Nos. 451-452.] AMERICAN SOCfETV OF ZOOLOGISTS. 495 



Commencing at the most eastern of tliese \-alleys and i)ro- 

 ceeding toward the northwest, we find that llie thsl sc\en or 

 eight miles brings us past about thai uuinl)er ol deej) \alleys, 

 occupied by shady groves and inhabited by many \aiu;lies 

 belonging to fifteen or more si)ecies. TIk-sc- siK'cies are all 

 of the larger type of the genus, being near an inch in length. 

 With some of them, all the individuals ot the spec k s are lounii 

 in the groves of a single valley, and in the ea.e n| mnM. .a en 

 stragglers arc not found at a distance of mmr ilun a nnK- h.-m 

 the metropolis of the species. 



Proceeding toward the northwest, we find that the -en us is 

 represented by many species, of progressivi-l\ dinnni-hui- >i/.e, 

 till at the other end of the mountain ran.ue the slulls have 

 dimensions not more than half those presented l»y the ^iiells 

 found in the valleys first explored. The habits ot Iceding are 

 also wholly different : for as we approach the northwest end of 

 the range we find that the larger trees have been deserted, and 

 the shrubs and herbs have become the homes of most of the 

 species. This diversity either in the inherited aptitudes, or in 

 the acquired habits of the isolated groups of one slock, re.^ultmg 

 in different methods of dealing with the same en\ ironmenl, un- 

 doubtedly results in different forms of selection helping to mould 

 the different groups into different species. We, however, note 

 that, in these cases, the i)r(Kess is not natural selection ; for it 



of divergent adaptations, but it is slightly divergent adai)tations, 

 protected by isolation from the swamping effects of tree-crossing 

 with the original stock, that ^initiates divergent method.^ ()l deal- 

 in the selecti(m which emphasizes the di\er-eiue. 1 hi^ loini 

 of selecti(m, determined by the aptitudes and hahitu(le> ot the 

 species in dealing with the environment. I call endononu^ 



A point of very gTcat interest is the important part filled by 

 isolation in producing divergent forms of endonomic selection. 

 If a pair of individuals from a plastic (or variable) species are 

 completely isolated from the original stock, the probability is 

 that in some way their method of dealing with the environment 



