NOTES ON THE ANATOMY OP DINOPHILUS. 185 



of the same side at a mucli earlier stage of development^ at a period, 

 indeed, wlien tlie entire ovary is composed of a mass of small, uni- 

 form, primordial ova. The relations of the investment of the ovaries 

 are further explained by this figure, in which it is seen that the 

 space between the anterior and posterior lobes is, as in the later 

 stage, devoid of any epithelium on its dorsal side. Ventrally, the 

 space is floored by a single layer of cells, sepai'ated from the skiu 

 by loose connective tissue ; the space itself contains (as was occa- 

 sionally observed in older stages) a few free cells of unknown 

 function. 



In the absence of any developmental evidence it is not easy to 

 say what is the nature of the interovarian cavity. From the analogy 

 of the male, as well as from a consideration of the general arrange- 

 ment of the ovaries, it would appear that the ovaries are primitively 

 paired bodies, and not merely lateral thickenings of a median cavity. 

 The interovarian cavity would thus be a specialised portion of the 

 general body-cavity, which conclusion would be supported by the 

 absence of any proper wall, the space being bounded partly by the 

 investment of the ovaries and partly by the wall of the stomach. 

 The conclusion is further strengthened by distinct evidence obtained 

 from sections, that the internal ends of the fourth nephridia project 

 into the space. 



In most females observed in section there was found to be a mass 

 of spermatozoa at the sides of the stomach and dorsal to the ovaries, 

 these masses of spermatozoa usually passing continuously into the 

 large central mass which is nearly always present in the inter- 

 ovarian cavity. The spaces in which these lateral masses of sper- 

 matozoa lie appear to be parts of the general body-cavity, which is 

 hence continuous with the interovarian cavity at those points where 

 the spermatozoa enter the latter. This continuity does not neces- 

 sarily prove that the ventral space is really part of the body-cavity, 

 as, from the method in which the spermatozoa are introduced into 

 the female, they must probably often have to make their way 

 through various obstructions in order to reach the ventral space. 



The layer of cells connecting the two ovaries (figs. 12 and 14) 

 across the middle ventral line of the body may thus be provisionally 

 interpreted as resulting from the median fusion of two originally 

 separate organs, and this process probably takes place at an early 

 stage of development, as in the case of the testes o£ the male. 



The interovarian cavity extends along the middle line of the 

 body throughout the whole of the region of the stomach, and there- 

 fore occurs, not only between the ovaries themselves, but also behind 

 and in front of the ovaries, which are lateral thickenings of the 

 walls of the cavity, projecting into it. In consequence of this pro- 



