2/6 E. H. HARPER. 



these and for dams of medium age can be seen by inspection of 

 Tables IX. XI. without any farther tabulation of the data. 



We find the following idea expressed by Redfield ('03) in re- 

 spect to the influence of age upon prepotency: "As between 

 two individuals of the same breed, the same rule probably holds, 

 that the individual which has had its characteristics more firmly 

 fixed by inbreeding will be prepotent. In the life of an indi- 

 vidual a character is more firmly fixed in comparative old age 

 than in youth. Consequently we may assume in the absence of 

 evidence to the contrary, that other things being equal, the older 

 individual will be prepotent over the younger one." Redfield's 

 work deals with human statistics and he aims to prove use-in- 

 heritance in the case of acquired mental powers. 



It is apparent of course that, color is not an acquired character 

 except to a limited extent in regard to which the present data 

 show nothing. It is rather obscure as to what may be meant by 

 a character becoming more fixed by age. But that the power 

 of a parent to influence the character of the offspring may be 

 correlated with the time of life and consequent vigor is a matter 

 seemingly independent of whether acquired or congenital char- 

 acters alone may be transmitted. On the whole these data show 

 that the influence of age is confined within narrow limits at least. 



CONCLUSIONS. 



1 . There is a secular change of color in progress in the breed 

 resulting from breeders' preference for black. 



2. Gray, the long-established color, is prepotent over black. 



3. The dam is prepotent over the sire in the ratio of about five 

 to four. Gray dams are more prepotent than black dams and 

 gray sires than black sires ; also gray dams are more pre- 

 potent than gray sires and black dams than black sires. The 

 dam's prepotency is partly due to association with the long pre- 

 dominant color. Gray dams and black sires are greatly in- 

 excess (nearly 75 per cent, of all). 



4. There is apparently a degree of correlation between age and 

 prepotency. 



5. There appears to be an optimum age of prepotency, occur- 

 ring in middle life. 



