Let us now try to take a short survey of all the regulations discovered relating to the substitution of one kind of food for another. We have said that food serves in the first place as building material, in the second place as fuel. It only deserves brief mention that, as all recent investigations have shown, fats, carbohydrates, and albumen are equally well able to serve as fuel.[103] [103] But nevertheless albumen is not to be replaced altogether in vertebrates by fat or carbohydrate; it probably serves some special function besides combustion, even in the adult. It is in the state of fasting, *i.e.* in the case of a real absence of *all* nourishing materials, that the organism has proved to be capable of regulations of the most marked nature, with regard to the combustion of its own materials. Respiration, we know, must go on if death is to be avoided, and now indeed it has been found that this process attacks the different tissues of the organism subjected to fasting in such an order that, after the combustion of the reserves, the most unimportant tissues with regard to life in general are destroyed first, the most important ones last. Thus in vertebrates the nerve cells and the heart are preserved as long as possible; in infusoria it is the nucleus; in flatworms, as the very careful studies of E. Schultz[104] have lately shown, it is the nerve cells and the sexual cells which longest resist destruction, whilst almost all the rest of the organisation of these animals may disappear. I should not say that we can do very much with these facts at present in our theoretical discussion, but they are certainly witness of very astonishing adaptive powers.[105] [104] *Arch. Entw. Mech.* 18, 1904. [105] To a physiological friend of mine I owe the suggestion that it is the permanently functioning tissues which stand hunger better than the others, at least if the sexual cells might be regarded as capable of a *sécrétion interne* in all cases. Then the adaptations in the state of hunger might be said to be reduced in some degree to “functional adaptation.” But it must remain an open question, it seems to me, whether such a view may indeed hold in the face of the facts observed in *Planaria* and infusorians.