Pearson writes: “As to the problem of evolution itself we are learning to see it under a new light. Natural selection, combined with sexual selection [by which Pearson means segregation of certain types through individual selection] and heredity, is actually at work changing types. We have quantitative evidence of its effects in many directions.” Yes! but no evidence that selection of this sort can do anything more than keep up the type to the upper limit attained in each generation by fluctuating variations. Pearson adds, “Variations do not occur accidentally, or in isolated instances; autogamic and assortive mating are realities, and the problem of the near future is not whether Darwinism is a reality, but what is quantitively the rate at which it is working and has worked.” This statement expresses no more than Pearson’s conviction that the process of evolution has taken place by means of selection. He ignores other possibilities, which if established may put the whole question in a very different light. Heredity and Continuous Variation It has been to a certain extent assumed in the preceding pages that both parents are alike, or, if different, that they have an equal influence on the offspring. This may be true in many cases for certain characteristics. Thus a son from a tall father and a short mother may be intermediate in height, or if the father is white and the mother black, the children are mulattoes. But other characters rarely or never blend. In such cases the offspring is more like one or the other parent, in which case the inheritance is said to be exclusive. Thus if one parent has blue eyes and the other black, some of the children may have black eyes and others blue. There are also cases of particular inheritance where there may be patches of color, some like the color of one parent, some like that of the other parent. The latter two kinds of inheritance will be more especially considered in the subsequent part of this chapter; for the present we are here chiefly concerned with blended characters. How much in such cases does each parent contribute to the offspring? This has been expressed by Galton in his law of ancestral heredity. This law takes into account not only the two parents, but also the four grandparents, and the eight great-grandparents, etc. There will be 1024 in the tenth generation. These 1024 individuals may be taken as a fair sample of the general population, provided there has not been much interbreeding. Are we then to look upon the individual as the fused or blended product of the population a few generations back? If this were true, should we not expect to find all the individuals of a community very much alike, except for the fluctuating variations close around the mode?