Nevertheless, the language of tone, like the language of gesture, clearly lies nearer to, and is more immediately expressive of the logic of recepts, than is the language of articulation. This is easily proved by all the facts at our disposal. We know that an infant makes considerable advance in the language of tone and gesture before it begins to speak; and, according to Dr. Scott, who has had a large experience in the instruction of idiotic children, “those to whom there is no hope of teaching more than the merest rudiments of speech, are yet capable of receiving a considerable amount of knowledge by means of signs, and of expressing themselves by them.”[67] Lastly, among savages, it is notorious that tone, gesticulation, and grimace play a much larger part in conversation than they do among ourselves. Indeed, we have some, though not undisputed, evidence to show that in the case of many savages gesticulation is so far a necessary aid to articulation, that the latter without the former is but very imperfectly intelligible. For example, “those who, like the Arapahos, possess a very scanty vocabulary, pronounced in a quasi-intelligible way, can hardly converse with one another in the dark.”[68] And, as Mr. Tylor says, “the array of evidence in favour of the existence of tribes whose language is incomplete without the help of gesture-signs, even for things of ordinary import, is very remarkable.”[C] A fact which, as he very properly adds, “constitutes a telling argument in favour of the theory that the gesture-language is the original utterance of mankind [as it is ontogenetically in the individual man], out of which speech has developed itself more or less fully among different tribes.”[69] In support of the same general conclusions I may here also quote the following excellent remarks from Colonel Mallery’s laborious work on Gesture-language:—[70]