14 



RUTLEDGE T. WILTBANK 



In pursuance of this method, there were twenty-five 

 maze-learnings, in twenty of which evidence might be 

 sought as to whether the positive transfer persists. Atten- 

 tion is drawn to the fact that the first group to learn a 

 given maze constituted a control-group for that maze. The 

 records of the groups upon these series of mazes are given 

 in tables 7 to 16. The average number of trials necessary 

 to learn a maze is given, and in connection with it the 

 mean variation of the individual rats from this average; 

 then the percentages of transfer for each maze are given, 

 these being computed by comparing the record of the 

 trial-group with that of the control-group in this particular 

 maze. The same method is followed with respect to errors 

 and time, and these are reckoned both on the basis of 

 the average total per rat and on that of the average per 

 trial. The percentages of transfer are shown graphically 

 in figures 6, 7, and 8. 



TABLE 7 



The Average Number of Trials per Rat in the Learning of a Series of Mazes 



Maze A. . 

 Maze B. . 

 Maze C. . 

 Maze D. . 

 Maze E. . 

 Maze A. . 

 Maze B. . 

 Maze C. . 

 Maze D. . 

 Maze E. . 



Group 1 

 26.62 ± 6.87 

 n.62± 5.28 



9.75± 3.31 

 25.56 ±12.67 



4.00 



6.62 ± 3.56 



Group 2 



28.'70±' 6.66 



12.20± 4.14 



27.60 ±10.93 



4.40± .56 



6.40± 1.92 



7.50± 3.91 



Group 3 



16.20±7.24 



18.00±8.03 



7.30± .49 



6.50±2.32 



13.20± .67 



6.20±1.84 



Group 4 



24.67±7.24 

 7.11±3.88 

 6.89±3.23 



11.44±3.06 

 9.00±2.64 

 8.89±3.96 



Group 5 



23.10±7.10 

 21.40±8.95 

 19.00±4.91 



7.00±1.51 

 22.10±4.87 



5.00± .83 



TABLE 8 



The Percentages of Transfer of Training in Trials through a Series of 



Mazes 



Group 2 



33 



—12 



81 



76 



74 



Group 3 



27 

 68 

 75 

 54 

 62 



Group 4 



69 



74 

 60 

 44 

 64 



Group 5 



20 

 34 

 57 

 4.33 



78 



