On the ninth day the chick was weak and would not choose 

 either passageway. When I dissected it a large intestinal cyst 

 was found in which there was much food and a fluid secretion. 

 Such a cyst could have formed in a few days. But the important 

 point is that the only sign of ill health in this chick for four days 

 was the decrease in the number of right choices. On the fifth day 

 physical signs of weakness appeared. 



In conclusion, it is evident that within the limits of the stimuli 

 which I used, the number of trials required by the chick to learn 

 to choose consecutively the darker of two unequally illuminated 

 screens, when discrimination is easy, decreases with an increase 

 of stimulus. Under medium difficulty of discrimination the above 

 law holds true only for the lower intensities of the stimuli which 

 were used, or, in other words, the optimal stimulus recedes 

 toward the threshold from 590 to 480 units. The above law for 

 the condition of easy discrimination holds true for that of 

 difficult discrimination if we consider only the records of 

 the chicks which succeed in learning to make the discrimin- 

 ation. If, however, we consider only the chicks which fail, the 

 optimal stimulus recedes once more to a point nearer the threshold 

 of stimulation than in the case of medium discrimination. In 

 other words, with the difficult condition of discrimination, strong 

 stimuli divide the chicks into two groups, those which succeed 

 in learning to discriminate by reason of more right choices at 

 the beginning of the training series and consequently fewer 

 pain stimuli, and those which fail because of fewer right choices 

 and more pain stimuli in the earlier trials. So far as I determined 

 the sensitiveness of the chicks it may be said that on the average 

 the more sensitive chicks learned more rapidly both for strong 

 and for weak stimuli. 



