54 NOTES ON THE ANATOMY OF DINOPHIHJS. 



jectiorij the posterior part of the cavity in fig. 12 is separated (in 

 the particular section in question) from that part which occurs 

 between the anterior and posterior lobes ; the posterior part of the 

 cavity is of course continuous with the anterior part. It will be 

 noticed from fig. 12 that the posterior part of the interovarian 

 cavity has an epithelial wall on its dorsal side as well as on its 

 ventral side, and the same is true of the anterior end of the cavity 

 (not involved by the section shown in fig. 12). The complete con- 

 version of the interovarian cavity into a tube which runs backwards 

 below the intestine takes place at the level of the posterior ovarian 

 lobes, and appears to be due to the fusion across the middle 

 line of the investments of the ovaries of opposite sides. The 

 tube thus formed runs backwards, becoming much smaller as it 

 approaches the end of the body. In one specimen examined, the 

 tube was distinguishable almost as far back as the anus, although 

 very minute in the hinder part of its course. 



In fig. 14, the eggs which are cut by the section are still outside 

 the interovarian cavity. Most of the large eggs in this individual 

 possessed two nuclei, as shown in one of those figured. They were 

 further provided with a somewhat shrivelled membrane, which is 

 probably the vitelline membrane. In the fresh condition, the only 

 case noticed in which the vitelline membrane was acquired before 

 the eggs reached the exterior was in a dead female, most of the 

 tissues of which were beginning to break up into fragments. 



In other sections of the series from which fig. 14 is taken, eggs 

 are found in the interovarian space. The posterior, tubular con- 

 tinuation of this space may probably be regarded as an oviduct, 

 although the process of egg-layiug was not directly observed. It 

 does not appear to me probable that the eggs are liberated by the 

 death of the female, as Weldon (13) supposes to be the casein D. gigas. 



In D. vorticoides (vanBeneden, No. 1) and in the species described 

 by Korschelt (-6) as D.apatris (probably identical with D.gyrocillatus), 

 the eggs are known to pass to the exterior by means of a minute 

 pore situated on the ventral side of the animal, at the base of the 

 tail. This pore is said not to be recognisable except when the eggs 

 are being laid; the eggs completely lose their shape in passing 

 through the aperture, but regain their spherical form on arriving in 

 the water. 



In Protodrilus, an animal to which Bmophilus is probably allied, 

 the eggs are said by Uljanin and Eepiachoff {v. Eepiachoff, No. 12^, 

 p. 29) to escape from the body in the same way as in the above-men- 

 tioned species of Dinophilus. According to the observations of Uljanin, 

 quoted and confirmed by Eepiachoff, the ripe eggs of Protodrilus 

 move about freely in the meshes of the network of connective tissue 



