488 



Journal of Agricultural Research 



Vol. IV. No. 6 



The correlation coefficient computed from the data of Table VII (A) is 

 r = o. 0770 ± o. 0074. 



This correlation is significant when judged by its probable error, and 

 it does not differ significantly from the correlation between dams and 

 offspring as found from Table 11(A). 



Table VI II (A) is a correlation table for maternal grandsires and 

 dams. This is the case of the correlation of sires with female offspring 

 with respect to birth as singles, twins, or triplets. The correlation co- 

 efficient is 



r = 0.0066 ± 0.0075. 



This result makes it somewhat probable that there is no correlation 

 between sires and their female offspring with respect to the character in 

 question. We shall investigate this matter further by the separation of 

 offspring with regard to sex. (See Tables XXV and XXXI.) 



Table VIII(A). — Correlation between dams and maternal grandsires 



r=o. 0066 ±0.0075. 



Table VIII(B). — Percentages of dams in states l, 2, and j to correspond to states 

 I, 2, and J of the maternal grandsires 



ANALYSIS OF DATA FOR SIRES AND OFFSPRING 



Tables IX(A) and IX(B) are analogous to Tables 11(A) and 11(B), 

 where the dams are replaced by sires and where certain abnormal cases 

 are excluded in accord with criteria given under " Repetition of sires and 

 paternal grandparents." 



