Friday, May 20, 2011

The Case of the Missing Extinctions

Or, The Pleistocene Extinctions Pattern Implications For Understanding Human Evolutionary Nutrition

Prior to fifty thousand years ago, before the Pleistocene extinctions, ancestors of modern humans lived in Africa.   Modern man, Homo sapiens sapiens, emerged in Africa, before these extinctions. 

As explained by geneticist Spencer Wells in the documentary film Journey of Man, by tracking the Y chromosome, we now know that all modern humans are descendants of a few men among the ancestors of the !Kung people currently living in the Khalahari desert. 

But in ancestral times, the Khalahari was not a desert. During the Pleistocene ice ages, “the great deserts of North Africa and Western North America today were mostly vast grasslands with large permanent lakes and abundant game animals.” [1]  

Many believe that humans evolved primarily by hunting those land animals, and take the Pleistocene extinctions as evidence that humans evolved as top predators of grassland animals in Africa.

In "Of mice, mastodons and men: human-mediated extinctions on four continents" [2], Lyons et al provide several convincing lines of evidence and reasoning that indicate that humans must have caused the Pleistocene extinctions.  To simplify, no other known natural phenomenon (such as climate change) could account for the sudden selective extinctions of the megafauna, without simultaneous extinctions of smaller species. 

But there is something very interesting about the pattern of the Pleistocene extinctions.

The following graph from Lyons et al [2] shows the pattern.  The hatched bars show the species exterminated in the late Pleistocene.  As you move from left to right, the bars represent increasingly larger species.   Notice anything unique about Africa?

On every continent where humans were an invasive species, the largest animals went extinct.  

But not in Africa.

There, most of the large body species survived.   


It appears that as humans invaded Australia and the Americas they succeeded in exterminating the fat megafauna, but humans who remained in Africa did not, and many, including giraffes, elephants, and rhinos,  have remained to this day.  Why?

Lyons et al have a suggestion:

“The lack of extinctions in Africa (Fig. 1) is especially notable given the long history of humans on this continent. The co-evolution of man with the African megafauna may have resulted in the evolution of effective anti-predator behaviours (Diamond, 1984; Martin, 1984).”[Italics added]

In other words, upper Paleolithic humans did not exterminate the megafauna of Africa because the African megafauna had plenty of experience with and knew how to avoid human predation.  Just the smell of humans probably sent them running.


In contrast, animals in Europe, Asia, Australia, and the Americas most likely evolved independent of human predation.  This meant that they were, for humans, relative to African fauna, ‘sitting ducks.’ 

Thus, considering the co-evolution of humans with African fauna, the fact that the extinctions did not occur in Africa supports the hypothesis that humans, as an invasive species, caused those extinctions on other continents.

It also suggests that the invasion of other continents allowed humans to pursue a subsistence strategy––hunting large, fat anmals––that would not have been as successful in Africa.

Moreover, we could reasonably take the statement by Lyons et al and modify it slightly:

The co-evolution of man with African fauna (mega or not) most likely resulted in those fauna evolving behaviors effective for avoiding human predation.

This further suggests that the human exodus out of Africa into other continents may have resulted in a significant change in human diet composition, after H. sapiens sapiens had already emerged.  In other words, it suggests that the later Paleolithic diet of humans outside of Africa may have contained much larger amounts of land animal meat and fat than would have been possible in the early Paleolithic among the far more wary wildlife of the homeland.

Let me put it this way.   An animal adapted to a diet obtained in the Africa invades a new ecosystem.  In this new ecosystem, this invasive species finds that hunting is a lot easier than it was in the homeland, because the animals are relatively oblivious to the danger presented by humans.  Consequently, in accord with optimal foraging theory, this animal goes for the easiest calories possible:  large animals that don’t know that they should run away from anything that smells, sounds, or looks like a human.

If this new diet supports health adequately to allow reproduction for a majority of individuals, but disrupts homeostasis just enough that it gradually induces metabolic disorders that emerge over time, past the time of reproduction,  or shortens the average individual’s lifespan by inducing chronic diseases, the new diet will not select against those not adapted to the new diet, nor produce a population better adaptated to the diet.

I might even call it the upper Paleolithic dietary revolution. 

Now, let’s consider Africa again.  As noted above, the great deserts of Africa were grasslands with large “permanent” lakes (well, permanent during those ages).  If hunting land animals was not as easy in Africa as in Eurasia and the Americas, what about hunting animals living in those lakes?  Would they all catch the smell of humans or run away easily?

Lacustrine, riverine, and wetland environments are very rich in food resources, both plant and animal, including animals that don't move very fast (shellfish).  Foraging in such environments might just provide greater return on investment than possible on a grassland.

Some people studying human evolutionary nutrition believe that the archaeological, cultural, nutritional, biochemical, and medical evidence points in the direction of humans evolving in those econiches, where shellfish, fish, and amphibious animals could have been the primary sources of animal foods.  These animals have nutritional properties significantly different from  those of savannah animals—the protein, fats, and mineral contents of water animals all differ significantly from land-based animals, in very interesting ways relative to specifically human nutritional requirements, particularly for the nervous and cardiovascular systems.

In fact, the earliest human fossils are consistently found associated with lacustrine or marine fossils indicating humans inhabiting niches incorporating a land-water interface:  wet woodlands, flood plains, wetlands, rivers, lakes, and coastlines.  In these two segments of the Journey of Man, we learn about archaeological evidence of humans living along the African coastline and consuming large amounts of seafoods, leaving large heaps of seashells as evidence (segment starts about 5:00 of first video):

 Also, and again, as explained in the  Journey of Man, the first exodus of humans from Africa appears to have followed a path along the marine coastline to Australia; it does not track through a grassland.   And, in this segment, we learn that some of the first humans in Australia lived in a lacustrine environment (now a desert), described as "quite a rich environment," eating fish (leaving hearths and fish bones as evidence):

Indeed, in the two segments below, Wells point out that, at the end of this journey, to get from Indonesia to Australia, it appears that people had to find a way across 150 miles of open ocean (segment starts at about 3:20 of the first video).  

The animal capable of this seafaring--150 miles of open ocean with only the most primitive of boats--would have had to have been very familiar with and physically very well adapted to a marine environment.  Not a likely accomplishment for a mammal specialized in exploiting the dry savannah environment. 

Which leads me again back to Michael Crawford and David Marsh, co-authors of Nutrition and Evolution.   Crawford, a biochemist, has devoted his life work to study of the effects of fats on human physiology, and has a bit to say on the subject and its relation to human evolution.  Some others have something to say about the differences between land and animal foods with regard to human brain nutrition.

But for now I have run out of time.  To be continued.


Timothy said...

I have been following the Venus figurine debate and then read this piece about African predation and extinction. I think these are linked. I would submit that a pre-agricultural diet would be sufficiently sporadic as to be occasionally insecure. That is, sometimes people feasted and sometimes they went hungry. In this context, hungry is bad and gorging is good. I think anyone can get obese by eating too much food. However, the specific dietary combinations that lead to modern obesity vs those that would have led to prehistoric obesity may have little to do with the obesity of the figurine.

The central point is that being an exceptionally well-fed woman was likely an aspiration. It was something to be desired and sought. It makes all the sense in the world that a goddess figurine or fertility figurine would be obese. As in "Have you heard the good news? The Goddess never goes hungry!!"

Hypothetically, then, her obesity could easily have been the form taken by a spiritual totem or relic used to ward off hunger through invocation of the well-fed goddess. This makes more sense than alternate explanations because it's the same kind of wishful and magical thinking religions have always produced.

selina.roesner said...

Well, again, nice post. Exactly what i stated in my post on your last article. Seafood as huge part of evolution...

Vitamins/minerals, amino acids, O3 - and you are healthy.

EL 66K said...

"The co-evolution of man with African fauna (mega or not) most likely resulted in those fauna evolving behaviors effective for avoiding human predation"

Is there proof for that from animal behavioral studies, etc.? Is it just a hypothesis?

Michal said...

Fish is much easier to get in Europe than in Africa thanks to having higher coastline / total land mass ratio and large number of permanent rivers (Europe has mountains where water can condense and lowlands where it can flow, Africa is flat). Even in Middle Ages rivers were full of fish. So if humans were predominantly fish eaters it would have been difficult for them to cause extinction of megafauna in Europe.

It even looks that we are badly adapted to eat tons of fish. Fish oil is great immune system suppresor in humans, so too much of it leads to increase in infections and even cancers! So up to few grams of it per day are great for the brain and the skin but bulk of calories have to come from somwhere else.

Speaking about Africa vs. norther countries you can't forget about the differences in intellectual capabilities. Northern hunter gatherers have IQ of ~90 for Inuits or Native Americans, while San hunter-gatherers have IQs of 54 or so. So if the extinctions was caused by humans it might be not because of animals being somwhat worse adapted to humans, but because humans there were more efficient at hunting. These animals had to do with Homo Erectus from 1.8 mln years ago, Neanderthals from 400k years ago - no way they wouldn't be able to recognize a human.

The only exception being Austraila which had its own, specific fauna. Weaker than the rest of the world.

selina.roesner said...

"It even looks that we are badly adapted to eat tons of fish. Fish oil is great immune system suppresor in humans, so too much of it leads to increase in infections and even cancers! So up to few grams of it per day are great for the brain and the skin but bulk of calories have to come from somwhere else."


Michal said...

To selina.roesner - I'm not a scientist, so I can't talk to you in the 'studies' language. You have to get used to such people, as we are the majority.

Of course I don't get my information from nothing. Here are just a few examples of information about negative effects of polyunsaturated fat:

On the other hand there's that Budwig cure for cancer, that uses fresh unsaturated oils (flaxseed) combined with right proteins (from cottage cheese) to actually cure cancer.

So it may be that the preparation is crucial, so the only proper way to eat PUFAs is to get them in unchanged (unheated, maximally fresh, protected by natural antioxidants) form and only combined with proteins?

Many notice this. For example Dr. Mercola stopped selling fish oil in partially because it oxidises too fast, and moved to krill oil that can last distribution and shelf time because it contains powerful antioxidant astaxanthin.

When it comes to fish today it's contaminated with mercury (recently you are also at danger of radioactive ones, and ones containing petrochemicals and oil dispersant) not fresh due to our food distribution system that keeps it for weeks from catching to consumption, and not consumed properly - fishing cultures also eat fish heads with thyroid full of metabolism increasing hormones, which balances some lowering effects of fish oils on metabolism.

Jay said...

Humans may have caused extinctions of other large animals earlier in the Pleistocene in Africa. From ref [2] Lyons, 2004:
The lack of extinctions
in Africa (Fig. 1) is especially notable given the long history of humans on this continent.
The co-evolution of man with the African megafauna may have resulted in the evolution of
effective anti-predator behaviours (Diamond, 1984; Martin, 1984). It is also likely, however,
that the entire Pleistocene fossil record for Africa would show corresponding ‘pulses’ of
human-caused extinctions of large mammals as human culture and hunting technology
developed. The diversity of very large mammals in Africa has declined. For example, during
the early Pleistocene at approximately the time when humans began developing more
complex culture and societies (i.e. the time of Homo erectus), 12 species of elephant-like
proboscids were extant in Africa (Coppens et al., 1978). By the Middle Pleistocene, only two
remained (Coppens et al., 1978).

Terry O. said...

What happened to the post on the melting points of various fats?

The post, as well as the various comments from the many high profile personalities in the health & fitness blogosphere, was very informative and entertaining.

Paleo Phil said...

It's true that there were fewer megafauna extinctions in Africa, though there were some and Jay is right that there is a minority view, which was first proposed by Paul Martin decades ago and that I found intriguing but doesn't seem to have developed much of a following, that H. erectus contributed to the African megafauna extinctions ca 1.4 mya. It's interesting that Lyons called this possibility "likely" in 2004. So the hypothesis is still alive.

See also:
> Martin, Paul S., Wright, H., E., Jr. (1967). Pleistocene Extinctions. The Search for a Cause. Yale University Press.
> Megafauna — First Victims of the Human-Caused Extinction, Baz Edmeades, (Edmeades was influenced by Martin)

There was a dropping off of African megafauna extinctions after the 1.4 mya event, so Don's hypothesis still has merit even if Martin's hypothesis re: H. erectus causing the 1.4 mya event is correct. It's a question of degrees--how much megafauna hunting did H. erectus do. Probably less than later Eurasian and North American hunters. However, is it possible that a period of megafauna slaughter by H. erectus could have had a crucial evolutionary effect? If I recall correctly, there was a jump in brain size around that time.

There also may have been more recent megafauna extinctions in Africa:
"even in Africa, the big game we see today is only about 70% of the genera that were present in mid-Pleistocene. About 50 genera disappeared about 40,000 years ago." (Biodiversity and Conservation: A Hypertext Book by Peter J. Bryant, 2006, Professor, Department of Developmental and Cell Biology,

Paleo Phil said...

P.S. If the Anthropogenic Predation Theory aka Prehistoric Overkill Hypothesis is correct even just in Eurasia and N. America, then it would seem that humans were able to subsist or perhaps even thrive for generations on meat-heavy diets.

Humans seem highly adaptable to varying macronutrient ratios, so macronutrient ratio seems less important than eating foods that we are biologically adapted to and avoiding harmful processing methods.

Joe said...

Don, this is a really interesting angle and I think I see where this fits in with your recent thinking.

Are there any predators which suffer detrimental health effects from overeating prey? My understanding was that the predator-prey models reflected a wave-pattern where overhunting reduced the population of prey, this food shortage then led to fewer predators. Barring extinction, the cycle then starts over. The ultimate consequence for an overeating predator is starvation.

The African model you present seems less an argument for macronutrient ratios and more of a case for intermittent fasting. Obviously people could thrive in situations were our preferred foods were more difficult to obtain.

One last thing, remember that the human foot has been essentially unchanged for nearly two million years. Human exploitation of coastal environments doesn't preclude earlier success at hunting large animals in grassland via persistence hunting prior to tool use.

jandro said...

Hey Don,

I've been a long time reader and I'm really enjoying your recent posts. I tried VLC for 2 months about 1.5 years ago and I was convinced it was not the way to go (it also didn't make sense scientifically for me). I am happy to see more people from the "paleo community" are looking more into this.

I found these two studies interesting. It seems that even though Okinawans do not have a high fat diet, those that eat the most fat have a higher life expectancy.

Obviously, this cannot be extrapolated to "high fat = good". But it does hint low-fat is not a necessity, and that it might be beneficial to add some fat.

What do you think?

Lila said...

Very interesting hypothesis. I am not sure that I agree entirely with the idea that humans hunted the large critters of northern climes into extinction, although to some small extent it may have been true. Another consideration is climate change. When the glaciers receded the land changed and it could no longer support mega-life.

Boiling Pot said...

This all hinges on one thing (in my opinion): just when does "ancestry" (for purposes of figuring out nutrition) start? Does our paleolithic period have as much importance as neolithic or recent period when it comes to figuring out what we should eat? What about what our mother ate (right or wrong) when pregnant? What about the condition of our father's reproductive cells? Why go back to Africa and Venus figurines?

It seems to me that there is no longer any one way to eat. We must eat what we need till it ceases to serve us. Then we do the opposite, or at least near-opposite. This lurching back & forth can do the trick, or at least it has for some people.

Purges - if we are suited to them - can help tune our bodies. Parasite purges, bowel cleanses, liver flushes, and so on and so forth. I'm not saying to get hung up on them, only to take a crack at it and see what they do for YOU and don't be bothered by what seems to be right for someone else. There've been too many influences in recent years and I would suggest that going way, way back in time is an exercise in futility.

Miki said...

Cordain published a paper in 2001 showing that obtaining DHA from animals brain is more efficient than obtaining it from fish. Another paper (Braun 2010) show that fish was consumed 2 million years ago along with meat.People can survive very well on fish and sea food. There are enough fishermen's villages to prove it. It doesn't mean that human evolution was based on it. The evidence for high meat consumption is overwhelming. N Isotope evidence show that fish entered the diet in the last stages of the Upper Paleolithic in Europe as addition to meat.

kyle said...

I was thinking. Its probably a good thing that when you said the new diet does not select against those not adapted to the new diet because the chronic diseases emerge over time until after the time of reproduction. If you think about it, if this were to happen natural selection would lead to humans eating more plants and changing physiology and in the end would result in humans losing their big brains.

Neonomide said...

A nice post, again. Hart & Sussman in Man the Hunted (2005) also discussed differences in megafauna extinctions in different continents, if only from the "being predated" perspective.

I recall Cavalli-Sforca's team's genetic studies made quite clear that african people are pretty much distinct from non-african people and that there exists a lot more variation between africans than between people of non-african heritage.

This could also mean that most promosing nutritional milieus should be looked upon that fact.

Starch-based diets also probably should be judged upon the view of 2,3-2,5 million years of meat eating omnivorism in the entire Homo family.

Don said...


Those would be the fauna that did not have behaviors effective for avoiding human predation. Those that survived (elephants, giraffes, rhinos, ostrich) through the Pleistocene were the ones that DIDN't evolved effective methods for avoiding humans.


The question is, how well adapted are humans to eating large amounts of meat. The fact (if it is a fact) that some hominids hunted some megafauna to extinction doesn't show that those hominids were well adapted to that level of meat consumption. Put otherwise, paleo proponents attack grain-based diets because presumably humans have been eating grains only 10K years, not enough time for genetic adaptation. So, if humans were only able to hunt some megafauna for 10-100K years out of 6 million, and intermittently, the question remains, did this brief period of hunting and its resultant diet exert enough selective pressure to produce a human genome highly adapted to meat-and-fat dominated diets?

Don said...


Think sweet spot. Also, from the second reference you gave:

"The OB subjects used threefold more antihypertensive medication than OO. Meat intake was 34% higher in OB than OO, whereas fish intake was sevenfold higher in OO than OB."

OB = Brazilian Okinawans. From the same abstract: "Japanese immigrants from Okinawa living in Brazil have a higher mortality from cardiovascular diseases and have their mean life expectancy shortened compared with their counterparts living in Japan."

Doesn't look to me like this supports the idea that increased meat and fat intake benefits the Okinawans.

jandro said...


you misunderstood me. The Brazil-Okinawa study is used to demonstrate their higher life-span is not a genetic mutation, but rather an environmental response.

"Meat intake was 34% higher in OB than OO, whereas fish intake was sevenfold higher in OO than OB"

In OO fish intake was 7x higher, how does that not support the idea that more "animal consumption" might have a benefit?(and again, that doesn't mean high fat is good, but that adding some fat to low-fat is a benefit) I don't think the 34% meat increase in OB outweighs the huge reduction in fish consumption (specially since fish is a consumed very often in Okinawa, probably more often than meat (although we have no data on that)).

I am very interested in knowing your opinion on this.

Don said...


Meat and fish are not the same thing nutritionally. We can't be sloppy and say OO are healthier from "animal" consumption when there are vast differences between salmon and beef.

I don't have the full data, but "sevenfold" can't give you any idea of actual volume. If BO consume 2 g of fish daily, and OO consume 15 g (half ounce), OO have seven times higher consumption...but the quantity is still small overall. Again, I am talking about sweet spots. Too low = not good, too high also not good. Every nutrient has a window of non-toxic benefits; too little = deficiencies, too much = toxicities. I believe that we will find this with animal products as well; and always relative to an individual (age, sex, activity level, etc. etc.). The ideal level of consumption will vary with individual, type of animal food (shellfish, fish, poultry, mammal's meat), and prevailing climate (hot, cold, wet, dry), among other factors.

jandro said...

I agree.