COLOR CHANGES AND ADAPTATION IN FISHES. 1 95 



of the season (fig. 62) the specimen had not had time to become fully adapted. This 

 specimen, taken from the dark-blue box, appeared dark blue in color when first put into 

 the green box and, although it had been in this box nearly four days and had changed 

 color considerably, it was still distinctly blue when it was autochromed. This is fairly 

 well seen in the reproduction. A comparison of this figure with figure 61, which repre- 

 sents the same specimen fully adapted to blue, and with figure 63, which represents one 

 fully adapted to yellow, shows fairly conclusively that if the light reflected from the 

 sides of the boxes or transmitted through the animals has any effect at all on their ap- 

 parent color it is, with the exception of that in the fins and tail, of minor consequence. 

 And even the fins and tail in Paralichthys are so nearly opaque that very little light is 

 reflected through them from the bottom, as can be seen clearly by examining the photo- 

 graphs on artificial backgrounds, especially figures 17 and 26. 



The best evidence in support of this contention, however, is found in the following 

 detailed account of the changes in color, observed during the process of adaptation in 

 the individuals on the various backgrounds. This account also throws some light on 

 the nature of these changes and on the rate of adaptation. It will be given essentially 

 in the form in which it was written at the time the observations were made. 



Light brown. — July 27, Paralichthys alhiguttus, 15 cm. long, taken from the white 

 aquarium and put into the light-brown box (considerably darker than Bradley's 

 orange, shade no. i) ; it rapidly assumed a brownish color. July 3 1 , simulation of back- 

 ground in shade and color, good. August 10, 12.07 P- ™-. color yellowish brown with 

 numerous small bluish spots, excellent color adaptation as seen from a distance of about 

 75 cm. Autochromed and photographed; put into dark-blue box and compared with 

 specimen fully adapted to blue, it still appears brown, st'iking contrast. August 11, 

 bluish gray but not nearly so blue as the specimen fully adapted. At 2 p. m. it jumped 

 into the brown box, changed back to brown almost at once. Returned to the blue box. 

 August 12, more blue than on preceding day. August 19, simulation of background in 

 color and shade remarkably good. 



Dark blue. — August 7, a. m., P. albiguttus, 19 cm. long, taken from white aquarium 

 and put on dark blue (darker than Bradley's blue, shade no. 2). Became dark gray 

 almost at once. August 8, 10.37 a. m., autochromed. (Fig- S^-) Simulation of back- 

 ground in color and shade excellent. August 1 1 , if any change, somewhat darker and 

 bluer; pattern more nearly uniform. Put into the light-blue box. 



August 1 1 , 5 p. m., a specimen of P. albiguttus which had been in the green box since 

 July 27 and appeared distinctly green was put into the dark-blue box. In this box it 

 appeared quite as green as it did in the green box. At 6 p. m. it was still distinctly 

 green, showing a marked contrast with the specimen autochromed on the blue. August 

 12, 2 p. m., much grayer but still has slight greenish tint. August 19, no longer greenish; 

 appears much like specimen autochromed on the blue, which is still in the box. There 

 is no evidence indicating that the color of the fish is due to colored light reflected from 

 sides of box or transmitted through the animals, except in the fins and tails of some 

 specimens. 



On August 19 a specimen of P. albiguttus, thoroughly adapted in color to a brown, 

 water-soaked, cypress board, was put into the dark-blue box. In this box it appeared 

 as brown as it did when on the cypress board. A bright yellowish-brown stripe about 

 0.5 mm. wide, extending along the entire margins of the ocular opening in the sldn, was 



