The fact that new individual organisms arise from both a paternal and a maternal influence, and from a line of ancestors every one of which had a similar bifold origin, modifies this first law of heredity only so far as to produce a more or less complex compound of hereditary reproductive tendencies in every individual, the effect of which must be analogous to that mechanical law of _the composition of forces_ resulting in the production of a new creature resembling its immediate and more remote progenitors in varying degrees, according to (1) the amount of force springing from each ancestral strain, and (2) the compatibility or incompatibility[26] of the prevailing tendencies, resulting in an intensification, perpetuation, modification, or neutralisation of ancestral characters, as the case may be. [26] Mr. Darwin tells us that two topknotted canaries produce bald offspring, due probably to some conflicting actions analogous to the interference of light. All such action is but “heredity” acting in one or other mode; but there is another and fundamentally different action which has to be considered, and that is the action of the environment upon nascent organisms—an action exercised either directly upon them, or indirectly upon them through its direct action upon their parents. That such actions produce unmistakable effects is notorious. It will be, I think, sufficient here to advert to such cases as the well-known brood-mare covered by a quagga, and the peculiar effects of a well-bred bitch being lined by a mongrel. These show how an action exercised upon the female parent (but with no direct action on the immediate offspring) may act indirectly upon her subsequent progeny. As a rule, modifications accidentally or artificially induced in parents are not transmitted to their offspring, as is well shown by the need of the repetition of circumcision, and of pressure of Indian children’s heads and Chinese girls’ feet, in each generation. Yet there is good evidence that such changes are occasionally inherited. The epileptic offspring of injured guinea-pigs is a case often referred to. Haëckel speaks of a bull which had lost its tail by accident, and which begot entirely tailless calves. With respect to cats,[27] I am indebted to Mr. John Birkett for the knowledge of an instance in which a female with an injured tail produced some stump-tailed kittens in two litters. [27] See _The Cat_ (John Murray, 1881), p. 7. There is evidence that certain variations are more apt to be inherited than others. Amongst those very apt to be inherited are skin affections, affections of the nervous system and of the generative organs, _e.g._ hypospadias and absence of the uterus. The last case is one especially interesting, because it can only be propagated indirectly.