Stockard’s results with guinea pigs, unlike those of Brown-Sequard, have been done under carefully controlled conditions. He has guarded against abnormalities in his stock by using pedigreed material. The malformations that reappear in successive generations are general rather than specific. Such organs as the eye are those hardest hit, but this is supposed to be rather a by-product of the general debility of the individual. Stockard points out that the alcohol has affected the germ cells, and it is through these that the effects are transmitted. Now if one or more genes had been permanently changed we should expect to have evidence of Mendelian inheritance. The results do not show convincingly that the inheritance is not Mendelian, but it does not appear to be so. There is another possibility. Recent results have shown that rarely entire blocks of genes—pieces of the chromosomes—may be duplicated (owing to imperfect separation) or pieces may be lost. Here the effects on the organism are more far-reaching than when a single gene is changed. It remains to be discovered whether, in some such way as this, Stockard’s remarkable results may be brought into line. Guyer injected the crushed lens of rabbits into fowls. From the blood of the fowl he obtained serum that was injected into pregnant rabbits. The offspring of these rabbits whether male or female often had defective eyes and lenses. The defect was even transmitted to later generations. Here also the germ cells of the embryo may be changed by serum that at the same time affects the development of the eyes of the embryo in utero. If this is the case we should expect, as Guyer pointed out, that the germ cells of the pregnant mother (into which the serum was injected) would also show effects. It should have been a simple matter to show this by a proper test. The test that Guyer made, namely by out-breeding the mother and finding no defective F₁ young, was quite inadequate if, as appears to be the case, the character is a recessive. It is important to keep clearly in mind that there are two distinct questions involved in these three cases. Genetics has to deal with only one of them. There is first the question of the action of environment on the germ cells. Genetics has nothing to do with this question. There is then to be determined whether, if variations may be induced in these ways, they fall into one or another of the Mendelian moulds. This is for the geneticist to determine, but he finds himself in a curious predicament, for it can not be claimed that any of these three cases have been shown to give a direct Mendelian result—but neither can it be denied that they may possibly come under the scheme, or some modification of it. There we must leave the matter at present.