COLOR CHANGES AND ADAPTATION IN FISHES. 1 89 



ground, as maintained by Pitkin and Loeb, pointed out in the introduction, or is it 

 the result of some other phenomena? 



By referring to table 11 and comparing figures 12, 26, and 27, all reproductions of 

 photographs of the same individual on backgrounds having black and white areas equal 

 in size but different in form, it will be seen that the pattern in the skin of the fish is 

 practically identical in all. The same dark spots and patches are found in all, and they 

 have the same form and relative position, although in the different backgrounds the 

 light and dark areas differ greatly both in form and in spacial interrelationship. 



The same will be found to be true in comparing figures 23, 24, and 25. In figure 



23 the background was changed every time the fish moved, so as continuously to keep 

 the long axis of the fish parallel with the short axis of the light and dark areas ; in figure 



24 it was so regulated as to keep this axis parallel with the long axis of these areas; and 

 in figure 25 the background was continuously but very slowly rotated. Thus, in figure 

 23 the light and dark areas, owing to foreshortening from the flounder's angle of vision, 

 appeared nearly square, in figure 24 very much elongated, and in figure 25 the appear- 

 ance changed continuously. Yet the pattern produced in the skin by these different 

 backgrounds appeared to be the same, even in details visible only under considerable 

 magnification. This may be seen by comparing with a lens the photographs repro- 

 duced in the three figures mentioned above, and even more clearly by studying those 

 reproduced in figures 31-40. 



Figures 33 and 34 represent the same individual on different backgrounds. In 

 figure 33 the specimen was in a shallow granite pan which was dark blue speckled with 

 white. The white spots, as shown in the photograph, varied in form and size and 

 were scattered promiscuously over the surface. Figure 34 represents the same indi- 

 vidual on an artificial black and white background. The white areas in this back- 

 ground covered about the same proportion of the whole as the white areas did in the 

 granite pan, and they were on an average about the same size on the two backgrounds; 

 but in the former they were regularly arranged and fairly, uniform in shape, while in 

 the latter they were irregularly arranged and varied much in form. 



By examining these two figures it will be seen, however, that there is a striking 

 similarity in the patterns produced in the skin. On the artificial background there are 

 in the skin a number of light areas which resemble the white spots in the background 

 quite closely in size but not at all in form and arrangement. In the pattern produced 

 in the pan these light areas are all present, moreover they are practically the same in size 

 and form and have precisely the same relative position, respectively, as those in the 

 pattern produced by the artificial background, although the form and arrangement of 

 the white spots in the two backgrounds differ greatly. This is well illustrated in figures 

 39 and 40, which are photographic enlargements of a small region taken somewhat 

 above the ventral ocellus in figures 33 and 34, respectively. The same characteristic 

 is shown in figures 31 and 32. In figure 32 the white areas in the background are some- 

 what larger than they are in the pan, represented in figure 31. The white spots in the 

 skin are likewise somewhat larger in the former, but they have precisely the same spacial 

 interrelationship in both. In neither, however, is there any similarity in arrangement 

 between the spots in the skin and those in the background. This is clearly shown in 

 figures 35, 36, 37, and 38. These figures are photographic enlargements of certain areas 

 in figures 31 and 32. The magnification in all is approximately nine diameters. Figures 

 97867°— vol 34—16 13 



