318 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



of the point where the lowest part of the long gullet turns off to enter the great 

 exterior coil. The walls of this long digestive tube are but slightly plicated. 



Histologically the digestive tube has been studied in greatest detail by Hamann. 



The epithelial lining of the gullet, which passes over uninterruptedly at the 

 mouth and anal openings into the general epithelium of the body wall, and on 

 longitudinal sections through the mouth opening is seen to be continuous with 

 that of the ambulacral furrows, is composed of hair-like cells 0.1 mm. in length, 

 of which the nuclei lie inwardly, toward the middle, or even at the bases. Egg- 

 shaped gland cells 0.1 mm. long, the contents of which remain unstained, lie toward 

 the lumen of the tube, while at the base of the epithelium there runs a nerve 

 fibrillar layer. 



When the cells are isolated it is easily seen that the fibers traversing the nerve 

 layer at right angles are basal processes from the supporting cells. The basal 

 processes from the other cells are much finer, break off easily, and behave like 

 the nerve fibrillae between which they can be followed. 



Below the epithelial layer there is a scarcely perceptible layer of connective 

 tissue, and in the gullet beyond this a well-developed ring muscle layer which is 

 overlaid by the ccelomic epithelium. 



The muscle layer is only recognizable in the commencement of the gullet 

 where the epithelium passes over from the disk into the lumen. 



The nervous layer becomes less and less noticeable as the distance from the 

 mouth increases, and in the hind-gut can not with certainty be demonstrated. 



Excepting within the anal tube, the cells of the whole alimentary tract glisten. 

 There is no cuticular lining. The cilia are attached to the ends of the cells by 

 short foot pieces. 



While about the commencement of the gullet the muscle fibrillse are in several 

 layers and form a more or less well-developed sphincter, about the remaining por- 

 tions of the gut there is only a single layer, which in Antedon is difficult to demon- 

 strate, though in the larger types easily seen. 



In longitudinal section the anal tube is seen to be a cylinder, more or less 

 swollen in the middle. In its formation the body wall, as well as the gut itself, 

 participates, the epithelium of each section passing over into that of the other at 

 the anal opening. 



Within the anal tube the wall of the gut is connected with the body wall 

 through connective tissue cords or septa at regular intervals. These have collec- 

 tively a definite order and structure which is the same in all comatulids. 



The wall of the rectum is composed of the same layers as the rest of the gut, 

 but these are here differently developed. The epithelial lining is provided with 

 cilia only in the lower half. 



While the connective tissue layer increases in thickness distally, the epithelial 

 layer correspondingly decreases, and in the terminal section in no way differs from 

 the epithelium of the body wall into which it passes over at the anal opening, 

 like the latter merging insensibty into the connective tissue layer beneath it. 

 Gland cells are here present in large numbers. If the gut is fully extended longi- 

 tudinal plications or folds are seen to extend from the anal opening about halfway 



