54-8 Journal of Comparative Neurology and Psychology. 



in August of that year and continued with many interruptions until 

 August, 1908. The time allotted to each day's test, even when the 

 work was conducted regularly, was very short and this fact accounts, 

 in part, for the irregular number of daily tests. 



The records of the learning of the sawdust box are shown in Table 

 I; Fig. 1 shows the learning process in graphic form. The records 

 obtained from the outside latch box are similar and are not given. 

 Both problems are far below the ability of the animals and demand 

 little intelligence beyond that required in the elimination of random 

 movements. 



Little need be said at this place by way of comment upon these 

 records. The squirrel's method of solving this problem is quite 

 similar to that of the rat. The only distinctive feature in the learn- 

 ing of the outside latch box was the possible use of vision and the 

 great eagerness with which the animals directed their attack upon 

 an apparently definite object {i. e., the bar). The food was visible 

 through the wires of the problem-boxes and seemed on many occa- 

 sions to awaken great activity by being thus visible. For example, 

 sawdust, shavings, etc., would more often be scratched away from the 

 side nearest the food, and at such places the fiercest and most per- 

 sistent attacks were more frequently made. Such movements were 

 made in the beginning of the experiments ; as soon as the method of 

 entering the box became in the least fixed, the food itself no longer 

 constituted the specific incentive for making the necessary move- 

 ments. All the energies of the animals were expended in getting into 

 the box in the quickest manner possible. The problem-box could 

 even be left empty and the squirrel would still make the run. How- 

 ever, unless very hungry, he could not be deceived in this way more 

 than twice; after that the box would lose its attraction.*^ 



(b) The Inside Latch Box. — Four animals learned this problem. 

 Two (Nos. 1 and 4) were new to any problem of the kind, though 

 one, E'o. 4, had previously learned the maze. N'os. 2 and 3 had 

 previously learned both the sawdust box and the outside latch box. 

 No. 1 had been in captivity six months, but had had no training of 



"The squirrels were trained to this point by Miss Chamberlain and Miss 

 Sprague. 



