158 Essay on the Interpretation of Milk Records. 



in the revised maximum, but this is not necessarily a point in 

 the latter's favour. The variation in skewness of physiological 

 characters is generally considerable, and at present seems more 

 or less mysterious. 



The value of the modal coefficient for totals shows that 

 about 25 per cent., or 1 in 4, of individuals conformed to the 

 dominant yield-type. 



When we come to compare the coeflBcients of variability 

 and the curves themselves, it is seen that, for normal lactation 

 periods, all the above figures, namely lactation totals, averages 

 fifth to twelfth week, revised maxima, and maxima show almost 

 identical variation, and may be taken to be of equal value as 

 indications of the capabilities of a cow. 



But a figure is required that shall hold good for abnormal 

 lactations also. 



In the following calculations all the 2,665 records pre- 

 viously examined were included, i.e., normal and abnormal, 

 and frequency curves plotted as before. 



Fig. 2.— Frequency Curves : 2,665 Lactations, normal and 

 abnormal. 



Lactation Totals - 

 Averages 

 Maxima 

 Eevised Maxima - 



Totals (gallons) 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 



Averages (quarts) 45 65 8-5 lOo 125 14-5 Wo ISS 205 225 24-5 265 28-5 30'5 32-5 



Maxima „ 1 g.j g.j jq-j jj-s 14o 16o 185 205 225 24-5 265 285 SOo 325 345 



Constants of variation, 2,665 lactations, normal and abnormal. 



