EXPERIMENTS ON THE RABBIT 151 



and half past three. At about nine in the morning the rabbits 

 were given a breakfast of oats. The food used in the tests was 

 always some form of green food, carrots, lettuce, apples, and in 

 season, clover. After the tests were over for the day the rabbits 

 were allowed to eat all the green food they wished. Usually, 

 they seemed to have excellent appetites for the food in the 

 experiment box, but there were variations, and since we had no 

 means of controlling and estimating the strength of our motive, 

 we do not wish to draw any conclusions regarding the speed 

 of the learning process involved. In case of a w r rong choice, the 

 only form of punishment involved was sometimes merely the 

 experience of pushing at a door that did not open. More often, 

 the experiment box was removed before the rabbit had had time 

 to go to the other door. But when a new discrimination was 

 being learned, for a short time the rabbit was allowed, after 

 pushing in vain at the wrong door, to go to the correct door and 

 get food. This was occasionally allowed, also, when an animal 

 had made a number of wrong choices in succession and seemed 

 in danger of becoming discouraged. 



Another factor which varied in different series, and whose 

 variations would prevent our drawing inferences regarding the 

 speed of the learning process, was the number of tests made on 

 a single animal in a day. In the earlier series only four were 

 made ; in some of the later ones eight, ten, or twelve. 



A choice was always counted as wrong if the rabbit went 

 near enough to the wrong door to touch it with the long whiskers, 

 for even the lightest touch on a fastened and wholly unyielding 

 door felt quite different from a touch on an unfastened door. 



Until the weather became very cold, the two rabbits tested 

 in the autumn of 1910 were kept in a hutch and small yard 

 built for them on the laboratory roof, and the experiment box 

 was placed in this yard for the tests. All the tests with the 

 rabbit used in the fall of 191 1 (Polly) were made in this manner. 

 Later in the winter of 1910-191:1 the rabbits, except Light Nose, 

 were kept and tested in one of the small rooms of the laboratory, 

 which was brightly lighted by a skylight and a north window. 

 For several hours a day during the winter they were allowed 

 the run of the large laboratory room. During the period when 

 the first series were being made, in the yard attached to the 

 hutch, the two rabbits used as subjects (Light Nose and Dark 



