304 Journal of Comparative Neurology and Psychology. 



of a time when a quick retreat would bring them under the 

 friendly cover of a tuft of grass or a little hillock. If the ex- 

 perimenter attempts to catch them they dodge under hay or any 

 cover at hand. No cover being forthcoming, the spot next 

 chosen for safety is a corner of the cage where they huddle up 

 and watch proceedings. Unfortunately laboratory cages were 

 not made to harmonize with their particular color effects, so 

 that the desired result of being invisible is not accomplished. 

 In order to study this point a cage was fitted up simulating the 

 scenery of their grassy South American home. To obtain the 

 best conditions the light was rendered dim by a high board 

 fence, while sticks, stones and mounds of earth completed the 

 realism. It is thought, from observations of both young and 

 old under these conditions, that they rely for escape, not so 

 much on protective coloration, as upon hiding under grass in 

 little inequalities of the ground. However, the English variety, 

 which is presumably nearer the original than any other in the 

 laboratory, is particularly harmonious in color with surrounding 

 grassy mounds. 



When undisturbed, the guinea pigs wander contentedly 

 around and nibble grass, but let them suspect that an enemy 

 may spring upon them and they approach their food only by 

 making a bold dash out of their retreat, and drag the food back 

 into a dark corner. In the experimental work this was almost 

 invariably the way food was seized from the boxes, even when 

 the environment was uniformly lighted, so that there could have 

 been no immediate advantage in snatching the food backward a 

 few inches. At first my own movements attracted much atten- 

 tion from certain individuals. This has to be taken into ac- 

 count in the first series of experiments, as will be noted. How- 

 ever, when the strangeness has worn off and the work becomes 

 habitual this factor is greatly reduced, if not entirely eliminated.^ 



At first perfect quietness in the room is apt to delay reac- 

 tion with all the individuals upon which I have worked. After 



' THORNDrrCE has discussed the fact that dogs pay more attention to the ex- 

 perimenter and less to the experiment than do cats. Animal Intelligence, p. 38. 



