was actually differentiated (say out of pre-existing neuroglia, though that is a hypothetical matter of detail) in the very act of performing the trick in question. There will be no doubt whether the new fibers are related to the arithmetical faculty or to the Sanskrit verbs, because they were actually rendered possible as nervous tracks in the act of learning decimal fractions. It is true, we may admit to the utmost the intense complexity of the existing brain, and the vast number of its elements involved in even the simplest muscular adjustment or the simplest visual perception. Nobody feels the necessity for admitting such complexity more fully than myself. One may allow with M. Ribot that every act of thought must be conceived rather as a vast dynamical tremor, affecting a wide plexus of very diverse nerve-elements, than as a single function in a single cell or fiber. One may acknowledge that what one ought really to picture to one's self (at the present stage of human evolution) is not so much the genesis of a new cell for governing the little finger, or of a new fiber for understanding a fact in decimal fractions, as the habituating an immense series of cells and fibers, perhaps in various parts of the brain, to thrill together in unison on the occurrence of a single cue. But let us thus purify and dematerialize our conception as far as we like, we must nevertheless come back at last to the fact that every gain implies a modification in structure, and that this modification in structure, if it is to have any functional meaning and value whatsoever, must be functionally brought about.
That such functional modifications are forever taking place in all of us is a matter of common observation, as evidenced by psychological facts. We are always seeing something which adds to our total stock of memories; we are always learning and doing something new. The vast majority of these experiences are similar in kind to those already passed through by our ancestors; they add nothing to the inheritance of the race. To use a familiar phrase in a slightly new and narrower sense, they do not help to build up "forms of thought"; though they leave physical traces on the individual, they do not so far affect the underlying organization of the brain as to make the development of after-brains somewhat different from previous ones. But there are certain functional activities which do tend so to alter the development of after-brains; certain novel or sustained activities which apparently result in the production of new correlated brain-elements or brain-connections, hereditarily transmissible as increased potentialities of similar activity in the offspring. If this is not so, then there is no meaning at all in the facts collected by Mr. Galton, or, indeed, for the matter of that, in the common facts of human experience as to hereditary transmission of faculties for acquired pursuits of any sort. If the children of acrobats make the best tumblers, if the descendants of musical families make the best singers and composers, if a great thinker or a great painter is usually produced by the convergence of