It may have occurred to some of you that the principles we have been considering are applicable to all forms of mental life-to the animal as well as to the human mind. The principles of recency, novelty, frequency, vividness are all employed, for example, in the training of a performing dog or cat. The canine or feline mind, too, learns undoubtedly to associate certain facts by which the animals habitually regulate their actions it is, in fact, upon a due observance of customary associations that their very existence depends. The dog, for example, can identify its master in a crowd (similarity), and at the sound of a threatening accent or at the appearance of a whip, either fact immediately suggesting punishment (contiguity), will cower with tail between its legs. How far any animal interprets causal sequence as anything beyond mere customary contiguity is a moot question.

We have already said that the human mind has a certain, if limited, power of controlling its associations. Now it is this power of controlling our associations that distinguishes the 'species man from the 'genus' animal: that constitutes man a rational' or a 'thinking' animal.

It will be your main business as teachers to teach your pupils to think. Now, what do we mean by the words 'to think'? From what we have already said it will be clear that 'to think' is not merely to have a succession of associated ideas. If this were true, then we would have to concede that many of the higher animals can think -for example, the dog, the cat, the elephant, the

horse. For undoubtedly these animals have associated ideas, and for that reason they can be trained to perform many actions that are reckoned intelligent. I have read somewhere about a dog which on hearing its master's whistle would immediately leave a room, race down a garden path, open a gate by raising the latch, and join its master in the field. Suppose we take the incident as an authentic occurrence. Let us consider its significance. You will agree that the dog seemed to do exactly what we might expect from a rational human being placed in a similar position-what, for instance, the son of the house might have done. From the analogy of our own minds, we might conceive the following to be the associative elements in the dog's mind. The sound of the whistle recalls the image of the master by simple contiguity. The dog knows by a frequent or habitual conjunction of these two ideas that a pleasurable experience is likely to be in store for him, and automatically he reacts or responds to the stimulus of the whistle, and bounds to meet his master just as the scent of a hare would send him in hot pursuit of the quarry. Now, if it be asserted that all the dog's actions are regulated by exactly the same kind of mental procedure, would you call the process thinking'? It is canine thinking— dog's logic, if you like—and is on the same plane as a good deal of human thinking: for example, of the Flora Finching type. It is what James calls 'habitual' thinking or associative thinking.

But, you ask, what about the latch-lifting?

Surely that was the action of a super-dog!-the result of a higher order of mentality. We are, of course, considering the case of what is commonly called a ' very clever' dog; but even the latch-lifting incident, like all the actions of the dog, must be regarded simply as an example of 'thinking by associative contiguity,' and may be quite simply explained. One day the dog, seeking to escape from the boredom of life in the garden, by a lucky chance hit the latch with his nose-and so released it from the catch! From that moment the excitation of the associated ideas-latch, nose-action, open gate-would result in freedom for the dog, and frequent repetition would soon establish the habit.

Now, is this action on the part of the dog exactly similar to what a normal boy would do under the same circumstances? Boy and dog would behave in a very similar manner in response to the stimulus of the whistle-they would do the same kind of 'habitual' thinking. But when we consider how the boy and the dog set about opening the gate, we find that they think in vastly different ways. The dog's action, we saw, was automatic; the boy's action is rational. The boy may never have seen the gate before, nor the particular sort of latch or bolt that keeps it shut. But former similar associations have taught him that a gate is meant to open. He therefore looks for the obstruction that keeps the gate closed. If the obstruction be a latch, he lifts it; if a bolt, he draws it. If there be both a latch and a bolt, and perhaps a stone laid against the gate, former experiences teach the boy to

remove all the obstacles that prevent the gate from opening. The dog, on the other hand, confronted with the new obstacles of a second latch or a bolt or an impeding stone, can only bark in impotent fury. The boy can reason the matter out; the dog cannot go beyond the bounds of his habitual thinking. The boy's associations are productive: they prove useful to him in many diverse situations; they help him to adjust himself amid the difficulties of a new situation. The dog's associations, on the other hand, are unproductive that is to say, they are useful to the dog in only a strictly limited measure or under identical conditions; they do not help the dog to solve a new problem that arises through some variation of the ordinary conditions.

I have been at some pains to show the difference between animal thinking and human thinking, because it is human minds which you are called upon to train. We have seen that in human thinking there are two degrees: the first (which it has in common with the animal mind), by which it simply judges or affirms that things are or by which it recognises the facts of sense; the second (which distinguishes it from the animal mind), by which it draws conclusions from the facts presented by sense. It is this second aspect of mind which we call reasoning. When we draw a conclusion from facts that are present to sense, or that we know, we are said to infer our conclusion. For example, when I refuse to budge from my bed on a cold morning when I had arranged

to go fishing, my action is due to the inference I make viz., that the east wind will prevent the trout from rising that day. One of the main tasks of the educator is to train the pupil to make true inferences, and it is of the utmost importance to realise that the way or method by which the mind learns to make inferences must indicate the method by which we should train the mind itself. I wish therefore to lead up to a discussion about method in teaching by first indicating the nature of inference. You will notice that I am now laying the emphasis upon logical rather than upon psychological factors; but it is impossible to separate them entirely. The psychological factors are always present-merged in the logical.

FILLING UP THE CONCEPT

It is important that you should understand first of all how a child comes to know and to increase its knowledge of any fact or thing. Let us consider, for example, how a child comes to know what an orange is. Long before the child is allowed to enjoy the delight of sucking an orange, he has probably seen one many times. When it first attracts his attention, he is conscious of it only as a vague, brightly coloured, nameless thing existing outside of himself. What corresponds in the child's mind to this first vague impression is the child's first 'idea of an orange. Possibly this first 'idea' or image may not survive beyond the moment of impression. If, however, without the presence of any actual orange, the idea of the