342 CHESTER A. STEWART 
conclusions, however, on account of physiological variability in 
the weight of the testes. . 
The weight of epididymides (table 2) in my test rats of three and 
six weeks of age is higher than for the corresponding controls. 
For the youngest group there is an apparent increase from a 
normal weight of 0.0078 gram to 0.0154 grain an increase of over 
95 per cent (uncorrected for slight difference in body weight). 
At six weeks of age the excess weight for the epididymides in 
the test rats is only 13 per cent. 
In the rats underfed to ten weeks of age and also in those at 412 
days, there is an apparent loss in the weight of the epididymi. 
The loss at ten weeks amounts to approximately 6 per cent, and 
at 412 days to about 32 per cent. However, in the test rats 
weighing 67.2 grams net at 291 days there is an apparent increase 
in the weight of the epididymi from 0.108 to 0.315 gram. In all 
cases the percentage losses above estimated are subject to 
correction for slight difference in body weight. 
In general, therefore, the data indicate that in the youngest 
rats there is a marked increase in the weight of the epididymides, 
but that later there is a loss of weight in most cases. 
Jackson (715 b) found an apparent slight loss in the epididymides 
in rats held at constant body weight from three to six and ten 
weeks of age. In adult rats during inanition Jackson (15 a) 
found that the testes and epididymides apparently lose weight 
in about the same proportion as the entire body. 
OVARIES 
‘The weight of the ovaries (table 2) in my controls at seven days 
of age (0.0018 g.) is very much lower than the Wistar norm 
(0.0033 g.) for corresponding body length. The difference 
amounts to more than 80 per cent. Jackson (’13) found the 
weight of the ovary to average 0.00121 g. at one week of age, 
which is even lower than my data. It therefore appears fairly 
certain that the weight of the ovary calculated from Hatai’s for- 
mula is too high for this early period. At later ages my data do 
not differ so greatly from the Wistar tables. 
