72 



rences between the two corresponding mitoses have been also observed 

 by Flemming and others in the case of Salamander. 



Several investigators have described the genesis of the chromo- 

 some from the structure of the resting nucleus in plants, the most 

 recent figures which especially give details of the process being those 

 of GuiGNARD ^). But certain important stages would seem to have 

 been overlooked. At an early stage in the process, the linin thread 

 is visible as a much convoluted filament. This becomes charged with 

 chromatin granules which are especially arranged along the parallel 

 edges of the somewhat flattened thread. As the linin thread thickens, 

 its coils at first are much tangled (Fig. 1) , and this appearance 

 is often specially noticeable in the vicinity of the nucleolus. There 

 seems to be no sufficient reason for regarding this phenomenon ^) a& 

 a mere elfect of reagents. It is best seen in the most successfully 

 preserved cells, and is quite unlike the appearance presented by badly 

 fixed material. In the latter case the tangle is driven up in the same 

 direction in neighbouring cells, but this is not true of the linin in 

 the phase of which we are speaking. The filament becomes thicker 

 and the skein unravels as the chromatin becomes more abundant in 

 it, and the distribution of this substance at the edges of the frame- 

 work is then very apparent. But in the meanwhile indications of the 

 transverse segmentation of the filament into its twelve portions have 

 become clearly visible the positions of such divisions being marked 

 by the poverty, and, gradually, by the absence of chromatin. The 

 linin thread now splits longitudinally along certain tracts, and as the 

 twelve chromosomes gradually separate transversely, the split is seen 

 to extend in some cases completely, in others only partly, through 

 the length of a segment (Fig. 2). The twelve segments are still 

 connected together by linin threads in which, however, chromatin is 

 absent, and three forms may now be distinguished: 1) Those in which 

 the longitudinal cleft stops short of the two ends (Fig 3 a). 2) Those 

 in which it only stops short at one end (Fig. 3 b). 3) Those in which 

 it traverses the entire length of the chromosome (Fig. 3 c). 



Further, the chromosome, as now formed, may be twisted on its 

 long axis so that its two sides cross like the strands of a rope, and 

 this is specially frequent in the forms 2 and 3 above mentioned (Fig. 2). 



1) GuiGNAKD, Nouvelles etudes sur la fecondation. Ann. 8ci. nat. 

 (Bot), 7ieme S^r., T. 14. 



2) To this phase the name of synapsis has been given by one 

 of us (Moobe). Science Progress, Vol. Ill, p. 333. 



