538 
R. RUGGLES GATES AND NESTA THOMAS. 
in fig. 14, in which the chromosomes in metaphase have 
begun their fission at one end. There are here sixteen or 
seventeen chromosomes, the extra ones being evidently due 
to a precocious completion of the fission by certain chromo- 
somes. This is an occasional source of apparent variation 
ill the chromosome number. One other observed case of 
16 chromosomes in the somatic metaphase is doubtless to be 
accounted for in the same way, and was not a real variation. 
This is further shown by the fact that counts in the prophase 
of mitosis have never been found to show any variation in 
number. 
In another lata plant (142 . 1 . 3) a similar case of the pre- 
cocious fission of certain chromosomes was found (fig. 15). 
In this group the chromosomes varied much in size and were 
strongly paired, both members of one pair (on the right) 
having undergone a precocious split, while a third chromosome 
is beginning the fission from both ends, the remaining chromo- 
somes showing no indication of a split. 
We regard these cases as a demonstration that certain 
chromosomes of a group sometimes divide prematurely, and 
there can be little doubt that this has led in some cases to 
an assumption of variation in the number of chromosomes 
where such variation does not actually exist. It seems 
probable that such premature splitting of chromosomes will 
be found to occur most frequently in hybrids and mutating 
forms. 
Meiotic Divisions. 
The behaviour of the 15 chromosomes during meiosis was 
especially studied in several plants. The normal features of 
meiosis in various CEnotheras are already well-known. Hence 
in the present paper we shall confine ourselves chiefly to 
pointing out various peculiar features in connection with the 
behaviour of the extra chromosome and the phenomena of 
chromosome degeneration. 
We may first point out that in the heterotypic mitosis in 
