EXPERIMENTS ON PROTECTIVE COLORATION 459 



endeavored to neither increase or decrease the contrast in any 

 case, but to reproduce as nearly as may be the effect as it ap- 

 peared to my eye. If I have erred in any direction, it has been 

 in minimizing the effect of the contrast and resemblance. In 

 every case the camera was placed much nearer (from 0.2 to 3.0 m.) 

 the backgrounds than were the birds at the commencement 

 of the experiments,^ thus diminishing the effect of the contrast 

 and resemblance. With the exception of figures 8, 11, 12, 13, 

 17, 18, 21, 22, 23, 49, 50, 51, 52, and 53, the photographs were 

 made with dry specimens. The color changes, if any, were not, 

 however, great enough to be appreciable in the photographs. 



Light and shade moreover in many instances altered the color 

 effect. Thus in figures 25, 27, 41 to 43, 46 the effect of the 

 reflected hght has been to make a dark insect (Gryllus, Silpha) 

 appear light against a dark background. The same effect of 

 course occurs in nature, so that the photographs are probably 

 reasonably true to life in this respect. 



SERIES I 



Plate 1, figures 12, 17 and 18 



In the first series of these experiments, I employed two individ- 

 uals of the common crow (Corvus americanus) and frogs (Rana 

 cantabrigensis and Rana pipiens). The crows were kept in 

 a cage 0.9 x 0.9 x 1.2 m. in size, on the top of which was a perch 

 0.7 m. from the floor. On this floor, I prepared two backgrounds, 

 one of mud or fresh grass, the former of which harmonized very 

 closely with the color of R. cantabrigensis and the latter with 

 that of R. pipiens, and the other of dry sand, with which the 

 color of the frogs made a decided contrast. Upon each of these 

 backgrounds a frog was placed, rendered insensible by sharply 

 tapping the skull. The crows soon learned to drop from the 

 perch and seize the frogs,"* and a record was kept of the number 

 of frogs taken from the mud and sand respectively, in the hope 



^ In many cases, as I shall explain later, the birds were close to the back- 

 grounds at the time the experiments became effective. 



^ When first fed frogs, the crows appeared suspicious Of them, but soon learned 

 to eat them readily. 



