There are also some carefully controlled experiments on mam mals in which malformations were produced and were inherited. The results can be explained as due to direct action on the germ cells of the young in utero, and in fact the experimenters have so interpreted them. It may be significant that these types are all de fective. It was reported by Stockard that, after prolonged treat ment of adult guinea pigs with the fumes of strong alcohol, the lenses of the eyes become opaque. Some of the offspring may have defective eyes, as well as other abnormalities, and defects may appear in some of their offspring; but again there is no specific relation between the defects in parent and offspring. Bagg has treated pregnant mice with radium emanations. Some of the young may have defective kidneys or abnormal extremities or other defects, and such defects continue to appear in some of their offspring. Amongst these, certain individuals that appear to be normal may also produce some defective offspring. Here again, however, it is not certain whether the defects may not have been latent in the untreated stock. All these results appear to be due to a direct effect on the germ-cells and not to be due to the transmission from the body to the germ-cells. Muller's recent ex periments with Drosophila treated with radium, in which defects, and even mutant types of a specific kind, are produced and in herited, appear also to be due to the direct action of the radium on the germ-cells. Some of the mutant types thus produced were identical with those which had occurred "spontaneously" in earlier experiments. (See MUTATION.) Finally, the most complete disproof of the inheritance of som atic influence is demonstrated in almost every experiment in gen etics. When an individual with a dominant character is mated to one with a recessive character, all the offspring show the dominant character, in some cases in full force, in others less completely.
When the hybrid is bred back to the recessive stock half of the offspring show the dominant character, half the recessive. This is the expected ratio if half the ripe germ-cells of the hybrid carry the dominant, half the recessive element. This result could not happen if the bodily characters (dominant) of the hybrid pro duced a sympathetic effect on the germ-cells. Furthermore, it is possible to breed continuously only from hybrid forms—a com mon procedure in certain Mendelian work—yet when after many generations the stock is tested, the dominant character has never been found to have affected the recessive elements in the germ material. It is surprising that this critical evidence is seldom re ferred to by the advocates of the inheritance of acquired char acters. Here, then, in the only field of heredity where we have really scientific evidence, the facts are positive and unquestioned, and contradict thoroughly the claim that the germ-cells are affected specifically by the character of the individual.
The social evolution of man has obviously come about by the transmission of the experience of one generation to the next by means of oral and written instruction. Our thinking is so saturated with this point of view that it was natural to extend it to the bodily structures and behaviour of lower animals. Instincts have, in fact, been sometimes defined as inherited habits, which implies the Lamarckian theory. Experience has shown, however, that it is unwise to apply the evidence from one domain to another where an entirely different set of relations is known to exist.