178 NEW YORK ZOOLOGICAL SOCIETY. 



its maximum among the various groups of caribou. Animals 

 in the same herd may differ widely in this respect. Even the 

 antlers carried in the successive years of an animal's growth may, 

 and often do, vary ; and the two antlers on the same animal may 

 not be symmetrical. Nevertheless, within the extreme limits of 

 this irregularity there are certain types of architecture which, 

 though clearly defined, are difficult to describe. This is true of 

 nearly all the genera of the deer family. 



The antlers of the Scandinavian elk and of the two species 

 of moose are generally distinguishable, as are also antlers of 

 the Virginia deer from certain localities. Furthermore, it is 

 often possible to pick out moose antlers from the Rockies from 

 those of Eastern Canada. 



The above remarks apply to the caribou with peculiar force. 

 The caribou from Newfoundland can be distinguished from those 

 of the mainland by those who have sufficient experience in this 

 matter, although the distinction might defy definition in words. 

 The typical eastern forms of antler could not possibly be mis- 

 taken for a typical antler from the Northwestern United States, 

 or from British Columbia. Selected specimens from each local- 

 ity might be found closely approximating, but, nevertheless, the 

 main statement remains true that one familiar with the modifica- 

 tions of caribou antlers could, in many cases, name the locality 

 of a typical set of antlers. 



When the antlers are distinguishable, and the color of the 

 pelage of two animals at the same season is in marked contrast, 

 and still further when there is added to these two characters a 

 third — size — sometimes extremely marked, as between the Bar- 

 ren Ground and the Woodland groups, sometimes less extreme, 

 as between the British Columbian forms and the Eastern Wood- 

 land, we have a group of variations clearly indicating that the 

 extremes of the genus in the dififerent portions of the range are 

 at least well on the road toward forming distinct species. 



When, in addition to the permanent variations in size, color, 

 and antlers, well-marked anatomical features are found in the 

 skulls or other part of the bony structure, the case in favor of 

 specific rank becomes greatly strengthened. 



When the ranges of two species adjoin or overlap, careful 

 search must be made for intermediate and annectant forms. 

 Such forms mav be found among the western Woodland Caribou 



