MUSEUM OF COMPAKATIVE ZOOLOGY. 397 



obscured, Biitschli is unable to say, but evidently inclines to the latter 

 opinion. 



ScHENK ('73, p. 369, Fig". 4) points out the existence of a small cavity 

 in the fecundated eggs of Raja quadrimaculata, which has a triangular 

 outline and opens by a narrow orilice at the surface of the formative 

 yolk. It occupies the place of the germinative vesicle. 



ViLLOT ('74, pp. 201, 202) informs us that the germinative vesicle 

 has apparently disappeared in the eggs of Gordius at the time of 

 deposit, but the subsequent contractions of the vitellus bring into view 

 a nuclear structure, which he insists is the original vesicle. In his 

 figures (PI. VII.) he represents only a single polar globule, but says in 

 the text that segmentation is preceded and accompanied (!), as in most 

 animals, by the formation of polar globules, the number, form, and 

 volume of which are variable. 



In his first studies on Anodonta, Flemming ('74, pp. 274 - 279, Taf. 

 XVI. Figs. 10, 11, 16) traced the appearance of polar globules, but was 

 in doubt as to whether the second of the two bodies resulted from a di- 

 vision of the first one, or was separately eliminated from the yolk. The 

 globules are expelled from the yolk at the pole diametrically opposite the 

 micropyle, and the process is introduced by the appearance of a hyaline 

 margin which in some cases is raised to a knoblike form. This is fol- 

 lowed by the pushing out of a rodlike projection having a conical apex. 



The formation of the first polar globule is of considerable interest, 

 inasmuch as its production is accompanied by the appearance of short 

 pseudopoda-like projections about the apex of the conical mass. This 

 observation stands without a parallel. The projections, although resem- 

 bling pseudopodia, were never observed to execute rapid motions. 



A layer of granules in the middle or under the apex of the projecting 

 body is doubtless to be referred to the elements of the external half 

 of the nuclear plate. (Compare sjol, Figs. 50, 40, and 67 of Limax.) 

 The yolk was observed to change its form periodically, during this pro- 

 cess of elimination, from a spherical to a more flattened condition, and 

 back again. At this time none of the yolks possessed a nucleus, but in- 

 stead a " clear place " was to be seen (especially if the egg were subjected 

 to pressure sufficient to flatten it) which was not sharply limited from 

 the rest of the vitellus ; it simply contained fewer and smaller granules 

 than the surrounding yolk ; it lay somewhat eccentric, nearer the pole 

 where the globules were eliminated. 



The polar globules persist only a short time (till the fourth segmenta- 

 tion) ; they stain more intensely than the yolk, and in this Flemming 



