228 Dr. W. B. Carpenter on the Structure [Jan. 20, 



somewhat resembling that of the human stomach, having a "large end" 

 into which the funnel-shaped oesophagus opens obliquely, and a " small 

 end" with a caecal termination {an, fig. 10), which is the original intestine. 

 To the middle of this gastric sac, in sections which exhibit it in situ, the 

 " axial prolongation " may be distinctly seen passing up from the base of 

 the calyx. The wall of the gastric sac is everywhere thick, and is chiefly 

 composed of a yellow substance which appears to consist of a layer of 

 columnar epithelial cells, the free surface of which layer is clothed with 

 cilia. 



The further development of the alimentary canal consists, in the first 

 place, in the prolongation of the intestinal extension, of the gastric sac — 

 this prolongation at first taking place horizontally, so that the intestine 

 coils round the stomach in the space left for it by the enlargement of the 

 calyx. Later on, as the second and third radials are developed and the 

 cavity of the calyx opens out, while the circlet of oral valves undergoes 

 very little enlargement, there comes to be a space between the outside of 

 that circlet and the inside of the circlet of incipient arms ; at one point 

 of this space a projection of the perisome shows itself, enclosing a single 

 thin oval calcareous plate ; and towards this projection the csecal termi- 

 nation of the elongating intestine directs itself upwards. At last it 

 reaches the prominence of the oral disk, and opens upon its surface, 

 forming the projecting vent. — The manner in which the axial canal is 

 formed by the gradual limitation of the space left by the coiling of the 

 intestine round the stomach {al, fig. 11) is shown very clearly in this 

 and the succeeding stages ; and I have been fortunate enough to trace 

 the axial prolongation {ax) upwards through this space, from the base of 

 the calyx to its subdivision, by the side of the oesophageal funnel, into 

 five radial branches. 



The extension of the oral ring into the radial and brachial tentacular 

 canals having been minutely described by Prof. Wyville Thomson, I need 

 not go over the same ground ; but I have now to speak of the origin of 

 the subtentacular and coeliac canals. 



From the time when the intestinal coil is formed, and its anal termina- 

 tion opens on the surface of the disk, the growth of the alimentary canal 

 goes on at a greater rate than the enlargement of the calyx ; and the 

 consequence is that the marginal portion of the intestine comes into such 

 close contact with the lining of the calyx, as to divide the coelom into an 

 upper or oral and a lower or dorsal portion* (fig. 12); and this division 

 is completed by the membranous fold, mf. From each portion there 

 passes a sort of digitate extension into every ray, each extension having 

 its own lining membrane. The communication of the lower extension, 



* This has been distinctly recognized by Metschnikoff, who also has pointed out 

 (what had long since occurred to myself) that the relation of the visceral and ten- 

 tacular apparatus to the two divisions of the perivisceral cavity remarkably corresponds 

 with that which prevails in many Polyzoa. 



