ON THE DEVELOPMENT OF LINGULA ANATINA. 67 



partition ; in the left of these compartments about twenty otoliths 

 were present, while in the right there were only three. 



By a study of the toto preparation of a larva at the 5 p. c. 

 stage (PI. VI., Fig. 81) we see that the walls of the otocyst 

 are composed of a small number of cubical or slightly flattened 

 cells. If we trace posteriorly a series of transverse sections of a 

 larva of this stage, the anterior edge of the otocysts is found in 

 the angles formed by the lateral and dorsal body walls in the 

 same section in which the nephridia appear, (PI. VII., Fig. 92, 

 and Fig. 108, which is a portion of the section next to Fig. 92, 

 drawn under a higher magnification). In the latter figure we see 

 that the otocyst is composed of a layer of epithelial cells. As 

 far as the origin of the otocyst is concerned, I think, we can 

 assume that it arises as an invagination of the lateral epidermis 

 of the body wall, and attains its definite position by pushing 

 aside the peritoneum. It is certain that the cells composing it 

 show no difference from the neighboring epidermal cells. 



By the 7-9 p. c. stage (PL VII., Fig. 94) the otocyst has 

 increased in size. Its walls are now composed of a number of 

 cubical epithelial cells, whose nuclei stain intensely. As the body 

 walls bulge out, the vesicles appear to be shifted toward the 

 median plane, but in reality they remain in their original position, 

 that is in the angles of the body walls, as we can see in transverse 

 sections. The dorsal wall of the otocysts is, as a rule, very thin, 

 and here one is apt to mistake an artefact for a normal openiug. 

 I should note that I have not observed otoliths in fixed prepara- 

 tions nor in sections. Of the chemical composition of the otoliths, 

 it must be confessed that little is definitely known. 



It has now been shown that the vesicles in question fulfill 

 all the characters of otocysts, and on the other hand we may 



