It is true that the student of Mendelian heredity does not often trouble himself about the nature of the character that he studies. He is concerned rather with its mode of inheritance. But the geneticist knows that opposed to each defect-producing element in the germ-plasm there is a normal partner of that element which we call its allelomorph. We can not study the inheritance of one member of such a pair of genes without at the same time studying the other. Hence whatever we learn about those hereditary elements that stand for defects, we learn just as much about the behavior of the normal partners of those elements. In a word, heredity is not confined to a study of the shuffling of those genes that produce abnormal forms, but is equally concerned with what is going on when normal genes are redistributed. This method of pitting one gene against the other furnishes the only kind of information relating to heredity about which we have precise knowledge. In man and in domesticated animals we find that individuals appear occasionally that are defective in one or another respect. Some of the defects are inherited. Rarely a new one appears that has not been seen before. But the majority of them are reappearances of characters that have been carried under the surface as recessive genes in the germ-plasm. Today we recognize that each of these modifications, if recessive, has first arisen as a mutational change in a single gene before it appeared on the surface as a character by the coming together of two such genes. Mendelism has furnished some information as to the way in which these hidden genes may get dispersed in the race. An example will serve to make this clear, Fig. 1. If a fly with vestigial wings, a recessive character, is crossed to a wild fly with long wings, all the offspring (F₁’s) will have long wings. If these are bred to each other the offspring will be of two kinds, like their grandparents, in the ratio of three long winged to one vestigial fly. The extracted vestigials will breed true to vestigial. The fact that the gene for vestigial has been carried by long winged F₁ parents has not affected the gene in any way, for the second generation of vestigials has wings as short as those of their grandparents. I have brought forward this case not so much to illustrate Mendel’s law of segregation as to use the facts for another purpose. [Illustration: FIG. 1. Cross between long-winged (wild type) _Drosophila melanogaster_ and vestigial-winged fly, producing long-winged offspring (F₁), which if bred to each other give in the next generation 3 long to 1 vestigial. In the middle of the diagram, the pair of chromosomes that are involved in this cross are represented. The chromosome with the factor (gene) for long wings is here black; that for vestigial is open (v).]