number from one to the other, a third. Hunting-dogs show great skill in threading mountain-paths and overcoming or avoiding obstructions, but no dog will remove a branch interposed in his way. When the associations by which instinct works come into play outside of or against their ordinary end, we may speak of their working as imperfect, and may say that the animal is mistaken.
We also have instincts that are characterized by the narrowness of their end. Among them are the reflex actions. The eyes wink when they are threatened with injury; but they also wink when a beneficial operation is performed upon them, to which the winking is an obstacle, the action going on all the same when it is useless or injurious.
I believe it can be shown that this type of instinctive action is also found in man, and that the origin of many types of errors may be found in the application to particular cases, but exceptional, of what is generally right. This proposition is confirmed by some errors of the senses. When a point on our retina is excited by an external pressure, we fancy we see something luminous in the ordinary field of vision of that point. Were it not for the experience of previous observations of objects and their reflections, we should localize as things behind the glass the reflections which we see in mirrors. In this and most like cases, we are acquainted with the mechanism of the phenomenon, and can distinguish between what is only the sensorial impression and what we owe to memory. The separation vanishes in the higher regions of psychic life. If we draw a line on a sheet of paper and cover the end of it with another sheet, an observer not in the secret will imagine it to be much longer than it is, because his conception is based upon the fact that when one object lies upon another, it usually covers a considerable portion of it. We are subject to a considerable number of illusions of this kind. The prestidigitator takes advantage of one form of them when, by a quick look to one side, he turns the eyes of the audience away from his manipulation and gains an opportunity to execute the trick without detection, although every one of his spectators had determined not to lose sight of his hands. He is aware that a glance and particular adjustments of the head and eyebrows and lids will usually suggest to the looker-on that he will see at a particular point something more interesting than anywhere else within his field of vision. At the same time the audience will not know why they looked in that direction, and may not even be conscious of having looked there.
We thus deal on this domain, remote from the physiology of the senses, with functions of the nervous system similar to what we have seen in the hen and the winking. Thought follows its course according to the usual process; with more or less of con-