98 Francis B. Sumner 
because we cannot always declare with confidence whether or 
not a given character is adaptive; the second, because the aver- 
age experimentalist is as little conversant with the literature and 
methods of the systematist as the systematist is with those of 
the experimentalist; and the last, for reasons so numerous that they 
cannot even be outlined within the limits of the present paper. 
Herein are presented some of the results of an inquiry into the 
effects of differences of temperature and humidity upon the post- 
natal development of the white mouse.* In all, upwards of 400 
individuals have been subjected to the experimental conditions 
during the past three winters,* though it must be confessed that 
the number of mice in any one series has been relatively small. 
Due allowance has been made for this fact, however, in consider- 
ing the probability of the various conclusions which are offered 
below. As will later be pointed out, differences which are great 
enough and constant enough are to be regarded as significant, even 
though the number of individuals is small. And it is scarcely 
necessary to remind my readers that the statistical treatment of 
even such small numbers as are here under discussion requires 
a great amount of extremely tedious work. 
It must be acknowledged at once that I have been actuated 
by ulterior motives in pursuing these experiments. My primary 
object has been to test the question of the transmission of certain 
characters, or, more correctly, of their reappearance in the off- 
spring. Thus far, it is true, no satisfactory or at all convincing 
test of this point has been made by the writer, though the experi- 
ments are still far from being ended. Large, obvious, and readily 
measurable changes have, however, been produced in the genera- 
tions immediately subjected to the conditions employed. ‘Thus 
one of the primary requirements for the fulfilment of such a test 
has been realized. It is not the purpose of this paper, accordingly, 
to present any evidence in favor of transmission. I shall content 
3 Some of the earlier results of this work were presented briefly before the American Society of Zodlo- 
gists, New Haven, December, 1907 (abstracted in Science, March 20, 1908. 
4To this number must be added nearly 300 others, the data for which were not available when 
the present paper was being prepared. A brief mention of this most recent series has, however, been 
inserted on a later page. 
