INTERFERENCE OF HABITS IN THE WHITE RAT 57 



splendid physical condition for experimentation. At the close 

 of the period, they were retested on h. c. to the right. It is 

 needless to give the data in detail. No one of these rats aver- 

 aged above 70% for any 50 trials although their retraining ex- 

 tended through from 34 to 45 days. Their behavior at the 

 beginning of the retesting indicated that the apparatus and 

 method were still familiar to them, but that was all. The re- 

 sults as a whole indicate that these rats had lost all measurable 

 traces of the original training. It may be well that in a habit 

 so difficult as the present one continued or retained familiarity 

 is too slight an aid to manifest itself in shortening the period 

 of relearning. The disintegration of this habit in the white 

 rat apparently takes place between 60 and 90 days. The 60- 

 day tests indicated practically perfect retention at the close 

 of that period, but the two sets of data are not strictly com- 

 parable. The rats in the 60-day tests had been retrained at 

 different intervals on h. c. to the right after the original learn- 

 ing. Hence the habit was considerably overlearned. 



VI 



Effect on retention of learned vs. unlearned habits. — It would be 

 interesting to know just what went on in the rats' nervous 

 systems during the 30 and 60-day periods of training. We seem 

 forced to assume that certain synaptic connections have per- 

 sisted in spite of the attempts of incoming stimuli to disintegrate 

 them. Inasmuch as either continued training (?) or the lapse 

 of time will result in the disintegration of these connections, 

 definite problems arise under each condition. We have indi- 

 cated that with the mere lapse of time, the dissolution of the 

 particular habit in our rats occurred between 60 and 90 days. 

 The present section contributes data throwing light upon the 

 comparative disintegrations brought about in the h. c. habit 

 by the 30 days' ineffective training on B and by a period of 

 training during which B was mastered. 



Of the 18 rats used on the 30-day test described above, 9 

 made the standard 87 \°/ immediately upon being re-tested on 

 h. c. to the right. Four others did essentially as well. Two 

 hundred and seventy trials was the maximum period of re- 

 learning and was found in two rats. Table 3 gave the data in 



