526 BULLETIN OF THE 



its boundary, no valid objection could be raised to considering them 

 (the centres) as localized portions of the nuclear substance, — the less 

 objection since their deportment under the action of staining fluids was 

 such as is exhibited by nuclear substance. Now the case appears some- 

 what changed. The fact that they may lie at some distance from the 

 nucleus while the membrane of the latter is still intact, seems to preclude 

 the possibility of any formal elements of the nucleus taking part in their 

 initiation. It does not, however, prevent the supposition that fluid por- 

 tions of the nucleus may have traversed its membrane, and have been 

 recondensed, so to speak, in the form of " areal corpuscles." Still these 

 corpuscles, when they exist, do not stain as deeply as the nuclear disks, ' 

 and are possibly only condensed portions of protoplasm. The nature of 

 their staining, however, indicates that they probably are composed ex- 

 clusively of neither nuclear substance nor cell protoplasm, but are pro- 

 duced by a fusion of the latter with fluid constituents of the nucleus. 

 The position of the centres of attraction constantly in the vicinity of the 

 nucleus, rather than in remote parts of the cell, is indirect evidence that 

 the nucleus exercises some influence in their production. But one is 

 incapable of saying why the asters appear at particular points, and why 

 there are just two of them. That they do not appear at the same instant 

 is evidence that they are to a certain extent independent of each other. 

 The regularity with which they arise in positions definitely related to 

 the main axis of the egg at the first division, and to the plane of the last 

 preceding division in subsequent stages, shows clearly that their location 

 is controlled in accordance with fixed laws, and it may be reasonably 

 conjectured that the distribution of the active protoplasm (or, what 

 amounts to the same thing, the position of the nutritive portion of the 

 yolk) is an important factor in determining the law. Yet it is not the 

 only factor ; since in the first division of Limax, for instance, it might 

 determine in w^hat one of an. infinite number of latitudinal planes the 

 asters should lie, but it probably could not influence the selection of any 

 one of the infinite number of diameters in that plane for the astral pair. 

 The latter might possibly be eff'ected by the direction from which the 

 male aster approaches the female, and thus its determination be ulti- 

 mately referred to an entirely fortuitous circumstance, — the location of 

 the point where the spermatozoon eff'ects an entrance into the yolk. 

 This, however, I doubt, since I have not been able to conclude that the 

 asters have any fixed position in relation to the two pronuclei or their 

 plane of contact. If, as Fig. 79 seems to show, one of the asters could 

 make its appearance before the contact of the pronuclei, it is difficult to 



