466 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



At first the vestibule extends over the entire ventral side, and encroaches 

 slightly on the posterior end. Later it migrates posteriorly until it leaves the 

 ventral side and becomes confined to the posterior end, which it entirely covers. 



When the anterior portion of the vestibule is still open the various walls of 

 the posterior closed part are seen to be very differently constituted. The whole 

 basal wall is extraordinarily thickened and lies close down upon the water vessel 

 so that the tentacular evaginations force themselves into it. The side of the 

 A^estibular wall turned toward the lumen is entirely smooth, and of the tentacles 

 there is nothing to be seen. 



Through the center of the as yet unclosed hydroccele ring a solid process runs 

 toward the gut to which it is attached. This represents the place on which in the 

 free-swimming larva a peglike process is formed on the primitive gut. 



As a rule after the closure of the vestibule the ectodermal portion can be 

 differentiated histologically from the entodermal. 



In the anterior part of the basal wall the nuclei lie almost exclusively very 

 close to the body cavity ; posteriorly they become more numerous, and not only are 

 not restricted to the innermost portion of the cells, but occur as far as the middle, 

 and occasionally in the outer halves. 



Seeliger was not able to demonstrate cell borders. Toward the posterior end 

 the basal wall gives the impression of a multinuclear plastic plasma mass, and 

 very often here a lighter middle zone can be differentiated between two darker 

 bordering zones. While in the latter the plasma usually appears granulated, in 

 the clearer portion the general impression is that of a fibrillar structure. 



The zones show individual variation in extent as well as in position, and in 

 many cases are quite lacking. In that event the whole base is composed of a 

 uniform finely granulated plasma. Since sections of the ciliated bands as well 

 as of the intermediate areas are involved in the formation of the vestibule, it is 

 perhaps possible that the differences are traceable to this. 



In addition to the nuclei, which are quite typical and generally poor in 

 chromatin, there are very numerous (in the light zone much less frequent) 

 smaller rounded bodies which stain very intensively in hsematoxylin and retain 

 the stain as tenaciously as the chromatin of the nuclei. Burj' believed these to 

 be phagocytes. Seeliger at first considered them chromatin granules from 

 degenerated nuclei, but later supposed them to be bodies arising in the plasma 

 which were later consumed. 



In the hinder part the lumen of the closed vestibule is crescentic in section, 

 with the concave side turned inwardly; in the anterior portion it is elliptical, 

 with the longer axis transverse. 



In the cuticular layer bordering the vestibular cavity a radial striping can 

 no longer be demonstrated. 



From the floor of the vestibule outward the walls decrease rapidly in thick- 

 ness, and in the median plane ventrally the lumen is separated from the exterior 

 only by a very fine layer of cells. This ventral wall is of only a single layer; in 

 the mid-zone it is composed of cubical and flat pavement cells, which laterally 

 become gradually higher while at the same time their borders become indistinct. 



