ON THE EMBRYOLOGY OF THE ECTOPROCTA. 11 



Entoprocta, is more probably the equivalent of the dorsal thicken- 

 ing of epiblast connected with the formation of the brain in 

 Alcyonidium. The cells e., identified by Repiachoff as hypoblastic 

 in nature, are perhaps part of the brain-tissue itself. 



In fig. 24 (Cyphonautes) of the plate just referred to, I 

 reproduced one of Repiachoff's figures which was in some respects 

 wrongly interpreted, owing to my previous inability to read Repia- 

 chotF's description. I now find that the structure z. c. is merely 

 the anterior part of the ciliated band, that x. is the pyriform 

 organ and that e. is the supposed endoderm-bud. This latter 

 structure is not said by Repiachoff to give rise to the first poly- 

 pide, as I formerly supposed. 



It has recently been proved by Ostroumoff (12) that the organ 

 shown in Repiachoff's figure just in front of the rectum ("I'organe 

 enigmatique de Schneider") is in reality the sucker or internal sac, 

 by which fixation is effected. The structure of Cyphonautes is 

 hence more similar to that of other Ectoproct-l&rvie than was 

 formerly imagined to be the case. 



It is difiicult to assert at present that the cement-gland of the 

 Entoprocta is represented in the Ectoprocta. Further investi- 

 gations are needed to show whether the calotte of the latter 

 is to be regarded as the homologue of the cement-gland, of 

 the dorsal organ or of any other structure possessed by the larvae 

 of the Entoprocta. 



The alimentary canal of the larvae of the Ectoprocta appears to 

 be functional in Cyphonautes alone (i.e. the larva of Memhranipcra). 

 It has, however, been shown to be present, more or less well de- 

 veloped, in Tendra (Repiachoff, 15), some Cyclostomata (Ostroumoff, 

 13), whilst I believe it to occur in Flustrella. Barrois (4) has more- 

 over suggested that the mouth is really shown as the depression ce 

 in PI. VII, fig. 13 and elsewhere in his large memoir (1), so that 

 there is a considerable amount of evidence in favour of the view 

 that the Ectoproct-larv^ were formerly provided with a distinct 

 alimentary canal. Even in those eases where a digestive tube 

 with a complete ventral curvature does not exist, the hypoblast is 

 formed in the embryo as in other Polyzoa with a better developed 

 alimentary can.al, but afterwards assumes the form of a mass of 



