That growling and barking cannot be called a very highly specialized language goes without saying; they are, however, so much diversified in character, according to circumstances, that they place a considerable number of symbols at an animal's command, and he invariably attaches the same symbol to the same idea. A cat never purrs when she is angry, nor spits when she is pleased. When she rubs her head against anyone affectionately it is her symbol for saying that she is very fond of him, and she expects, and usually finds that it will be understood. If she sees her mistress raise her hand as though to pretend to strike her, she knows that it is the symbol her mistress invariably attaches to the idea of sending her away, and as such she accepts it. Granted that the symbols in use among the lower animals are fewer and less highly differentiated than in the case of any known human language, and therefore that animal language is incomparably less subtle and less capable of expressing delicate shades of meaning than our own, these differences are nevertheless only those that exist between highly developed and inchoate language; they do not involve those that distinguish language from no language. They are the differences between the undifferentiated protoplasm of the amoeba and our own complex organization; they are not the differences between life and no life. In animal language as much as in human there is a mind intentionally making use of a symbol accepted by another mind as invariably attached to a certain idea, in order to produce that idea in the mind which it is desired to affect--more briefly, there is a sayer, a sayee, and a covenanted symbol designedly applied. Our own speech is vertebrated and articulated by means of nouns, verbs, and the rules of grammar. A dog's speech is invertebrate, but I do not see how it is possible to deny that it possesses all the essential elements of language.