That somewhere, and somehow, such nuclear-protoplasm as induces the changes in the characters of hybrids, etc., has been influenced by its immediate environment--the cell-protoplasm of the plant--appears to be a conclusion from which there is no escape. We may obtain similar evidence from the experience of grafting. It is relatively easy to influence the cell-protoplasm of a scion by a suitable stock, obviously because the latter, while handing on to the former all necessary materials from the soil, presents the indispensable elements and compounds in somewhat different proportions, dilutions, etc., from those which its own roots would have done, and probably mingles with them a certain amount of its own peculiar products, as well as affects the modes of working and interaction of both by the molecular impetus impressed on them. Consequently the cell-protoplasm of the scion, while obtaining from the stock all it needs within the limits of its own variations of structure and activity, nevertheless builds up and breaks down in ways or at rates slightly different from those hitherto normal to it, and perceptible variations result when the sequences and correlations of these material and mechanical changes have affected a sufficiently large mass for the accumulation of visible effects. The limits to grafting suggest not that an inappropriate stock does not offer to the protoplasm of the scion the right materials, but that it presents them in proportions and in forms which are unsuitable for the assimilable powers of the latter, or, possibly, mingled with substances poisonous in themselves or capable of becoming so in conjunction with bodies in the scion. What has been said of the action of stock on scion, will also be true, _mutatis mutandis_, of the reciprocal action of scion on stock. Here again we may have causes for disease, or predisposition to disease. It occasionally happens, however, that the nuclear protoplasm of the stock or scion _is_ affected in grafting, and we infer from the difficulty of modifying it in any other way in ordinary reproduction than by means of other nuclear protoplasm--_e.g._ in hybridisation--that in such cases a fusion of the nuclei of stock and scion has occurred during the grafting, and a graft-hybrid has resulted--_e.g._ _Cytisus Adami_. It is not impossible however that the nuclear protoplasm has in such graft-hybrids been subsequently modified by the differences in nutrition to which it has been subjected, in the modified cell-protoplasm affected by the mingling of the juices, etc., of scion and stock; for it is quite conceivable that such materials may affect the protoplasm far more profoundly than anything derived directly from the environment.