464 
GEOFFREY SMITH AND EDGAR SCHUSTER. 
Thus, in his figure 3, referring to a frog castrated in 
.January, 1910, implanted with testes “ in autumn/’ and 
killed on October 21st, 1910, the thumb is in the state of a 
normal January frog. In Fig. 4, referring to a frog 
castrated in September, 1909, implanted with ovaries “ in 
autumn/’ and killed on October 21st, 1910, the papillae on 
the thumb are far less developed than in fig. 3, and are in 
about the condition of a normal frog during September. 
This difference is attributed by Meisenheimer to the less 
effective action of the implanted ovaries compared to the 
testes, but according to our interpretation it is simply due to 
the fact that the frog referred to in fig. 3 was castrated in 
January, when the papillae were comparatively well de- 
veloped, while that referred to in fig. 4 was castrated in 
September, when the papillae were less well developed. It is 
true that fig. 5, referring to a male castrated in January, is 
in an intermediate condition between figs. 3 and 4, but we 
know that the individual variability in the times and degrees 
of development of the papillae in normal frogs is very great. 
Therefore, without fostering any preconceived hostility to the 
theoretical interpretations of these authors, or without in any 
way casting a doubt on their perfect good faith, we are 
utterly unable to regard the conclusions they draw from their 
experiments as in the least degree convincing. Any active 
influence on the thumb, either of testis injections or the 
implantation of pieces of testis or ovary, is according to our 
view entirely unproven. 
The fundamental point in which these authors appear to 
have been mistaken is in supposing that the castration has 
the effect of always reducing the papillae present on the 
thumb, and of causing them to disappear. The effect of 
castration, except actually during the breeding season, is to 
make the papillae on the thumb-pads remain essentially in the 
condition in which they were at the time of castration. The 
presence of functional testicular tissue is necessary for the 
normal cycle of changes in the thumb-pad, and an 
essential part of this normal cycle is the acquisition of the 
