Sept. IS, I9I9 Variation in Milk of Ayrshire Cows 303 



eliminated. The observed variation must, therefore, be due to the com- 

 bined action of all the external environmental influences which affect in 

 greater or less degree the milk yield of every cow. 



On the other hand, the constants of variation for milk yield determined 

 in this paper are based upon the diversity or variation in weekly yield 

 exhibited among a large number of different cows. Here one primary fac- 

 tor in the causation of the observed variation must be the individuality of 

 the animal with respect to milking ability. By individuality in this sense 

 is meant the genotype of the individual with regard to the character named. 

 But in the causation of the variation in milk yield as here discussed there 

 must be involved the combined influence of the individuality of the 

 animal plus that of all the environmental factors which act in producing 

 variation in the mixed milk of the herd, since each of these causes influ- 

 ences every individual animal while it is making its individual record. 



It is therefore possible to make comparison here between observed 

 variations (as measured by the coefficient), due, on the one hand, to en- 

 vironmental influences alone and, on the other hand, to genotypic differ- 

 ences plus environmental influences. The difference should represent 

 in a general way that part of the observed variation due to genotypic 

 differences. 



The figures as they stand suggest that roughly about one-half of the 

 variation (measured by the coefficients of variation) in milk production 

 results from the varying genotypic individuality of the animals with re- 

 spect to this character, and the other half results from the varying exter- 

 nal circumstances to which cows are subjected during lactation and 

 which have an effect upon the flow of milk. Or, to put the matter in 

 another way, if the conclusion just stated were true it would mean that 

 if a large number of cows were placed in environmental circumstances 

 which were at once ideal and uniform we should expect the variation 

 exhibited in milk production to be roughly about one-half of that which 

 we actually find when we measure this variation under ordinary cir- 

 cumstances. 



Another point of interest in connection with Table VI is the com- 

 parison of the coefficients of variation for milk yield with that for the 

 weight of the albumen of the &%g of the domestic fowl. Both of these 

 are secreted products. The weight of the shell of the &gg is another 

 character falling in the same category. The figures here given indicate 

 that the variation in these characters, taken in relation to their respec- 

 tive means, is greater for milk secretion than for albumen or shell secre- 

 tion. Or, put in another way, the oviduct as a secretory organ appears 

 to work truer to type than does the udder of the cow. This result is 

 what would be expected from all that is known of the physiology of the 

 two organs. The secretory activity of the cow's udder is apparently 

 very much more easily influenced by external circumstances and by 

 nervous impulses than is the oviduct of the fowl. 

 122502°— 19 5 



