§ 64. Deductive interpretation of the phenomena of Repair, is by no means so easy. The tendency displayed by an animal organism, as well as by each of its organs, to return to a state of integrity by the assimilation of new matter, when it has undergone the waste consequent on activity, is a tendency which is not manifestly deducible from first principles; though it appears to be in harmony with them. If in the blood there existed ready-formed units exactly like in kind to those of which each organ consists, the sorting of these units, ending in the union of each kind with already existing groups of the same kind, would be merely a good example of Segregation (_First Principles_, § 163). It would be analogous to the process by which, from a mixed solution of salts, there are, after an interval, deposited separate masses of these salts in the shape of different crystals. But as already said (§ 54), though the selective assimilation by which the repair of organs is effected, may result in part from an action of this kind, the facts cannot be thus wholly accounted for; since organs are in part made up of units which do not exist as such in the circulating fluids. We must suppose that, as suggested in § 54, groups of compound units have a certain power of moulding adjacent fit materials into units of their own form. Let us see whether there is not reason to think such a power exists. "The poison of small-pox or of scarlatina," remarks Mr. (now Sir James) Paget, "being once added to the blood, presently affects the composition of the whole: the disease pursues its course, and, if recovery ensue, the blood will seem to have returned to its previous condition: yet it is not as it was before; for now the same poison may be added to it with impunity." ... "The change once effected, may be maintained through life. And herein seems to be a proof of the assimilative force in the blood: for there seems no other mode of explaining these cases than by admitting that the altered particles have the power of assimilating to themselves all those by which they are being replaced: in other words, all the blood that is formed after such a disease deviates from the natural composition, so far as to acquire the peculiarity engendered by the disease: it is formed according to the altered model." Now if the compound molecules of the blood, or of an organism considered in the aggregate, have the power of moulding into their own type the matters which they absorb as nutriment; and if they have the power when their type has been changed by disease, of moulding materials afterwards received into the modified type; may we not reasonably suspect that the more or less specialized molecules of each organ have, in like manner, the power of moulding the materials which the blood brings to them into similarly specialized molecules? The one conclusion seems to be a corollary from the other. Such a power cannot be