Probably the arboreal progenitor of man was originally compelled to abandon his arboreal habits by some slow change in the flora of his habitat, resulting in the diminution and eventual disappearance of trees suited for his movements. He would thus be compelled to adopt, at first, some such course as the chimpanzee—making huts of such branches and foliage as he could conveniently use for the purpose. The habit of living in large companies would (as in the case of the chimpanzee) become before long necessary, especially if the race or races thus driven from their former abode in the trees were, like the gibbons, unapt when alone both in attack and in defence. One can imagine how the use of vocal signals of various kinds would be of service to the members of these troops, not only in their excursions, but during the work of erecting huts or defences against their enemies. If in two generations the silent wild dog acquires, when brought into the company of domestic dogs, no less than five distinct barking signals, we can well believe that a race much superior in intelligence, and forced by necessity to associate in large bodies, would—in many hundreds of generations, perhaps—acquire a great number of vocal symbols. These at first would express various emotions, as of affection, fear, anxiety, sympathy, and so forth. Other signals would be used to indicate the approach of enemies, or as battle-cries. I can see no reason why gradually the use of particular vocal signs to indicate various objects, animate or inanimate, and various actions, should not follow after a while. And though the possession and use of many, even of many hundreds, of such signs would be very far from even the most imperfect of the languages now employed by savage races, one can perceive the possibility—which is all that at present we can expect to recognize—that out of such systems of vocal signalling a form of language might arise, which, undergoing slow and gradual development, should, in the course of many generations, approach in character the language of the lowest savage races. That from such a beginning language should attain its higher and highest developments is not more wonderful in kind, though much more wonderful, perhaps, in degree, than that from the first imperfect methods of printing should have arisen the highest known developments of the typographic art. The real difficulty lies in conceiving how mere vocal signalling became developed into what can properly be regarded as spoken language.