180 Proceedings of the Royal Physical Society. 



An interesting blind diverticulum (Figs. 3, 5, d.v.) from one of the spaces 

 communicating with the central chamber falls to be described. It invaginates 

 the wall of the dorsal sac, and, expanding within this cavity, acquires 

 comparatively thick walls, becomes lined by elongated columnar cells, 

 and contains free in its lumen very numerous corpuscles apparently of 

 leucocytic nature. The walls of the diverticulum are richly infiltrated 

 with leucocytes. 



So far as I know this diverticulum has not been previously described, 

 though Durham's figure 2 of Cribrella oculata (4) shows a somewhat similar 

 cavity. I have seen an indication of its mode of development in young- 

 specimens of Porania and of Solaster papposa. It has some similarities 

 to the ampulla of the stone canal as described by Ludwig, but then 

 Ludwig did not recognise the dorsal sac as quite separate from the 

 axial sinus, and believed that stone canal and axial sinus were shut off 

 from one another. 



From the work of various observers, it may be taken as fairly certain that 

 there is an outward current from some of the pore canals of the rnadreporite. 

 That an important part is played in excretion by wandering cells is well 

 known, and also that they can pass with freedom through the connective 

 tissues and through lining membranes. It was noted above that very many 

 of the intramadreporic collecting trunks contained wandering cells. Probably 

 these have come from the axial sinus, and are waiting their turn to be swept 

 out from the rnadreporite. I would suggest that the diverticulum above 

 described serves as a channel for the escape of wandering cells from the 

 dorsal sac. This would explain the nature of its walls, as well as the presence 

 of so many leucocytes lying free within its lumen. 



2. Position of Anus (Fig. 6). 



In the adult Solaster the anus is very small, and lies near the centre 

 of the aboral surface. Indeed, it was put down as being central by 

 Joh. Miiller. So far as I can find, its position since his time has not 

 been the subject of particular investigation. In the Klassen u. Ordnungcn, 

 and in text -books generally, there is no indication that it occupies 

 any other position than that which is normal for ordinary starfish, namely, 

 a sub-central one in the first interradius to the dextral side of the 

 madreporic interradius. 



I have lately had the opportunity of investigating the question, by three 

 different methods, in a number of specimens both of Solaster endcca and of 

 Solaster (Crossaster) papposa. 



