ber of repetitions necessary to commit a paragraph or a string of words too long to be retained after a single bearing or a single reading; here, as everywhere, care must be taken to have the paragraphs of equal difficulty, and to repeat the test a number of times. Dr. Ebbinghaus[1] has made a valuable study of the memory, tracing a curve of forgetfulness, and establishing many interesting conclusions by this method; while Mr. Joseph Jacobs[2] and others have used the maximum number of sounds repeatable after a single hearing, which they call the "span," as a test of the growth of mental power with the increase of years, and as a mark of the narrow intellect of idiots. The successive corrections and improvements, until a perfect repetition is possible, are often full of interest. The "auditaire" reaches this stage sooner by having the passage, etc., read, the "visionaire" by reading it; in addition I find that the former has all along (both in I, II, and III) a tendency to remember the last words best, while the latter retains the first most readily. One must also observe by which method the sense is best retained when the exact words are forgotten; moreover, it may be noted that the one confuses words allied in sound, the other words are in appearance, and so on.
II. All the various processes described under I can be repeated with the list of words, numerals, paragraphs, and the rest, so long that error is sure to arise. It is not necessary to give details. These errors are often highly amusing as well as instructive. The fleetness with which an impression which you feel perfectly sure of firmly possessing while listening or reading, suddenly disappears with a blank in its place, is very startling. After an interval, only the most prominent words or ideas are left. Of three persons subjected to a variety of tests, one retained most and more of what the eye had taken in, the second nearly equally of each, with a preponderance of the visual, while the third (myself) was a decided "auditaire." This suggests the remark that a type of mind to which all the avenues of perception are almost equally attractive is doubtless common. In fact, M. Binet,[3] who has much interesting matter to offer on this topic, regards this indifferent type as the normal type, representing a harmonious development of all the sensory faculties. But here, as elsewhere, specialization has its advantages; and, moreover, if the tests are carefully made, I suspect a noticeable superiority in favor of sight will be the most usual result. It is not impossible to imagine the tests so arranged as to give roughly quantitative estimates of the relative impor-
