SPECIAL, REPORT FOR YEAR. 21 



sex determination in lower animals like the frog and the toad 

 have tended to show that in its essentials Thury's idea was 

 probably right. In these lower animals it has in some cases 

 been possible to control sex absolutely, that is, to produce all 

 the offspring of one sex. 



The statistics regarding sex determination in cattle collected 

 at the Maine Agricultural Experiment Station are the most 

 extensive which have yet been available to test this matter. 

 Putting together all the authentic evidence which has been col- 

 lected at the Maine Station and at several experiment stations 

 in Germany the relations are as shown in the following table : 



Number of Male 



Calves to every 



Time of Service. Sex of Calf. lOO Female Calves. 



Early in heat 134 male 178 female 75.3 



Middle of heat 67 male 58 female ii5-5 



Late in heat .......... y] male 44 female i75-0 



Totals 278 280 



These figures comprise 558 distinct breeding operations to 

 test the matter. They show that when the service was early 

 in the heat there were 133 heifers to every 100 bull calves (or 

 put the other way about there were 75 bull calves to every 100 

 heifers). When the service was late in the heat this relation 

 was reversed. There were then 175 bulls born to every 100 

 heifers. These figures have been subjected to the most refined 

 mathematical tests, applied to determine whether they are to be 

 regarded as accidental or as representing a real and definite 

 law of sex determination. The results point with a high degree 

 of probability to the latter conclusion. 



The position of the Experiment Station in regard to the 

 results set forth above should be clearly understood. It is not 

 contended or supposed that the time of service in relation to the 

 period of heat absolutely controls the sex of the subsequent 

 offspring. It is believed, however, that the facts show, with a 

 considerable degree of probability, that the sex ratio in cattle 

 can be to some extent modified by controlling the time of ser- 

 vice. But the amount of such observed modification is not so 

 great that the matter can be tested with a few individuals. 



