424 BULLETIN 82, UNITED STATES NATIONAL MUSEUM, 



columnar cells, here consists of a layer of very minute cells, such as line the whole 

 of the canal at an earlier date, and identical with the cells covering the general 

 surface of the disk. In thickness this inner layer is very irregular, and the cells 

 composing it appear to be rapidly proliferating. At its extreme end they block up 

 the lumen of the canal so that the latter appears solid and the end of the canal 

 fuses indistinguishably with the layer of similar cells covering the general surface 

 of the disk. At the point of fusion there is a thick mass of these minute round 

 cells forming a little elevation on the surface of the disk. This indicates the posi- 

 tion of the future anal cone. Dendy was unable to find a distinct anal opening in 

 the specimen. He believes that the facts support the view that the alimentar}' 

 canal is formed by invagination from the oral end. 



At 12 days no great advance has been made on the condition presented by a 

 nine days' specimen. In one specimen the pentagon around the mouth, formed by 

 the lips of the ambulacral grooves, has become notched into lappets all the way 

 around, and in sections the alimentary canal is seen to have become very consider- 

 ably complicated and folded upon itself. In sections of another specimen of this 

 age the ambulacral epithelium and canals can be traced across the disk very nearly 

 to the lip of the mouth ; they present the same features as in the normal specimens ; 

 thus tlie canals exhibit the transverse muscle fibers found in ordinary specimens. 

 The anal cone is still very small. 



At 19 days little advance is to be noted except in the further growth of the 

 anal cone, which is now fairly well developed. Sections show a large body cavity 

 in the anal cone, surrounding the terminal portion of the intestine, which latter 

 is attached to the body wall by strands of connective tissue. 



In sections of a specimen of 21 days a blind diverticulum is visible, given 

 off from the alimentary canal at the junction of the stomach with the esophagus, 

 as described by Ludwig in the normal animal. There is little to distinguish a 

 regenerated specimen of this date from a normal adult excepting the smaller size 

 of the visceral mass and the want of pigment upon it. 



The 43-hour stage is described from specimens of Antedon bifida taken at 

 Millport on the Clyde, the other stages from specimens of A. mediterranea from 

 Naples. 



Dendy suggests that, as the crinoids can have no possible power of food selec- 

 tion, the facility of regeneration of the visceral mass may be an acquired provision 

 for the rejection of poisonous or obnoxious food which has found its way into the 

 alimentary canal. 



DETAILS OP ABM EKaBNEEATION. 



The histological details of arm regeneration have been studied by Perrier, 

 and more recently by Eeichensperger. The former in 1872 recorded the outward 

 appearances of regenerated parts, and the inner structure, so far as it could be 

 made out in stained and cleared specimens, while the latter worked along the 

 most modern lines. 



Closure of the wound and degeneration of the cell elements. — Immediately after 

 the fracture a certain amount of fluid escapes from the water vessel and from the 



