Quick viewing(Text Mode)

Late Quaternary Biogeography and Extinction of Proboscideans in North America

Late Quaternary Biogeography and Extinction of Proboscideans in North America

Late Quaternary biogeography and of Proboscideans in

R.Wm. Graham Department of Earth and Space Sciences, Denver Museum of Nature and Science, Denver, USA - [email protected]

North America supported one of the highest overlapped temporally and geographically with diversities of proboscideans in one of the small- the woolly with which it may have est areas during the late (10-20 ka). been a potential competitor. There were at least four different genera (Mammut, Mammuthus, Haplomastodon, and The most widespread (latitudinally, longitu- ). All of these genera were mono- dinally, and altitudinally) North American pro- specific except for Mammuthus with four dif- boscidean at the end of the Pleistocene was M. ferent (M. columbi, M. jeffersonii, M. columbi (). It dominated primigenius, and M. exilis). None of these pro- the US west of the Mississippi River and south boscideans had a continental wide distribution of the continental ice sheets. It ranged well into at the end of the Pleistocene. In fact, during full . The Columbian mammoth also glacial times when continental ice covered occurred in between 10-20 ka where it most of northern North America, all four pro- may have been isolated from the western popu- boscidean genera inhabited areas south of the lations. Columbian are known from ice sheet that were smaller than the geographic coastal sites to more than 3000 m (10000 feet) range of either of the living genera above sea level. This species of mammoth is (Loxodonta and ). predominantly associated with mam- mals like camels (), horses (Equus), Based upon geographic distribution, dental (Bison), and big headed morphology, isotopic analyses, and associated (); it appears to have preferred a paleoenvironmental indicators, all of these pro- steppe//parkland environment. boscideans probably had preferred habitats. M. primigenius () was the north- Mammut americanum (American mastodont) ern most taxon and it was a late Eurasian has been in North America for at least the last immigrant, appearing in North America some five million . During the time after 200 ka. It occurred in , the (10-20 ka), the American mastodont inhabited upper midwestern and northeastern United most of the eastern US and southeastern States (US) and along the northern Atlantic south of the ice sheet and ranged into coastal plain. The woolly mammoth preferred Mexico. Some of the richest and latest finds of or , “tundra”, and the Mammut occur in the region of the forest/woodland ecotone between these two US and Canada, the Mississippi River valley, “biomes.” M. jeffersonii (Jefferson’s mam- and Florida. There are a few isolated terminal moth) evolved in North America and it is con- Pleistocene records in the western US. Its dis- vergent with the woolly mammoth by having a tribution is almost a mirror image of the distri- large number of plates and thin enamel. bution of the Columbian mammoth. Also, in Jefferson’s mammoth occupied the upper mid- well-stratified sites in the midwestern US, western US at the end of the Pleistocene and mammoth and American mastodont almost

707 The World of - International Congress, Rome 2001

never co-occur at a site suggesting habitat climate trigger, the Younger Dryas. exclusion. The American mastodont preferred woodland to forest habitats, especially the late Several factors probably contributed to pro- Pleistocene boreal forest. It is frequently found boscidean extinction and these same factors are with browsing like giant critical to the survival of modern elephants. (Castoroides), stagmoose (Cervalces), wood- Geographic range reduction was probably the land musk ox (Bootherium), and stout-legged dominant factor. The pre-Wisconsin distribu- (Palaeolama). tion for all North American proboscideans, except for M. exilis, demonstrates that their Haplomastodon and Cuvieronius are two ranges were reduced throughout the Quater- New World genera with pre-Quaternary origins nary. Among mammals in general, it is believed in North America but they primarily inhabited that the probability of extinction increases southern North America, Central America, and exponentially with reduction in geographic South America during the late Quaternary. At range. Futhermore, the size of the home range the end of the Pleistocene, Haplomastodon and of a species is a direct function of Cuvieronius only extended as far north as its body size. As the largest ungulates, pro- Mexico. There may have been an isolated pop- boscideans would have required the largest ulation of Cuvieronius in Florida. These two geographic ranges. As already documented, the mastodonts were geographically sympatric inclusive geographic range of all of these pro- with Mammut and Mammuthus in central boscideans during the late Quaternary was less Mexico in the terminal Pleistocene, although than the range of either of the living elephant their temporal and ecological overlaps are species historically. poorly known. The teeth of Haplo- and Cuvieronius have similar mor- Habitat destruction was probably another phology and they both presumably preferred contributing factor. The Pleistocene pro- open woodland habitat as well as mesic tropical boscideans had preferred habitats as docu- lowland habitats farther to the south. mented by their geographic ranges, dental morphology, isotopic composition of their Mammuthus exilis had the smallest geo- , and association with other envi- graphic distribution of any of the North ronmental proxies. The extinction of pro- American proboscideans as it was restricted to boscideans and other is highly cor- the Channel Islands off the coast of . related with the disappearance of certain It was presumably derived from the Columbian Pleistocene habitat types (open boreal forest, mammoth. parkland, and savanna).

All of these proboscideans became extinct One of the challenges for environmental sometime after the (18 models of extinction is the question: What was ka) and it appears that most, if not all, survived unique about the end of the Pleistocene envi- until the terminal Pleistocene (10-12 ka). ronmental change? If geographic range reduc- Extensive dating of M. columbi and M. ameri- tion was one of the major drivers, as proposed canum remains indicate that both of these taxa here, then the environmental change does not survived until 10.8 ka while most other large need to be unique but must simply serve as a Pleistocene mammals were extinct by 11 ka. trigger once a threshold has been reached. For These data suggest that the Keystone Species geographic range reduction, the threshold is Model of Pleistocene extinction may not be critical minimum population size or critical valid for these taxa since they were the last, minimum size for geographic range. Until rather than the first, to go extinct. These dates these limits are reached, environmental change are concordant with a Clovis model of extinc- will not cause extinction but once these limits tion but they are also in agreement with a major are exceeded then extinction will be triggered. 708 Late Quaternary Biogeography and Extinction of Proboscideans in North America

The environmental change does not need to be humans may have been involved in their unique but merely drive populations and geo- demise. However, the actual impact of humans graphic ranges below viable limits. is not really known because it is hard to esti- mate predation pressure and population sizes Human involvement in proboscidean extinc- for proboscideans and humans. tion in North America is hard to evaluate. Evidence for human association with late Finally, the extinction of North American late Quaternary North American proboscideans has Quaternary proboscideans underscores the sig- only been found for M. columbi, M. ameri- nificance of geographic range and habitat for canum, and Haplomastodon. In these cases, the survival of the modern elephant species.

709

Top View