238 BULLETIN OF THE BUREAU OF FISHERIES. 



Fig. 57. P. albigtittus, 15 cm. long, on green from August 11 to 26. 



Fig. 58. P. albigutius, ig cm. long, on dark blue from August 7 to 8. This specimen became bluish 

 exceptionally rapidly. It remained on dark blue four days longer and became but little more blue 

 than it was when autochromed. 



Fig. 59. P. albigutius, 20 cm. long, on vermilion from July 27 to August 26. 



Plate XXXIV. 



Fig. 60. Ancylopsetta quadrocellata, 14 cm. long, on pink (Ridgivay 's alizarin pink) from August 3 

 to September 11. Excellent reproduction. 



Fig. 61. .4. quadrocellata, 17 cm. long, on dark blue from August 22 to 20. 



Fig. 62. Same specimen on green from August 26, 4 p. m. to August 30, 2.30 p. m. When first 

 transferred, this fish appeared even bluer in the green box than it did when it was in the blue box, and 

 there was no evidence of green whatever. During the following four days the blue became fainter and 

 the fish gradually assumed a greenish tint. But at the end of this time there was still considerable blue 

 in the skin, as the autochrome shows. This demonstrates conclusively that the color of the skin as repro- 

 duced in all of the figures on this plate is due to internal structures and not to colored light reflected from 

 the siu-face or transmitted through the fish, as has been suggested by a number of those who have seen the 

 original autochromes. This is more conclusively demonstrated in the reproduction of autochromes taken 

 later (figs. 64, 65, and 67). 



Fig. 63. A. quadrocellata, 16 cm. long, on chrome yellow from July 27 to August 26. 



Plate XXXV. 



Fig. 64. P. albigutius, 15 cm. long, on blue from August 20 to September 1 1 ; autochromed in a green 

 box. This proves that the color shown in the autochromes is due to the structure of the skin and not to 

 light reflected from the background as many who saw the originals maintained. 



Fig. 65. Four specimens of P. albigutius, adapted to blue, green, yellow, and pink backgrounds, 

 respectively, autochromed on green. Color in all but the yellow ore is quite faithfully reproduced. 



Fig. 66. P. dentatus, 14 cm. long, in a variegated blue and white granite pan from August 19 to 

 September 10. The brown spots are due to particles of rust. Colors faithfully reproduced. 



Fig. 67. Paralichthys, same individuals adapted to the same backgrounds as in figure 65. Auto- 

 chromed on a white background. The color in all but the smallest specimen is fairly faithfully repro- 

 duced. The smallest specimen appeared much greener in life and also in another autochrome taken at 

 the same time than it does in this figure. 



Plate XXXVI. 



Figs. 68-76. Photographs taken with orthonon plates of specimens adapted to colored backgrounds. 

 Fig. 68, brown; 69, dark blue; 70, maroon; 71, green; 72, vermilion; 73, light blue; 74, dark blue; 

 75, green; 76, chrome yellow. Note the remarkable simulation of the background in shade and the 

 absence of conspicuous patterns, especially in Paralichthys. The conspicuous dark areas and light spots 

 showTi in figure 70 were present only after abnormal stimulation. 



Plate XXXVII. 



Fig. 77. Photograph of a group of Paralichthys on a white background containing considerable debris 

 scattered over it. A'l of the specimens are albigutius except the large one above and to the left, which 

 is dentatus. Note that this specimen has five dark areas, two rather inconspicous ones near the head and 

 three near the posterior end. In some specimens there are several others nearly as conspicuous as these. 

 In albigutius there are but three, the two near the posterior end At the base of the fin being absent. 



The difference in shade in the different individuals is largely due to difference in their past 

 experience. 



