﻿270 BOTANICAL GAZETTE [april 



certain striking resemblances to that in the pollen mother-cell. 

 In all probability they are exactly the same. In the chromatin 

 spirem no indication of a longitudinal splitting Is to be seen. 

 The nuclear membrane disappears as such, and the spindle fibers 

 penetrate the nuclear cavity before the spirem is segmented into 

 chromosomes {Jif^^ 43). The complete segmentation seems to be 

 effected later than in the pollen mother-cell. Within the multi- 

 polar complex of spindle fibers the partly segmented spirem, as 

 in the pollen mother-cell, is generally greatly entangled and 

 kinked. As the spindle matures, the arrangement of the chro- 

 mosomes becomes more regular, but, as a rule, less so than \nfig' 

 44. When the spindle has reached the typical bipolar stage, 

 the chromosomes are oriented in the nuclear plate, some radially 

 and some tangentially [Jig. 44). 



Each consists of two rather long and closely applied segments 

 that may be twisted upon each other. It seems that the spirem 

 may not be entirely segmented even when the typical bipolar 

 stage of the spindle is reached {fig^ 44^, and as the apparently 

 unsegmented portions are double nothing seems more reasonable 

 than the conclusion that the double nature is due to the longi- 

 tudinal splitting of the chromatin spirem, and it was upon this 

 evidence that my conclusion of 1898 was based, namely, that the 

 spirem of the daughter nucleus undergoes a longitudinal fission. 

 Since the second longitudinal fission takes place during the 

 first mitosis, it is not difficult to understand, especially when one 

 bears in mind the entangled or kinked condition of much of the 

 spirem during the development of the spindle, and the fact that 

 portions of it are very slender, appearing as if stretched by the 

 fibers, how parallel portions of the spirem may be so brought m 

 contact as to look like a longitudinal splitting had occurred. 

 Further evidence in favor of the doctrine that the two segments 

 of each chromosome are brought in contact side by side, is the 

 fact that in both pollen and embryo-sac mother-cells one very 

 frequently meets with chromosomes that do not consist of two 

 .rather straight and readily distinguishable segments, but rather 

 short, thick, and lumpy chromosomes which show that the seg- 

 ments are kinked or twisted up into a short spiral or coil. It 



