Miscellaneous. 403 



The parablast extends beneath the germ, and forms the floor of 

 the germinal cavity : it is more abundant at the periphery than in 

 the central region ; so that it forms a sort of cupule in which the 

 germ is enshrined. There is a canal with a triangular section sur- 

 rounding the germ and included between the corneous lamina, the 

 parablast, and the point of inflection of the sensorial layer. 



When the embryonal shield begins to appear, the blastoderm is 

 thicker at this level than in the rest of its extent, and the reflected 

 portion advances further into the germinal cavity than that of the 

 opposite side. 



In the fresh state, in the ovum of the perch, thanks to its ex- 

 treme transparency, I have been able to see the reflection of the 

 margins of the blastoderm ; and I was easily able to ascertain, by 

 slightly compressing the ovum, the presence of the fissure which 

 separates the sensorial lamina from its reflected part. 



In trout-ova of which the blastoderm had covered rather more than 

 half the vitelline globe, I have found, at the posterior part of the em- 

 bryo, beneath the point at which the dorsal cord stops, a small vesicle 

 lined with cylindrical cells. This vesicle, by its position, its form, 

 and the constitution of its walls, appears to me to be identical with 

 that described by Kupffer in the stickleback under the name of 

 allantoid. In this last fish, in which I have been able to verify its 

 existence, Kupfier's vesicle projects into the interior of the vitel- 

 lus, and has the form of a hemispherical cap, the convexity of 

 which is turned towards the vitellus, whilst its floor looks towards 

 the ventral surface of the embryo. In the trout the vesicle does not 

 project into the vitellus ; and although it presents the same form, 

 its convexity is fixed in the embryo, and it rests by its flat part upon 

 the parablast. 



Hitherto I have been unable to ascertain the presence of a canal 

 placing the vesicle in communication with the exterior, either in 

 transverse or in longitudinal sections. This vesicle has only a tem- 

 porary existence ; I have been unable to detect it in more advanced 

 embryos. 



By making sections of ova of which the blastoderm had just closed 

 behind the posterior extremity of the embryo, I have been able to 

 see at this point a canal placing the surface of the vitelline globe 

 in communication with the dorsal surface of the embryo. This 

 canal therefore traverses the posterior extremity of the embryo ; for 

 the blastodermic pad (bourrelet), as demonstrated by His, has just 

 soldered itself to the embryo to become subsequently the extremity 

 of the tail ; it is completely independent of Kupffers vesicle, which 

 has long since disappeared. 



In the perch, as Lereboullet was the first to observe, the embryo 

 forms slowly ; it does not appear until the blastoderm has almost 

 entirely covered the vitelline globe. The blastodermic pad, corre- 

 sponding to the reflected part of the margins of the germ, however, 

 some time before the closure of the blastoderm, presents a widened 

 part at the spot where the embryo will be formed. When the 

 blastoderm closes, there remains for some time at the posterior part 



