30 

 speot to clays. 



Both the distribution ourve Plate III and Curve VI 

 indicate clearly that not all perfectly normal rats 

 used \lve trials were given. The num- 



ber of days and the total time required by these rats 

 is sho'.vn on the distribution curve. Those of three 

 trials daily requiring 69,78 and 89 trials, and of five 

 trials requiring 150,165 and 175 trials, before reaching 

 the norm, must be considered abnormal. The average 

 number of trials required for one, three and five trials 

 daily when normal rats ^d was for one trial £4, 

 for three trials 40 t and for five trials 56. These dif- 

 ferences in the number of trials correspond more closely 

 with the results of the latch box. 



likewise Ourve VI shows, because of the presence of 

 these abnormal rats, different points of Interest. The 

 general appearance of the individual curves is the same, 

 however, those representing throe and five trials being 



