The remarkably short time needed in both of these investigations to regain the former skill, after the long interval without practice, shows the persistence of neuromuscular memory. A still more significant fact, however, is disclosed by the curves for ball-tossing given above. It will be observed that in eleven days, after six years of cessation of practice, the experimenter acquired much greater skill than that with which he closed his experiments of forty-two days, six years earlier.

It is clear that the effects of activity upon the nutrition of the organ exercised and the dropping out of interfering associations do not account for the persistence and rapid improvement of skill after such long periods of inactivity. The astonishingly rapid gain in skill in ball-tossing beyond that originally acquired indicates not only a "fixing" of nervous and muscular associations, but also, during the long interval, some sort of integrative nervous activity by which the skill was further improved. At present there is no other tenable explanation. Batson, in his investigation,1 also observed improvement in skill during intervals of no practice. "After a long rest period," he says, "the subject is found to be in a condition to improve very rapidly. In some cases the results show that they have actually gained power during the rest period."

The matter of "rest periods" has wide application. Jost's investigations, as well as the others to which reference has just been made, emphasize the value of repetitions, but always with an interval between them. A book, a legal opinion, an investigation in medicine or in science, which one wishes to remember, should be read again, but not at once. It is always important, however, that the meaning be as clear as possible in the first reading, because that gives the memory a freer field; and, when it is a matter of elaboration rather than of memory alone, the reconstruction of the thoughts, as a result of cerebral proc

1 Op. cit.

esses, has a better start. Error may be corrected, but mere confusion is rarely if ever clarified without further study.

These investigations also make conscious and thoughtful a method which has been followed instinctively, or, at any rate, unconsciously. Children who "go over" the lesson just before class are acting upon this principle of distributed study. To older persons, a book, or chapter, may seem quite unintelligible when read for the first time. Upon second reading, however, after the lapse of a day or two, it is often amazing how clear the misty statements become. One now sees meaning in what before was unintelligible. Had the reader stopped with the first reading nothing would have been remembered, but now the confusion takes an orderly form. Something is indicated, however, in addition to reinforcement of memory impressions. Deepening impressions by repetition might account for the greater clearness, but this is hardly sufficient to explain the increased significance that a second reading produces after an interval of a day or so. And an intervening day gives better results than an immediate rereading. It seems to be another illustration of the integrative nervous activity to which reference has just been made.

As a practical deduction from these investigations and observations, it may be said that students would save time and effort by keeping at least one day ahead in their studies; and the last reading, like the earlier ones, should be as rapid as following the thought permits. "Cramming" is clearly unpsychological; for accumulated repetitions within a short time have only a temporary effect. That which is learned in this way vanishes quickly.

One reason for the meagre results of cramming is that association of ideas is reduced to their sequence in the text. The wider meanings do not have a chance to assert themselves. The bearing of associations upon memory may,

perhaps, be best illustrated by comparing those at the two extremes of usefulness.

Associations may be strictly artificial and mechanical or they may be significant and interpretative. Memory systems offered for sale are of the first sort. One of these systems which has been long in the market teaches the purchaser to remember numbers by translating them into letters of words. Zero, for example, is represented by s, z, or c soft; I by t, th, or d. "All the other letters" [of the word] "are simply to fill up. Double letters in a word count only as one." This or dizzy would then stand for the number 10, and catch or gush for 76. "Now," continues the writer of the book from which we are quoting, "suppose you wish to memorize the fact that $1,000,000 in gold weighs 3,685 pounds. You go about it in this way: "How much does $1,000,000 in gold weigh?

"Weigh―scales.

"Scales statue of justice.

"Statue of justice, image of law.

"The process is simplicity itself” [mirabile dictu!]. "The thing you wish to recall, and that you fear to forget, is the weight. Consequently, you cement your chain of association to the idea which is most prominent in your mental question. What do you weigh with? Scales.

"What does the mental picture of scales suggest? The statue of justice, blindfolded and weighing out award and punishment to man. Finally, what is this statue of justice but the image of law? and the words image of law translated back from the significant letters m, g soft, f, and l, give you 3-6-8-5, the number of pounds in $1,000,000 in gold."

Another of these mechanical memory systems is based upon visualization. "We must not impress upon the memory mere words, but turn our attention and train our minds to see the objects or ideas they represent, associated or combined in mental pictures," the writer of this system

says. A list of a hundred "code" or "index" words is given, of which hat and hen are the first two. "See the hat and the hen in the picture together," we are told. "Do not see a hen of ordinary size, but use one strengthening principle of imagination, that of exaggeration. See a large hen four feet high. Put the black silk hat on the hen's head; now put motion into the picture, and see the hen strut about. . . . Knowing our index words will enable us to remember a number of varied items, such as errands." Suppose, now, that we wish to buy stamps. "We may form a picture of a large hat decorated with postage-stamps, or any clear combination of stamps and hat. The more conspicuous and striking the picture, the easier will it be to recall it." But there are also other associations in this system. "Trustees is similar in sound to rusty, so if we make trustees suggest rusty, and visualize a rusty can, the gap is bridged. . . ." Letters are to stand for figures, but it all goes back to visualization of the objects represented by the "code" words. "A combination picture of a mop and a chair will represent 3964. The danger of transposing figures, by recalling the picture as chair-mop 6439, instead of 3964, can be avoided by having the first object much larger than the second. In the case of 3964, picture the mop larger than the chair."

Of course there is no fault to find with any of this except with the whole of it. Its associative machinery is so crude that the rattle is distressing. Yet these systems are

fairly representative of those on sale. That they continue and multiply indicates buyers. Consequently, some of the purchasers must feel that they receive value for the price. The explanation seems to be that if a man is sufficiently interested in improving his memory to pay an exorbitant price for a "system" he will continually keep the thought of remembering in mind. He will also observe details and repeat what he wishes to remember as he never did before the expenditure made it worth while.