182 BOTANICAL GAZETTE [uakch 



a new sister centrosome with new radiations. The two centrosome 

 in the prophase of the first division in the oogonium in Fucus seem 

 to be entirely independent, one appearing after the other. Often the 

 second centrosome has not yet appeared even at a late prophase, when 

 the chromosomes are well organized (figs. 48a, b). The radiations 

 seem to increase in number and elongate, probably at the expense 



ot the cytoplasm, as the mitotic process proceeds from prophase to 

 metaphase. 



The spindle fibers at first are clearly seen developing from the 

 area surrounding the centrosome, w r here the nuclear membrane 

 seems to be so thin as to allow the intrusion of the achromatic sub- 

 stance. The rest of the membrane holds its contour very sharply, 

 so that the mitotic figure is intranuclear. Thus the intranuclear 

 spindle of Fucus seems extranuclear in origin. 



In late prophase the bivalent chromosomes are arranged in the 

 equator. The nucleolus often remains as a vacuolate structure. 

 The axis of the figure of the first division is variable, either parallel 

 (fig- 50) or perpendicular (fig. 51) to that of the oogonium. 



The 



rule, dissolves after metapl 



intrusion 



(fig 



somes 



is 32, each being bivalent (figs. 48a, b). The same number is counted 

 from the polar view of early metaphase (fig. 52b). In metaphase, 

 the two halves of the bivalent chromosomes separate. These two 

 halves are not formed by the splitting of one chromosome, but are 

 two independent chromosomes which were two arms of one loop- 

 Later metaphase (fig. 53 ) an d anaphase (fig. 54) follow; in late 

 anaphase the chromosomes near the pole are straight rod-shaped 

 without any apparent indication of partition (fig. 55). Probably the 

 initiation of the splitting which provides for the second division may 

 be very much delayed in this form. After telophase there are organ- 



J 



a short rest. 



e centrosomes become unrecogni**" 1 • 

 oogonium follows the first after only 



- #..._ .a,,* racked 



chromosomes 



chromatin reticulum and the appearance of centrosomes seem essen- 



mitosis. The mitotic process in 



