54 ‘TENOPYR: CONSTANCY OF CELL SHAPE 
is directly proportional to the number of chromosomes. Accord- 
ing to this law, one would expect to find here a direct relation be- 
tween the amount of chromatin, the nuclear size, and the size of 
the cell. Gates (1909) does not find that this holds equally for 
all the tissues. While the chromosome number of Oenothera gigas 
is double that of Oenothera Lamarckiana, the cells of the former are 
not uniformly twice as large as those of corresponding tissues of 
Oenothera Lamarckiana. 
Gates estimated the size of rectangular cells, as those of the 
epidermis, by multiplying the three dimensions. The length and 
breadth of the cells he estimated from measurements made on 
camera-lucida drawings, magnified about 1,380 diameters, the 
third dimension being considered as identical with the length of 
the cell. By length of the cell he means the measurement along 
the long axis of the organ, though this is not necessarily the longest 
dimension of the cell; the width being the measurement at right 
angles to the surface of the organ, that is in the direction of the 
thickness of the organ in which it occurs. The epidermal cells of 
the petals of O. gigas were 1.9 times as large in volume as those 
of O. Lamarckiana, which is closely in accordance with Boveri's 
law. However, other tissue cells depart widely from the expected 
ratio of 2:1. Thestigmatic cells, the cells of the anther epidermis, 
and the inner wall of the anther, were more than three times as 
large in O. gigas as in O. Lamarckiana, while the pollen mother 
cells of the former, both during synapsis and in the reduction divi- 
sions, were only 1.5 times as large. : 
Moreover, the increase in size of the O. gigas cells was not 
equally great in all dimensions. The epidermal cells of the anther 
increased 72.8 per cent. in length, and only 28.4 per cent. in width; 
the stigma cells increased 51.9 per cent. in length, and 32.2 per 
cent. in width. The anthers of O. gigas are approximately twice 
as long as those of O. Lamarckiana. Gates believes that both have 
approximately the same number of cells, ‘‘and that the greater 
length of the O. gigas anthers is accounted for by the greater length 
of the individual cells.” Gates concludes that there is evidently 
some regulating factor determining that the increase in length o 
the epidermal cells shall be greater than the increase in thelr 
in the anther epidermis, but less in the petal epidermis. 
