blackman: the spermatogenesis of scolopendra. 67 



nucleus. However, by the time it has grown to the length shown in 

 Figures 91 and those following, it has become nearly perpendicular to the 

 nuclear membrane and its proximal end is much closer to the membrane 

 than previously. While in the process of formation the axial filament is 

 never perfectly straight, but, on the other hand, it is never extensively 

 curved or coiled. However, while fairly direct in its general course, it 

 is often more or less sinuous, as shown in the figures. 



Meantime the nucleus has undergone some interesting changes. At 

 the stage last described the chromosomes had broken down into diffuse 

 granular bodies, which were scattered irregularly over the inner face of the 

 nuclear membrane (Fig. 79). This condition persists during the rapid 

 growth of the nuclear vesicle. The connecting strands of linin and 

 chromatin become even more prominent. However, at the time when the 

 centrosome leaves the cell membrane and begins its migration toward the 

 nucleus, the chromatin bodies become rearranged (Fig. 84). They now 

 congregate upon the side of the nucleus which is nearest the centrosome, 

 and retain this position until the nucleus begins to elongate to form the 

 head of the spermatozoon (Figs. 84-93). The normal number of chro- 

 mosome bodies is not always to be recognized, for in the cells contain- 

 ing the accessory chromosome some of the chromosomes unite with 

 this to form a dense mass, — a karyosphere (Figs. 84, 85, 86, 93). The 

 position of the nucleus has also changed, for it now (Fig. 91) lies upon 

 one side of the cell a short distance only from the cell membrane. 



The change which now begins is one of the most interesting in the 

 transformation of the spermatid into the spermatozoon. The spermatid, 

 hitherto nearly spheroidal, now begins to undergo changes of form which 

 finally result in its extensive elongation. The first observable indication 

 of this change of form is noticed in Figure 92, where the cell on the side 

 farthest from the nucleus shows a wavy contour. This is true, not only 

 of the membrane in the immediate vicinity of the distal end of the axial 

 filament, but of the whole periphery except the third nearest the nucleus. 

 This condition continues to become more and more marked (Fig. 93), 

 until soon the whole cell membrane, except a small portion in the 

 nuclear region, is involved. Projections like pseudopodia are protruded 

 in various directions, causing the cell to assume the remarkable amoe- 

 boid shape shown in Figure 94. The "pseudopodia" are directed not 

 only in what might be thought the most natural plane of elongation, 

 i. e. parallel to the course of the axial filament, but in many other direc- 

 tions as well (Figs. 93, 94). There seems to be an attempt on the part 

 of the ceil to seek out and follow the path of least resistance (Fig. 94), the 



