290 THE TREND OF THE RACE 
made to separate the effect of the children’s use of tobacco and 
liquor, which he claims are deplorably prevalent, from the effects 
possibly due to heredity. The papers of MacNicholl belong to 
that very large class of literature on the hereditary influence of 
alcohol which neglects nearly all of the elementary precautions 
which are absolutely essential for attaining reliable results. 
From the kind of data we have on the hereditary effects of 
alcohol in human beings it is difficult to come to any positive 
conclusion. And there is a much less confident tone in the utter- 
ances on this subject among more recent authorities on heredity 
than there was several years ago. It is commonly recognized that 
in certain families there is a bent toward alcoholism. This no 
more proves that such a trait is the result of the liquor habit than 
the reappearance of kleptomania proves that this failing is the 
result of parental thieving. What caused the original appearance 
of the bent toward alcoholism we do not know. Neither do we 
know in most cases what causes the first appearance of feeble- 
mindedness and the hereditary forms of epilepsy and insanity. 
When the attempt is made to follow the history of these maladies 
we usually uproot a strain of defective inheritance which runs 
back and back farther than we can traceit. The Jukes, the Tribe 
of Ishmael, the Kallikak family, the Zero family and the Nam 
family all have much the same melancholy sort of history. All 
show alcoholism and degeneracy going hand in hand. It is 
reasonably certain that much alcoholism is the product of degen- 
eration. That it is a common cause of the first appearance of 
degenerate strains is of course possible, if not probable. But 
our present knowledge of the subject does not justify us in assert- 
ing that such a conclusion is anything more than a good working 
hypothesis. 
There is no question in eugenics more important than that of 
the origin of defective strains of human beings. How much light 
might be thrown on the problem by statistical investigation, if 
undertaken in the right way, I shall not presume to predict, but 
so far as the hereditary influence of alcohol is concerned the most 
promising method consists in experiments on animals. In this 
