BLACKMAN. — SPERMATOGENESIS OF THE MYRIAPODS. 333 



but is in the form of a number of granular segments, one half as numer- 

 ous as the ordinary chromosomes of the spermatogonia. 



These chromosomes are produced by the union in pairs of the chromo- 

 somes of the division period. That this is an end to end union is very 

 strongly indicated, although not definitely proved, by the appearance of 

 the chromatin segments at this stage. In Figures 4 and 5 the nuclear 

 vesicle is occupied by a number of slender granular chromatin segments, 

 each of which is bent near its middle, forming a V-shaped figure. At 

 the angle of the V, which doubtless represents the point of union of the 

 two component elements, the chromatin is often interrupted in such a 

 manner that the halves of the segment seem to be connected by only 

 a strand of linin. These conditions in Scolopendra are quite similar to 

 those described by Montgomery (:00) in Peripatus, and later (:01) in 

 Hemiptera, and they lead to the same conclusion as did the more con- 

 vincing observations made by Sutton (:02) on Brachystola. The latter 

 author has shown conclusively that the synapsis of the chromosomes 

 occurs during the telophase of the last spermatogonium, and that 

 it is brought about by an end to end union of entire spermatogonial 

 chromosomes. 



Thus in S. subspinipes the pseudo-reduction of the chromosomes is 

 accomplished in exactly the same manner as in S. heros. Up to this 

 stage (Figure 4) the appearance of the accessory chromosome has been 

 entirely similar to that in S. heros, but from now on the behavior of all of 

 the chromatin is quite noticeably different. This is evidently due to the 

 introduction within the nucleus of a new structural element, — the true 

 nucleolus, or plasmosome. 



The first indication of the nucleolus in the spermatocyte is seen at the 

 stage when the nuclear membrane has just been re-formed (Figure 5). At 

 this time a transparent, non-stainable structure appears upon one side of 

 the accessory chromosome. This fundament of the nucleolus at first has 

 very much the appearance of a vacuole occupying one side of the acces- 

 sory chromosome (Figure 5). 



In the telophase of the last spermatogonial division the accessory chro- 

 mosome, when it first becomes distinguishable from the other chromosomes, 

 is of an oval shape (Figures 3, 4), but after the origin of the plasmosome 

 the compound element thus arising is at first spherical. The chromatin 

 representing the accessory chromosome assumes a crescent shape and 

 nearly surrounds the unstained part — the plamosome. In this manner 

 the accessory chromosome forms one half or more of the outline of the 

 sphere (Figures 5, 8 b). 



