1897.] Psychology. 915 
the requirement, endeavors to repeat the figures without error and in 
the order in which he heard them. The experiment is repeated several 
times, beginning with a small number of figures, e. g., four, which any 
adult can give correctly ; it is then increased to five figures, then to 
six, and so on, until a number is reached which the subject can no 
longer repeat correctly ; care is taken to repeat each trial, and to al- 
low sufficient intervals of rest to avoid fatigue and the confusion of 
figures in’ the memory. This procedure, adopted by Jacobs, Galton 
and many others, has already borne fruit. It is not, properly speak- 
ing, a test of the memory alone ; it is extremely diffcult, be it said in 
passing, to experiment on any isolated psychological phenomenon ; the 
experiments taken together show, on the contrary, that the subject em- 
ploys not only his memory but also his powers of voluntary attention ; 
this explains why children retain fewer figures by this method than 
adults ; their inferiority is certainly due to the fact that they have less 
control over their attention. The average educated adult retains seven 
figures; a child from 6 to 8 retains five; a child of 10 retains six. A 
difference of one single figure is of considerable importance in the 
results, and it is one of the drawbacks of this method that we cannot 
Operate with fractions of figures. I have had occasion to measure the 
retentive memory of Jacques Inaudi, the celebrated lightning calcula- 
tor; he is able to commit more than 40 figures at one trial ; it will be 
seen from this how far his memory is above the average. 
Instead of finding out the number of figures, letters or words that 
can be retained by one person after a single hearing or reading, a dif- 
ferent procedure may be adopted ; we may endeayor to find the time 
required by different individuals to learn a given number of figures, 
Say twelve ; further, we may try to find the time necessary to learn 
again a series once learned and afterwards forgotten. For details in 
regard to these rather complex methods I refer the reader to the work 
of Ebbinghaus. ( Ueber das Gedachtniss). ; 
A rather curious question, which is closely related to that of the 
measurement of memory, is the simulation of the memory for figures. 
emory can be simulated as well as other things. This is done by 
means of mnemonics, a process which consists in associating arbitrary 
ideas with figures; I have indicated, in a study undertaken with V. 
Henri, how real memory can be distinguished from simulated memory, 
by measuring the time required to learn and reproduce. 
3. Fi orgetfulness— We now reach a question that has an important 
Pedagogical bearing: the problem of forgetfulness. In what does it 
consist? What is its course? What memories are attacked first? 
