26G HARVEY CARR 



per cent of the trials. The disturbance due to the alteration 

 persisted to some extent on the subsequent day's test in normal 

 conditions. The number of errors per rat ranged from 9 to 15. 

 A repetition of the test for each of the three novel situations 

 exhibited a pronounced tendency toward adaptation, but the 

 experiment was not continued until complete adaptation was 

 secured. 



Position of the Maze. After being learned, the maze was 

 removed to a new position in the laboratory but its original 

 cardinal orientation was preserved. The maze was shifted about 

 twelve feet in position but the shift was of such a character 

 that the maze was now situated in practically a new environ- 

 ment. This procedure involved two alterations; a change in 

 the objective environment while running the maze, and a new 

 route from the living cage to the maze. The latter factor has 

 been shown to be non-effective and may thus be disregarded. 

 Six rats were given three tests in the new position, and four 

 were affected. These made errors in 55 per cent of the trials, 

 and gave an average error record of 2.08. The number of 

 errors per rat ranged from 4 to 10. The animals adapted 

 quickly to the novel conditions, and in some cases a slight 

 disturbance was evident on a return to the old position. 



Rotation of Maze. In this experiment the canvas top was 

 not used, and as a consequence the maze was rotated in reference 

 to a stationary heterogeneous environment. This experiment 

 was first performed by Professor Watson and our results are in 

 harmony with those secured by him. Unless otherwise specified, 

 the three novel positions utilized were 90, 180, and 270 degrees. 

 The tests were conducted on different mazes and with different 

 procedures and thus need to be described separately. 



1. The glass covered maze was used and tests were given 

 for the three novel positions on successive days followed by a 

 return to the original position on the fourth day. This pro- 

 cedure was now repeated to determine the effect of adaptation. 

 Ten rats were employed and all were disturbed. In the first 

 shift, errors were present in 65 per cent of the trials, and an 

 average error record of 6.95 was secured. The induced effect 

 was occasionally carried over to the subsequent day's trial in 

 the normal position. A repetition of the shifts disclosed a 

 pronounced adaptive tendency. All members of the group were 



