Baffled over bison

Two centuries ago, bison and elk were more common to the east of North America's Rocky Mountains than to the west. A look into why this was so may hold lessons for modern conservation biologists.

In 1804, President Thomas Jefferson sanctioned and funded one of the most important expeditions in North America's history. Meriwether Lewis and William Clark set off from St Louis, Missouri, with a commission to survey the upper Missouri River basin and to seek a means of crossing the Rocky Mountains to reach the west coast of the continent (Fig. 1). Their geographical and biological discoveries were of great consequence for the development of the region, but they also brought to light some difficult questions for future biogeographers. One that is still not fully answered is why two large ungulates then common on the east side of the Rockies, namely the bison (Bison bison) and the elk (Cervus canadensis), were scarce on the western side. Writing in Conservation Biology, Lee Lyman and Steve Wolverton1 take up the 200-year-old challenge posed by Lewis and Clark's observations and conclude that a range of factors, involving both climate and natural and human predation, were involved.

Figure 1: North American explorations.


Meriwether Lewis and William Clark's famous expedition brought to light questions that tax biogeographers to this day. (Painting by Alfred Russell.)

The journals of the expedition, written by both Lewis and Clark, supply data on the contemporaneous distribution and abundance of bison and elk. These include qualitative observations, such as the fact that bison became scarcer as they proceeded west along the upper Missouri River in Montana, and also quantitative data in the form of records of game killed to supply food for the expedition. Their figures relating to game kills have been analysed previously by Martin and Szuter2, who came to the conclusion that the lack of game on the western side of the Continental Divide was a consequence of intense hunting by native American peoples, resulting in a 'game sink'. The intensity of such human 'predation' was less on the east of the mountains, they claimed, because this was disputed territory that lay on tribal boundaries, resulting in high game populations within this 'war zone'. But Lyman and Wolverton1 have re-examined the kill data, and express their results with greater temporal and geographical precision. They supply the numbers of animals killed in each area per day, providing a more accurate picture of the relative abundance of game species over the area covered.

The recalculated data1 present a rather different picture from that of Martin and Szuter in two respects. First, the boundary between game-rich and game-poor areas is much more diffuse than was originally proposed; there is no sharp division between game sink and war zone. Second, there is high spatial and temporal variability even within the game-rich region. Some sites that were game-poor on the westward journey proved rich on the return leg. But, as the original journals give no indication of the effort made by the expedition in hunting, even the variability may be spurious. The idea of a clear distinction between rich and poor areas cannot be sustained. Nonetheless, the relative scarcity of game to the west of the Rockies compared with the east is still apparent.

Why should this be? Would elk and bison numbers have been substantially higher in the west if human predation in that region had been lower? This, like all historical hypotheses, is difficult to test. Martin and Szuter2 claim that horses and cattle have since managed perfectly well west of the mountains, so grazing resources for herbivores such as elk and bison were evidently available. This is not a very robust argument, however, because large herbivores do not all have the same dietary or ecological requirements3, and Lyman and Wolverton1 also show that cattle have been successful in the area only when supplied with supplementary food.

Historical theses may be best tested by historical methods. Archaeological data on bison kill sites in the past 10,000 years show a much greater abundance of such sites east of the Rocky Mountains, although there is a scattering to the west. So the mountains were not an insurmountable physical barrier to the bison, and archaeology seems to indicate that there was, in fact, higher human predation in the east than the west.

So factors other than human predation must have limited the success of these large herbivores in the west coast region, and a climatic factor such as high snowfall is one possibility. Daubenmire4 has observed that bison can remove light snow (up to 75 cm deep) from buried vegetation with their heads, but that they are defeated by deep snow. As the western region has heavy snowfalls, this could adversely affect bison populations in the area. Under such circumstances, any additional stress, such as human predation, accompanied by limited recolonization through the mountain passes following local extinction, could well account for the bison's lack of success in the west. The Rocky Mountains would thus act as a filter rather than a complete barrier to animal migrations.

Archaeological data also confirm the presence of elk to the west of the Rocky Mountains throughout the past ten thousand years or so, and indicate a sustained level of human predation. But within the last century, elk abundance has greatly increased. Lyman and Wolverton1 account for this in terms of lower natural predation rather than less human predation. The extirpation of wolves and cougars from much of the region in the early twentieth century is the likely cause of growing elk populations, so natural predation was probably a constraint on population growth in earlier times.

The conclusion of this piece of research into the historical biogeography of the Pacific Northwest is that a combination of factors, rather than simply human predation, produced the past pattern of abundance of these two ungulates. The historical approaches used here also have much to offer conservation biologists. If the factors that have limited biogeographical distributions and population levels of organisms in the past are better understood, then current programmes of reintroduction and management can be more effectively evaluated. The past, for the modern ecologist, often provides a key to the present.


  1. 1

    Lyman, R. L. & Wolverton, S. Conserv. Biol. 16, 73–85 (2002).

    Article  Google Scholar 

  2. 2

    Martin, P. S. & Szuter, C. R. Conserv. Biol. 13, 36–45 (1999).

    Article  Google Scholar 

  3. 3

    Menard, C., Duncan, P., Fleurance, G., George, J-Y & Lilas, M. J. Appl. Ecol. 39, 120–133 (2002).

    Article  Google Scholar 

  4. 4

    Daubenmire, R. Ecology 66, 622–624 (1985).

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Peter D. Moore.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Moore, P. Baffled over bison. Nature 416, 488–489 (2002).

Download citation


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing