168 K. S. LASHLEY 



them. If correct, they go far toward clearing up the problem. 

 The hyper-emotionalism of the caffeinized animals suggests an 

 increased output of adrenalin in agreement with Sajous' con- 

 ception. It is possible, then that the retarding effects of caf- 

 feine on learning are due to endocrine activity, which, in the 

 case of strychnine is obscured by the action of the drug upon 

 the central nervous system. 



Since only the larger dose of strychnine affects learning and then 

 only when other symptoms indicate its attack upon the central 

 nervous system it seems probable that its acceleration of learn- 

 ing is correlated in some way with the reduction of synaptic 

 resistances and the increased ease of spread of nerve impulses. 

 This condition should result in a greater diversity of activity, 

 a readier change from one type of action to another, than is 

 possible in a normal nervous system, and might therefore mili- 

 tate against the appearance of stereotyped errors. As this is 

 the sort of error appearing in the later stages of practice it seems 

 possible that the acceleration of learning by strychnine may be 

 effected in this way. 



SUMMARY 



Albino rats were trained in the circular maze after subcutane- 

 ous injection of strychnine sulphate and of caffeine. Others 

 were trained after similar manipulations but without the drugs, 

 as controls. Analysis of the data obtained gives the following 

 results : 



1. Small doses of strychnine sulphate are without effect upon 

 the rate of habit-formation. 



2. Large doses of strychnine sulphate, sufficient to produce 

 tremor and incoordination of movement, accelerate learning. 



3. Caffeine, in doses of 0.5 mgm. or more, retards learning 

 in direct proportion to the size of the dose. 



4. Strychnine sulphate in large doses increases the accuracy of 

 performance of a perfected habit. 



5. Large doses of caffeine result in increased activity and 

 reduced accuracy of performance. 



