208 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



fact, as in the oyster egg, it may be entirely removed from the nucleus 

 and left only in a part of the nucleolus. The methyle green, ou the 

 other hand, does not tend to stain the eggs, but rather the spermatozoa 

 and the cells from which they are derived, and it is one of the most as- 

 tounding facts known to histological chemistry that, although both of 

 these dyes, to begin with, are intimately mixed together in the staining 

 fluid, the different histological elements of the section exert some kind 

 of selective power by which they absorb and hold mainly the one color 

 only. This peculiar property of the two colors, even when mixed to- 

 gether, enables one to distinctly map out the relations of the sexual ele- 

 ments in the reproductive follicles, the nuclei of the ovarian ova being 

 stained red by the safranin, and the heads of the spermatozoa bluish 

 green by the methyl green. The foregoing is maiuly the method to which 

 I have had recourse in working out the sexual characteristics of Ostrea 

 edulis. Simpler staining methods suffice in the case of Ostrea virginica 

 and Ostrea angulata. A single color used in staining sections of 0. edulis 

 is liable to lead to error in consequence of the peculiar mode in which 

 the spermatozoa are packed together in oblong clusters, which are often 

 of about the size of the ovarian ova. This egg-like appearance of the 

 masses of unripe spermatozoa in the follicles of the reproductive organs 

 of the common oyster of Europe misled me when examining sections 

 stained only with eosin or carmine. The monochromatic effect produced 

 by one color only gave no hint as to the real relations of ova and sper- 

 matozoa in the follicles until high powers were used with special manip- 

 ulation of the light. 



The characteristics of the reproductive organs of Ostrea edulis, 0. vir- 

 ginica, and 0. angulata are sufficiently marked to be very precisely de- 

 scribed and figured so as to enable any person to appreciate the differ- 

 ences, especially between the first and last two. 0. edulis is essentially 

 hermaphroditic in the structure of its reproductive organs, while the 

 othej? two are as distinctly monoecious or unisexual. A marked difference 

 is also to be noted in the relative size or caliber of the reproductive fol- 

 licles in the hermaphroditic and in the unisexual species. In 0. edulis 

 the caliber of the generative tubules appears to be relatively much 

 greater than in 0. virginica and 0. angulata, nor are the tubules so 

 densely crowded together as in the latter species. Up to this time this 

 difference appears to me to be so marked that I think it would be pos- 

 sible to distinguish sections of 0. edulis from those of the other two 

 species by means of this one character. In other respects the history 

 of the development of the reproductive tissues in both species appears 

 to be similar. In all the sexual tissue arises as a linear, interstitial 

 differentiation between the coarse, connective-tissue cells of the animal, 

 only that in 0. edulis the rudimentary network does not form quite so 

 close a mesh work as in the other two forms here considered. The tub- 

 ules have a more extensive anastomosis with each other in the unisex- 

 ual species than in the hermaphroditic. In all the forms fine vessels 



