362 Recent Sttidles of the Inherltante Factor in Insanity 
A quite striking instance of tlie fallacy, if it be snch, would be to consider the antedating of 
" violent deaths." Fully a quarter of such deaths in males, nearly a half in females, occur before 
the age of twenty years. Consider now the parents and offspi'ing who die from violent deaths ; 
clearly there would be no representative of death from violence under twenty in the parent 
generation, and we should have a most marked case of antedating, because the offspring 
generation would contain all the infant deaths from violence. 
In the case of insanity, is the man or woman who develops insanity at an early age as likely 
to become a parent as one who develops it at a later age ? I think there is no doubt as to the 
answer to be given ; those who become insane before twenty-five, even if they recover, are far 
less likely to become parents than those who become insane at late ages — many, indeed, of them 
considering the high death-rate of the insane, will die before they could become parents of large 
families. Now Dr Mott took 508 pairs of parents and offspring, "collected from the records of 
464 insane parents whose 500 insane offspring had also been resident in the County Council 
Asylums," and ascertained the age of first attack. As at present advised, it seems to me that 
his data must indicate a most marked antedating of disease in the offspring, but an antedating 
which is wholly spurious. There is, I think, a further grievous fallacy involved in this method 
of considering the problem, but before discussing that I should like to see if my criticism of this 
method of approaching the problem of antedating can be met. 
KARL PEARSON. 
BioMETRic Laboratory, 
University College, London, 
November 11, 1912. 
Dr Mott has referred to this letter in his Report for 1912*, but it will be more 
convenient to deal with his reply after we have examined the method by which his 
data have been collected and the use made of the data. Let us consider first of all 
how the data were obtained. Dr Mott in desciibiug his material says that it 
consists of a collection of cases in the London County Asylums where two or more 
persons are related to one another. Thus Dr Mott has dealt — not with a series of 
complete pedigrees in which every member is included, whether insane or normal, 
but with a series of cases in which two or more members of a family are known to 
have been in London County Asylums. No notice is taken of those who are 
normal throughout their lives and no allowance is made for those who are normal 
at the time the record is made but who may afterwards become insane. 
Do cases selected in this way provide a complete or impartial view of the facts? 
Some of Dr Mott's own comments on his data throw a considerable amount of 
light on this point. In his Report for 1909f he says: "From all the Asylums 
I have received valuable reports, but in the case of the older asylums it has been a 
matter of the utmost difficulty to trace the records of so many years back," and in 
his Report for 1910| he says, "Some of the asylum authorities have gone through 
their case books for a number of years back, but the results have not be6n 
satisfactory owing to the difficulty of obtaining particulars without a living repre- 
sentative of the family being resident in the asylum — for instance, 110 old cases 
* Annual Report of the London County Council for 1912, Vol. ii. p. G2. 
t Ttventieth Annual Report of the Asylums Committee of the L.C.C., p. 90. 
+ Twenty -first Report of the Asylums Committee of the L.C.C., p. 94. 
