352 BOTANICAL GAZETTE [NOVEMBER 
occurs, the nucleus develops lobes and divides up into a number of 
daughter nuclei varying from two to a dozen or even more (jig. 41, 4, ¢). 
One cannot tell by observation whether a nucleus is about to divide 
into two or many daughters, for there is no apparent segregation of 
chromatin into parts corresponding in number with the number of 
small nuclei to be formed. Nor are the small nuclei necessarily 
equal in size or chromatin content. Sometimes the variations are very 
great; thus fig. 36 shows a cyst in which the primary nucleus has 
divided, leaving the nucleolus undivided in one of the daughters, 
while the other and smaller daughter nucleus has constricted or 
budded again to form a relatively minute nucleus at one side. It is 
difficult to believe that these three nuclei had their origin in equational 
divisions of the chromatin previous to the actual constriction of the 
mother nuclei. All appearances go to indicate rather that the factors © 
controlling the division of these nuclei are entirely disconnected from 
the behavior of the chromatin, and favor CHILp’s hypothesis (2) that 
amitosis is merely a physical process. But whether this be the case 
or not, nuclei so formed are normal and sometimes pass into mitosis. 
When this happens they show the four chromosomes characteristic of 
the species (fig. 35, a). More commonly, however, amitosis by con- 
striction is but an incident in the division of the chromatin, for the 
constricted nuclei are soon completely converted into groups of small 
nuclei by gemmation (see 6, fig. 3). Since the constitution of these 
nuclei is of great importance in this connection, I have introduced 
here a series of three drawings of nuclei from the same cyst. Fig. 
38 shows a case of amitosis of the type most commonly observed, 
though it is not common in Synchytrium; in jig. 39 the two nuclei 
are completely separated but still touch each other; in fig. 4° 
one of the granules similar to those on the periphery of the other 
nuclei has formed a small nucleus by gemmation. The individuality 
hypothesis leads us to rather startling conclusions regarding the con- 
stitution of these nuclei. Each of the large nuclei must have four 
chromosomes. And since each of the granules on the periphery as 
the power of organizing a new small nucleus with four chromosomes; 
it also, if the chromosomes have a material continuity from generation 
to generation, must contain four chromosomes. In other words, the 
large nuclei must at one and the same time have four and 74 chromo” 
