160 



K. S. LASHLEY 



As in the case of the tests on strychnine, the number of ani- 

 mals trained is small, but the differences brought out are so 

 pronounced in every case that we seem justified in concluding 

 definitely that caffeine has a retarding effect upon the rate of 

 learning the maze. 



H 2 0o.iocc. 



Ca.o.Somy. 



Hx.0 0.10 c.c. 



Ca o.somq. 



Ca. i.oomg. 



100 



200 



300 



400 



500 



600 



700 



800 



FIG. 4. A comparison of the total excess distance over the shortest path 

 through the maze traversed by animals under the influence of caffeine and by 

 normal animals. Arranged as figure 2. 



THE EFFECT OF THE DRUGS UPON RETENTION 



When each animal in experiment 1 had made a perfect record 

 it was kept without further training for twenty days and was 

 then tested for retention without further administration of the 

 drugs. The criterion used in testing relearning was the same 

 as that for learning: three successive trials without error. The 

 average numbers of trials required by the different groups to 

 reach this standard are given in table 9. The range of varia- 

 tion in each group was very great, from 1 to 30 or more trials 



