BIGELOW : NUCLEAR CYCLE OF GONIONEMUS MURBACHII. 315 



of synapsis; moreover it has been maintained by most authors that it is 

 during this stage that the numerical reduction, or pairing, of the chro- 

 mosomes takes place. 



Such appearances occur in Gonionemus in great numbers among the 

 primary spermatocytes (Figs. 50-54), and at first I had little question 

 that I was dealing with a true "synapsis" zone. Further study, 

 however, threw doubt on this conclusion, and I may at once state my 

 conviction that, in the present species at any rate, they are purely acci- 

 dental, and in no way connected with the essential features of synapsis. 

 This may seem rather a sweeping statement, but it is, I think, sup- 

 ported by all the facts of the case. The starting point in the formation 

 of such artifacts is represented in Figure 48. Here the karyoplasm has 

 entirely disappeared, leaving the nucleus empty and transparent except 

 for the chromatin net, which has lost its connection with the nuclear 

 membrane and contracted away from it toward the centre of the nucleus. 

 All possible gradations are found between this stage and such a condition 

 as is shown in Figure 53, where the entire chromatic structure is so 

 densely packed that it appears to have fused into a single mass. As is 

 shown in Figure 48, a massing of part of the reticulum may take place, 

 while the remainder continues normal in aspect. The chromatin mass 

 thus formed usually lies in contact with the nuclear membrane, but oc- 

 cupies no constant position in relation to the axis of the oval nucleus. 

 Viewed from certain positions it is, of course, difficult to tell whether or 

 not it is in contact with the membrane, though in some cases it is certain 

 that it occupies a less peripheral position. In almost all cases a more or 

 less complete resolution of the mass into its component strands is pos- 

 sible, and careful study has shown that these represent many very 

 different stages. For example, the cell represented in Figure 48 is evi- 

 dently in a stage immediately after the disappearance of the nucleolus ; 

 Figure 50 shows a contracted chromatin net, while Figures 54 and 55 

 show still later stages. This is not due, I think, to any progressive 

 change of the nuclear structures during the stage of chromatic conden- 

 sation, but rather, in all probability, to a nearly instantaneous conden- 

 sation, which may occur at various periods — a conclusion supported by 

 the observations of Lee ('97) on Helix. Indeed, we may safely say 

 that in Gonionemus there is no stage in the prophase of the primary 

 spermatocyte between the disappearance of the nucleolus and the dis- 

 solution of the nuclear membrane when it may not occur. This is 

 of itself nearly conclusive evidence that the " pseudosynapsis " phase is 

 not to be considered as representing a definite stage of development, and 



