CERTAIN LAWS OF VARIATION. 205 



Quite recently, Professor Pearson * has shown that 

 in man also the diminution in variability with growth is 

 very marked. Working on measurements of 1000 

 male and 1000 female new-born babies, he obtained 

 the following values for the weight and length : 



MEAN MEAN PER CENT. VARIABILITY 



WEIGHT. LENGTH. WEIGHT. LENGTH. 



Females, 7.073 Ibs. 20.124 in. 14.228 5.849 



Males, 7.301 Ibs. 20.503 in. 15.664 6.500 



Here we see that the male infant at birth is slightly 

 heavier and slightly longer than the female, and also 

 that its relative variability (error of mean square ex- 

 pressed as a percentage on the mean value) is distinctly 

 greater. On working out similar data for 160 female 

 students (mostly from 19 to 25 years of age), and for 

 1000 male students of the same age, Pearson obtained 

 the following values: 



MEAN MEAN PER CENT. VARIABILITY 



WEIGHT. LENGTH. WEIGHT. LENGTH. 



Females, 125.605 Ibs. 63.883 in. 11.170 3.696 



Males, 152. 784 Ibs. 68.863 in. 10.830 3.662 



The woman is now slightly more variable than the 

 man, both in weight and length, but the variability of 

 both is verv much less than at birth. It is, in fact, on 



i/ 7 / 



an average, 26.4 per cent, less as regards weight, and 

 40.4 per cent, less as regards length. 



As variability undergoes such marked diminution 

 with age both in man and the guinea-pig, we may ven- 

 ture to agree with Minot that probably this diminution 

 occurs in the growth of all mammals, and in fact that 



"Proc. Roy. Soc., Ixvi. p. 23, 1900. 



