MUSEUM OF COMPARATIVE ZOOLOGY. 573 



author, however, doubts the genetic connection, since these granules are often 

 entirely wanting. Subsequently the amphiaster is completed by the welding 

 of the intranuclear rays end to end, and the "granules de Biitschli" make 

 their appearance as enlargements of the bipolar filaments. But the relation of 

 these enlargements to the grains presented by the still isolated rays remained 

 obscure. The amphiaster elongates, and at the same time stretches the mem- 

 brane of the vesicle. The vitelline rays have increased in extent, and the 

 centre of each aster is occupied by a few granulations, around which is a space 

 occupied by homogeneous protoplasm. Meanwhile the membrane of the ger- 

 minative vesicle assumes indefinite contours and entirely disappears. The 

 amphiaster moves toward the periphery; at first oblique, it becomes perpen- 

 dicular to the surface, with which the centre of one of the asters becomes 

 almost "flush." Then the surface is raised into a dome, and the enlarge- 

 ments of the bipolar rays divide ; the first polar globule, composed of half the 

 "amphiaster de rebut," is detached. The internal half undergoes the same 

 modifications as in Asterias, but the second amphiaster is smaller than the 

 first. Portions of the bipolar filaments and their enlargements are readily dis- 

 tinguished at, and some time after, the formation of the globule. The enlarge- 

 ments all lie at the same height ; at the time of segmentation the polar globule 

 has assumed the appearance of a cell with a large nucleus, and one or several 

 nucleoli. They decompose, and have no part in the development of the egg. 

 The views of the author, it will be observed, seem to have been modified in 

 some particulars since the publication of his earlier paper on Heteropoda. See 

 pp. 429, 430. 



The principal events of fecundation as described for Asterias have already (pp. 

 480, 486) been given. It is necessary to add only a few particulars. The " cone 

 of attraction" may extend to half the thickness of the mucilaginous layer 

 if the spermatozoon advances slowly, but is much shorter and more rounded 

 when it approaches quickly, for as soon as the contact between the two is 

 effected, the cone commences to retract. Most spermatozoa enter the nutritive 

 hemisphere, but one often sees a penetration in the formative half, even up to 

 the immediate vicinity of the polar globules. At the moment when a space 

 appears ufider the vitelline membrane around the point of fecundation, the 

 differentiation, but not the elevation, of the membrane has extended quite 

 around the vitellus. From this instant the egg is inaccessible to every sper- 

 matozoon which reaches the membrane ; for the vitellus is no longer able to 

 produce a " cone d'attraction," and in Asterias a spermatozoon is hardly 

 capable of penetrating without the aid of this excrescence. The space embraced 

 between the elevated membrane and the yolk is occupied by a transparent sub- 

 stance, which cannot be a liquid, but must be a very clear jelly, since, if it were 

 a liquid, the vitellus would change position, and the space could not remain of 

 uniform thickness all around. Does this substance arise exclusively as a secre- 

 tion from the yolk, or is there at the same time an imbibition through the 

 vitelline membrane ? If the former, the vitellus should suffer a diminution of 

 volume. It is difficult to determine whether this is so, on account of the 



