us.” The Lord Steward squinted at the line of trees.

Psychological examination of sense-perception has been going on since Heraclitus. Most of the mistakes discovered have been used for various purposes, from sport to science. They are surprising and attract and sustain public attention; they have, hence, become

familiar, but their influence upon other phenomena and their consequences in the daily life have rarely been studied. For two reasons. First, because such illusions seem to be small and their far-reaching effects are rarely thought of, as when, e. g., a line drawn on paper seems longer or more inclined than it really is. Secondly, it is supposed that the influence of sensory illusions can not easily make a difference in practical life. If the illusion is observed it is thereby rendered harmless and can have no effect. If it is not observed and later on leads to serious consequences, their cause can not possibly be sought out, because it can not be recognized as such, and because there have been so many intermediate steps that a correct retroduction is impossible.

This demonstrates the rarity of a practical consideration of sense- perception, but does not justify that rarity. Of course, there are great difficulties in applying results of limited experiments to extensive conditions. They arise from the assumption that the conditions will be similar to those which the scientist studies, and that a situation which exhibits certain phenomena under narrow experimental conditions will show them, also, in the large. But this is not the case, and it is for this reason that the results of modern psychology have remained practically unproductive. This, of course, is not a reproach to the discipline of experimental psychology, or an assault upon the value of its researches. Its narrow limitations were necessary if anything definite was to be discovered. But once this has been discovered the conditions may be extended and something practical may be attained to, particularly in the matter of illusion of sense. And this possibility disposes of the second reason for not paying attention to these illusions.

Witnesses do not of course know that they have suffered from illusions of sense; we rarely hear them complain of it, anyway. And it is for this very reason that the criminalist must seek it out. The requirement involves great difficulties for we get very little help from the immense literature on the subject. There are two roads to its fulfilment. In the first place, we must understand the phenomenon as it occurs in our work, and by tracing it back determine whether and which illusion of the sense may have caused an abnormal or otherwise unclear fact. The other road is the theoretical one, which must be called, in this respect, the preparatory road. It requires our mastery of all that is known of sense-illusion and particularly of such examples of its hidden nature as exist. Much of the material of this kind is, however, irrelevant to our purpose, par-

ticularly all that deals with disease and lies in the field of medicine. Of course, where the nature of the disease is uncertain or its very presence is unknown, it is as well for us to consider the case as for the physician. But above all, it is our duty to consult the physician.

Apart from what belongs to the physician there is the material which concerns other professions than ours. That must be set aside, though increasing knowledge may require us to make use even of that. It is indubitable that we make many observations in which we get the absolute impression that matters of sensory illusion which do not seem to concern us lie behind some witnesses' observations, etc., although we can not accurately indicate what they are. The only thing to do when this occurs is either to demonstrate the possibility of their presence or to wait for some later opportunity to test the witness for them.

Classification will ease our task a great deal. The apparently most important divisions are those of ``normal'' and ``abnormal.'' But as the boundary between them is indefinite, it would be well to consider that there is a third class which can not fall under either heading. This is a class where especially a group of somatic conditions either favor or cause illusory sense-perceptions, e. g., a rather over-loaded stomach, a rush of blood to the head, a wakeful night, physical or mental over-exertion. These conditions are not abnormal or diseased, but as they are not habitual, they are not normal either. If the overloaded stomach has turned into a mild indigestion, the increase of blood into congestion, etc., then we are very near disease, but the boundary between that and the other condition can not be determined.

Another question is the limit at which illusions of sense begin, how, indeed, they can be distinguished from correct perceptions. The possibility of doing so depends upon the typical construction of the sense-organs in man. By oneself it would be impossible to determine which sensation is intrinsically correct and which is an illusion. There are a great many illusions of sense which all men suffer from under similar conditions, so that the judgment of the majority can not be normative. Nor can the control of one sense by another serve to distinguish illusory from correct perception. In many cases it is quite possible to test the sense of sight by touch, or the sense of hearing by sight, but that is not always so. The simplest thing is to say that a sense-impression is correct and implies reality when it remains identical under various circumstances, in various conditions, when connected with other senses, and observed

by different men, with different instruments. It is illusory when it is not so constant. But here again the limit of the application of the term ``illusion'' is difficult to indicate. That distant things seem to be smaller than they are; that railway tracks and two sides of a street seem to run together are intrinsically real illusions of sense, but they are not so called--they are called the laws of perspective, so that it would seem that we must add to the notion of sense- perception that of rarity, or extraordinary appearance.

I have found still another distinction which I consider important. It consists in the difference between real illusions and those false conceptions in which the mistake originates as false inference. In the former the sense organ has been really registering wrongly, as when, for example, the pupil of the eye is pressed laterally and everything is seen double. But when I see a landscape through a piece of red glass, and believe the landscape to be really red, the mistake is one of inference only, since I have not included the effect of the glass in my concluding conception. So again, when in a rain I believe mountains to be nearer than they really are, or when I believe the stick in the water to be really bent, my sensations are perfectly correct, but my inferences are wrong. In the last instance, even a photograph will show the stick in water as bent.

This difference in the nature of illusion is particularly evident in those phenomena of expectation that people tend to miscall ``illusions of sense.'' If, in church, anybody hears a dull, weak tone, he will believe that the organ is beginning to sound, because it is appropriate to assume that. In the presence of a train of steam cars which shows every sign of being ready to start you may easily get the illusion that it is already going. Now, how is the sense to have been mistaken in such cases? The ear has really heard a noise, the eye has really seen a train, and both have registered correctly, but it is not their function to qualify the impression they register, and if the imagination then effects a false inference, that can not be called an illusion of sensation.