The Stature and Chromosomes of Oenothera gigas, De Vries. 535 
be true, the question remains whether tbe size of tbe individual 
chromosomes determines the size of the nucleus and cell in different 
tissues, or whether, as seems more probable, some common regulative 
factor dependent upon the nature of the metabolic processes and the 
extent of the surfaces of exchange between chromosomes, nucleus, 
and cytoplasm, determines all three in the different tissues of a 
given species. 
In the outer wall layer, or epidermal layer of the anther, there 
is the greatest difference in the cell size that has been observed 
between the two forms, the relation approaching 4 : 1. The cells 
designated the “inner wall of the anther” beloug to the first row 
inside the epidermal layer. They are regularly placed and fairly 
rectangular in outline. The relationship here hetween gigas and 
Lamarckiana works out exactly 3 2 / 3 : 1. In the case of the tapetum, 
the cells being multinucleate, it might be expected that their size 
relationships in the two forms would be considerably disturbed, but 
this is not markedly the case; for while the ratio is 1.44:1 in the 
multinucleate tapetum, it is 1.50 : 1 in the pollen mother cells both 
in synapsis and during the reduction divisions. 
It is noteworthy that in the pollen mother cells, although there 
is considerable growth between the time of synapsis and the two 
reduction divisions, yet at both these periods the ratio of the cell 
volumes in the two forms is not 2 : 1 but 1.50 : 1. 
In a single case the sizes of the nuclei were measured. Nuclei 
in the typical synapsis state were compared, wlien the spirem is 
closely compacted into a ball. The surfaces of the nuclei, estimated 
from the formula 4-r 2 , were found to be in the ratio l 2 3 : 1. The 
volumes of the nuclei, however, using the formula -f-r 3 , were found 
to be very nearly 2 :1. 
Thus it will be seen that while in every case the nuclei and 
cells were undoubtedly larger in 0. gigas than in the corresponding 
cells of 0. Lamarckiana, yet the ratio varied within wide limits, 
falling in with Boveri’s law in some cases and departing rather 
widely from it in others. Some of these departures may be due in 
part to errors in the method of observation, oblique sections of cells, 
etc., but they cannot all be accounted for in this manner. The 
writer hopes to eliminate some of these sources of error in a more 
extensive series of measurements which will include various indi- 
viduals and a wider ränge of tissues in several of the mutants. But 
it is believed that even here the error is not large. 
