EFFECTS OF STRYCHNINE UPON RATE OF LEARNING 147 



other than those which are intentionally varied in the experi- 

 mental procedure. Careful selection of the subjects, and 

 rigorous duplication of the experimental conditions may go 

 far towards the elimination of chance variations in the results. 

 The ideal control of small numbers of subjects is to train all 

 at first in exactly the same activity and then with the results 

 of this training as an index of individual variability, to introduce 

 experimental modifications of the training methods in teaching 

 a second activity. But where learning is relatively slow, as in 

 the white rat, where the age of the subjects may be an important 

 item, and where there may be unknown seasonal differences in 

 the rate of learning, the demands upon the time of the experi- 

 menter become such that he cannot alone supply all the requisite 

 controls. As I have pointed out in a previous paper (Lashley, 

 '17), such experiments should be either carried out by the co- 

 operation of several experimenters, or the subjects should be 

 trained in the simplest possible tasks so that as many as possible 

 may be trained. 



In the present series of experiments the controls employed 

 seem to cover most of the variable factors save that of individ- 

 ual differences in the animals. They were, briefly, the follow- 

 ing: 



(1) Genotype. The animals trained were all taken from large 

 litters, one animal from each litter being assigned to each of 

 the differentially treated groups. (2) Sex. The animals chosen 

 from any given litter were all of the same sex so that each ani- 

 mal in any group was controlled by a sibling of the same sex in 

 each of the other groups. (3) Age. Except in the second 

 experiment, all animals were beween sixty-five and seventy 

 days of age at the beginning of training. In the second experi- 

 ment they ranged from seventy-five to eighty-five days. Train- 

 ing was begun with all members of the same litter on the same 

 day. (4) Daily variations. For convenience in injecting the 

 drugs all members of each group were trained at about the 

 same time of day, morning, afternoon, or night, but on succes- 

 sive days the order in which the different groups were taken up 

 was varied so that there was no constant difference in the time 



PSYCHOBIOLOGY, VOL. I, NO. 2 



