random and not to constitutional causes." He arrived at the conclusion3 that 60% of the deaths were selective, in the Quaker families which he was then studying. The exact proportion must vary in accordance with the nature of the material and the environment, but as A. Plœtz found at least 60% of the deaths to be selective in the European royal families and nobility, where the environment is uniformly good, there is no reason to think that Professor Pearson's conclusion is invalid.
Dr. Plœtz4 investigated the relation between length of life in parents, and infant mortality, in about 1,000 families including 5,500 children; half of these were from the nobility and half from the peasantry. The results were of the same order in each case, indicating that environment is a much less important factor than many have been wont to suppose. After discussing Professor Pearson's work, he continued:
It seems to me that a simpler result can be reached from our material in the following way. Since the greater child-mortality of each of our classes of children (divided according to the ages at death of their parents) indicates a higher mortality throughout the rest of their lives,