BLACKMAN. — SPERMATOGENESIS OF THE MYRIAPODS. 339 



During the early prophase, up to a time when the tetrads have begun 

 to he formed, the accessory chromosome remains apparently unchanged. 

 It is still a small, deeply staining, homogeneous hody closely apposed to 

 oue side of the plasmosome. Later, when the other chromosomes have 

 already assumed the tetrad shape and have begun the process of conden- 

 sation, it leaves this position, and thereafter may lie in any part of the 

 nucleus. Owing to the ordinary chromosomes being of irregular form 

 during the late prophase (Figure 20), and to their being approximately 

 equal in size to the accessory chromosome, the latter is in many cells 

 not distinguishable from them. Consequently it has been impossible to 

 trace the later history of the accessory chromosome in S. subspinipes with 

 as much certainty and in such detail as has been done in S. heros 

 ( Blackman, :05). However, I can say that the accessory chromosome, 

 unlike the other chromatic elements, undergoes but one — a longitudi- 

 nal — division. 



Thus it is certain that the behavioi of the modified chromosome in the 

 two species of Scolopendra which I have studied is similar, at least in all 

 essential particulars. In both species the accessory chromosome is pro- 

 duced not by the union of two spermatogonial chromosomes during syn- 

 apsis, as the other chromosomes are, but is derived from one of the 

 spermatogonial elements directly, being in no way altered by synapsis. 

 Since the object of the reduction division is the separation of the chromo- 

 somal elements which conjugated during the telophase of the last sper-' 

 matogonium, it naturally follows that the accessory chromosome, not 

 having taken any part in this conjugation, does not undergo a reduction 

 division. In Scolopendra heros (Blackman, :03, :05), as well as in many 

 iusects, Pyrrhocoris (Henking, '91), Anasa (Paulmier, '99), Orthoptera 

 (McClung, :02, Sinety, :01, Sutton, =03), the absence of this element in 

 one half of the spermatids has been demonstrated. This is also true for 

 S. subspinipes. 



At the end of the growth period, during the vesicle stage, the inner, 

 lightly staining portion of the karyosphere, i.e., the nucleolus, often con- 

 tains several small clear spaces. As these show no reaction whatever to 

 the stains, and as they are bounded by a definite spherical outline, it is 

 probable that they are vacuoles filled with a fluid or gaseous substance. 

 Before the beginning of the prophase the conditions are such that these 

 structures are not often visible. It is only in such cases as that repre- 

 sented in Figure 10, where a part of the karyosphere has been removed 

 in sectioning, that they may be observed. Later, however, in the pro- 

 phase, when the chromatic portion of the karyosphere has become de- 



