Page:Popular Science Monthly Volume 71.djvu/402

396 the convicts in his Majesty's prisons, serve to increase the total data already available or nearly so. While all eugenics workers crave for more material, and for better quality of material, yet there already exists ample material upon which to base the beginnings for our science.

If we turn from material to method, we note that except in so. far as results for animals have application to man, we can not experiment on individuals, and our methods must, therefore, be those applicable to mass-observations—that is to say, those actuarial methods applied to biological data which we now term the methods of biometry. It is not needful for me to enlarge now or here on those methods. Suffice it to say that they appear to measure effectively the relationship between factors which are not causally linked together. For the explanation of what follows I would state that the arithmetical value of a certain quantity—the so-called coefficient of correlation—is chiefly used to measure this relationship. Starting when the quantities are absolutely independent with zero value, it rises with their complete causal relationship to unity. Table I. shows the sort of values taken by this coefficient for various kinds of association, when the variates lack the absolute dependence of pure causation.

From method I turn finally to illustrate the nature of the conclusions which have already been reached by eugenic inquiry. As a preliminary, I must picture for you what I think evolution means in the case of human societies.

There was a time when, thinking over the marvelous intellectual,