224 RAYMOND PEARL AND H. M. PARSHLEY. 



preting the results of this paper. A trial on a half dozen indi- 

 viduals will not in any sense whatever adequately test the 

 accuracy of the results set forth in this paper. Nor will a breed- 

 ing experiment with any other animal than the cow. If any 

 person chooses to generalize from the data of this paper that in 

 animals in general, or indeed in any other animal than the cow, 

 time of service and sex of offspring are causally related, the 

 responsibility for such generalization must be entirely his, not 

 ours. We have studied statistics for cows in regard to this 

 problem, and as yet no others, and have tried to make clear just 

 what these data show. Beyond this solid ground of fact we do 

 not care to venture, particularly so far as concerns the practical 

 application of these results by the breeder. 



Time of service is very evidently no absolute determining 

 factor for sex in cattle. On the other hand the probability that 

 the sex ratio can be changed by careful attention to this matter 

 of time of service is sufficiently great, in our judgment, to warrant 

 any man in modifying his breeding practice in accordance with 

 it, particularly since in so doing he will be incurring no added 

 risk of any kind. In the every-day affairs of life in regard to 

 business, investment of funds, and the like, practical men every 

 day undertake courses of action on the basis of probabilities 

 much smaller than that in favor of getting an increased number 

 of males if cows are served late in the heat. The practical cattle 

 breeder in most cases would like, if he could get it, an excess of 

 female calves. All the evidence at hand warrants the belief 

 that by taking care that cows are served as soon as possible after 

 the onset of heat there will be some reduction in the proportion 

 <of male calves born. In short, the facts set forth in this paper 

 warrant the breeder in paying attention to the time of service in 

 his cattle breeding operations, but he must not suppose that by 

 so doing he can absolutely control the sex of the offspring, or 

 even approach measurably close to absolute control. He can 

 at best merely modify, over a period of years, the sex ratio in 

 greater or less degree, in the direction which he desires. 



SUMMARY. 



In this paper statistics collected some years ago at the Maine 

 Agricultural Experiment Station in regard to the relation between 



