conclusions which tended to show that this visualising power varied according to age and occupation in life. But the quantitative value of these observations was not very great. There is obviously considerable variety in breakfast tables, and not only would this circumstance affect the accuracy of any results attained, but it had, in point of fact, been tolerably well known before these experiments were made that great personal differences existed in this matter. Dr. Wigan, for example, in a work published in 1844,* tells us of a painter who, from the result of only one sitting, was able to paint a portrait with the utmost accuracy. He could at will, by fixing his eyes on the chair which the sitter had once occupied, call up at all future times his simulacrum, and paint from it as from the original. And we have it upon good authority that Charles Dickens so completely realised the favourite characters of his novels, that he felt towards them almost as if they were his intimate acquaintances.

A more valuable experiment was made at another time by Mr. Galton in the following manner. He arranged a list of words, and placed it before him in such a way that by a slight effort he could just see one word at a given moment. As soon as he had seen this word, he allowed about a couple of ideas to form themselves in his mind, and then by means of a stop-watch measured the length of time which these two ideas had occupied. He then set himself to take a memorandum of everything he could remember with regard to the character and idiosyncrasy of the ideas which had thus presented themselves. Mr. Galton made use in this manner of 75 words. He went through them on four successive occasions in the manner prescribed, making thus a total of 800 separate trials. These he found to have given rise to 505 ideas in the space of 660 seconds. There were, however, many cases of recurrence, that is to say, of the same idea * The Duality of the Mind; A. Wigan, M.D., 1844.

having arisen more than once upon the suggestion of different words, and thus the number of different ideas was only 279.*

These observations of Mr. Galton's came under my notice some time after a somewhat similar course of enquiry had engaged my own attention, and, whilst I appreciate the value of his novel and striking results, it appears to me that there are flaws in his methods of procedure which detract considerably from their value. As regards his mind pictures, it is evident that anything like exactness in comparing results is not to be thus attained. A form in which such experiments can be attempted by any one, and that with results of higher quantitative value, is to take a list of names of persons whom you have known well enough to be able to produce at will a mental picture of them. It is not requisite that the picture should be vivid; it is sufficient that it shall be recognisable. Then pass through the mind a number of these recognisable likenesses as rapidly as possible, and observe how many pass in a given time, which, as a matter of convenience, should not be more than a minute or two. It will be found that in this way something like fifty mental sketches per minute can be taken,† and that any increase in rate is attended with an immensely increasing ratio of difficulty. The one advantage in this species of mental presentation over that of Mr. Galton, is that you are making use, in the shape of portraits, of units of measurement which may fairly be taken as of equal value. The range of quantitative variation between faces is far less than that between breakfast tables.

A similar consideration applies again mutatis mutandis to Mr. Galton's second process. You will recollect that, like a member of a gun-club shooting at pigeons, he slipped a

"Psychometric Facts," Nineteenth Century, March, 1879.

+ This coincides almost exactly with the result obtained by Mr. Galton in the process last described.

couple of ideas from a mental trap, knocked them over when fairly on the wing, and then proceeded to anatomise them. But here, as before, the unit of measurement is extremely indefinite. It is impossible to assign to ideas taken promiscuously in this way any direct quantitative value. All that we are in a position to assert is that they vary very greatly in this respect. And in so far as it is an object of the experiment referred to to ascertain the rate at which the formation of ideas goes on, it appears to me that more satisfactory results can be arrived at in a manner about to be described, but with regard to which one or two prefatory remarks are needed.

As we have already said, ideas themselves cannot be accepted as convenient units for the measurement of mental operations, simply because they are incongruous and dissimilar, some more complex than others, and few, if any, bearing the character of true psychical molecules at all; but fortunately for our purpose all ideas possess in common a derived function, into the medium of which they may and do pass. It is clear that language is a function of this character. It is also evident that although we are certainly not in a position to say that equal quantities of language contain always equal quantities of ideas, we may nevertheless assume that such a proposition would become more and more nearly true as we take a larger and larger number of words as the basis of investigation. In the final limit, using the term in its mathematical sense, it may therefore be said that words and ideas vary simultaneously and in a constant ratio to each other. There is, of course, a colloquial sense in which we might be disposed to say that words and ideas in some men's use of them stand in an inverse ratio, but it will be found upon reflection that what we are really estimating in making such a statement is, the novelty, the originality, the logical and rhetorical value of the ideas propounded, and not their actual

inherent number or efficiency as mental counters. It follows from these considerations that an estimation of the mind rate may be obtained in terms of the number of words which can pass through the mind in a given time, and that such an estimation might be made as accurate as we please by increasing the number of words operated upon.

One further proviso it is also necessary to make. Velocity or rate is, under all experimental conditions, a thing which varies constantly. If, therefore, we wish to measure the rate of anything, it is clear that we can only do so satisfactorily by taking it at its maximum. The only way, for example, by which you can for any purpose of comparison define the velocity of a shot fired from a cannon, is by taking its initial or greatest velocity. Or again, if you wish to express the speed of a particular race-horse, you take its best performances as the criterion, although it is clear that both the cannon-ball and the race-horse have, at other times than those taken as representative, other rates of speed, observation of which would be useless for the special end supposed to be in view.

It evidently follows that, in order to arrive at any useful measure of the velocity with which words pass through the mind, the result must be obtained under the most favourable circumstances. The words themselves must be so well known as neither to raise the friction of novelty nor in any other way to divert the mental force employed. The most convenient form, therefore, of making such an experiment is to take a poem, or other piece of composition, known, as the saying goes, by heart, and to read it as rapidly as is possible to the observer, taking care, however, not to slip a single word. If this is done under suitable conditions, and the time occupied by the process accurately noted, it will be found that a very constant result for each individual is arrived at, and that any attempt to go beyond this rate is most arduous, and

almost certain to involve a break in the exact continuity of the reading process. About eight hundred words per minute, or thirteen to fourteen per second, is the approximate number which can be thus consciously read.

But the experiment may now be carried a stage further, by removing the book and passing the same words through the mind, in the same way as when read, that is to say, as rapidly as they can be rushed through consciousness without, however, omitting anything. A very striking result will at once present itself. However perfectly familiar the lines may be, and although there is no conscious effort whatever in recalling them, which last is of course a necessary condition for the purpose of any useful comparison, it will be found that the same matter now takes much longer to pass through the mind, the retardation being nearly equal to onehalf. Thus a person who in the former experiment could pass seven hundred to eight hundred words per minute through his mind, would in the latter find himself reduced to about four hundred and fifty, and no effort whatever would enable him to increase this rate to any appreciable extent.

cause.

It is clear that this comparison supplies a differential result of importance, for which there must be some adequate The first solution of it which occurred to myself was, that in the act of reading, the double character of the brain rendered possible a duplex action, whilst in the pure act of thought, the whole brain co-operated and acted singly. And this, which would accord with a theory propounded by Dr. Wigan of the duality of the mind, in a work already referred to, might seem at first sight to be a satisfactory explanation of the phenomenon. But it so happens that such an explanation can be at once disproved by a crucial experiment. It is clear that if there is any independent action in the two halves of the brain in the act of reading at a forced rate, such independent action must be ministered to