The Development of Phascolosoma. 133 



Protobranchia) possess a highly modified velum. If they are 

 really primitive, the modification of the velum is to be regarded as 

 of cenogeuetic rather than of palingenetic significance, which I believe 

 to be true also of Sipunculus. In view of the differences between 

 Sipunculus and Phascolosoma as regards the prototroch, I would 

 venture to predict that typical trochophores, similar to that of 

 Chiton, may yet be found among the Solenogastres. Sipunculus 

 accordingly is to be regarded neither as a degenerate Mollusc nor 

 as an Annelid, but as a form that is closely allied to the primitive 

 Molluscs and Annelids. 



The trochophore of Phascolosoma resembles in general the typical 

 trochophore of Chiton, Teredo, Patella, Dentalium and other Molluscs, 

 without possessing, so far as I know, any purely molluscan features. 

 On the other hand^ the large prototroch cells of Phascolosoma, and 

 the postoral circlet, are distinctly annelid-like. 



Taking Phascolosoma into account, the Sipunculids resemble in 

 their development the Annelids, particularly the Archiannelida 

 and the Chaetopoda, more than thej^ do the Molluscs, though 

 their relationship to the latter is not distant. 



Comparison with the Vermidea. 



Delage et HÉROUARD (1897), in their excellent text book, ex- 

 press the opinion that the Sipunculids are to be regarded as the 

 point of departure of the entire series of worm-like forms which 

 they designate as the Vermidea, whereas the other branch of 

 the "Gephyria", the Echiurids, lead directly to the Annelids. The 

 Sipunculids, with their crown of tentacles, are connected with the 

 Brachiopods, with the Rotifers, and finally with the PJiahdopleura, 

 Cephalodiscus, Balanoglossus, and the Chor data. 



Lang (1894) includes the Sipunculids y^'ith Phoronis, the Bryozoa, 

 and the Brachiopods, in the class which he calls Prosopygii. 



It is not my purpose in this paper to venture far into the 

 fields of comparative anatomy, or to launch forth upon the treacher- 

 ous waters of phylogenetic speculation, but I wish merely to express 

 such opinions as I have derived from a comparison of the develop- 

 ment of the Sipunculids with that of the other Vermideans. 



In the first place my studies seem to show that the Echiurids 

 and the Sipunculids should be separated, the former being Chae- 

 topods, the latter, primitive Annelids, closely allied to the Chaetopods. 



It cannot be denied that the Sipunculids show marked resem- 



