HEREDITY IN MAN 



107 



a 1-5 per cent, basis to the number of insane sons dealt 

 with in the table. 



The column ' numbers to be expected if there were 

 no inheritance ' shows the number to be expected if 

 insane fathers were no more likely than sane fathers to 

 have insane sons, and is calculated on the same 1-5 per 

 cent, basis. Thus there were 198 insane fathers, of whom 

 49 had insane offspring and 149 had sane ones. But if 

 there were no inheritance- that is to sav, if it were a 



^ 7 



pure matter of chance as to whether insane fathers had 

 sane or insane sons, we should expect only 1-5 per cent, 

 of the 198 sons i.e. only 3 to be insane, instead of the 

 49 actually found. 



We are now in a position to consider the Table witfr a 

 view to discovering whether insanity is inheritable or not. 

 If inheritance is at work we should expect to find an 

 excess of cases where father and son are alike, and a 

 deficiency of cases where they are unlike . and on examining 

 the Table we find that this is the case. By the application 

 of appropriate mathematical methods we can measure 

 the intensity of inheritance in the same way as we measured 

 it for stature, and hence compare it with this as a standard. 

 If this be done the correlation is found to be -6 that is 

 to say the records of the Perth asylum disclose an intensity 

 of inheritance of insanity 20 per cent, stronger than the 

 inheritance of stature, though too much reliance must 

 not be placed upon the exact figure. 



A large number of human characteristics can be 

 examined on the same principle. Let us take another 

 mental characteristic, namely Ability. 



In this case a slight modification of the usual method 

 has to be employed. It is very difficult to estimate the 

 relative ability of adults. It is easy, however, to get 



