An Experimental Study of Somatic Modifications etc. 
341 
of the same weight and of the same sex. Taking into account the 
three characters under consideration, there are thus 45 pairs of con- 
trasted figures to be reckoned with. The figure for the > warm « -room 
descendants is greater in 26 of these cases, smaller in 19. The pre- 
ponderance is thus not large. Furthermore, for one character (foot 
length) there is an actual majority in favour of the cold-room descen- 
dants, i. e. there are only 7 cases in which the »warm« figure is 
larger, 8 in which the »cold« figure is larger. Indeed among the 
males alone, there are only 3 positive cases and 6 negative ones. 
Fig. 11. 
-11 IZ 13 15 1 6 17 1 8 19 20 21 2 2 23 24 25 26 2 7 
Foot and ear length of same mice (see legend for Fig. 10). 
When all three of the characters are considered, the males show 
12 positive cases to 15 negative ones, i. e. the >coldc figures is larger 
in an actual majority of cases. For the females, however, the rela- 
tions are quite different. We have 14 positive cases and 4 nega- 
tive ones. 
Thus, a consideration of this last table, taken by itself would 
not be very convincing, though it must be remembered that the ge- 
neral tendency of the figures, viewed as a whole, is distinctly in 
accordance with the results of the earlier analyses. As may be seen 
