September 10, 1886.] 



SCIElSfCE. 



239 



tion is to repeat the process by vs^hich the throw- 

 ing time was obtained, with the diiference that the 

 card is not to be depoisited before the thrower has 

 appreciated the color, whether red or black, of 

 each card to be thrown. The time necessary for 

 this process, minus the throwing time, would be 

 the time which it took the person to distinguish 

 red from black. But this method is really not 

 valid at all, and for the following reason. While 

 throwing one card, one can in the indirect field of 

 attention, so to say, be preparing to decide or 

 already deciding what the color of the following- 

 card is ; so that the two operations of throwing 

 and distinguishing partly overlap. A distinction 

 time gotten by such a proceeding would be en- 

 tirely too short. Several ways of avoiding this 

 difiSculty were suggested, of which I used the fol- 

 lowing one. The cards were held with the backs 

 towards the thrower. The operation consisted, 

 first in simply turning the card with its face up- 

 ward, and depositing it on a heap ; and, second, 

 in not depositing it before its color has been seen. 

 In this way the person cannot see the following 

 card, because it has its back towards him ; and all 

 the cards may be placed on a single heap. The 

 average difference between the time required for 

 the first operation and that for the second, divided 

 by the number of cards, will give the distinction 

 time for distinguishing red from black. ' 



I have described the simplest type of a distinc- 

 tion time. The process can be indefinitely com- 

 plicated by having three, four, or more colors to 

 distinguish, using the backs of variously colored 

 cards, or by distinguishing the four suits of one 

 pack. By having several packs of cards, one can 

 vary the experiments in very many ways. One 

 can distinguish as many of the spot-cards as one 

 pleases, from two to ten ; can, in addition to this 

 distinction, distinguish between the suits ; and so 

 on. Before giving the results I have obtained in 

 this way, I will anticipate the question whether 

 the number of cards used will not affect the result. 

 It probably will ; for the mind, being once set on 

 the habit of making these distinctions, can keep up 

 the process with less energy, and thus with greater 

 rapidity. This question I hope to solve by a special 



1 In another method the forty cards are spread out upon 

 a table, say, in five rows of eight each. The subject first runs 

 his eye along each row, going forward on one row and back- 

 ward on the next, dwelling on each card only long enough 

 to bring it into distinct vision. The operation is very rapid 

 (being faster than counting), but is rather uncertain. Next, 

 he ' reads ' the color of each card in the same manner. The 

 difference between the times necessary for these operations 

 evidently, again, gives the distinction time. Here, too, 

 reading ahead in indirect vision is possible, but not to any 

 great extent. The method is of value only as a means of 

 checking the results of the first method, but is inferior to 

 it. Doubtless some of my readers will invent a method 

 better than this or the one described in the text. 



set of experiments. From what I have done I am 

 able to say that the variation will be extremely 

 slight. It is recommended to use forty or sixty 

 cards, as it is easy to hold that number in one's 

 hand, and these numbers are divisible by 2, 3, 4, 5, 

 and the latter by 6. Moreover, 5L or J^ of the 

 error involved in neglecting fractions of a second 

 less than a quarter is a small error indeed. 



The persons whose reaction times were taken 

 were, I., a girl of ten years ; II., a young lady 

 and. III., myself. In all the experiments in 

 which II. and III. took part sixty cards, and in all 

 in which I. was the subject forty cards, were used. 

 In the following table the time is always given in 

 seconds. 



The column headed 5's from 9's, 2's from 4's, 

 etc., indicates that the pack of cards was divided 

 equally between two-spots and four-spots, or five- 

 spots and nine-spots, or some similar combination 

 of two kinds of cards ; and that the subject had to 

 distinguish by the method above described the 

 denomination of each card. It thus appears that 

 it took I. .058 of a second to make this distinction, 

 and II. and III. .045 and .043 of a second respec- 

 tively. In other words, it takes ^^ of a second to 

 tell whether a card is a five-spot or a nine-spot, 

 or to make any similar distinction. The only 

 experiment performed by the usual laboratory 

 methods, which I could find, comparable with this, 

 was one by Professor Wundt, undertaken in his 

 psychological laboratory at Leipzig, in which the 

 distinction was made between a black cross on a 

 white background, and a white cross on a black 

 background. He gives .0485 of a second as the 

 distinction time, which agrees well with .044, the 

 average of the times of the two adults in the above 

 table. The distinction between the green and blue 

 backs of cards, as is shown by the last column 

 of the table, is more rapidly made. Perhaps part 

 of the difference is due to the fact that the card 

 did not need to to be turned so completely around 

 to see the color as to see the denomination. 



In the other columns of the table is shovvn the 

 result of a series of experiments in which the cards 

 were divided among three, four, five, or six kinds, 

 as indicated in the heading. It is seen, that 

 though the thing to be done remains the same, 



