176 POLYTROPA. 



Dr. P. P. Carpenter was the first who attempted to classify 

 the lapillus group of the West Coast of North America ; he 

 separates them as follows : 



" P. xaxicola, Val. Some varieties run into the New England 

 form of P. lapillus sufficiently nearly to justify the identifica- 

 tion; but the bulk of the specimens are easily distinguished by 

 the excavated columella. They pass by insensible gradations to 

 P. ostrina, Gould, which is a rare and extreme variety. Many 

 of the shells called P. Fruy<;inelii by Midd., are certainly refer- 

 able to this species. Some forms pass towards the true P. Frey- 

 cinetii* Desh., while others are equally close to the very different 

 P. emarginata, Desh. 



" Var.y%w&fo, Forbes. 

 " Var. emarginata, Desh. 

 " Var. oxtrina, Gld. 



" P. canaliculata, Duel. = decpnifostata . Midd., attenuata, Rve.. 

 analoga, Forbes. 



" P. crispata, Chemn. = plicata, Mart., laHuca. Esch.. septen- 

 trionalis, Rve.. etc." 



E. von Martens (Mai. Blatt.. xix, 86, 1872) does not hesitate 

 to refer all these forms to P. lima, Martyn, but does not speak 

 of their relationship to lapillus except as ' N. W. American 

 species of Purpura of the lapillus group." 



Doubtful Species referred to Polytropa. 



P. EXILIS, Dunker. 



Narrowty ovate, yellowish, whorls four, with a number of 

 revolving riblets and striae, decussated by narrow growth-lines ; 

 bodv-whorl double the length of the spire, with three revolving 

 costulse, next whorl bicarinate, upper one unicarinate ; aperture 

 ovate; columella sinuous ; labrum thickened, crenulate. 



Length 7 mill., diam. 3 mill. 



Upolu. 



Not figured. Evidently a young shell, and appears to be 

 related to P. hdemastoma. 



