m 



dorsally in pairs until a maximum of about seventeen pairs may- 

 be formed. But metamorphosis may take place when the 

 tentacles number twelve pairs. The variation in this respect 

 may be said to be due to a variation in the length of the planktonic 

 period. The tentacular ring is thus incomplete dorsally as it is 

 in the adult. It is evidently homologous with the postoral cihated 

 band of many invertebrate larvae. During development a median 

 conical cihated papilla appears in front of the apical plate, and 

 this may be held to represent the pre -oral cihated band. The 

 anal band of cilia is borne on immensely thickened ectodermal 

 cells, and has the same disposition as that of the Tornaria larva. 

 The larva measures when fully grown about 1-5 to 2-5 mm. 



As is well known, the metamorphosis and the preparation 

 for it are characteristic and unique. Adult tentacles appear as 

 rudiments posteriorly to the larval tentacles, and on the mid 

 ventral surface just behind the adult tentacles a small patch of 

 the ectoderm is folded inwards, traversing the ventral mesentery 

 of the body cavity, and forms a winding tube within the body. 

 When this is completed the larva has reached the bottom as a 

 rule. Metamorphosis takes place through the evagination of 

 this tube. It is rolled out as will be seen from Fig. 2 in such a way 

 that the blind end of the tube is the last part to appear. It is not 

 until the blind end of the tube has left the larval body that the 

 alimentary canal of the larva is pulled into it, the larva at the same 

 time suffering contraction. At this period of the changes the 

 hood is withdrawn or may be sw^allowed hterally together with the 

 larval tentacles, and as a last effort the anal ring is pulled in, the 

 cells either before or after sinking to a size more hke those on 

 either side. All this brings the tentacles and the oral field into 

 the horseshoe shape so characteristic of the adult. Much varia- 

 tion accompanies the metamorphosis, especially as to the fate of 

 the tentacles. They may be cast off before the tube begins to 

 appear, or they may be retained until the process is completed. 

 When the tentacles are cast off they are sw^allowed, for their 

 remains are to be seen in sections of the stomach, as de Selys 

 Longchamps has aheady pointed out. The whole process only 

 lasts about a quarter of an hour. 



I have had, thanks to Professor Herdman, an opportunity 

 of examining the specimen of Actinotrocha which he captured at 



