We may here notice that, in most of the tricks which animals are taught to perform, the action is suggested by a form of words (or the tone and manner in which they are uttered). Mr. John G. Naish, J.P., of Ilfracombe,[GQ] has taught his cockatoo the following trick (I quote Mr. Naish's own words): "I give him a shilling, which he puts into the slit of a money-box. This is 'enlisting.' After that, I say to him, 'Will you die for the queen, like a loyal soldier?' Then he lies on his back, with his paws together, for as long as I hold up my finger. 'Now live for your master!' He takes hold of my finger and resumes his erect posture. Last year I took him into the street near my house, and collected on our 'Hospital Saturday.' He worked for more than an hour before he became impatient. And then he would do no more, but flung the coins over his head or at the giver in the funniest way. He went to sleep for a long time after that performance; and when he awoke and I took him, he covered my face with kisses, as if he was glad to find his bad dream was over." The weariness and failure to perform the trick when tired, and the long sleep which succeeded, are interesting points. What I wish especially to notice is, however, that the actions are suggested by certain forms of words; but that there is no evidence that the form of words is in any sense understood. When the onlooker sees a bird lie on its back when asked if it will die for the queen, and get up again when told to live for its master, he is apt to think that, since _he_ understands the form of words, the bird must understand them too. But I am convinced that Mr. Naish's intelligent cockatoo could have been taught with equal ease to lie down at the command "Abracadabra," and to stand up again at "Hocus pocus." Tricks taught to animals involve the performing animal and the human onlooker. The form of words introduced is _for the sake of the latter_, not for the sake of the former.