130 PROCEEDINGS OF THE ACADEMY OF [Feb., 



ward a short distance under the hinder portion of the stomach. Weldon 

 describes no peritoneal investment of the sex glands, it is true, but 

 this might easily have been overlooked, inasmuch as it is probably 

 very thin. Indeed, in the anterior portion of the testes of D. meta- 

 meroides, Schimkewitsch (1895) states that the peritoneum becomes 

 reduced to a mere memhrana propria. For this reason, and because 

 of the occurrence of a peritoneum enclosing the sex products in at 

 least two other species, it is not unwarrantable, to say the least, to 

 assume its presence in D. gigas. The condition obtaining in this species 

 would then be almost precisely intermediate between that found in 

 D. vorticoides and tccniatus, on the one hand, and D. conklini (and prob- 

 ably also apatris and gyrociliatus) on the other. In other words, there 

 has been within the Dinophilidce a progressive reduction in the shape 

 and extent of the sex glands, and also of the coelom, represented by the 

 cavity within them; this reduction beginning at their anterior ends, and 

 its stages exemplified by the three species named. These are repre- 

 sented diagrammatically by text figs, V, A, B, and C. In the first 

 stage, (A), represented by D. vorticoides and tceniatus, the sex glands 

 are paired, extend throughout the greater portion of the trunk, and 

 are united only at their posterior ends. In the second, (B), repre- 

 sented by D. gigas, the paired portions are much shortened, while 

 the unpaired median part, which joins the former, has increased some- 

 what in length. In the third and last stage, represented by D. conklini, 

 the lateral paired portions have disappeared altogether, leaving only 

 the posterior unpaired part. That the paired condition, represented 

 in A, is probably the more primitive one need scarcely be pointed out ; 

 the unpaired is therefore specialized. It is also to be noted in this 

 connection that those species possessing the unpaired sex gland are 

 also sexually dimorphic. 



In D. vorticoides, tceniatus, and gigas the reproductive cells appear 

 to be produced directly from the wall of the sex glands (coelomic 

 epithelium) by transformation of its cells. Moreover, the production 

 of germ cells appears not to be definitely localized, but to take place 

 throughout the length of the glands. The condition in D. conklini is 

 evidently quite different. Here a circumscribed portion of the coelomic 

 epithelium is, as in the higher Vermes, differentiated to form a well- 

 defined gonad, in the shape of the pyriform mass of oogonia. 



Korschelt (1882) was of the opinion that the germ cells arose in the 

 walls of the alimentary canal. Yon Malsen (1906) has recently thought 

 to have found confirmation of this view and has described and figured 

 the germ cells as arising in the ventral wall of the stomach, and migrat- 



