575.2 115 



V 



On the Scientific Measure of Variability 



a HO review the reviewer is always a profitless task, and yet I 

 am tempted to repeat what must be more or less of a failure. 

 In this case, however, the reviewer happens to be a man whose 

 opinion deservedly carries weight, and many readers may consider 

 that he must have fairly epitomised the statements made in my 

 paper on " Variation in Man and Woman." This does not seem to 

 me to be the case ; and, in justice to myself, I wish to distinctly 

 repudiate one or two opinions Professor Weldon fastens upon me 

 {Natural Science, vol. xi., pp. 50-54). 



In the first place, Professor Weldon states that the object of 

 my paper " is to support the contention that women are, on the 

 whole, more variable than men." I wish to entirely disclaim any 

 such object. The paper was written with the purely scientific 

 aim of comparing the variation of man and woman, and was due 

 to the fact that a study of numerous writers on the subject had led 

 me to believe that there was as yet no evidence to show greater 

 variation in one sex than the other ; that most of the reasoning on 

 the subject was invalid and nearly all partizan. I may safely say 

 that the two friends who undertook with me the lengthy arithmetic 

 involved had no " contention " and no bias. We simply thought 

 that no evidence of a satisfactory kind was forthcoming, 

 in the case of man, for Darwin's law of the greater variability 

 of the male ; and we determined, so far as was possible, to 

 undertake a thorough investigation of the question. And what is 

 the general conclusion reached ? That the female is more variable 

 than the male — which is the impression any reader must form 

 from Professor Weldon's review ? Not at all. In the summary 

 I distinctly state that, in the material considered, there is no 

 evidence of greater male variability, but rather of a slightly 

 greater female variability. In the body of the paper it is stated 

 that the less civilised races have nearly equal variability for the 

 two sexes, while, in the more highly civilised, woman — probably 

 owing to the lessening of her struggle for existence as compared 

 with man — has apparently greater variability. I conclude : — " I 

 would ask the reader to note that I do not proclaim the equal 

 variability of the sexes, but merely assert that the present results 

 show that the greater variability often claimed for man remains as 



