In man there are no certain cases known of lethal dominants unless some of the short-fingered types come under this heading. Dominant and recessive characters have been so much discussed in modern Mendelian literature that it is popularly supposed that all Mendelian characters must be either dominant or recessive when bred to the type. This is not the case. The hybrid (or heterozygote) is frequently intermediate. In fact, it might be said, almost without exaggeration, that the heterozygote nearly always shows some traces of its double origin. Sometimes the hybrid character is nearly midway between the parent types, sometimes more like one, or like the other. The important fact, however, is that in the germ cell of such intermediate hybrids, there is the same clean separation of the parental genes. In consequence, we find in the second generation the two grandparental types in pure form and an array of intermediates connecting them. [Illustration: FIG. 4. Yellow mouse (YB) crossed to yellow mouse (YB) produces here black and yellow offspring, in the ratio of 2:1. These yellow are again hybrid, and if bred to each other give the same result again. Pure yellow (YY) offspring die at early stage. They constitute one quarter of all the offspring.] In connection with the question of spreading of mutant genes in the race there is another consideration, seldom referred to, that may occasionally have some weight in accounting for the dispersal of genes. In _some_ combinations the hybrid may be more vigorous and more fertile than either parental race. Hence it may have a better chance of survival than an individual of either parent stock. It is a difficult question, that we cannot answer at present, whether a mixed strain has a better chance of survival than one or another of the strains of which it is made up. The possibility that some hybrid strains may be better than either pure strain is enough to put one on his guard against the popular doctrine of racial purity so-called. Whatever advantages some kinds of pure races of mankind may have, from a political, religious or militaristic viewpoint, this should not blind us to the possibility of the biological advantages that certain mixtures may bring about. I emphasize the statement that _certain_ mixtures of races _may_ have a biological advantage. It is equally possible that other combinations may have a biological disadvantage. We are far from being able to state at present what combinations are beneficial and what are biologically injurious. It is an interesting problem, one of deep significance I think for the future of the human race, but mixed up as it is at present with difficult social and political questions it is a problem that only a light-hearted amateur or a politician is likely to be dogmatic about.