BLACKMAN : THE SPERMATOGENESIS OF SCOLOPENDRA. 63 



from the cell substance, one or more of the chromatin bodies are forced 

 out through one of the openings in the nucleus and there form a vesicle 

 about themselves, just as do the chromosomes in the telophase of the 

 mitosis in many egg cells. This gives rise to such appearances as were 

 noted in my former article and are shown in Figures 81-83 of the 

 present paper. In Figures 81, 82, this vesicle is shown as still connected 

 with the nucleus, and probably the material enclosed by it is later with- 

 drawn within the nucleus of which it continues to form a part. In 

 Figure 83 the vesicle, though smaller, is entirely separated from the 

 nucleus, and the nuclear membrane on the side nearest it is ruptured, 

 showing whence the "bud" is derived. Whether in such cases there 

 is later a fusion of the nucleus and the extruded portions, I cannot say 

 positively. This may be the case, or the bud may be extruded from 

 the cell entirely, as I suggested in my first paper ; or, finally, the sub- 

 stance may be dissolved in the cytoplasm. At any rate, no such struc- 

 tures have been observed at later stages in the spermatid. 



As a result of the rupture of the nuclear membrane and the imbibition 

 of material from the cytoplasm, the diameter of the nucleus is increased 

 about one half. From this stage (Fig. 84) until the time when the 

 nucleus has begun to condense and to elongate to form the head of the 

 spermatozoon there appears to be no definite continuous nuclear mem- 

 brane. There is, to be sure, a sharp line of demarcation between nucleus 

 and cytoplasm, due to the differences in staining reaction which they 

 exhibit, but nothing more. 



In the meantime the cytoplasmic structures have undergone some 

 slight modifications. The reticulum is considerably coarser, and the 

 meshes more transparent, than in preceding stages in this transparency, 

 which had not yet reached its highest point of development. This is due to 

 the fact that the archoplasm is being withdrawn from general distribu- 

 tion throughout the cytoplasm and is again becoming aggregated into 

 specialized masses in various parts of the cell. These aggregations may 

 at first appear in any part of the cytoplasm (Fig. 84), but in later stages 

 they are collected principally in the region of the centrosome (Figs. 85 

 et seq.). 



The centrosome at the stage when it was last described still preserved 

 its position in contact with the cell membrane (Fig. 79). At that time 

 it was a simple spherical granule without any enveloping layer of special- 

 ized cytoplasm. Shortly before the stage represented in Figure 84, it 

 leaves its peripheral position and migrates through the cytoplasm toward 

 the nucleus. The centrosome during its migration through the cytoplasm 



