Abstract
30 exptl. rats (injected subcut. with 0.5 mg. of benzedrine sulphate) and 30 non-injected rats were given one trial on the Stone Maze. The exptl. animals showed little decrease in the number of errors made per day, while the controls learned the maze quite readily. The critical ratios between the number of errors for each day for the benzedrine and control groups were significant. On the 23rd day of running the injected animals were given physiological saline instead of the drug; the errors dropped quite rapidly. Resumption of the drug caused the errors to increase again to a chance level; as the author points out this increase can hardly be regarded as "unlearning." Benzedrine evidently retards general performance levels on a maze.