532 BULLETIN OF THE 



was nearly as extensive as the whole " area " ; and, finally, at an early 

 stage in the formation of the first segmentation amphiaster, by a few 

 scattered (Fig. 85), or more definitely grouped (Fig. 82), refringent cor- 

 puscles. Since these various conditions exist in eggs that were sub- 

 jected to nearly the same treatment, it is not possible to account for the 

 differences as due to the action of the acid. Still, the nearly homoge- 

 neous condition is the one by far the most prevalent with the employ- 

 ment of acetic acid. ' With osmic acid, which according to Hertwig is 

 the most satisfactory to demonstrate the existence of nuclear substance, 

 I have not uniformly succeeded in showing a central body, but with 

 chromic acid a small, lustrous, sharply limited corpuscle is almost always 

 distinguishable (Figs. 44, 52) exactly in the o^ntre of the radiation. A 

 comparison of Fig. 52 with Figs. 73, 79, and 80 will illustrate the dif- 

 ferences which result from treatment of the same stages with different 

 reagents. It may possibly appear significant of the accuracy of Plert- 

 wig's view of the nature of this areal corpuscle, that it is already differ- 

 entiated at the earliest stages in the formation of the aster which I have 

 seen. I cannot deny, in those cases where the nucleus elongates and its 

 poles are observed to occupy the centre of the aster, that the most natu- 

 ral inference is that the corpuscles are segregated portions of the nuclear 

 substance, but in the case of Limax the assumption seems impossible. 

 These areal corpuscles lie outside the sharply marked territory of the 

 nucleus, often at a considerable distance, and there is no evidence of a 

 direct connection between the two. From theoretical considerations it 

 may be difficult to explain the activities of the astral centres without 

 admitting a fusion of nuclear and vitelline substances, since it is not 

 plausible that a chemical process should be initiated in a homogeneous 

 substance without the presence of a second material differing in compo- 

 sition from the first ; but the mingling of vitelline and nuclear matter 

 does not necessitate the appearance of the latter in the form of discrete 

 corpuscles; besides, the detachment of portions of nuclear substance 

 seems in Limax irreconcilable with the early relations of nucleus and 

 aster. It seems unsatisfactory to consider the areal corpuscle as un- 

 modified " nuclear substance," exercising an attractive influence on the 

 surrounding protoplasm, since the remaining and major portion of that 

 substance gives no evidence of exercising a like influence on the yolk. 

 I am therefore inclined to regard the corpuscles as a product of the 

 fusion of nuclear and vitelline substances. 



Their ultimate fate is as uncertain as their origin. There is reason 

 for believing, from evidence given elsewhere, that in Limax the corpuscle 



