]^28 John H. Gerould, 



the whole internal organization. The larva of Sipunculus, however, 

 should be compared with a larva of FhascoJosonia immediately after 

 the loss of the prototroch ( cf. Hatschek's fig. 51 with Fig. 54, 55, 95, 

 of this paper). The mouth in the larva of Sipunculus is wide, 

 capacious and nearly terminal, the apical plate thus being crowded 

 slightly backward, whereas in Phascolosoma at this stage the mouth 

 is distinctly ventral, the apical plate is terminal, as is generally the 

 case also in Chaetopods, and there is a well-defined prostomium. 



Comparison with Chaetopods. 



If we compare the trochophore of Phascolosoma with that of 

 Amphitrite, in which the prototroch cells have about the same 

 relative size as in the former, the resemblance between the two 

 forms is most striking. The prototroch in both consists of sixteen 

 primär}^ cells of exactly similar origin; hence they may be regarded 

 as homologous structures. In addition to them there are three 

 secondary prototroch cells in Phascolosoma and nine in Amphitrite, 

 but, according to Mead (1897), there is a wide variation in this 

 respect among the Chaetopods. It is thus evident that, since the 

 pi'imary cells of the prototroch in these forms are homologous, the 

 cilia which cover the entire prototroch in a broad adorai band also 

 may be said to correspond. Difficulties arise when we attempt to 

 make comparisons between the preoral and postoral bands of cilia 

 in Sipunculids and in Chaetopods, because we know so little about 

 the exact relations of the cells which bear them. Without this 

 knowledge, comparisons between circlets of cilia must be merely 

 tentative. 



The trochophore of Ph. vulgare, in respect to the cilia, is probably 

 typical of the Sipunculids. In addition to the adorai cilia of the 

 prototroch cells, it is provided with a postoral circlet of strong cilia 

 almost immediately behind the mouth, and a delicate preoral circlet 

 around the edge of the apical plate. In the latter region only is 

 a prominent circlet developed in Ph. gouldii. In Sipunculus, on the 

 other hand, the postoral circlet is conspicuous, as it is in Ph. vulgare, 

 but according to Hatschek's description no preoral band is present. 

 In the trochophore of another Sipunculid, probably Phymosoma, which 

 I examined at Naples, I found a postoral circlet of exceedingly long 

 and closely-set cilia. These facts appear to justify the conclusion 

 that the Sipunculid trochophores are directly comparable as regards 

 the cilia to mesotrochal forms of Chaetopods, like Psygmobranchus. 



