SNAKES OF THE GENUS PITUOPHIS 207 



is obvious that however variations may arise they cannot become 

 established in one or the other of two groups unless they are separated 

 by an environmental difference, however slight. Such an environ- 

 mental difference may be due either to an actual change in the environ- 

 ment in the original habitat, or by the migration of the animals to a 

 different habitat. In the latter case the animals affected will be those 

 that have migrated and are found on the periphery of the range, since 

 the animals in the center would be in the original habitat and already 

 adapted to the environment, and whatever variations arose would be 

 unlikely to persist; while in the periphery of the range such variations 

 as proved favorable, whether induced by the changed environment 

 according to the Lamarckian idea, or selected by it according to the 

 Darwinian theory, could become established. If the change were in 

 the environment itself, it seems probable that such animals as migrated 

 m order to retain their association with the original environment 

 woiUd remain little changed, while those which remained in the area 

 of environmental change would evolve in correlation with the changing 

 conditions. In such a case, the "most advanced stages," as Matthew 

 says, "should be nearest the center of dispersal, the most conservative 

 stages farthest from it." In its application to this group, where 

 several distinct evolutionaiy lines radiate from a common center, this 

 principle is of questionable value, since it is evident that in this genus 

 environmental change has not advanced in concentric circles from a 

 central point, but has rather been encountered along several different 

 lines of dispersal or migration routes, which are represented by the 

 four evolutionary series radiating from the range of affinis in the south- 

 western United States and northern Mexico. 



There remains for consideration the other possibility, that the 

 animals migrate to a different environment, with the result that the 

 most conservative and generalized forms occur at the center of the 

 range in the original habitat, while the more specialized forms are 

 found at the periphery. This is undoubtedly the case in Pituophis, 

 where the most generalized form, affinis, is in the geographic center of 

 the genus, where the various evolutionary trends and lines of dispersal 

 originate, and the most specialized forms, such as lineaticollis and the 

 three most eastern subspecies of melanoleucus, are found on the 

 periphery of the range of the genus. 



We may conclude, therefore, that the genus arose in or near the 

 range of affinis, presumably in northern Mexico, where affinis inter- 

 grades with s. sayi and overlaps d. deppei in range. From this central 

 point the genus spread over every possible migration route, and as a 

 result, the melanoleucus group was developed in the east, with m. 

 ruthveni ancestral to the three eastern subspecies, and 6?. sa.yi phylo- 

 genetically as well as geographically intermediate between ruthveni 



