Yoakum, Behavior of Squirrels. 549 



any sort, beyoud learning to run in the wheel common to squirrel 

 cages. All were in good condition. 



Table II shows the number of trials, time of each trial for each 

 squirrel, the average time for the four animals, the average time for 

 squirrels Nos. 2 and 3, which had been previously trained, and for 

 the untrained animals, Nos. 1 and 4. Fig. 2 represents the last 

 two sets of averages. A is the curve for squirrels 1 and 4, and AB, 

 the curve for I*^os. 2 and 3. 



The two curves are strikingly dissimilar. Trials 1 and 2, of curve 

 AB, as shown below, are not the average trials of the two animals 

 Nos. 2 and 3. Table II shows that at both of these trials squirrel 

 1^0. 2 failed to open the problem-box. The curve AB, showing the 

 averages of those animals that had learned the sawdust and outside 

 latch hexes, is very irregular and does not fall below one minute 

 until after the fourteenth trial. The curve A of the two animals, 

 which came to the problem without previous training on problem- 

 boxes of the type, is a typical learning curve. ^^ 



In all the early trials, both experienced animals were decidedly 

 uncertain and irregular in their movements and in their mode of 

 attack. The notes taken at this time are definite in their explanation 

 of this difference in the learning process. The following extracts 

 from the diai-y record will serve to show how the previously acquired 

 habits were "carried over" to the new problem : 



Aug. 1, '07, 12 M. Female No. 3 (trained), second trial, ran immediately 

 to door, nosed it and then walked around box ; nosed string ; climbed on top 

 of box, then off and went to south side ; back to door ; then back of cage ; then 

 to south side, next to door and then completely around again ; scratched at 

 bottom of cage ; went to south side ; next to string, nosed it and then mounted 



""X" in the table and curve shows where squirrel No. 4 suffered a severe 

 fright during the night preceding the thirteenth test. A cat had been inad- 

 vertently fastened in the basement where the squirrels were kept, though not 

 in the same room. She had evidently attempted to get out, and failing, kept 

 up a continuous "meowing" for a large part of the night and was still making 

 considerable disturbance in the morning. At each cry of the cat the squirrel 

 would tremble and crouch, and frequently give the peculiar cry of a squirrel 

 when cornered or badly frightened. The cat was removed early, but at the 

 noon hour, though apparently recovered from fright, the squirrel made abortive 

 and unsuccessful attempts to enter the problem-box. 



