284 i NATHAN FASTEN. 



the so-called "synaptic pole" and arrange themselves in parallel 

 pairs. This paired arrangement is very distinct and has been 

 observed in a great many cells. While this is going on, two 

 large round bodies, of more or less equal size, and surrounded 

 by clear spaces (Fig. 19, k) make their appearance in the cyto- 

 plasm. These bodies were first observed in a smear preparation 

 which was heavily stained. They stained exactly like the chro- 

 matin. In preparations which were strongly destained no trace 

 of them was found. However, when these latter preparations 

 were restained, the bodies loomed up with exceptional clearness. 

 It is thus seen that although these structures stain like chromatin, 

 yet they differ from chromatin in their affinity for nuclear dyes, 

 and, in all probability, they are chemically different from chro- 

 matin. Wilson ( '13), first described a similar body in Pentatoma, 

 and called it a "chromatoid body." In Cambarus virilis, 

 I ( '14) have found a pair of chromatoid bodies appearing at about 

 the same stage in the developing spermatocyte as in Cancer 

 magister. After the leptotene stage, the chromatoid bodies of 

 Cancer magister persist within the cells, and their subsequent 

 history will be outlined when the respective stages of the matura- 

 tion are dealt with. 



Synapsis soon sets in. The parallel threads become more 

 closely paired at the synaptic pole and the cell enters the pachy- 

 tene stage, in which the pairs of leptotene threads fuse into thick 

 gemini (Fig. 20). Here the union appears complete as no traces 

 could be found of the longitudinal furrows which separated the 

 original pairs of parallel threads. 



After remaining fused for some time, the components of each 

 geminus begin to unravel. Along every geminus a longitudinal 

 split makes its appearance (Fig. 21), and simultaneously with 

 this there is an opening of the paired threads at one end, while 

 remaining attached at the other end, thereby producing figures 

 which appear like 8, V, or less commonly like U (Figs. 21 and 22). 



Another longitudinal split soon occurs along the arms of each 

 geminus. In Fig. 22 the beginnings of this second longitudinal 

 furrow are clearly visible in the two arms of the 8 which is located 

 in about the center of the nucleus. By means of these two longi- 

 tudinal cleavage planes, four chromatin threads are formed, 



