be? The colleges would be sure to meet with the readiest sympathy and the most hearty co-operation from schoolmasters. It will probably be admitted that a great advance has been made in the amount and quality of the work of secondary schools within twenty years. Of this gain, surely no inconsiderable part has been brought about under the stimulus of improved and increased admission requirements. The promptness and energy with which the schools have responded to greater demands will in coming time be regarded as an astonishing fact in the history of education. Mr. William F. Bradbury, headmaster of the Cambridge Latin School, in summing up the increase of the Harvard requirements, says:

"There have been added to the requirements of 1870, French (or German), to which in the Cambridge Latin School not less than four hundred recitations are given; physics, not less than two hundred; English, not less than three hundred; that is, in all, an increase of not less than nine hundred recitations; or, counting two hundred recitation days in a year, with three recitations a day, this makes just a year and a half's increase in the time of preparation. There is no question that, with the greater difficulty in the Latin and Greek composition, in the algebra, and geometry papers, in the addition of French, physics, and English, Harvard College has herself made such demands as, with an equally thorough preparation, ought to increase the age of the candidates not less than two years."

Mr. Bradbury even understates the increase. Till within a few years Harvard designated the amount of reading in Latin and Greek; so much Anabasis, so much Cæsar and Vergil; nay, the books were indicated from which passages would be chosen for the examination. Now the mode of examining, setting passages previously unseen, amounts to a great increase in the requirements.

Formerly the first three books of the Iliad, about 1500 lines, was the requirement in Homer; now the candidate must be able to translate average passages selected at random from the Iliad and the Odyssey. It is not safe to send up boys now,

who have not read double the former amount. Moreover, concurrently with this prodigious increase in the college requisitions, has gone on a diminution of the school year, from 258 days in 1860, to 200 days at present; so that, as Mr. Bradbury observes, there were more schooldays in four years then than in five years now.

From all that has been said, it ought to be apparent that an immense responsibility rests upon the colleges in respect to secondary schools. It is right that the colleges should consider first their own needs; it is not right that the highest interests of the schools should be disregarded. But there is no incompatibility. On the contrary there is a unity of interests, and it is only necessary that there should be a full recognition of the natural and necessary relations of the two classes of institutions greatly to advance the good of both. A great gulf has separated them. Recently the subject of shortening the college course by a year was discussed with great vigor and earnestness, at least in New England. Such a step was certain to react upon the schools directly and powerfully; but who, in all that discussion, saw so much as an allusion, or at most, more than an allusion from any college source, to its effect on the schools? College presidents and professors joined in the debate, t each and all utterly ignored the schools. To their mina evidently college education is a thing apart, distinct in its nature, scope, and aims, not a continuation of secondary education. Hardly was any voice heard from the schools. There was no recognition that this was a subject of common interest. I repeat, a great gulf has existed between the schools and the colleges. They ought to begin to draw together. The schools have greatly benefited and greatly suffered from the action of the colleges; but the injury has been accidental, and the good has been mostly unforeseen and unpurposed. One important and promising step has been taken toward closing up the chasm between the two classes of institutions, by the formation of the New England Association of Colleges and Preparatory Schools. Why should not other associations of the same kind be formed all over the country?

66

I find in a report on Uniformity in Requirements for Admission to College" presented to the National Council of Education, at Toronto, to which reference has already been made, a suggestion that "educators should make the question [of uniform requirements] a national one." This is much too ambitious a scheme. The partial exhibition, in this paper, of the enormous range and variety of requirements, coupled with the extreme diversity of standards of admission, shows that such a plan is as impracticable as it is undesirable. But one sees no reason why groups of colleges having somewhat similar standards should not unite themselves for conference with preparatory schools, to the better understanding and advancement of both.

ROXBURY LATIN SCHOOL, BOSTON, MASS.

WILLIAM C. Collar.

III.

A STATISTICAL STUDY OF MEMORY AND

ASSOCIATION.

The statistical method has been successfully applied to various departments of mental science; and, indeed, not a few of its problems seem most promising of results when thus studied. General tests of the various mental faculties, and circulars requesting descriptions of personal mental traits, are no longer a rarity, and their use will probably be extended as the experimental basis of psychology becomes more generally recognized. With the double object of interesting the members of my class in psychology in their own mental processes, and of gaining material for such statistical study, I have at times tested their processes of memory and association. On one occasion they each wrote one hundred words as rapidly as possible; fifty such lists I studied with reference to the community of ideas, the nature and time of the associations therein exemplified, as well as to the comparison of the masculine and feminine traits of mind. The results of that study will be brought into comparison with those of the present study. My present data are some seventy lists of words obtained as follows: The withdrawal of a screen revealed a word upon the blackboard, whereupon each member of the class wrote upon a slip of paper the first word suggested by the word upon the board, and folded the paper so as to conceal what had been written; another word was then similarly shown and the process repeated until each student had written ten words. The ten original words were all very common monosyllables, selected for various points on the basis of my former study. The words were: I. BOOK; II. MAN; III. TREE; IV. CAT; V. HAND; VI. HAT; VII. BREAD; VIII. PEN; IX. WRITE; X. BLUE. Secondly, just two days later, and without the slightest expectation on their part, the students were asked to write as many as possible of the words

they had written forty-eight hours previously, and to regard the order. I then wrote the original ten words upon the board and again asked them to write in order as many as possible of their associations recorded two days previously. Let us call the results of the first test the "original lists," of the second, the "A lists," of the third, the "B lists." The first show the most accessible associations to ten common words; the “A” lists show to what extent such can be recalled by memory alone, upon short notice, and when written with no expectation of future use; the "B" lists show to what extent the recollection is aided by the presence of the suggesting words.

First as to the unaided memory. I have 70 original lists, or 700 words, written by 42 men and 28 women, or in the proportion of and; of A lists I have but 64 serviceable ones, contributed by 39 men and 25 women, on which appear but 411 words; that is, 35.8 per cent. of the words have been so far forgotten as to be beyond voluntary recall after 48 hours interval. This includes 72 cases in which wrong words (often of kindred meaning to the correct ones) were written, so that but 339, or 53 per cent., were correctly remembered, and not a single student correctly remembered all ten words. One student recalled nine words correctly; 4 eight words; 9 seven words; 17 six words; 13 five words; 12 fqur words; 4 three words, and 4 only two words. The attempts at recalling the original order of the words were very unsuccessful; in the best record six words were written in their original order.

In the B lists, in which the original suggesting words are furnished, I find a considerable increase of words correctly recalled; for in 67 lists there appear 640 words; or only 4.5 per cent. are entirely forgotten, and 536, or just 80 per cent., are correctly recalled. On the average each student recalled 2.7 more words with the aid of the suggesting words than without them. Sixteen recalled all ten words correctly; 16 recalled nine words; 14 eight words; 6 seven words; 7 six words; 5 five words; 2 four words, and I three words-a very marked improvement upon the A lists. Roughly speaking, the average