1 86 



BULLETIN OF THE BUREAU OF FISHEMES. 



not shown in the photographs, was fully as striking as the adaptation in shade and 

 pattern. In the specimens on the shells the skin was very distinctly brownish yellow; 

 in those on the gray sand it was yellowish gray; while in those on the black sand there 

 was no trace of yellow or brown, nothing but black and white. The adaptation in color, 

 however, was more clearly evident in specimens kept in boxes painted on the inside. 

 These experiments we shall consider in detail later. 



Thus by changes in the shade and color, and in the pattern in the skin and by partly 

 burying, these animals become very effectively concealed on a variety of different bot- 

 toms. But there is still another peculiarity in their reactions that serves this same pur- 

 pose. When they lie on the bottom the central portion of the body is usually consider- 

 ably raised in such a way as to form a channel from the lower gill to the posterior end, 

 and while in, this position they frequently breathe only through this gill, the exposed gill 

 being closed and perfectly quiet. Thus when buried in sand the water enters the mouth, 

 passes through the lower gill, back under the body and up on either side of the caudal fin 

 (fig. 5), where it can be seen oozing up through the sand, moving the grains slightly. 

 This is the only movement that can be detected, except a very slight and inconspicuous 

 movement in the mouth. As to the function of the concealment we have all sorts of 

 suggestions, but as yet no direct e\'idence. I hope, however, to deal with this matter 

 experimentally in the near future. 



BLACK AND WHITE BACKGROUND, ARTIFICIAL. 



While the study of Paralichihys and Ancylopsetta on natural bottoms shows conclu- 

 sively that the skin tends to assume a pattern such as to make the fish inconspicuous, it 

 shows but little concerning the relation between the pattern assumed and that found in 

 the background. It does not tell us precisely why the pattern in the skin resembles the 



bottom. It gives but little knowledge con- 



s-.. 



I 



ceming the degree of accuracy with which 

 the configurations in the bottom are re- 

 produced. For this purpose it is essential 

 to be able to test the effect of figures which 

 can be varied as desired with reference to 

 size, form, and interrelationship. This can 

 readily be done by means of artificial back- 

 grounds. Such backgrounds were made 

 by painting, with India ink, various figures 

 on bristol board. 



The fish to be tested were in all cases 

 putinto2i cm .glass crystallizing dishes and 

 these, containing water about 2 cm. deep, 

 were set in water of the same depth on the 

 painted boards, which were always put between two thin clear glass plates so as to pro- 

 tect them against the action of the water. To change the background it was only neces- 

 sary to remove one board and push under another with a different figure. In every case 

 all bubbles of air were removed from under the dish. This is of the greatest importance, 

 for if not removed there is so much refraction and reflection that the figures may appear 

 much distorted, and if the air is sufficient to form a continuous layer many of them can 



FlG. I. — Vertical section of apparatus used in studying adapta- 

 tion to artificial backgrounds. A. aquarium; CD, crystalliz- 

 ing dish; S, siphon; I. inlet for water; O, outlet. 



