154 F. B. SUMNER, M. E. MCDANIEL AND R. R. HUESTIS. 



Newcomb ( 1904) likewise finds that " the first born child of 

 any mother is more likely to be a male, in the proportion of about 

 8 to 7" (P- 28). 



NUTRITION. 



Our data on the much-discussed subject of the relation of nutri- 

 tion to the sex ratio are extremely meager, though it seems worth 

 while to include them in this report. Any such extreme effect 

 as has been alleged by certain writers might have been expected 

 to manifest itself, even in the limited numbers here considered. 



For the purpose of testing the possible beneficial effect of in- 

 cluding meat in the dietary of the stock, certain lots of mice, in 

 three successive generations, were divided into two sections, one 

 being given a rather liberal supply of chopped, boiled meat (com- 

 monly liver), in addition to the regular ration which was strictly 

 vegetarian, the other (control) lot being restricted to the latter. 

 The experiment was commenced at about the time when the first 

 of the three generations referred to was mated. The meat diet 

 was continued until the mice of the third generation were fully 

 grown. 



In all, there were 237 young of determined sex among the off- 

 spring of meat-fed parents (second and third generations). 

 These gave a sex ratio of 104.31 9.13. In the control lot were 

 228 young, giving a ratio of 96.56 8.6r. Here, again, it is obvi- 

 ous that the difference is not significant. The transfer of a very 

 few individuals from the male to the female column (or vice 

 versa) would be sufficient to reverse it. 



Cuenot and Schultze (both cited by Morgan, 1907, pp. 385, 

 386) performed experiments upon rats and mice, respectively, 

 with a view to testing the possible effects of feeding upon sex 

 determination. In neither case were significant differences found 

 between the sets of animals under comparison, but the numbers 

 were very small and the results almost valueless statistically. 

 More recently, Slonaker and Card (1918) have computed sex 

 ratios for omnivorous and strictly vegetarian white rats, finding 

 these to be .113.6 and 107.6 respectively. Since the number of 

 individuals and other essential matters are not stated in the brief 

 preliminary communication referred to, we must suspend judg- 



