﻿264 BOTANICAL GAZETTE • [april 



observer. In case the daughter segments are in contact for their 

 entire length the chromosomes may be fastened to the spindle 

 at one end and arranged radially upon it; but, as a rule, the 

 point of attachment is at the place of bending, whether that be 

 near one end or at the middle of each segment. The X-, Y-, and 

 U-shaped chromosomes, although of less frecjuent occurrence in 

 Podophyllum, are formed in the same manner as described for 

 Lilium. 



During the metaphase, or immediately following it, each daugh- 

 ter chromosome is seen to be double or composed of two similar 

 granddaughter segments, the result of the second longitudinal 

 splitting i^figs. 22y 2j), The shape of the pairs of grand- 

 daughter segments will, of course, depend upon that of the 

 daughter chromosome. In Jig. 21, a, it is evident that this chro- 

 mosome before metakinesis was similar to those in Jig, 20, In 

 Jig, 21, 5, the daughter segments may have been in contact 

 throughout and the chromosomes fastened to the spindle at the 

 end. The retreating pairs of segments in this case may present 

 the familiar Vs. The chromosome to the right in Jig, 22 was 

 either in the form of a ring or a U when in the nuclear plate. 

 This condition was observed at the time of my earlier study upon 

 Podophyllum, but such conditions were interpreted as two simi- 

 lar but different chromosomes lying close together, one beneath 

 the other. 



The double nature of the daughter chromosomes can be 

 clearly distinguished until they reach the poles. Having arrived 



-h- 



at the poles, they generally become closely crowded together so 

 that the individuals are difficult to distinguish. It can be seen, 

 however, that the segments of each pair tend to separate, and 

 there was nothing observed which might lead one to conclude 

 with any certainty that a double spirem is formed. The grand- 

 daughter segments in forming the spirem tend to reticulate some- 

 what so that they become irregular in shape and their exact 

 identity is lost. Those observers who assert without reserve 

 that the identity of the chromosomes can be followed at all times 

 from one division to the other, certainly leave this fact and 

 other steps of the process out of consideration. 



