140 CHAS. B. WILSON. 



angles to the first plane of cleavage, and forms au amplii- 

 aster. Its ends are not enlarged in each blastomere, but 

 there is instead a centrosome at either end, which becomes 

 the centre of activities very similar to those manifested by 

 the enlarged ends of the polar bodies (fig. 58). 



Is it too much to conclude that this marked similarity 

 between external and internal phenomena cannot be the 

 result of chance, but that there must be a close relation 

 between the two ? These considerations lead naturally to 

 our second and third conclusions. 



2. In Cerebratulus we have an instance where 

 oogenesis approaches more closely to spermato- 

 genesis, and thereby shows more clearly the mor- 

 phological equivalence of the two. 



Adopting the view first put forward by Mark in 1881 (32), 

 that the polar bodies are to be regarded as abortive eggs, we 

 find in Cerebratulus an instance where they are less abortive 

 than usual. In spermatogenesis the spermatocyte usually 

 divides into four sperms, each of which is in every way the 

 equal of all the others. But in the final division of the 

 oocyte the perfect egg ordinarily appropriates to itself all 

 the activities of the three rudimentary ones, as well as the 

 entire mass of the yolk. In Cerebratulus, however, it con- 

 tents itself with taking all the yolk, and leaves quite a share 

 of the activities for the polar bodies, especially the second 

 one. 



3. The egg-cytoplasm, and probably the female 

 pronucleus also, exhibit a well-marked polarity at 

 right angles to the first plane of cleavage before 

 the two pronuclei have joined to form the segmen- 

 tation nucleus. 



This can be easily seen by reference to figs. 34 and 56. 

 Even in those exceptional instances where the whole polar 

 surface is covered with ridges and papillae the same trans- 

 verse polarisation is manifest in the increased size of certain 

 of the papillae (fig. 37 ; cf. also 2, fig. 12). 



