SEX-RATIO OF THE DEER-MOUSE. 127 



might be led to suppose that the young males were subject to a 

 mortality about twice as great as the females. But aside from 

 the extremely limited numbers here concerned, there are other 

 reasons for believing that no such marked differential mortality 

 exists in this case. 



Separate computations have been made for the broods in which 

 no deaths were recorded and for the broods in which deaths are 

 known to have occurred. Of the former there were 1,301 broods, 

 containing 4,081 individuals, 2,020 being males and 2,061 fe- 

 males. The sex ratio here is 98.01 2.07. 



Of the broods which are known to have sustained losses prior 

 to marking there were 297, originally comprising 1,079 individuals. 

 The number dying (or missing) was 438, of which 398 were of 

 unknown sex. The surviving 641 mice comprised 312 males and 

 329 females, giving a sex ratio of 94.83 5.06. Thus we do 

 find a slight difference between the complete and the depleted 

 broods, though the difference is a quite non-significant one, sta- 

 tistically speaking. The reasonable inference is that the mortality, 

 during the period here concerned, is approximately equal for the 

 two sexes. In any case, there can be no such disparity in their 

 respective death rates as the meager record of identified dead 

 might lead one to suppose. 



Seven possible influences, which have been held by various 

 writers, to affect the sex ratios of animals, have been considered 

 in our treatment of the data at hand. These are: (i) season, 

 (2) size of the litter, (3) race, (4) hybridization., (5) inbreeding 

 and outbreeding, (6) order of birth, (7) diet. As will appear in 

 the ensuing pages, we have some evidence that the first, second 

 and fourth of these influences are actually effective in the case of 

 Peromyscus, though perhaps in no case can this evidence be 

 regarded as wholly conclusive. As regards the other four pos- 

 sible influences, the evidence is inconclusive or is quite negative. 



To the foregoing list of possible factors affecting the sex ratio 

 we may add an eighth, namely the year. Surprising and inex- 

 plicable as the fact may be, we have found large and sometimes 

 significant differences between the sex ratios for certain of the 

 years covered by our observations. Indeed these differences are 



