Mr. Spencer finds it hard to believe that the modifications conveyed to offspring are not identical in tendency with the changes effected in the parent by altered use or habit (pp. 23-25, 34). But it is perfectly certain that the two sets of effects do not necessarily correspond. The effect of changed habits or conditions on the individual is often very far from coinciding with the effects on the reproductive elements or the transmissible type. The reproductive system is "extremely sensitive" to very slight changes, and is often powerfully affected by circumstances which otherwise have little effect on the individual (_Origin of Species_, p. 7). Various animals and plants become sterile when domesticated or supplied with too much nourishment. The native Tasmanians have already become extinct from sterility caused by greatly changed diet and habits. If, as Mr. Spencer teaches, continued culture and brain-work will in time produce lessened fertility or comparative sterility, we may yet have to be careful that intellectual development does not become a species of suicide, and that the culture of the race does not mean its extinction--or at least the extinction of those most susceptible of culture. The reproductive elements are also disturbed and modified in innumerable minor ways. Changed conditions or habits tend to produce a general "plasticity" of type, the "indefinite variability" thus caused being apparently irrelevant to the change, if any, in the individual.[14] A vast number of variations of structure have certainly arisen independently of similar parental modification as the preliminary. Whatever first caused these "spontaneous" congenital variations affected the reproductive elements quite differently from the individual. "When a new peculiarity first appears we can never predict whether it will be inherited." Many varieties of plants only keep true from shoots, and not from seed, which is by no means acted on in the same way as the individual plant. Seeing that such plants have _two_ reproductive types, both constant, it is evident that these cannot both be modified in the same way as the parent is modified. Many parental modifications of structure and habit are certainly not conveyed to neuter ants and bees; other modifications, which are not seen in the parents, being conveyed instead. Many other circumstances tend to show that the individual and the transmissible type are independent of each other so far as modifications of parts are concerned. It may seem natural to expect the transmission of an enlarged muscle or a cultivated brain, but, on the other hand, why should it be unreasonable to expect that a modification which was non-congenital in origin should still remain non-congenital? Why should the non-transmission of that which was not transmitted be surprising?