time and by this sober husband again produced sound children."
Although such evidence is at first sight pertinent, it lacks much of being convincing. Much must be known about the ancestry of the drunken husband, and of the woman herself, before it can be certain that the defective children owe their defect to alcoholism rather than to heredity.
We can not undertake to review all the literature of this subject, for it fills volumes, but we shall refer to a few of the studies which are commonly cited, by the believers in the racial-poison character of alcohol, as being the most weighty.
Taav Laitinen of Helsingfors secured information from the parents of 2,125 babies, who agreed to weigh their infants once a month for the first eight months after birth, and who also furnished information about their own drinking habits. His conclusion is that the average weight of the abstainer's child is greater at birth, that these children develop more rapidly during the first eight months than do the children of the moderate drinker, and that the latter exceed in the same way the children of the heavier drinker. But a careful analysis of his work by Karl Pearson, whose great ability in handling statistics has thrown light on many dark places in the alcohol problem, shows9 that Professor Laitinen's statistical methods were so faulty that no weight can be attached to his conclusions. Furthermore, he appears to have mixed various social classes and races together without distinction; and he has made no distinction between parents, one of whom drank, and parents, both of whom drank. Yet, this distinction, as we have pointed out, is a critical one for such inquiries. Professor Laitinen's paper, according to one believer in racial poisons, "surpasses in magnitude and precision all the many studies of this subject which have proved the relation between drink and degeneracy." As a fact, it proves nothing of the sort as to