D. Heron 
367 
by selecting the latter we can alter tlie proportion of any particular type of 
insanity. Dr Mott lias obtained his material in such a way that, in the younger 
generation, cases of insanity coming on late in life are much less likely to be 
recorded than those which appear in early life, and hence the early cases are in a 
majority, but the change in age of onset, and consequently of the type, is entirely 
spurious and arises solely from the way in which the material has been obtained. 
We can now deal with the reply Dr Mott has made to Professor Pearson's 
criticisms. In his Annual Report for 1912 (p. 62), Dr Mott says: "Professor 
Karl Pearson, writing to Nature, November 21, 1912, 'On an apparent fallacy in 
the statistical treatment of " Antedating" in the inheritance of pathological con- 
ditions,' criticises on mathematical grounds the evidence of anticipation. I do not 
feel myself competent to reply to the opinion of such an eminent authority on 
mathematics applied to biometrics, but it does not militate against my conclusions, 
nor explain away the fact that a large proportion of the insane offspring of insane 
parents are affected with imbecility or adolescent insanity ; for granting the 
assumption that there is no antedating at all, we might rightly expect the ages 
at onset of insane offspring of insane parents to be comparable with the ages at 
onset of all the admissions to the asylums during the same period*. This is by no 
means the case, for amongst the insane offspring there is a far greater proportion 
affected early in life, as is shown in the following figures and curves" (they appear 
here as Fig. 4 and Table V). 
According to these figures the onset of insanity among the recorded insane 
offspring of insane parents is considerably earlier than among the general admis- 
sions to asylums, but it has already been shown that this is due to the fact that 
the data have been selected in such a way that the eai-ly cases in the younger 
generation are the most likely to appear. Further, if Dr Mott's argument be a 
valid one, we might also expect the ages at onset of the insane parents of these 
insane offspring to be comparable with the ages at onset of all the admissions to 
asylums during the same period. This is by no means the case as is shown in 
Fig. 5 below (see also Tjibles I and V). We see here that the insanity of the 
parents comes on at a much later period than among the general admissions to 
asylums and that there is a far less proportion affected early in life. If Dr Mott's 
method of argument be sound, he has not only to deal with an antedating of 
insanity among the offspring but also a post-dating of insanity among the parents. 
Both are of course spurious and arise from the peculiar selection of the data and 
from the fact that, owing to differential death-rates, the ages at onset of " admis- 
sions " will never be the same as the ages at onset of the admitted — i.e. the asylum 
population — at any time. 
* "We might rightly expect" these ages to be different, because "admissions" are not the same as 
the population in the country who have at one time or another been insane. The percentages of total 
cases of acute mania, of senile insanity, of congenital idiocy, and of melancholia, who reach the asylums, 
are not the same. The reader has to distinguish between the population of admissions, the population 
of admitted, and the insane population of the country. A sample of the latter may be reached from 
completed family histories, but not from records on admission or from records of an asylum population. 
47—2 
