A STUDY OF THE BEHAVIOR OF THE CROW 85 



A significant incident is the following. In one of the trials* 

 number 4 accidentally fell through a five inch space which had 

 been left before the entrance doors in order that the crow should 

 not too closely approach a compartment unless it intended to 

 enter it. The bird fell to the ground beneath the apparatus, 

 finding there some pieces of bread which had been dropped 

 earlier in the day. Naturally enough, it ate them before it 

 could be induced to return to the fly. Ever thereafter, until 

 this crack had been closed, this bird, as it approached the com- 

 partments, would look through the crack to the ground. Several 

 times it flew down in search of food. 



RESULTS, PROBLEM 1 



With the final series of trials given on June 30th, regular 

 experiments were initiated. The problem which the birds were 

 required to solve was that of learning to select the first open door 

 at the right. 



Ten settings as we shall call them, were chosen by the exper- 

 imenters. These are given below, numbered 1 to 10. After 

 each number appears the series of open doors ; in the next column, 

 the total number of doors open; and finally in the last column, 

 the number of the right compartment in which the reward of 

 food might be obtained. 



Problem 1. First door at the subject's right to be chosen 

 Settings Doors open No. of doors open No. of right door 



1 7.8.9 3 9 



2 2.3.4 3 4 



3 3.4.5.6.7 5 7 



4 1.2 2 2 



5 2.3.4.5.6 5 6 



6 6.7.8 3 8 



7 3.4.5 3 5 



8 2.3.4.5.6 5 6 



9 1.2.3 3 3 



10 7.8.9 3 9 



In this series of ten settings, a total of thirty-five doors were 

 open, of which number, ten were of course "right doors." Con- 

 sequently, the chance of a selection of the right door, without 

 previous experience or trial, is one to two and one-half. 



In general, it was the purpose of the experimenter, as far as 

 possible, to follow through this series of settings from 1 to 10, 

 and then to return to the beginning and repeat the series. No 

 matter how many trials in succession could be given, the exper- 



