268 BULLETIN OF THE BUREAU OF FISHERIES. 



CLASSIFICATION AND NOMENCLATURE. 



In conformity with my "Review of the Cephalopods of Western North America" 

 (Berry, 1912a) I have followed the general lines of cephalopod classification laid down 

 in the various works of Hoyle, with certain modifications adapted from Pfeffer's "Synop- 

 sis" (1900) and the masterly monograph of the Valdivia CEgopsida by Chun (1910). 

 The present material has afforded few bizarre forms and, although highly important 

 from a distributional and local standpoint, accomplishes very little toward the elucida- 

 tion of the wider evolutionary problems. Of decided importance, however, is the 

 discovery of an otherwise typical member of the family Cirroteuihida, which is remarkable 

 not only for being the second species in the entire group known to possess an odontophore, 

 but also for its apparently total lack of the alternating rows of cirri along the arms, 

 hitherto thought to be a characteristic as invariable as it is puzzling. The occurrence 

 of a Pacific Scmirgus and a new sepiolid having affinity with the Atlantic Stoloteuthis 

 leucoptera is also noteworthy. The presence in the collection of a large series of young 

 ommastrephids comprising at least two Rhyncoteuthion-like forms specifically distinct 

 from one another indicates that the occurrence of the curious Rhyncoteidhion stage is 

 an exceedingly widespread ontogenetic character in this family and one by no means 

 confined to the typical genus. 



The primary division of the Dibranchiata of Owen into the Octopoda (devilfishes) 

 and Decapoda (squids ), as proposed by Leach in 181 7, is here adhered to, although 

 the latter term is very confusing in view of the prior use of the name for a well-known 

 group of Crustacea (Decapoda Latreille 1806). In many respects the restoration of 

 Blainville's Octocera and Decacera would be much more satisfactory, an excuse for 

 discarding the complementary term Octopoda being possibly available because of the 

 existence of Octopodia Schneider 1784, which was proposed as a general term for the 

 entire class over a decade before Cuvier called them Cephalopoda. 



I have found myself quite unable to accept the various major divisions into which 

 some authors have grouped the Octopoda, but on the other hand I have been unable to 

 formulate any more natural arrangement of my own which could be utilized in their 

 stead. In regard to the Decapoda, however, the anciently recognized bifurcation into 

 Myopsid and CEgopsid forms seems very convenient and on the whole a natural separa- 

 tion. Chun's further division of the latter group into CEgopsida libera and CEgopsida 

 consuta appears to me cumbersome and but little superior to the nearly or quite synony- 

 mous Teuthidea and Taonidea of Verrill. For the purposes of the present paper it is 

 not, however, necessary to adopt either system. 



In the arrangement of the families themselves scarcely any two authors are in 

 entire agreement, so that the sequence herein followed can be regarded as in no way 

 more than provisional. 



A word should be said in regard to the sort of morphological characters chiefly 

 depended upon by the author in distinguishing species; that is, beyond such conspicuous 

 differences in bodily structure and general form as obviously require no explanation 



1 1n the Hawaiian Islands the term "squid " is colloquially employed to include nearly all cephalopods. especially theedible 

 forms such as Polylms, even though the usage is not scientific nor accurate English. 



