D. Heron 
357 
data by men who have clearly had no adequate training in statistical science. The 
craniologist, the anthropologist, even the biological student of heredity and evolu- 
tion are recognising that a statistical training is needful for the true interpretation 
of many of the facts in their special fields of research. The physiologist still 
appears to believe that he can deal with the average effects of diverse dietaries or 
the pathologist with the " mass-phenomena " of the hereditary factor in insanity 
without any training in statistical method. A physicist might just as logically 
assume that without mathematical training he could give an adequate mathe- 
matical account of a physical phenomenon, or a cosmic theorist suppose that he 
was effectively furnished for asti'onomical research by the perusal of a popular 
primer on the stars ! The statistical calculus cannot be mastered by any easier 
road than the differential calculus, or, to put a more apt illustration, statistical 
training is as needful a preliminary to the handling of statistics, as time spent 
in a physiological laboratory to the effective handling of tissues. In twenty years 
it will be unnecessary to insist on these points, they will be universally recognised 
in the courts of science ; but at present it is not only necessary to reiterate 
unpleasant truths, but to emphasise their validity by illustrations which bring 
home forcibly to scientist and layman alike the danger of amateur statistical 
handling. To state that a man is in error is not sufficient, if he continues time 
after time to repeat his assertions, apparently under the belief that incessant 
repetition will convince the world of the value of his theories. 
In the case of the inheritance factor in insanity we are not dealing with any 
purely academic question of science. We are up against one of the most difficult 
problems of modern life, where true advice is of urgent importance to the nation 
as well as to the individual. It is not only the medical man but the layman who 
seeks guidance in the question of the marriage of members of insane stocks, and a 
laboratory like the Galton Laboratory knows how often advice on such points is 
sought. It is disheartening when help is rendered to the seeker to be faced with 
the criticism : " But Professor says I may marry if I take a wife of sound 
stock," or "Dr recommends marriage, although my father was insane, because 
I am over twenty-five and still sane myself." When teaching of this kind, arising 
solely from false interpretation of defective data, is spread widecast in a dozen 
different papers or journals, it is not sufficient to issue a brief statement of its 
futility. It is needful to give it the coup de grace by a more lengthy criticism of 
its fallacies and their illustration in a form more likely to impress the imagination. 
The attempt is made in this paper to deal with only one of the authors, who have 
contributed fallacious eugenic rules to those seeking knowledge on the influence of 
the hereditary factor in insanity. 
In a long series of papers Dr F. W. Mott, Pathologist to the London County 
Asylums, has stated that when the children of insane parents become insane, they 
do so at a much earlier age than did their parents, and on the basis of this assertion 
he has drawn some very sweeping conclusions for practical conduct. Thus in the 
Biometrika x 46 
