BIGELOW: NUCLEAR CYCLE OF GONIONEMUS MUEBACHII. 375 



membranes of the two pronuclei — one of the two chromosome groups 

 consists of twelve separate bodies, while in the other there are only six, 

 which from their shape are evidently bivalent. At a slightly later stage 

 shown in his Figure 44, there are in all twelve bivalent chromosomes. 

 From these same figures there appears one very significant fact ; the 

 pairing takes place between members of the same chromosome group, 

 so that both members of each bivalent chromosome are either paternal 

 or maternal in origin. 



In attempting to explain this temporary reduction in the number of 

 chromosomes in the early cleavage spindles, we are on very uncertain 

 ground. E. B. Wilson (: 00, p. 274) maintains for Cyclops that " we have 

 here a wholly new light on the historical origin of reduction ; for the 

 pseudo-reduction of the germ nuclei seems to be in this case a persistence 

 of the embryonic condition." But this conclusion will not apply at all 

 to Gonionemus. In this animal all the embryonic nuclei acquire the 

 full somatic number of chromosomes after the fourth cleavage. More- 

 over the germ cells are not differentiated until after a long series of cell 

 divisions. And they also, when first formed, exhibit the full somatic 

 number of chromosomes. Therefore in this animal the pseudo-reduction 

 in the cleavage nuclei has no connection whatever with the pseudo- 

 reduction in the germ nuclei. 



4. The Metamorphosis op the Spermatid. 



The recent discussions of this subject by Meves (:02), Korschelt und 

 Heider (: 02), and \Valdeyer (:0l) have been so thorough as to make 

 any detailed consideration of it unnecessary here. I shall therefore limit 

 mvself to a brief comparison between the process in Gonionemus and 

 that which has been described for other coelenterates. 



a. The centrosome. — The only author who has traced the fate of this 

 structure in the coelenterate spermatid is Gorich (: 03\ :04). In Aurelia 

 and Sycandra, according to Gorich, there are two centrosomes in the 

 spermatid. These at first lie side by side on the cell margin, but one 

 soon migrates inward to the nucleus, remaining connected, however, 

 with the outer one by an axial strand or filament. According to Down- 

 ing (:05), metamorphosis of the spermatid in Hydra follows a different 

 course, for he found only one centrosome, which lay at the cell margin. 

 This result agrees with the earlier studies of Aders (:03) on Aurelia. 

 But Gorich (:04) has shown that Aders's observations were very incom- 



