igo BULLETIN OF THE BUREAU OF FISHERIES. 



35 and 38 represent the background and figures 36 and 37 a small portion of the surface 

 of the fish shown in figures 31 and 32, respectively. The section of the skin enlarged in 

 these figures is found in the same relative position as that enlarged in figures 39 and 40. 



A glance at these figures shows at once that there is no indication of an actual 

 reproduction of the background. In the granite pan the adaptation of the fishes to 

 the background in color and shade, as well as in pattern, was, as previously stated," 

 most remarkable. So closely did they resemble the bottom, in most instances, that 

 strangers rarely saw them until they were pointed out, and yet a comparison of figures 

 35 and 36 shows conclusively that the configuration of the light and dark areas in the 

 skin is strikingly different from that in the background. The pattern in the skin, on 

 the other hand, is almost identical, even in minute details, with that produced by the 

 artificial background. The different black and white areas in one are practically 

 identical in form and spacial interrelationship with those in the other, although some 

 are considerably larger on the artificial background than they are in the pan. This 

 is especially true of the prominent white spots, two of which are included in the areas 

 enlarged. 



The simulation of the background in Paralichthys , in so far as the pattern in the 

 skin is concerned, is due merely to the formation in the skin of light and dark areas 

 similar in size to those in the background. On a background with large figures the 

 light and dark areas in the skin become relatively large. On one with small figures 

 the pattern breaks up into small areas, black dots appearing in the larger white areas 

 and white ones in the larger black areas. The location of these dots and their form 

 are morphologically fixed. They are essentially the same in fishes over a background 

 consisting of alternate black and white stripes as they are over one consisting of alter- 

 nate black and white squares, or one consisting of black spots on a white field, or white 

 spots on a black field, or one consisting of light and dark areas irregular in form and 

 arrangement, as found in granite pans and on natural bottoms. 



All our evidence supports the general conclusion of Sumner with reference to 

 Rhomhoidichihys , stated in the following words (191 1, p. 468): "Squares, crossbands, 

 circles, etc., were never copied in any true sense," and contradicts the contention of 

 Pitkin and Loeb that there is an actual reproduction of the figures in the background. 

 It supports only with certain limitations Sumner's contention (p. 472) "that there 

 may be very specific relations between the distribution of light and shade in the back- 

 ground and the pigment pattern assumed by the fish." The size of the light and dark 

 areas in the background and the relative amount of surface covered by them have a profound 

 effect on the pattern produced in the skin, but the form and arrangement of these areas have, 

 at least within rather wide lim,its, none. 



Individuality of tJie pattern. — In some individuals adaptation is very much more 

 precise and is attained much more rapidly than in others. This, however, I think is 

 largely due, as I shall show later, to difference in experience. 



The characteristic markings in the skin of different individuals of the same species 

 appear very much alike, provided they are on the same background. This is clearly 

 seen by comparing figures 10, 19, 20, and 30, all photographs of different individuals of 

 the same species on similar backgrounds. In detail, however, there is considerable 

 variation in the patterns of these individuals, as a comparative examination with a lens 



» See the autochrome, fig. 66. 



