HISTORY OF THE EGG FROM FERTILIZATION TO CLEAVAGE. 23 



at or near the centre of the blastodisc, but not invariably so (Fig. 36) 

 In most preparations of artificially fertilized eggs, their position is decidedly 

 eccentric. We shall presently see that the centre of the blastodisc is 

 determined independently of the position of the polar spindles. As the 

 centre of the blastodisc does not always coincide with the position of the 

 polar spindle, one or both of these points must be inconstant. We are 

 able to explain variation in the former, but we can find no satisfactory 

 grounds for doubting the constancy of the latter. The position of the 

 spindle is determined in the regular coarse of development and matura- 

 tion, under conditions that are strictly uniform both within and without. 

 There is every reason, therefore, to suppose that its position is predeter- 

 mined, and that it marks the pole of the egg with the accuracy of a 

 magnetic needle. 



We have then to distinguish a constant egg pole and an inconstant 

 germ centre. It is quite possible that the germ centre, very early in the 

 formation of the blastodisc, becomes coincident with the egg pole, as the 

 result of the establishment of a new nuclear centre. The polar globules 

 would become eccentric, and their polar distance would vary as that of 

 the germ centre prior to the readjustment. 



The first polar spindle is usually found in a vertical position imme- 

 diately after the egg clears sufficiently to enable us to find it. In one 

 case ( Fig. 1, PI. XXII.) it was found strongly inclined to the axis of 

 the egg, but lacking considerable of being tangential. The equatorial 

 plate has already divided, and the short chromatic elements are arranged 

 more or less regularly in two plates at the truncated ends of the spindle. 

 The whole spindle is in many cases strongly stained with osmic acid, and 

 the chromatic elements are almost black. The outer plate lies at the 

 very surface of the egg, and we have failed to detect with certainty astral 

 rays. In surface views (Figs. 3, G, 7) there are sometimes indications of 

 rays. Beneath the inner plate (Figs. 10 and 11) a clear circular area is 

 usually recognizable, and sometimes feeble rays converge to it. There is 

 invariably a clear area around the spindle, with boundary lines scarcely 

 indicated. The spindle rays are evident, but feebly expressed. The 

 spindle is barrel-shaped, with evenly convex contours. In Fig. 9 its 

 length is only about 10 p, and this does not much exceed its breadth. 

 The spindle shown in Figs. 10-12 is a little longer, and a little more 

 advanced. The number of chromatic elements was not precisely deter- 



