754 



SCIENCE 



[N. S. Vol. XL. No. 1038 



TABLE I 



The Selation of Initial Ability in Mental Multi 



Students 



to Improvement: 76 College 



exercise of the function. We may then com- 

 pare the improvement made in the course of 

 approximately Y5 minutes of practise (I count 

 from the mid-point of the first row's time to 

 the mid-point of the time of the row such as 

 makes this time from mid-point to mid-point 

 as near to 75 minutes as possible), by those of 

 initially high and those of initially low scores. 

 Doing this, we find the facts of Table I., 

 (I.) for 18 individuals whose average score for 

 the first row was at the rate of .61 examples 

 per minute, (II.) for 18 individuals whose 

 average score for the first row was at a rate 

 of .36 examples per minute, (III.) for 18 

 whose average score for the first row was at 

 the rate of .24 examples per minute, and 

 (IV.) for 18 whose average initial rate was at 

 the rate of .14 examples per minute. As the 

 table shows, the initially high-scoring indi- 

 viduals made an equal gain in speed and some- 



what less gain in accuracy, the net results 

 being that they made about as much improve- 

 ment as the others. 



The same result appears in the case of addi- 

 tion where data from some 6Y0 individuals 

 give the facts of Table II. 



These experiments add one more corrobora- 

 tion of the general result, so far uniformly 

 attained, that equalizing opportunity does not 

 reduce individual differences. Such experi- 

 ments furnish a very strong argument against 

 referring individual differences of unknown 

 causation to differences in training, and in 

 favor of referring them to original inherited 

 characteristics in cases where they follow 

 family relationships. We are unable experi- 

 mentally to equalize training in such gross 

 complexes as scientific achievement, literary 

 fame, or reputation as a monarch. But we 

 can easily do so with various minor capacities 



Improvement Made in 1,800 Seconds of Practise at Adding Columns, Each of Ten Digits 

 Early refers to the ability estimated for the mid-point of the first day. Late refers to the 

 ability shown after 1,800 seconds of practise, counting from the mid-point of the first day. 



