TRAINING IN WHITE RATS UPON VARIOUS SERIES OF MAZES 3 



inches long, four inches wide and five inches deep. In 

 the C maze they were thirteen inches long, four inches 

 wide and four inches deep. The mazes were built of half- 

 inch boards, the partitions between the runways in each 

 section being of metal, although the partitions between 

 the sections themselves were of wood. By a section is 

 here meant one of the four main divisions of the maze, 

 in which in all the mazes except C the blind-alleys run 

 in the same direction relative to the observer. By a run- 

 way is meant one length of the maze, such as a — b in 

 maze A, whether it forms part of the true pathway or 

 constitutes a blind alley. The terms, " true pathway " 

 and " blind alley," are open to no ambiguity. The mazes 

 were painted black without and within, and were covered 

 with glass tops. 



The rats were between two and three months of age at 

 the beginning of the particular experiment in which they 

 were used. About one third of them were born in the 

 laboratory where the experiments were carried on, and the 

 remainder were obtained elsewhere. The rats were allotted 

 to the various groups in a thoroughly promiscuous way, so 

 that the groups approximated unselected collections. They 

 were fed in the food-box of the maze every day for a week 

 before they were introduced into the maze proper. They 

 were given one trial a day for the first five days, and after- 

 wards two trials a day, one in the forenoon and one in 

 the afternoon. They were fed for two minutes after the 

 trial in the forenoon and for five minutes after the trial 

 in the afternoon. The food was bread and milk with an 

 occasional allowance of sunflower seed. The rats were 

 transferred from one maze to another by groups. Those 

 individuals in a group which learned a maze earliest were 

 still given one trial a day upon the same maze until all 

 had learned it. 



Records were kept of the number of trials necessary 

 to learn to run a maze four times out of five without an 

 error, of the errors made during the course of learning 

 and of the time required for each trial. The only errors 

 that were counted were those made by the rat running 

 into blind alleys while moving in the forward direction, 



