Phylogeny of Animals 



521 



Fig. 23 is a somewhat composite diagram that illustrates 

 details common to some rotifers and to gasteropodous, pele- 

 cypodous (lamellibranch), pteropodous, and chitonoid organ- 

 isms. One featm*e alone first deserves consideration, viz.: 

 the primitive or cuticular molluscan shell. This at once sug- 

 gests correlation with the dorsal cuticular lorica of shelled 

 rotifers, and equally with the prodeltidium of brachiopods. 

 Its dorsal position, even in lamellibranchs, as a median body 

 is noteworthy, while the fact that in some existing loricate 

 rotifers there is an expanded dorsal plate only, or a dorsal 

 and ventral plate, or two lateral plates, might indicate that 



— soc 



— mc 



s9ac 



Fig. 23. — Diagram of Rotifer to left, of gasteropodous embryo to right, 

 a. p., apical area; 6.p., an.j)., basal area; m, mouth; a., an., anus; d.h , d.g., 

 dorsal brain; v.b., v.g., ventral ganglia; e, eyes; nep., r.t., nephridia; I, liver; t, 

 tentacles; s, shell; s.g., shell gland. Compare also with Fig. 25 a., p. 534. 



some ancient rotifers formed a dorsal, later also a ventral, 

 and still later two lateral plates that fused as the tubular lorica 

 of various genera. Of these four, the dorsal and ventral have 

 been retained and expanded in Brachiopoda, the dorsal may 

 be the cuticular shell, and the two laterals the future calcified 

 shells of Lamellibranchiata, while the dorsal shell alone has 

 persisted in Gasteropoda. Such theoretical points suggest 

 the need for a careful reexamination of lorica formation in 

 rotifers, and particularly a more detailed study of the character 

 and cytological nature of the cells that secrete the lorica. 



As the free-swimming veliger matures and then passes into 

 the creeping or sedentary or free- swimming ^(as in pteropods 



17* 



