A. E. Yerrill — Molluscan Archetype, 97 



distinctly spiral (figs. 7, 8, 9) ; in the Pteropoda ^ it is some- 

 times spiral and sometimes conical or cornucopia-shaped (figs. 

 (13, 14). Therefore some of the fundamental distinctions 

 between these classes are clearly developed in the early veliger 

 stages, at a time when the gills and heart have not appeared, 

 and when the central nervous system is represented only by 

 rudiments of the cerebral ganglions (fig. 13, n)^ and when the 

 foot, itself, has no definite or characteristic form. 



These facts naturally lead to the conclusions that all mol- 

 lusks have been derived from free-swimming forms similar to 

 modern veligers and proveligers, and that each of the great 

 classes, and perhaps some of the subclasses, had already become 

 differentiated while still having veliger-like forms and modes 

 of life, or in other words, that the primitive bivalves, scapho- 

 pods, and gastropods, when adult, were all small, free-swim- 

 ming forms, furnished with a ciliated locomotive organ, similar 

 to the velum of modern veligers. Some of these swimming 

 species may have developed directly a swimming foot adapted 

 to a continued free-swimming existence, and thus the primi- 

 tive pteropods may have originated. In others the primitive 

 foot may have developed directly into a creeping disk, or into 

 a sucker-like disk for adhesion while resting, from which the 

 transition to a creeping disk would be easy. Many modern 

 veligers {Natioa^ fig. 11, etc.) acquire a well-formed creeping 

 foot, with an operculum, before they lose the velum, so that 

 they can either swim or creep at will. In this way the creep- 

 ing Gastropoda may have arisen. In the case of the hetero- 

 pods, which are free-swimming Gastropoda, with the foot in 

 the form of a median fin, we often find a small cup-shaped 

 sucker on the foot, which probably is used for adhesion to 

 floating objects while resting. This may be the survival of a 

 condition that was common among primitive gastropod veli- 

 gers. Indeed the heteropods, as a group, may be of very 

 ancient origin, and not derived, in comparatively recent times, 

 from normal gastropods, as is often stated. Associated with 

 the remains of other forms of pelagic animals, in paleozoic 



of the pull of the retractors (figs. 14, 15, r, r^), which are here attached far back, 

 near the end of the shell. When they are attached farther forward, the tendency 

 is to cause a stronger curvature, as in gastropod veligers. The primary one- 

 sided and spiral form of the shell is probably due to the inequality in size of the 

 nutritive yolk-masses (figs. 6, 12, 13, u, u). 



* For my present purposes I prefer to treat the Pteropoda as a distinct class, 

 but yet admit their close relations to the opisthobranchiate gastropods. The 

 latter may have been derived from primitive pteropods, instead of the reverse, as 

 claimed by Pelseneer. who holds that they are a modern group, not earlier than 

 the Tertiary, and derived directly from two different families of living opistho- 

 branchs. In the paleozoic rocks of America there are. however, an abundance 

 of pteropod shells having all the shell-characters, including the peculiar nucleus, 

 present in the common living genus Styliola or Creseis, even if the Tentaculites be 

 not considered as pteropods. There is no good reason for denying the latter. 



