BJGELOW : NUCLEAR CYCLE OF GONIONEMUS MURBACHIL 307 



in essentially the condition shown in Figure 23. Furthermore, there is, 

 at x, a single dumb-bell-shaped chromosome fully formed. 



Changes now occur resulting in the destruction of the nuclear mem- 

 brane, the formation of the chromosomes, and of the achromatic spindle 

 figure. The actual steps in the formation of the chromosomes are traced 

 only with great difficulty, owing to the fact that with the disappearance 

 of the membrane, the chromatin structures become as a rule so densely 

 crowded together that, although stages in the process are fairly common, 

 it is difficult to correlate them. Fortunately, however, in a few cases I 

 have been able to find cells at this stage whose nuclear contents were 

 more loosely arranged, and one such is represented in Figure 26. The 

 nuclear membrane has wholly broken dowu, and the former nuclear area 

 is occupied by a substance which by its homogeneous or slightly granu- 

 lar nature is readily distinguished from the surrounding cytoplasm. 

 The chromatic structures present a condition not essentially different 

 from that seen (Fig. 23) just prior to the disappearance of the mem- 

 brane, and consist of a considerable number, probably forty-eight, spheri- 

 cal, sharply outlined and deeply stained chromomeres connected with one 

 another by a loose and irregular network of linin strands. The occur- 

 rence of cells exhibiting this condition is very strong evidence for the 

 view that the condensation of the chromatic segments into chromomeres 

 twice as numerous as the chromosomes, and not directly into the chro- 

 mosomes themselves, is a normal and typical step in the mitotic process 

 of the spermatogonia. The formation of the definitive chromosomes, 

 which, in the early metaphase, are dumb-bell-shaped structures, probably 

 takes place by the conjugation of adjacent pairs of chromomeres. Figure 

 27 represents a stage in this process, several of the chromosomes showing 

 the mature conditiou, while many of the chromomeres are still inde- 

 pendent. 



This type of chromosome formation, which differs very markedly from 

 the process occurring in somatic cells and in oogonia, agrees in general, 

 as we shall see later, with the course of events in the first generation of 

 spermatocytes, except for the number of the bodies involved. Yet the 

 ensuing mitosis results, as it does in oogonia and somatic cells, in an 

 equatorial division of chromosomes in the somatic number. 



The succession of stages leading from the resting nucleus to the chro- 

 matin-segment (modified spireme) stage, is so close and the conditions 

 are so clear as to allow of but one interpretation. Unfortunately this is 

 not the case with the processes involved in the formation of the chromo- 

 meres from these segments. Indeed, were it not for the later stages, it 



