The discovery of Gruber and others, that Protozoa are only capable of restitution if they contain at least a fragment of the nucleus, has also been used occasionally as a proof of the morphogenetic importance of the nucleus. But might not this absence of restitution where nuclear material is lacking be understood equally well on the hypothesis of Loeb and R. S. Lillie that the nucleus is a centre of oxidation in the cell? Remove the heart from a vertebrate and the animal will not digest any more; but in spite of that the heart is not the organ of digestion. And so we lay stress once more upon this point: that the experimental results of hybridisation and the analytical results obtained by the discussion of the complex-equipotential systems are of greater value to the theory of heredity than all speculation about the importance or unimportance of special constituents of the cell, of whose organisation, chemistry, and physics, scarcely anything is known at present.[137] [137] Boveri (*Ergebn. üb. d. Konstitution etc. des Zellkerns*, Jena, 1904; and “Zellen-Studien VI.” *Jen. Zeitschr.* 43, 1907) has made it highly probable by experiments that the different chromosomes of the nucleus of the sexual products play a different part in morphogenesis, though not in the sense of different single representatives of different single organs. This doctrine, of course, would not alter the whole problem very much: the chromosomes would only be *means* of morphogenesis and nothing else, no matter whether they were of equal or of different formative value. It only is with regard to the problem of the determination of sex (see page 107, note 3), that the morphogenetic singularity of *one* certain specific chromosome can be said to be proved. VARIATION AND MUTATION Heredity, it has been said, may be understood as resting upon the fact that each organism forms its own initial stage again, and that this initial stage always encounters conditions of the same kind. If this statement were quite correct, all the individuals of a given species would be absolutely alike everywhere and for ever. But they are not alike; and that they are not alike everywhere and for ever is not merely the only real foundation of the so-called theory of descent we possess, but also forces us to change a little our definition of heredity, which now proves to have been only a sort of approximation to the truth, convenient for analytical discussion.