212 BULLETIN OF THE BUREAU OF FISHERIES. 



surface becomes reddish brown. This marked difference in effect must be due to the 

 difference in the illumination from above." How can this be explained? 



Under normal conditions adaptation to a colored background, as demonstrated 

 elsewhere, involves two processes. One results in adaptation in shade, the other in 

 adaptation in color. Adaptation in shade, as the tests with the gray in particular 

 indicate, is, at least to some extent, dependent upon the ratio between the intensity 

 of the direct and the reflected light. A decrease in this ratio causes concentration and 

 an increase causes distribution of the black pigment in the melanophores. The distri- 

 bution of the guanin crystals may also be affected but in the reverse manner, so that 

 when the dark melanin granules become concentrated the white guanin crystals spread 

 out and cover them over. When the red background was illuminated only from below, 

 this ratio became very small, and therefore adaptation in shade required concentration 

 of the melanin and distribution of the guanin; but adaptation in color required just 

 the reverse, together with distribution of the red and yellow pigment. Thus under the 

 conditions of the tests there was an antagonism between the two processes involved; 

 and this accounts for the resulting limitation of the brown color to the ocelli and dark 

 spots, in which it requires less stimulation than elsewhere to cause an expansion of 

 the melanin. Distribution of the red and yellow pigment was no doubt masked in all 

 other regions by the distribution of the guanin which lies nearer the surface. Hence 

 the extreme whiteness in these regions. 



This leads directly to a discussion of the whole question concerning the relation 

 between the different structures in the skin and the production of the different shades, 

 colors, and patterns that have been observed in fishes. This question, however, I 

 hope to treat more fully in a later paper. 



Experiments (B), Light from Above Relatively Stronger than Normal. 



In these experiments a specimen of P. alhigutius 20 cm. long was used. This 

 specimen had previously been transferred frequently from a white to a black back- 

 ground and vice versa; and at the time the experiments were made adaptation in shade 

 to either of these backgrounds occurred very rapidly. 



The fish was put into a 22 cm. crystallizing dish on a pure white background and 

 covered with the black cylinder, which was lined with white cloth extending from the 

 bottom to a line several centimeters above the surface of the water. A piece of white 

 wire screen was so bent and adjusted in the glass dish as to prevent the fish from turning. 

 The top of the cylinder was nearly closed with opaque cloth, making the intensity of 

 the light within, so low that the white background appeared decidedly gray. A tube 

 3 cm. in diameter and 45 cm. long was fastened so as to extend down through an opening 

 in the opaque cloth, directly over the eyes of the fish. Through this tube a beam of 

 light from the sky was reflected, by means of mirrors (fig. 3, p. 213). This tube was 

 so adjusted that the beam of reflected light illuminated the eyes and the skin about 

 them, but not the background, and it was readjusted whenever the fish moved. In 



a The fact that the fish became nearly maximum white on the red background illuminated from below indicates that the 

 melanophores, groups of black pigment-bearing cells, respond to monochromatic light much as they do to white light; that 

 the stimulus which afiects these organs bears a definite relation to the quantity of light regardless of the quality. It would be 

 interesting to ascertain whether or not this relation is the same for all colors — that is, if a given amount of Ught energy has the 

 same stimulating effect regardless of the length of the waves. 



