166 SYSTEMATIC PALEONTOLOGY 



the latter in bulk. The tooth-ridge formula is also intermediate in size, 

 there being less than two cross ridges to the inch, although the ridges are 

 not so widely separated as in the Imperial mammoth. 



It may well be termed the Southern mammoth, for it is the common 

 species of the southern United States, and while in places the northern 

 limit of E. columbi slightly overlaps the southern boundary of E. primi- 

 genius, yet, as a whole, its range is to the south of that species. This 

 range extended from the State of Washington southward to Texas and 

 eastward to Maryland and Florida. The occurrence in Maryland of ex- 

 amples of E. columbi and E. primigenius within a few miles of one an- 

 other is of great interest, as it definitely marks the overlapping of the 

 range of these two species on the Atlantic coast. 



The probable course of the Southern mammoth northwards into Mary- 

 land was along the coast from Florida together with the Megatherium 

 and accompanying animals. Since the number of specimens that have 

 been found increase as we go south remains of the animal may be called 

 fairly abundant in parts of Florida. It may have been found even far- 

 ther north than just noted, as a tooth discovered at Edmonton, Alberta 

 is ascribed to this species, and the specimen from Long Island, Hudson 

 Bay, is said to be intermediate in character between E. columbi and 

 E. primigenius. The occurrence of the Columbian mammoth so far 

 north is, however, extremely doubtful and the presence of even the 

 northern species so far to the east of its known range is very likely due 

 to accidental causes. 



Occurrence. — Teeth of the Columbian mammoth have been reported 

 from Pennsylvania, but the most unimpeachable evidence of its north- 

 ern range is a fairly complete skeleton from Indiana in the American 

 Museum of Natural History. 



From the modern point of view it would be singular if all the ele- 

 phants over such a considerable range of territory should belong to one 

 species and, just as Professor Matschie has recently separated the African 

 elephants into four species, so if we could have entire specimens instead 

 of detached parts of skeletons, it would undoubtedly be an easy matter to 

 separate the fossil elephants ascribed to E. columbi into several well- 



