to a family, a society, or a nation. These reference points form for each of us different series corresponding to the events that make up our life: daily occupations, domestic incidents, professional work, scientific investigations, etc., the series becoming more numerous as the life of the individual is more varied. These reference points are like mile-stones or guide-posts placed along the route, which, starting from a central place, diverge in different directions. There is always this peculiarity: that the series may, so to speak, be placed in juxtaposition and compared one with another.

It remains for us to show how these reference points permit us to simplify the mechanism of localization. The impression which we call a reference point returning, by hypothesis, very often to consciousness, is very often compared with the present according to its position in time -that is to say, intermediate states which separate them are more or less completely revived. As a result, the position of a reference point is, or seems to be (for we shall see further on that all recollection implies an illusion), better and better known. By repetition this localization becomes immediate, instantaneous, automatic. The process is analogous to the formation of acquired states (habits). The intermediate terms disappear because they are useless. The series

is reduced to two terms, and these two terms suffice, since their distance in time is known. Without this abridged process and the disappearance of a prodigious number of terms, localization in time would be very long and tedious, and restricted to very narrow limits. By its aid, as soon as the image is formed, its primary localization is instantaneous; it is placed between two landmarks-the present and some given point of reference. The process is concluded after a few trials, often laborious, sometimes fruitless, and perhaps never precise.

If the reader will study carefully his own recollections, I do not believe that he will raise any serious objections to what has just been said. He will, moreover, note how this mechanism resembles that of localization in space. Here, also, we have our reference points, abridged methods, and well-known distances which we employ as units of measurement.

It will not be unprofitable to show in a few words that localization in the future is executed in an analogous manner. Our knowledge of the future can only be a copy of the past. I find only two categories of facts: those which are a reproduction, pure and simple, of what has occurred at similar epochs, in the same places under like circumstances; and those which consist of inductions, deductions, or conclusions, drawn

from the past, but produced by the logical working of the mind. Outside of these two categories everything is possible, but everything is unknown.

Evidently the first class most nearly resembles memory, since it is a simple reproduction of what has been. A man is in the habit of going every year to pass the month of September at a country house. In the middle of winter he sees it with all its surroundings, inhabitants, and characteristic activity. This image is at first indeterminate; it is equally an object of remembrance and of the future. Then it glides away from the present through winter, spring, and summer; finally it is localized. The course of the year, with its succession of seasons, fêtes, and changes of occupation, provides reference points. The mechanism differs from that of memory only in one respect: we pass from the termination of the present to the beginning of the following state. We do not proceed, as in recollection, from beginning to end, but from end to beginning. Theoretically, we traverse in this invariable order all intermediate states; in fact, only the reference points. The mechanism is the same as that employed in memory, only it acts in a different direction.

To recapitulate: setting aside verbal explanations, we find that recollection is not a "facul

ty," but a fact, and that this fact is a result of aggregate conditions. As with recollection, localization in time varies through every possible degree according to the conditions. At the highest stage of development are the reference points; below those, rapid and precise recollections, located almost as quickly; one degree lower, those which cause hesitation, requiring an appreciable time; lower yet, laborious recollections, only attained by trial and stratagem; finally, in some instances, the labor is useless, and our indecision is translated into such phrases as, "It seems to me that I have seen that form!" "Did I dream that?" One step more, and localization is entirely wanting; the image, denied an abiding place, wanders in devious mazes, incapable of rest. There are many examples of this last case, and they are found in the least expected forms. Through the effects of disease or old age, celebrated men have been unable to recognize cherished works of their own production. Toward the close of his life Linnæus took great pleasure in perusing his own books, and when reading would cry out, forgetting that he was the author, "How beautiful! What would I not give to have written that!" A similar anecdote is told of Newton and the discovery of the differential calculus. Walter Scott as he grew old was subject to similar forgetfulness. One day

some one recited in his presence a poem which pleased him much; he asked the author's name; it was a canto from his "Pirate." Ballantyne, who acted as his secretary and wrote his life, has related in the most circumstantial manner how the greater part of "Ivanhoe" was dictated during a severe illness. The book was finished and printed before the author was able to leave his bed. He retained no remembrance of it, except the main conception of the romance, which had been thought of prior to the attack.

In a case cited by Forbes Winslow, the image is apparently waiting to be seized and localized; it is on the edge of recognition; the smallest aid would suffice, but is wanting. A lady was driving out with the poet Rogers, then ninety years old, and asked him after an acquaintance whom he could not recollect. "He pulled the checkstring, and appealed to his servant. 'Do I know Lady M.?' The reply was, 'Yes, sir.' This was a painful moment to us both. Taking my hand, he said, 'Never mind, my dear, I am not yet compelled to stop the carriage and ask if I know you.'"*

A much more instructive instance is recorded

*Laycock, "Organic Laws of Personal and Ancestral Memory," p. 19; Carpenter, op. cit., p. 444; Ballantyne, "Life of Walter Scott," ch. xliv; Spring, "Symptomology," vol. ii; Forbes Winslow, op. cit., p. 247.