HYATT: GENEALOGY OF ACHATINELLID^E. 381 



of any particular pattern to the species. There are no 

 known enemies which could be attracted by the colors of 

 these shells nor do they seek protection from any so far as 

 known. The only possible cause of limitation has been sug- 

 gested by Mr. Gulick. He thinks that the colors of each 

 species are made up by the crossing of a limited number of 

 varieties which migrated across the ridges and mingling in 

 new localities of neighboring valleys necessarily produced a 

 distinct combination of patterns and colors from those of 

 the parent species. 



This may or may not account for the differences observed 

 between closely related species that occur in contiguous val- 

 leys which are separated by high ridges, but the fact that 

 there is parallelism in the color patterns in different species 

 occurring in different valleys, is easily observed. This 

 parallelism is also obviously divisible into homoplastic simi- 

 larities and genetic differences as shown in the predomin- 

 ance of some definite pattern or color in the same species ac- 

 companied usually by differences in proportion, size and 

 often direction of the spire. 



The tooth-fold is also more constant in arboreal forms, and 

 no case of absence or of the presence of additional folds 

 has been observed. This greater permanency in the generic 

 and family characters is obviously similar to what "Williams 

 has observed, and which he considers as a secondary condi- 

 tion acquired through inheritance whereas, the variable 

 ones noticed in the tooth-fold and columella of Amastra 

 would he explained as due to the greater force of inherent 

 variability in this more primitive genus. 



This is true, as we have said, of all such primitive forms, 

 and still more remarkable examples of this law of acceler- 

 ated divergence are present, in these islands, in genera that 

 are more primitive than Amastra, but there is this significant 

 qualification: They occur in one island, Kauai, which is the 

 most ancient of the group ; they are survivals of a still more 

 ancient and variable fauna composed wholly of ground shells 

 which are divisible into several distinct genera. They occu- 

 pied a perfectly free field and evolved several large robust 



