one has ever tried to teach an animal to do this, we know that trained dogs will display an even better understanding of words by means of appropriate gestures.[135] The other point which has to be noticed in connection with these early stages of indicative sign-making in the young child is that, sooner or later, they begin to overlap the earliest stage of articulate sign-making, or verbal denotation. In other words, denotative sign-making never begins to occur until indicative sign-making has advanced considerably; and when denotative sign-making does begin, it advances parallel with indicative: that is to say, both kinds of sign-making then proceed to develop simultaneously. But when the vocabulary of denotation has been sufficiently enriched to enable the child to dispense with the less efficient material furnished by indication, indicative signs gradually become starved out by denotative, and words replace gestures. * * * * * So far, then, as the earliest or indicative phase of language is concerned, no difference even of degree can be alleged between the infant and the animal. Neither can any such difference be alleged with respect to the earliest exhibitions of the next phases of language, namely, the denotative and receptually connotative. For we have seen that the only animals which happen to be capable of imitating articulate sounds will use these sounds with a truly denotative significance. Moreover, as we have also seen, within moderate limits they will even extend such denotative significance to other objects seen to belong to the same class or kind—thus raising the originally denotative sign to an incipiently connotative value. And although these receptually connotative powers of a parrot are soon surpassed by those of a young child, we have further seen that this is merely owing to the rapid advance in the _degree_ of receptual life which takes place in the latter—or, in other words, that if a parrot resembled a dog in being able to see the resemblance between objects and their pictures, and also in being so much more able to understand the meanings of words, then, without doubt, their connotative extension of names would proceed further than it does; and hence in this matter the parallel between a parrot and child would proceed further than it does. The only reason, therefore, why a child thus gradually surpasses a parrot in the matter of connotation, is because the receptual life of a child gradually rises to that of a dog—as I have already proved by showing that the indicative or gesture-signs used by a child after it has thus surpassed the parrot, are psychologically identical with those which are used by a dog. Moreover, where denotation is late in beginning and slow in developing—as in the case of my own daughter—these indicative signs admit, as we have seen, of becoming much more highly perfected, so that under these circumstances a child of two years will perform