168 CORA D. REEVES 



Discussion of results. — The large number of trials (700) neces- 

 sary to establish this discrimination seems to indicate how small 

 a part the visual stimulus, has in the daily life of a rat. Had 

 it not been for the records of the first week and especially of 

 rat C (fig. 3) and the lack of any adequate explanation other 

 than ' ' learning to discriminate ' ' which would account for the 

 improvement both in choice and in time, the conclusions (after 

 each had had 200 trials) would have been that rats cannot dis- 

 criminate a moving and still light. (See table.) The rats came 

 in time to select the food box more accurately. In the fact that 

 the rats halted at the doorway of the boxes where they had 

 not been receiving food is an indication that they used other 

 criteria than the lights when they reached the food box. They 

 were, however, not able to select the right box when the con- 

 dition of movement of the lamps was changed. It seems pos- 

 sible that some laboratory failures to find that animals possess 

 as acute sensory mechanisms as have been popularly ascribed 

 to them may be from the fact that the problem presented was 

 not fitted to the animals tested. This incident will illustrate. 

 At the close of this series of experiments an old rat which had 

 been handled continually for some months was running about 

 the room when the writer chanced to be winding up a piece of 

 cord. As the end of the cord was drawn along the floor the rat 

 followed, patted the end, for some eighteen inches, as a kitten 

 would. This rat was tested several times with a cord but would 

 never repeat the behavior. The rat became familiar with a 

 new situation quickly but the stimulation afforded by a moving 

 object could not be doubted. The effectiveness of movement 

 in controlling reactions is shown in the difference in the curves 

 (fig. 5) and in the path of the rat as shown in fig. 6. The length- 

 ened average time when 20 trials per day were used instead of 

 10 confirms the evidence already presented that a large number 

 of daily repetitions is not the most advantageous method of 

 establishing a given discrimination. 



CONCLUSIONS 



1. Rats can and do discriminate a stationary from a moving 

 light. 



2. Rats show some tendency to approach a moving rather 

 than a stationary light. 



