MUSEUM OF COMPARATIVE ZOOLOGY. 567 



of them are distinguishable. When the germinative vesicle is about to dis- 

 appear, the network is no longer to be found, even after employing reagents. 

 This is proof that it is not an artificial product. The germinative dot is highly 

 refringent, embraces one or several vacuoles, but no granulations, and is not 

 surrounded by any layer different from the rest of its substance. 



It will be unnecessary to give the details of the metamorphosis of the ger- 

 minative vesicle and dot, which have been briefly reviewed elsewhere (p. 436). 

 I will only mention some points of particular interest. 



With the flattening of the space (lacuna) which represents the disappearing 

 vesicle, the substance of the dot changes form, and becomes pale. In some cases 

 the granular substance of the yolk was seen to encroach upon the " lacuna," 

 especially on the side nearest the surface of the ovule, and to unite with the 

 remnant of the germinative spot ; in other cases the remnant approached the 

 periphery of the lacuna, but without penetrating into the vitellus, and some- 

 times its position appeared to follow no law. The results from hardened eggs 

 were conflicting as to the connection of the nucleolus with the forming amphi- 

 aster. In some cases the bipolar rays, emerging from an aster lying outside the 

 lacuna, abut upon a cluster of granules which are the disintegrated germinative 

 spot ; occasionally a remnant of the spot is joined by a refringent filament to 

 one of the asters of an amphiaster ; more frequently the fragments of the nu- 

 cleolus are widely scattered, and then one is certain that there is only a single 

 aster, though subsequently the amphiaster is seen occupying a nearly horizon- 

 tal position, and those few fragments of the nucleolus which remain visible are 

 distant from it. The former cases would indicate that a minimum portion of 

 the nucleolus contributes to form the amphiaster ; the last, that they have no 

 connection. The author "therefore prefers to consider the participation of the 

 germinative spot in the formation of the ' amphiaster de rebut ' as improbable, 

 but without venturing to deny it absolutely." 



Respecting the metamorphosis of the membrane of the germinative vesicle, 

 he is unable to assert that it is entire, because of the folds which appear 

 when treated with acids ; but whether entire or not, it always forms the 

 limit between the clear substance and the granular vitelline substance ; it 

 does not expel its contents, as Van Beneden erroneously thought. Whether the 

 equatorial zone of fibre thickenings in the first spindle arises from the nucle- 

 olus or from the membrane, is left undecided. 



The author still holds to the probability that the amphiaster first formed is 

 not the " amphiaster de rebut." This opinion seems to rest on the oblique or 

 horizontal position of the amphiaster seen in the earlier stages, as compared with 

 the radial position observed later, and especially on certain preparations made 

 with osmic acid, in which an oval corpuscle, embracing vacuoles and having a 

 denticulate border, occupies the place of the horizontal amphiaster. This he 

 thinks probably indicates a period of inactivity, during which the amphiaster, 

 without ceasing to exist, masses itself together. I am not certain how this is 

 to be harmonized with the subsequent statement that this corpuscle is probably 

 only an amphiaster little accentuated and disfigured by the osmic acid. Both 



