86 



The opposite view which has been supported by Farmer aud Moore, 

 Mottier, Shaffner, Juel and others, holds that two spirems are not formed 

 previous to synapsis, but that the spirem is formed singly and the chromo- 

 somes are arranged tandem in the spirem. The bivalent chromosomes 

 are formed by the spirem segmenting into pieces the length of two chromo- 

 somes. These pieces have often formed loops, but this is not always the 

 case. Some of the pieces become approximated after cross segmentation. 



In the two genera investigated there are no signs of a complete spirem 

 previous to synapsis. The chromatin content of the nucleus always exists 

 as rather large granular chromatic lumps connected by delicate anasto- 

 mosing strands of linen. The number of these lumps is quite large, much 

 exceeding the number of somatic chromosomes typical of the .species. There 

 is, therefore, no support in these genera for the theory of the individuality 

 of chromosomes based on the idea of prochromosomes as suggested by 

 Rosenburg for Drosera and by Overton for Thalictrum. 



The first indication that the synaptic condition is approaching is the 

 withdrawal of the nuclear net work toward one side of the nuclear cavity. 

 There is no change whatever in its structure or staining reaction. Some 

 of the chromatin bodies may be seen lying together in close proximity, but 

 this is not interpreted as an indication that they are preparing for subse- 

 quent fusion. It is likewise clear that an occasional linen thread follows 

 for some distance the same course as one lying near it, but this does not 

 occur with anything like the regularity figured by Allen, Cardiff and 

 others and is regarded as occurring too irregularly to have any significance. 

 The synaptic condition is reached while the nuclear content is still in the 

 form of a reticulum. There is no pairing of chromomeres or of the linen 

 threads connecting them. 



The chromatin emerges from the synaptic knot in the form of a rather 

 broad spirem or skein which frequently reveals a double nature and this 

 is interpreted as due to a longitudinal fission. The spirem becomes quite 

 evenly distributed throughout the cavity. It may frequently happen, how- 

 ever that the more regular spirem gives rise to a very irregular and some- 

 what lumpy reticulum before cross segmentation. Formation of the chromo- 

 somes now takes place. The chromosomes consist of two somatics, which 

 have either become approximated together or were not separated from 

 each other during segmentation. 



The chromosomes now become arranged in the spindle plate by the 

 action of the fibers and their distril)ution to the daughter nuclei takes 



