]^22 John h. Gerould, 



portion of the coelom. This middle part consists of a descending- 

 and an ascending division. The descending part of the coil passes 

 backward in a left-spiral direction around the ascending portion, 

 which it surrounds loosely. 



Large yellow chloragogue cells (Fig. 72—74, 76 d. fiil) are dis- 

 posed in somewhat regular rows along the inner side of the body 

 wall, to which they are attached, each by a small area of its surface, 

 and from which they project into the coelom. Each is filled with 

 yellowish brown granules of various sizes. The function of these 

 cells is probably excretory, the granules being shed into the coelom 

 and taken up by the nephridia, which at this stage have a yellowish 

 or yellowish brown color. I did not observe such cells upon the 

 surface of the intestine, nor did I find any cells of this nature in 

 the larvae of Ph. vulgare, although they very likely occur. The 

 nephridia of the latter species are yellowish in color at this period, 

 which is probably due to the presence of similar excretory granules. 

 The reproductive cells first become evident in specimens between 

 two and three weeks old (19 hours, Fig. 78) as a group of com- 

 paratively large cells situated upon the posterior extremity of each 

 of the ventral retractor muscles, and projecting into the coelom. 



The chief further modification, which the oldest larvae (of 

 14—30 days) undergo, is the completer separation of the ventral 

 nerve cord from the body wall and the development of the lateral 

 nerves. These are first developed, and are always most in evidence, 

 at the anterior end of the body. There are no indications of 

 ganglionic enlargements of the nerve cord, nor are the lateral nerves 

 in the larvae given off at regular intervals. Local currents in the 

 coelomic fluid, due to isolated ciliated cells, were observed in the 

 oldest larvae. 



The rudiments of epidermal organs, which I have described in 

 connection with the youngest larvae, increase greatly in number 

 during the larval development. Those which have the cuticula 

 above them raised into papillae are still found scattered over the 

 entire body, though they are more abundant and larger at the 

 posterior end. In the latter region in Ph. vulgare the papillae are 

 so large and prominent, that they are plainly seen even in the 

 living larvae (Fig. 56, 57). In the larvae of Ph. gouldii I have seen 

 the papillae only in whole preparations and in sections (Fig. 78, 79). 

 In larvae of nineteen days two or three slender bipolar nerve cells 

 can be seen in the midst of a group of undifferentiated supporting 



