The Evolution of an Ecosystem:
Pleistocene Extinctions

Elin Whitney-Smith Ph.D.


It is generally assumed that evolution is an issue of looking at how the species fits into its environment. This paper suggests that we over-constrain our thinking on evolution we should be thinking of how the species and the ecosystem evolve together.

The current theories of the Pleistocene extinction (Climate change and Overhunting by H. sapiens) are inadequate. Neither explains why: (1) browsers, mixed feeders and non-ruminant grazer species suffered most, while ruminant grazers like bison generally survived, (2) surviving mammal species including both subspecies of bison were sharply diminished in size; and (3) vegetative environments shifted from plaid to striped (Guthrie, 1980.)

Climate change theories do not explain why mammoths and other megaherbivores survived changes of similar magnitude.

Although flawed, the simple overkill hypothesis does link the extinctions and the arrival of H. sapiens. However, they omit the reciprocal impact of prey decline on H. Sapiens; standard predator-prey models, which include this effect, demonstrate that predators cannot hunt their prey to extinction without themselves succumbing to starvation.

An alternate scenario and computer simulation (available on the internet from characterized by a boom/bust population pattern is presented. It suggests that H. sapiens reduced predator populations, causing a herbivore population boom, leading to overgrazing of trees and grass, resulting in environmental exhaustion and extinction of herbivores.

If true, bison survival through the Pleistocene may be accounted for thus: herbivore population explosion created a condition of scarcity in which there was selective pressure favoring animals that could extract maximum energy from low quality forage to survive and reproduce. As a new Holocene stability emerged Holocene bison split into two sub-species (B. bison bison [plains bison] and B. bison athabascae [woodland bison]) that fit and maintained the emerging "striped" environments.

1.1 Background

Eighteen thousand years ago at the height of the last ice age in North America the land not ice covered looks like a park with mixed trees and grass

There are mammoths being killed by massive lions and sabertooth cats. Bison are almost as big as elephants. Beavers are as big as bears, the short faced bear stands more than 5 foot at the shoulder, almost twice the size of a grizzly bear. These big animals merit their name - mega-fauna.

Eight thousand years later the ice caps are wasting away. Paleo-indians are using beautiful fluted points to hunt the bison the mastodons and other mega-fauna in park lands of North and South America. They work little - about 8 hours per week - and live well. Game and vegetation is available for the taking. Theirs is a world of plenty - an Eden.

Just a thousand years later, the huge ice caps, horses, sloths, other giant animals, the carnivores that ate them, and the paleo-indians, all are gone. Climate is like it is today. The parklands are gone. There are belts of closed canopy forest on the coasts and a vast grassland in the center of the North American continent.

The few people we find are living in small isolated bands. Their stone technology is far less developed than that of the paleo-indians. Who are these people? What happened to the Paleo-Indian culture?

Why did the pattern of vegetation change? Why are bison and beavers so small? Where are the mastodon, the mammoth, the horse and the big cats who hunted them?

These are the mysteries of the extinctions at the end of the last Ice Age.

Two theories have been put forward to explain them - climate change and overkill by people. (See Pleistocene Extinctions: The Search for a Cause edited by P.S. Martin and H. E. Wright, Jr., published by Yale Univ. Press in 1967 and Quaternary Extinctions: A Prehistoric Revolution edited by P.S. Martin and R. G. Klein published by Arizona Univ. Press in 1984.)

Vastly simplified, the climate change hypothesis suggests that as the ice age waned climate changed so drastically that animals were not able to adapt. The overkill hypothesis suggests that humans swept into the New World killing everything in their path.

1.2 The problems with these theories is:

Both theories are like Escher drawings, correct in detail but inconsistent over all.

1.2.1 Climate change

Mammoths, sloths, mastodons and other animals that went extinct had survived similarly warm periods during previous interglacials.

New World horses, which went extinct at the end of the Ice Age, are thriving in that same climate today.

1.2.2 Overkill

Predators cannot wipe out their major food source and establish its own population. As animals become scarce either predators turn to other prey or they can't find enough food to get the nutrients necessary to reproduce. The prey is able to re-establish itself.

The huge ground sloths and other animals not hunted by people, nonetheless went extinct.

2.1 A new theory is needed - Second Order Predation.


Consider the following:

  • Paleo-indians realized that eliminating their competition would increase the amount of game available to them.

  • New World carnivores were probably easier to kill than those of Africa and Asia because they had no instinctive fear of humans.

  • Also without instinctive fear New World carnivores may have killed people.

  • People are dangerous game - they remember and establish policies like, "The only good carnivore is a dead carnivore."

  • Reducing carnivore populations by a relatively small amount changed the ecological balance. Herbivore populations boomed.

  • Herbivores ate everything in sight. When the vegetation was gone they died of starvation. The ones who depended on the slowest growing resources (trees) died first (browzers and mixed feeders like mastodons). They were followed by grass eaters who were less efficient (non-ruminants like horses, mammoths and sloths).

  • Specialist carnivores who depended upon the mega-fauna went extinct.

  • The survivors had less game to hunt. They had to deal with other carnivores (human and non-human) who could not find enough game. Those who made it were those who could deal with scarcity.

Ruminant herbivores like bison, which extract the maximum nutrition from the minimum amount of poor quality forage. Small animals who require less food to survive and reproduce.

Generalist carnivores and scavengers like wolves and smaller cats.

People who were willing to change their life style and earn their living through primitive planting, fishing, foraging, and scavenging. People living in small isolated groups wary of those who might take away their subsistance.

3.1 Modeling Extinctions

To see if the relationships specified in this hypothesis are internally consistent I made a system dynamics computer model of a simple ecosystem (see Whitney-Smith 2001).

The ecosystem consists of two kinds of plants - trees and grass, four kinds of herbivores - browzers, mixed feeders, ruminant and non-ruminant grazers. I simulated Overkill (Figure 1) and Second Order Predation (Figure 2 ). The graphs show that hunting herbivores has far less impact than does killing carnivores.

Graph H. sapiens increasing significantly carnivore & herbivore populations decline slightly

Figure 1. People enter the New World and hunt herbivores (Overkill Hypothesis) Herbivore and Carnivore populations decrease, plants increase because there are fewer herbivores eating them.

graph of second order predation

Figure 2. People kill carnivores (Second Order Predation) herbivores and human populations boom then crash. Plants decline, crash and then rebuild after the crash of herbivore populations

If we look behind the scenes in the Second Order Predation simulation, browzers, mixed feeders and non-ruminant grazers go extinct (Graph 3). The immediate cause of their extinction is overgrazing (Graph 4). The browzers and mixed feeders eat all the trees and then starve.

graph of second order predation- herbivores

Figure 3. Herbivores (Second Order Predation), non-ruminant grazers decline and all other herbivores boom, then mixed feeders and browsers crash. Ruminant grazers recover and out compete non-ruminants for scarce resources. Ruminants take a predation when non-ruminant populations crash.

graph of second order predation- plants

Figure. 4. Plants (Second Order Predation) Trees decline and eventually crash due to over uses by herbivores. When browsers and mixed feeders go extinct trees recover and thrive because they have no predators.

If this model is correct for the New World, we know why the pattern of vegetation changed, why animals not hunted by people went extinct, and why horses went extinct even though they can live in our climate.

Comparing the Overkill Hypothesis and the Second Order Predation Hypothesis suggests new insights into the peopling of the New World:

Under the original Overkill Hypothesis, human populations establish themselves slowly and the herbivores and carnivore populations are only slightly reduced. Plants do well because there are fewer herbivores eating them.

Under the Second Order Predation or Boom/Bust Hypothesis, human populations establish themselves more quickly because there is less competition from carnivores and more herbivores to hunt. When the herbivore populations begin to overgraze the environment there is a crash in human populations.

This sequence gives us a new insight into both the spread of Paleo-Indian culture and its disappearance. During the initial boom period there was enough, and more than enough, for everyone. Human populations could expand as fast as breeding allowed. People had no reason not to share food or ideas and technology. It was a world of plenty.

Then, with the crash, people starved. Smaller animals and the denuded vegetation could only support small bands widely spaced out over the landscape. They were isolated. It was a land of scarcity. In that harsh world there was less leisure to create beautiful points and many of the people with the skills to do so had starved - paleo dark ages. As in other dark ages, skills were lost and the quality of life reduced.

4.1 Evolution of ecosystems

Thus, it is possible for horses to live in modern environments and yet to have gone extinct during the time of scarcity. The scenario and model results presented above suggest that the entire ecosystem evolved from the stability of the Pleistocene through the boom/bust time of instability until the Holocene stability was established. Animals, like horses, are fit to live in both the Pleistocene and the Holocene environments but did not survive the "bottleneck" of the transition.

Unless we understand the dynamics of the time of extinction we get a distorted picture of what survival of the fittest means. We assume that it means some absolute quality of fitness. This in turn, may get translated, in our minds with some sense that success is related to virtue. Whereas it may simply mean that events at the time of extinction were not fortuitous for survival at that particular time and place.

5.1 References:

Martin, P.S. & Wright, H.E. (eds)

    1967. Pleistocene extinctions: The search for a cause. Yale Univ. Press, New Haven.

Martin, P.S. & Klein, R.G. (eds)

    1988. Quaternary extinctions: A prehistoric revolution. Univ. Arizona Press, Tucson.

Whitney-Smith, E.

    2001 Second Order Predation and Pleistocene Extinctions:A System Dynamics Model PhD dissertation available from University Microfilms or at