§ 304. These interpretations, which have already occupied too large a space, must here be closed. Of course out of phenomena so multitudinous and varied, it has been impracticable to deal with any but the most important; and it has been practicable to deal with these only in a general way. Much, however, as remains to be explained, I think the possibility of tracing, in so many cases, the actions to which these internal differentiations may rationally be ascribed, makes it likely that the remaining internal differentiations are due to kindred actions. We find evidence that, in more cases than seemed probable, these actions produce their effects directly on the individual; and that the unlikenesses are produced by accumulation of such effects from generation to generation. While for all the other unlikenesses, we have, as an adequate cause, the indirect effects wrought by the survival, generation after generation, of the individuals in which favourable variations have occurred--variations such as those of which human anatomy furnishes endless instances. Thus accounting for so much, we may not unreasonably presume that these co-operative processes of direct and indirect equilibration will account for what remains. * * * * * [NOTE.--After having dismissed this revised chapter as done with, and sent it to the printer, further thought concerning those differentiations which produce bone, has reminded me of a fact of extreme and varied significance named in the first volume. I refer to the formation of adaptive structures round the ends of dislocated bones, and to the formation of “false joints.” These are ontogenetic changes of which phylogeny yields no explanation. They do not repeat the traits of ancestral organisms, and they cannot be ascribed to either of the recognized evolutionary factors. If a humerus be broken across and, failing to set, presently comes to have its two loose ends so modified as in a measure to simulate the parts of a normal joint--the ends becoming smooth, covered with periosteum and supplied with fibrous tissue, and attached by ligaments in such ways as to allow of restrained movements--it is impossible to think that natural selection has had anything to do with the power of adjustment thus shown. No survival of individuals in which adaptations of this kind, now in one place and now in another, were better and better effected, could account for acquirement of the ability. Nor can it be supposed that the ability might result from a functionally-produced habit; since it is scarcely conceivable that the number of cases in which individuals profited by it (at first a little and gradually more) could be such (even did they survive) as to affect the constitution of the species. Both of the alleged causes of structural modifications are out of court. It is manifest, too, that the foregoing hypothesis respecting bone-formation yields us not the slightest help.