SMALLWOOD: MATURATION OF HAMINEA SOLITAKIA. 291 



to the cortical layer of the spliere. When they begin to be formed 

 (Fig. 47), they are not strictly radial, but often are more or less curved 

 in various directions independent of the form of the prospective spindle. 

 Sometimes (Fig. 48) the fibres between the nucleus and each of the 

 asters are so curved in relation to one another as to produce a short trun- 

 cate or cask-like spindle, such as is seen in plant cells, and sometimes in 

 animal cells, during nuclear division. Each of these two cask-like bundles 

 of fibres later contributes to the formation of the spindle. These fibres 

 are purely cytoplasmic in origin, and continue to be formed until the new 

 spindle is complete. As the fibres increase in number, the nuclear mem- 

 brane breaks down, and the fibres become attached to the chromosomes. 

 During this process the fibres do not push before them the nuclear 

 membrane, as in cleavage, but the outline of the nucleus remains 

 regular as long as it is discernible. When the two asters and the nucleus 

 do not lie in a straight line (Plate 7, Fig. 49 ; Plate 11, Fig. 79), the 

 direction of the forming fibres is noticeably modified. 



In Figures 55 (Plate 8) and 80 (Plate 11) the mitotic figure is in the 

 metaphase. The spindle at this stage is very long, but it becomes much 

 shorter during the anaphase, when the separated chromosomes move 

 toward the poles of the spindle (Fig. 50). This shortening of the spindle 

 sometimes occurs before the chromosomes have begun to move apart. 



Centrosome. Nearly at the same time with the beginning of these 

 changes in the chromosomes, and during the initial stages in the forma- 

 tion of the spindle, the centrosome has become differentiated into centro- 

 plasm and centriole (Plate 7, Fig. 47). The lack of correlation, not only 

 between the progressive changes in the centrosome and those in the 

 chromosomes, but also between the two centrosomes of a single spindle 

 figure, is evident from the drawings. 



As s"en in Figure 46 (Plate 7), each centrosome at the time of the 

 disappearance of the central spindle is a solid, spherical, deeply staining 

 body. The centrosome at the inner pole of Figure 47 shows a thin layer 

 of centroplasm surrounding an irregular centriole, but the one at the 

 outer pole of the same egg remains undifferentiated. In Figure 48 the 

 two centrosomes exhibit similar differences, while in Figure 49 both 

 centrosomes are solid or undifferentiated. In Figure 54 (Plate 8) 

 centroplasm and centriole are already clearly distinguishable in both 

 centrosomes ; but in Figure 55 the outer centrosome is still homo- 

 geneous. The centroplasm arises from the substance of the centrioles 

 precisely as in the case of the centrosome of the first polar spindle. 

 During these changes the centriole remains irregular in outline until the 



VOL. XLV. — NO. 4 3 



