ALCOHOLISM AND BEHAVIOR OF WHITE RATS 219 



Rat 1265: Time record for twenty-first trial lost; averages 

 for second half of training, all training, training and retention 

 exclude this trial. In obtaining total tinie for the seventh day 

 of training, there was substituted for the twenty-first trial a 

 time based on its distance record. 



Rat 1211: Showed very unusual behavior on the second day 

 of retention, taking five minutes when it might have been ex- 

 pected to take five seconds. The rat was obviously sick, so its 

 records have been omitted from the average for retention. For 

 training and retention its exclusion makes such a slight difference 

 that only the averages including it have been given. 



Rat 1312: Failed to reach the center on the first trial on the 

 second day and did not succeed until the first trial on the fourth 

 day. 



Rat 1169: Failed on the first trial on the second day and suc- 

 ceeded on the first trial on the third day. These last two rats 

 are the only cases where a w^hole day was lost through failure; 

 to give the arbitrary maximum in these cases seemed to make 

 the result depend too heavily upon the arbitrary figures. In 

 one case only half of the average would be based on actual records 

 and in the other, three-quarters would be based on actual records. 

 These two rats are so outstandingly different from the other 

 'failure' rats that it has seemed more fair to omit them from the 

 averages including the first three days. They are included in 

 the averages for the second half of training and for retention 

 when the other 'failure' rats are included. 



Rats 1193, 1168, and 1173 had no perfect trials in all of the 

 thirty-six trials of training and retention. In the summaries 

 on the number of trials before the first perfect trial, these rats 

 have been given a score of 36, as though they had had a perfect 

 trial on the thirty-seventh trial (which was not given). To 

 give averages free from this difficulty, these three rats are omitted 

 from the alternative averages. 



