ON THE EMBRYOLOGY OF THE ECTOPROCTA. 5 



embedded in a small quantity of protoplasm, with a few nuclei at 

 intervals, or it may have the form of a very thin protoplasmic 

 layer, in which nuclei are sparingly developed ; but it is in any 

 case extremely unlike an ordinary secreting epithelium, and this 

 taken in conjunction with the facts (1) that the lumen of the 

 stomach becomes progressively smaller as development proceeds, 

 and (2) that there is probably, in the later stages, no communica- 

 tion between the stomach and the exterior, leads me to the belief 

 that in Alcyonidium the alimentary canal is a rudimentary 

 structure. A reference to fig. 3 will show that food could hardly, 

 by any possibility, pass through the oesophagus at this stage of the 

 development. Owing to the large supply of food-yolk in the eggs, 

 to the fact that development proceeds within the body-wall of the 

 adult, to the extremely short free larval life and to the degenera- 

 tion of many of the embryonic organs during the metamorphosis, 

 the alimentary canal is no longer required in its functional form. 



The mouth {m.) is unmistakeable in fig. 2 ; the oesophagus, 

 whose walls contain large numbers of yolk-spheres, has, however, 

 no obvious lumen except near its junction with the stomach ; the 

 supposed anal region is now provided with a few cilia. The 

 sucker is loaded with less yolk than in the preceding stages, 

 although a few spheres still remain in its walls. A deep groove 

 (m.c), running round the aboral region of the embryo, has 

 appeared on the dorsal side of the ciliated ring, with which it 

 is concentric. This groove is already distinguishable in fig. 1, and 

 is the structure which has been described by Barrois and others as 

 the mantle-cavity. Its function is probably to render possible the 

 revolution of the ciliated ring into the interior of the vestibule 

 which is formed on the ventral side of the larva during the process 

 of fixation. 



The changes which subsequently take place in the alimentary 

 canal, sucker and mantle-cavity may be understood by referring to 

 fig. 3. The oesophagus is, at this stage, somewhat difiicult to 

 distinguish at the middle part of its course, whilst the stomach has 

 thicker walls and a less conspicuous cavity than before. 



The sucker has a very small lumen, and is further characterized 

 by the almost complete absence of yolk from its long, columnar 



