50 JOHN LIXCK ULRICH 



changes occur during this excitable state. It has been vshown 

 by Cannon (14) that during any emotional excitement there is 

 a secretion of adrenalin. It seems that this substance in the 

 blood causes the activities of the alimentary canal to cease; 

 there is a change in the circulation from the great vessels of the 

 abdomen to the lungs, limbs, and central nervous system, and 

 the heart action increases in vigor. There is in every way a 

 bettering of the circulation of the blood, and the animal can 

 stand the strain resulting from emotional excitement during the 

 early trials in learning. Moreover the increase in circulation 

 in some way obviates fatigue in the muscles. Thus there are 

 metabolic changes going on in the organism under stress of 

 learning which tend not only to sustain it, but also to foster 

 movements. 



There must necessarily be some indication of these changes 

 on the learning curves. The irregularities of the curves during 

 the early trials represent the entire process of learning during 

 the excitable period. It is most prominent when several trials 

 are given daily, and when problems are given concurrently. 

 After the early trials, when the movements have become par- 

 tially integrated, that is, just after the rapid descents in the 

 curves are made, the -excitable state, to a great extent, dis- 

 appears. The curves become less complex and hunger there- 

 after mainly dominates the action of the body as a whole. 

 Learning is smoother, and this clearly means that there is no 

 more clashing among opposing activities of the different parts 

 of the body. When antagonistic reactions, the kind that pro- 

 duce errors, cease, and the attending excitations subside, learn- 

 ing can be said to be complete. 



CONCLUSIONS 



1. Considering first the latch box and the maze when they 

 were learned alone, we find that one trial a da}^ is a more efficient 

 procedure than either three trials per day or five trials per day. 



2. Tests upon the latch box show that a more infrequent 

 practice distribution is even more efficient — one trial given 

 every third day giving better resuhs than one trial every day 

 or one trial every other day. 



3. Again, when three ]jroblc>nis are learned abreast two things 

 a])]jear clearly: First, infrctiuenl ])ractice again appears to be 



