We shall finish this part of our studies by mentioning a little more explicitly one fundamental fact which has already entered incidentally into our considerations, viz. *retro-* or *back-differentiation*.[74] We know that it occurs in *Clavellina* and in *Tubularia*; we may add that it also happens in *Hydra*, and that in the flatworm *Planaria* the pharynx, if it is too large for a piece that is cut out, may be differentiated back and be replaced by a new pharynx, which is smaller. [74] “Retro”-differentiation, of course, is not “Re”-differentiation (“Umdifferenzierung,” see p. 111), though it may help it to occur. It is not death and sloughing of parts that occurs in these cases,[75] but a real process of active morphogenesis; not, however, a process consisting in the production of visible manifoldness, but the opposite. Loeb was the first to lay much stress upon this topic, and indeed, there may appear a very strange problem in its wake: the problem, whether *all* morphogenesis might be capable perhaps of going backwards under certain conditions. [75] Of course such a real decay of parts may happen in other cases. It is important to note that in most[76] cases retro-differentiation occurs in the service of restitution: it goes on wherever restitution requires it. This fact alone would show that not very much could be explained here by the discovery of modern chemistry, important as it is, that one and the same “ferment” or “enzyme” may affect both the composition and the decomposition of the same compound. We could regard what is called “catalysis” solely as an agent in the service of entelechy. But this point also will become clearer in another part of the work. [76] Certain cases of retro-differentiation occurring under conditions of strict fasting will be described in a later chapter. *C.* ADAPTATION INTRODUCTORY REMARKS ON REGULATIONS IN GENERAL