382 HYATT: GENEALOGY OP ACHATINELLIDyE. 



types of shells; but these all died out. Whereas, the small 

 insignificant Amastra and another still smaller genus of 

 ground shells, Leptachatina, that arose in the same fauna, or 

 migrated from island to island, persistently held their own. 



Leptachatina is a glassy shell, smaller than Amastra, and 

 adapted for 'both living on the ground and very rarely on 

 plants in places unfavorable for the existence of other genera 

 of this family, such as in open country, and is now found on 

 all of the islands. This genus is also most abundant in a 

 fossil state, and is comparatively invariable. It maintains 

 its characteristics without any marked changes in all loca- 

 tions. Its characteristics are those of a specialized type and 

 there is no indication in its structure that it is the ancestor 

 of any other genus. On the other hand, there is strong 

 evidence that it is a modified descendant of the Kauaiian 

 species, Amastra nucleola Gould. We can therefore neglect 

 this series as having no further meaning in the phylogeny. 



In Oahu, the island next in geologic age and geographic 

 succession to the southeast, the only shells of this ground- 

 living group found as fossils are Amastra and Leptachatina, 

 the genera first evolved on Kauai. On this island the mi- 

 grations extended in two directions ; first geographically along 

 the surface of the ground. Here they met with more varied 

 conditions than in Kauai but their habitat as long as they 

 occupied the surface alone was practically similar to that of 

 their ancestors on Kauai. They therefore remained the same 

 dark-colored ground-living shells, but evolved a number of 

 connected species to fill this field. 



The trees were, however, open to them, and when they 

 ascended them we find that different series and genera arose 

 marking distinct concordant stages of evolution. 



The genus Amastra as limited and defined by Gulick has 

 several series of forms on Oahu. I divided these from each 

 other by following the genetic lines that connected one species 

 with another through their similarities and intermediate gra- 

 dations and this was done without reference to any other 

 class of facts. Among these series of Amastra, or Amastra- 

 like forms, one was remarkably distinct in the possession 



