400 JOHN LINCK ULRICH 



other mechanisms are functionally developed. The degree of 

 the efficiency of the organism depends upon the functional con- 

 dition of the mechanism for the production of reflex excitability. 

 In some rats reflex excitability is quite absent, and in some, it 

 is manifest for only a short time. The duration of its existence 

 is with most rats 5 or 10 minutes at the most, yet some rats show 

 excitability a greater part of the 30 minutes given to solve the 

 problem. Usually on each of the 6 days reflex excitability is 

 manifest at the beginning of each trial and lasts a shorter time on 

 each additional trial, or disappears completely after the third 

 or the fourth day. At these times or on the sixth day, some rats 

 very often appear less vigorous or are totally exhausted. Later in 

 this paper, we shall have occasion to mention the appearance of 

 exhaustion, and suggest conditions which probably determine it. 



The foregoing descriptive results from observations of rats in 

 the first trial are from 24 litters, a total of 156 rats. If reflex ex- 

 citability determines the readiness with which the problem is 

 solved on the first trial, other conditions in addition must de- 

 termine the rapidity and the effectiveness of learning. The 

 learning records of 14 litters, 80 rats, are presented in table 2. 

 The results from 10 litters are not given for the problem was un- 

 solved after 6 daily trials of 30 minutes each. 



This table gives the number of individual rats in a litter, in 

 how many instances the problem was solved, in how many it 

 remained unsolved, and the day on which solution occurred. 

 Learning was thought to be complete when for 15 times direct 

 progression from the entrance box to the latch occurred, and the 

 neck reflex was produced to raise the latch. Slight, imperfect 

 responses made near the latch were not considered in this reck- 

 oning. This determination of the point when learning was 

 thought complete was experimentally established when in the 

 preliminary work with 10 rats or more, it wag seen that 15 abso- 

 lutely perfect records could not be obtained when from 60 to 

 125 trials were given. 



Of the 80 rats recorded in table 2, solution of the problem oc- 

 curred only in 40 per cent. On the first day solutions took place 

 in 47 per cent of rats, on the second and third days in 19 per 



