138 John H. Gerould, 



If, on the other hand, we accept the view that the Sipuiiculids 

 are degenerate Annelids, how extraordinary it is that they should 

 have lost to such an extent the metamerism of their ancestors! It 

 seems more probable to me that the paired lateral bristles of 

 Selenka's larva of Ph. elongatmn, and the indications of internal 

 metamerism which I have found in Ph. goiddii, are an expression 

 of an incipient, rather than of an atavistic tendency. I have been 

 unable to find any traces of segmentation in the trochophore of 

 Ph. vulgare, and it is by no means as prominent in Ph. gouldii as 

 it is, for example, in the larva of EcMurus. The transitory meta- 

 merism in Ph. gouldii possibly may be due to the compression of 

 the growing trunk within the j^olk membrane. When the latter is 

 shed, and the pressure upon the mesoblastic bands and the nerve 

 cord is relieved, segmentation disappears. 



(2) The answer to the question, whether the Sipunculids are 

 to be classed with Annelids or as an independent phjdum of the 

 Trochozoa, is to be found in the presence or absence of metameric 

 segmentation. The only indications of segmentation in the group^ 

 so far as I know, are the three pairs of lateral bristles of Selenka's 

 larva, and the transitory metamerism in the rudiment of the nerve 

 cord, and in the mesoblast bands in Ph. goiddii. On the basis of 

 these evidences of segmentation, we may include the Sipunculids 

 among the Annelids. 



I have already shown that they are pi'obably not degenerate 

 Chaetopods, but are more primitive and trochophore-like in their 

 adult structure than the Chaetopods, or even than Polygordms and 

 other Archiannelida, which they resemble in the simple structure 

 of the ventral nerve cord. Thus I hold the opinion that the Sipun- 

 culids form an oifshoot of the annelid stock, that stands nearer the 

 ancestral coelenterate ancestor than do the Archiannelida,^) and 

 near the molluscan branch. 



1) In my preliminary note (1904) I stated the conclusion that the 

 "Sipunculids are to be regarded as forms that have recently sprung from 

 the ancestral Trochozoon". In using the word "recently" I had no in- 

 tention of referring to the length of time that has elapsed since their 

 differentiation from the hypothetical ancestor of the Trochozoa, but had 

 in mind merely the small amount of modification from the ancestral form 

 which they have experienced. It was an unfortunate expression , since 

 this slight differentiation may have been accomplished during a relatively 

 remote period. When it occurred could, of course, be determined only by 

 the discovery of fossii remains. 



