Correlation o/ Brain Capacity and Intelligence. 241 



We have already shown, as fairly as we can, that on this point 

 there is a marked diverj^ente of opinion, and we now proi)ose to 

 examine the facts from both the medical and the biometric side with 

 a view to determining how far the present research tends to 

 harmonise the undoubtedly conflicting opinions on the subject. Witli 

 this object in view we shall first submit the results of the present 

 work and the selected objects of comparison in a table wherein are 

 shown the true means of the estimated cubic capacities with their 

 probable errors, the standard deviation of the same with their 

 probable errors, as also the extreme minimum and maximum figures 

 in every class where they are known to us. 



Concerning this last, Udny Yule (15) has written, " The simplest 

 possible measure of the dispersion of a series of values of a variable 

 is the actual range, i.e., the difference between the greatest and 

 least values observed. While this is frequently quoted, it is as a 

 nde the worst of all possible measures for any serious purpose. 

 There are seldom real upper and lower limits to the possible values 

 of the variable, very large or very small values being only moie 

 or less infrequent; the range is. tliei-cfore, subject to meaningless 

 fluctuations of considerable magnitude according as values of greater 

 or less infrequency happen to have been actually observed." 



In the table which follows, Yule's objection, the very pioper one 

 jc)f the mathematician, is met by the inclusion of the standard devia- 

 tion, and the individual range of variation is retained for reasons 

 which appeal strongly to the medical man on medical grounds alone. 



2'ahli'. of true mfiam^, standard deviations, probable errors and in- 

 dividual range of variation of 355 criminals and other classes oJ 



comparison. 



