10 THOMAS WILLIAM BROCKBANK 



the " redintegration " trials. These trials were so desig- 

 nated because, following the norm of learning in the redin- 

 tegration trials after the period of disuse of the habit, the 

 rat did actually reestablish the integrated movements of 

 the original learned habit. It was seen further that, by 

 completing the redintegration trials, one could arrive at a 

 more definite conclusion as to what movements of the 

 habit had been lost and what retained. Retention here is 

 to be understood as the potential perservation in the 

 organism of those certain precise integrations of movement 

 necessary in the habit. The word " retention " will be used 

 in reference also to the period of disuse of the habit — the 

 time between learning and redintegration. 



After the completion of learning, all rats were held over a 

 retention period for redintegration. During the retention 

 period, which according to the group was respectively 70 

 days, 45 days, or 30 days, the rats were required to run 

 through a runway about 900 cm. in length, which was 

 considerably more than the exercise they received in the 

 last few trials of learning. The amount of exercise thus 

 given was found to be sufficient to avoid the evils that 

 arise from allowing the rats to remain inactive during the 

 retention period. The rats were thus kept in normal con- 

 dition throughout the whole experiment, comprising the 

 successive periods of learning, retention, and redintegration. 



The first trial in the redintegration series was not pre- 

 ceded by any preliminary feeding in the center of the maze. 

 On the day the redintegration trials were to begin, instead 

 of being exercised in the runway the rat was taken directly 

 to the entrance box of the problem and the procedure as 

 in learning was followed. It was in this stage of the work 

 that our opinion concerning the time norm as a secondary 

 criterion was confirmed. More than one rat was observed 

 to travel through the maze in relatively slow time, but 

 without loss in integration of movements, while others, 

 moving more quickly, a certain and definite loss. The 

 redintegration trials followed to the end the same norm 

 as the learning. 



This is an adopted and modified system, taken from the theory which Dr. Ul- 

 rich recently advanced, that nearly all " errors " in the maze, after those in the first 

 few trials, are consequent upon the rat's making an imperfect turn (integration), or 

 not making a turn where one should be made. 



