132 BOTANICAL GAZETTE [febrvaby 



28). This statement is based on the fact that the naked karyosomt 

 are, in the cases observed, larger than those of any other nuclei in the 

 same cysts, and that the resultant clusters of small nuclei are greater 

 in mass than any single nucleus in the cyst at this stage. The varia- 

 tion in the size of the nuclei, however, is so great in other stages that 

 it is not impossible that these may have been larger nuclei in the 

 beginning. Lobes (fig. 29) now appear on the margin of the karyo- 

 some, each of which rounds off and becomes the karyosome of a small 

 nudcus. When these karyosomes have separated, vacuoles of 

 nuclear sap appear around them; surrounding membranes are next 

 formed in the meshes of the cytoreticulum bounding the cavities, thi 

 completing the process. The membranes, however, do not appear 

 s im u ltaneously around all the nuclei of a cluster. There is usually 

 sufficient difference to allow some observation of the process of mem- 

 brane formation. The vacuoles which become the nuclear cavitie 

 are at hrst indistinguishable from those between the meshes of the 

 cytoreticulum which are filled with cell sap, but they are gradually 

 >urrounded by membranes which are apparently precipitated from 

 the cytoplasm next the cavity. Neither in heteroschizis nor in nuclear 

 gemmation is there evidence of any connection of the centrosome 

 with membrane formation such as occurs in the reconstruction of 

 the nucleus after mitosis (Kusano 8, Griggs 7). 



Besides the two sorts of amitosis just described, a third method ha 

 been observed a few times. In this process, which has been seen only 



when the nuclei were in spirem, the nucleus becomes strongly lobcd; 



each lobe contains a portion of the original unchanged spirem; the 

 lobes become more pronounced and are cut apart by continued con- 

 striction. There may be only two lobes, as in ordinary amitosis, or 

 there may be several, as in heteroschizis. Although even a single 



(fig 



ot the process, there is much concerning it which is doubtful, and its 

 occurrence is rare in my slides. I refrain, therefore, from more than 

 mention of the matter at this time. 



LATER HISTORY OF THE SMALL NUCLEI 



If amitosis leads to degeneration and death, as has been held 

 almost universally until recently, we should expect to find a large 



