Studies on the Development of Larvai Nephridia. 507 



gut Willi. I have shown tbat in the larvae I have studied the first 

 rudinieuts of the eoelom occur much later, aad are as far as I can 

 determine nnpaired; this by no meaus precludes the possibility of 

 their paired nature at an earlier date thoug-h I have never observed 

 any indications of this. 



Ikeda (ll) whose careful work has done so much towards 

 eluddating difficult points in the developnient of Pkoronis unfor- 

 tunately passes over the orig-iu and early history of this cavity, and 

 only treats of it when it is already well formed, passing from a stage 

 in which there is no septum in the trunk region to one in which 

 this septum is well advanced. Cowles (6) states tbat in larvae of 

 Ph. architecta, he has uoticed an arrangement of mesoderma! cells 

 on the dorsal side of the gut which may be the beginning of the 

 trunk eoelom. He says in larvae ''with two tentacles I have found 

 an arrangement of mesoderma! cells on the dorsal side of the in- 

 testine which seems to be the beginning of a sac; this however is 

 not paired. Whether or not this sac and its cavity give rise to the 

 lining and cavity of the trunk, I cannot say for I have found but 

 a single specimen in which this condition exists". 



The views of De Selys LongchAxMps on the nature of the trunk 

 cavity I have already discussed. I should however like to point 

 out again, that if the body cavity of the trunk is in free communi- 

 cation with tbat of the coUar region (the haemocoel) how is it that 

 only the posterior cavity develops into a real eoelom, and the 

 other forms haemocoel? Both these cavities are according to his 

 account formed from the original space of the blastocoel, yet why 

 if they are all one, do the mesoderma! cells of one form a coelo- 

 mic lining, and tlie same cells in the other form blood vessels? 

 I have tried to show that this is a mistake, that these two cavities 

 are essentially different; from the first the trunk eoelom in normal 

 larvae is completely closed. By the growth of this closed sac in a 

 forward direction the septum of the trunk-collar region is formed. 



5. The development of the nephridia in late stages. 



I have little to add to the account given by Goodrich (11) of 

 the structure of the nephridia in the Actinotrocha larva; this account 

 has been confirmed in all its essential details by De Selys 

 L0NGCHAMP8 (25) and Cowles (6). The shorteniug of the nephri- 

 dial canals in the larva, once it has begun to assume the 

 form of the Actinotrocha, is very rapid, and is caused by the 



MittheilQDgen a. d. Zool. Station zu Neapel. Bd. 17. 34 



