - at least if such changes were, in a sense, antagonistic to the majority.

Peripheral and Systematic Evolution in Adaptation

Micro evolution is the name given to peripheral genetic variations and explains such phenomena as darkly coloured moths surviving well in smokey cities, while light counterparts of the same species do better in the wild. In certain scenarios just one variation in an allele can lead to a selective advantage. More generally however, such changes must be of a more neutral effect with little change of a significant benefit or cost leading to selection. Under this banner we could include various minor mutations including hair or skin colour, freckles, length of fingers, height and so forth. These mutations do not become selective issues, unless they are extreme, because they have no downstream impacts on the essential functionings of other genes in the body. Micro evolution may also explain why some people continue to be able to produce lactase (the milk sugar digesting enzyme) into adulthood, or why the Inuit people have adaptations that allow them to assimilate fats more successfully. These small adaptations, probably dependent on certain childhood genes remaining turned on, do not affect the fundamental nutritional requirements of our metabolism, even if they do offer survival benefits.

Macro evolution is how a species differentiates into 2 new species and includes gross systematic changes in genetic makeup such that cross breeding is eventually impossible, thus speciation occurs. Macro evolution encompasses a chain of compatible advantageous genetic mutations. Clearly macro evolution takes far longer than micro evolution and has dramatic consequences when even closely related species are compared.

Metabolism, particluarly in connection to diet, involves huge 'teams' of genes that have to work together within certain limits, to energise and repair the organism. There can certainly not, as has been popularly believed, be a rapid (i.e. thousands of years) and significant 'adaptation' to novel foods, if their nutritional or physical characteristics fall drastically outside of what an animal has been experiencing in its diet, for the previous few million years. If such items have characteristics too far wide of the natural diet, then such items are maladaptive. In a population where selective pressure is low however, such maladaption may be well tolerated, having little significant bearing on reproductive potential.

Mutations that deviate from the norm often interfere with biochemical processes and are rigorously removed by selection. Much of the body is made up of the building blocks of cells, which must fit accurately with each other. This is true for the fibrous protein - collagen - that makes up much of our skin, the crystallin proteins in the eye and the cellular factories of metabolism in mitochondria and ribosomes. Any mutation that changes the shape of such molecules will almost always be at a disadvantage as it reduces their ability to interact with each other or with their substrate. Stabilising selection of this kind can be a very conservative force.

Stabilising selection is a powerful agent that leads to genetic homogeneity. To explain how genetic variation is maintained in the face of such widespread censorship by natural selection is one of the main problems of population genetics.

Natural Selection in Humans, p. 286, The Cambridge Encyclopedia of Human Evolution, 1992

Comparative Lactobiology

What are the essential biochemical properties of human metabolism which distinguish us from our non-human primate relatives? One, at least, is our uniquely low protein requirement as described by Olav T. Oftedal who says:

Human milk has the lowest protein concentration (about 7% of energy) of any primate milk that has been studied. In general, it appears that primates produce small daily amounts of a relatively dilute milk (Oftedal 1984). Thus the protein and energy demands of lactation are probably low for primates by comparison to the demands experienced by many other mammals.

The nutritional consequences of foraging in primates:
the relationship of nutrient intakes to nutrient requirements, p.161
Philosophical Transactions: Biological Sciences vol 334, 159-295, No. 1270

One might imagine that given our comparatively 'low protein' milk, we would not be able to grow very fast. In fact, as the image on the right shows, human infants show very rapid growth, especially of the brain, during the first year of life. Human infants are born a full year earlier than they would be projected to, based on comparisons with other animals. This is because of the large size their brains reach. A human infant grows at the rate of 9 kg/year at birth, falling to 3.5 kg/year a year later. Thereafter its growth rate is about half that of a chimpanzees at 2 kg/year vs. about 4.5 kg/year. Humans are relatively half as bulky as the other great apes, thus allowing nutrients to be directed at brain development and the diet to be less demanding. The advantages of such an undemanding metabolism are clear. Humans delay their growth because they 'catch up' later, during puberty as seen on the graph. Even so, the growth rate never reaches that of a newborn infant who grows best by only eating breast milk. By about 2 years of age the human brain is about half way along in its postnatal growth, but which toddler could eat the massive portions of meat, nuts or other such items held to be necessary for brain growth during our evolution?

According to Exequiel M. Patiño and Juan T. Borda 'Primate milks contain on the average 13% solids, of which 6.5% is lactose, 3.8% lipids, 2.4% proteins, and 0.2% ash. Lactose is the largest component of the solids, and protein is a lesser one'. They also say that 'milks of humans and Old World monkeys have the highest percentages of sugar (an average of 6.9%)' and when comparing human and non human primate milks, they have similar proportions of solids, but human milks has more sugar and fat whereas the non human primate milks have much more protein. They continue 'In fact, human milk has the lowest concentration of proteins (1.0%) of all the species of primates.' Patiño and Borda present their research in order to allow other primatologists to construct artificial milks as a substitute for the real thing for captive primates. It is to be expected that these will have similar disasterous consequences as the feeding of artificial bovine, and other false milks, has had on human infants.

Patiño and Borda also present a table which compares primate milks. This table is shown below and identifies the distinctive lower protein requirements of humans.


Total Solids





Human (1)






Apes (2)






Old World
Monkeys (3)






New World
Monkeys (4)






Prosimians (5)












The composition of primate milks. (1) Homo sapiens, Packard, 1982; (2) Pongo pygmaeus, Pan troglodytes, Ben Shaul, 1962; Gorilla gorilla, Tailor & Tomkinson, 1975; (3) Cercopithecus talapoin, Buss & Cooper, 1970; Papio anubis, Papio cynocephalus & Papio papio, Buss, 1968; (4) Saimiri sciureus, Buss & Cooper, 1972; Leontopithecus rosalia, Buss, 1975; (5) Lemur spp., Buss et al., 1976.

The Composition of Primates' Milk and Its Importance in Selecting Formulas for Hand-Rearing
Exequiel M. Patiño and Juan T. Borda

Undoubtedly these gross metabolic differences between humans and other mammals must have system wide implications for our metabolism. They allow us to feed heavily on fruits, and may restrict other species from choosing them. Never the less, many nutritional authorities suggest that adult humans need nearly double (12% of calorific value) their breast milk levels of protein, although it is accepted that infant protein requirements for growth are tripple those of adults. The use of calorific values might also confuse the issue since human milk is highly dilute (1% protein), and clearly eating foods that might be 25 times this concentration, such as meat, are massive excesses if constantly ingested. Certainly the body might manage to deal with this excess without suffering immediate problems, but this is not proof of any beneficial adaptation. It also needs to be pointed out that berries, such as raspberries, may yield up to 21% of their calorific value from protein, but are not regarded as 'good sources' of protein by nutritional authorites. There are millions of fruits available to wild animals, and blanked generalisations about the qualities of certain food groups, need to be examined carefully, due to some misconceptions arising from the limited commercial fruits which we experience in the domestic state.

The weaning of a fruigivorous primate would clearly demand the supply of a food with nutritional characteristics similar to those of the mothers milk. We must realise that supportive breast feeding may continue for up to 9 or 10 years in some 'primitive' peoples, and this is more likely to be representative of our evolutionary history than the 6 month limit often found in modern cultures. This premature weaning should strike any aware naturalist as being a disasterous activity, inflicting untold damage. However, what we do know of the consequences is that it reduces the IQ and disease resistance of the child, and that the substitute of unnatural substances, like wheat and dairy products, is pathogenic.

Finally we need to compare some food group compositions with human milk in order to establish if any statistical similarity exists. This would demonstrate that modern humans have inherited their ancient fruigivourous metabolism. This data is examined below in the final sections of the article.

Neural Network Analysis of Food Groups

The purpose of this section is to establish the affinity between human breast milk and other raw foods by grouping according to similarity of their nutrient profiles. Unlike any other food which mammals consume human breast milk has specifically evolved to feed the baby with the correct proportions of nutrients essential for the rapid growth during the first few years of life. Milk, unlike most other things eaten by mammals, has evolved specifically as an ideal nutrient source. One would have difficulty suggesting with credibility, for example, that a fish has evolved itself to be the ideal nutrient set for fish eaters, or that bread is a suitable food for any living animals metabolic requirements as controlled by genes. The genes of any species evolve for their benefit alone and not to provide their body as perfect food for other species to prey upon. Similarly non-biological foods, such as bread, have no significant evolutionary history for humans to have adapted to them, if indeed such is possible.

Since animal foods, and plant foods have not evolved genes to make them ideal foods for the species which feed upon them, we see that the species which depend heavily on these foods have, by selection, evolved suitable biochemistry and physiology to deal with these foods. Thus, carnivores have adapted with strong gut acidity (i.e. shark gastric pH is 1) to rapidly digest flesh and bones, water conserving methods (i.e. panting instead of sweating to conserve water), and with herbivores have fermenting chambers in different regions of the gut for the bacterial fermentation of plant foods.

The demands of any living cell are for a fairly constant flow of nutrients, on average, and the genes determine the ideal quantites for successful life. This is why breast milk is of consistant nutrient characteristics even when such factors as race and diet are taken into account. The mammary gland will attempt to control the nutrients for the baby and optimise towards evolutionary success by giving baby cells (genes) what they need but no more.

We would expect the relative proportions of nutrients of an animals milk to reflect in the nutritional value of the foods eaten by a species as an adult, otherwise a smooth transition during weaning would be too difficult, and cellular requirements would not be sufficiently matched for optimum cellular metabolism and therefore survival. Thus dogs and rats, for example, have more proteinous milks than primates since their adult diet contains more proteinous foods. Whale milk would be quite unsuitable for an infant ape and the opposite would also hold true because the metabolic requirements of these species are grossly different. Whale milk has far more fat and protein than primate milk. Where a particular food source cannot be guaranteed, then more generalised adaptation to foods would provide an essential survival strategy. Many different feeding strategies will evolve, but whether they will cause changes in a species metabolism will depend on the selective pressure. A species may eat a particular food often, but in insignificant quantity, thus no tendancy to evolve specialisation towards that food will occur because such an advantage is more costly that the benefits derived.

Analysis with statistical tools has proven unsatisfactory because such methods are not specifically designed to perform pattern recognition. Therefore a more appropriate method is to use a self-training pattern recognising neural network. The Kohonen self-organising map (SOM) is a neural network designed to categorise data sets into alike groups based on corresponding similarities in data magnitudes. The SOM neural network finds typical application in voice and image pattern recognition where inexact input patterns can be matched to a generalised pattern for the group.

Based on source code provided in C++ Neural Networks And Fuzzy Logic (V. B. Rao & H. V. Rao, MIS Press, 1993) a Kohonen SOM program was compiled. Nutrient data sets for 22 nutrients of 230 edible raw foods were extracted from the USDA SR14 database and normalised. These were then used as input data for the Kohonen SOM neural network program. The final output from the program was then assigned back to the correspoding foods and USDA food groups for comparison. A number of runs were performed in accordance with suggested guidlines in order to establish input parameters that would allow for a reasonable amount of generalisation, while still keeping the food group categories reasonably exclusive.


Raw Data

Normalised Data

Output Data

Data As Spreadsheet

Table A

Table B

Table C


A number of runs with different input parameters were performed and mature human milk was consistantly assigned to categories with fruits and vegetables. The best output data obtained are presented in Table C above.

To test that the network was performing reasonably, other animal milks were added to the input data, and the network created a milk category and correctly exclusively assigned all the milks to it. Finally water was removed from the original input data, which resulted in human milk being placed in a category with just coconut milk.

These results seem to support the suggestion that human milk and a variety of fruits have statistically similar nutrient proportions. This in no way indicates that they are a suitable replacement for each other or that any particular fruit is an ideal food on its own. These data could also support the hypothesis that commercial fruit is of similar nutritional value to the ancestral fruits that may have occured in our evolution, though we now know that wild fruits are generally richer in protein and fats, and in minerals.

One must also challenge the commonly held view that a "balanced diet" can be created by combining foods that are "good sources of" certain nutrients as most of such foods seem to be far from ideal individually. A suitable analogy is to the common phrase "two wrongs do not make a right." Even if a combination of foods really did form a "balanced diet" it would be unreasonable to assume that it can better the results obtained through genetic selection such as occurs between a human mother/baby and a human/fruit evolutionary relationship.


We must treat anthropological arguments for selecting certain foods with great caution. Entrusting your health to the food choices made by our ignorant ancestors in hostile conditions, while we live in a modern environment, is absurd. A rational, and far more scientific approach, is a thorough analysis of the effects of meat, and other more recent foods, on health based on biochemial research. This has revealed that cooked meats contain carcinogenic substances and pose many other health risks.

We must also challenge the suggestions that humans have 'evolved' for, or 'adapted' to, any food items during our recent evolution. We share about 99% of our DNA in common with chimpanzees, the bulk of the difference being genes that control morphology and not metabolism in general. Even so, it appears that our metabolism and morphology are substantial less demanding that the chimpanzees. Since our metabolism is so undemanding, we must ask how might selection favour a more demanding metabolism, one requiring a move away from fruits? If such selective pressure did occur, the resulting advantageous mutations (if any) would have to rewrite our metabolism all the way back to something like a chimps, and perhaps far beyond.

If humans have become something other than a highly fruigivorous primate, a so called 'carnivorous' ape, then where are other such examples in Nature? Will we see the carnivorous cow or perhaps a herbivorous hyaena?

Last Updated on 06/03/02
By John Coleman ©