COLOR CHANGES AND ADAPTATION IN FISHES. 237 



Plates XXXI, XXXII. — Limits of vision, keenness of discrimination, and adaptation to 



gray background. 



Plate XXXI. 



Fig. 48. P. albigutius, 12 cm. long, fully adapted to a very light gray background. This specimen 

 continuously appeared somewhat darker than the background. Note that the three ocelli have almost 

 entirely disappeared. 



Fig. 49. Individual (B) (fig. 9), on black and white background (dots 0.5 mm. in diameter) from 

 August 16, 2 p. m., to August 17, 9 a. m. Dots of this size, no matter how numerous, had no effect except 

 perhaps to cause the skin to become uniformly slightly grayer. 



Fig. 50. Individual (B) on black and white background (dots i mm. in diameter) on August 5, 9.43 

 a. m. to 4 p. m., after having been fully adapted to i cm. squares. These spots have no appreciable 

 effect on the skin unless they are more numerous than they are in this background. 



Fig. 51. Individual (B) on black and white background (dots i mm. in diameter) on August 20, 

 10 a. m. to 4.4s p. m., after having been fully adapted to dots 2 mm. in diameter. Note that the posterior 

 ocellus is much darker than on white, but that the two anterior ones are not. 



In the production of a pattern in individuals which are maximum white tlie posterior ocellus always 

 becomes dark first, then the two anterior ocelli and then a row of ten dark spots appears along the base 

 of the fins. Following this two dark spots appear in the tail and two in the central part of the body, 

 after which others appear in the fins, the tail, and elsewhere. 



Plate XXXII. 



Fig. 52. Individual (B), on black and white background (dots 3 mm. in diameter) on August 14, 

 7 a. m. to 1.30 p. m., after having been fully adapted to white. 



Fig. 53. Individual (B), on black and white background (dots 3 mm. in diameter) from August 16, 

 10 a. m., to August 19, 12 m. The black spots are clearly somewhat larger and denser than those pro- 

 duced on dots 2 mm. in diameter (fig. 54). 



Fig. 54. Individual (B), on black and white background (dots 2 mm. in diameter) from August 19, 

 3 p. m., to August 20, 10 a. m., after having been fully adapted to white. On this background the same 

 number of black spots are produced as on that in figure 52, but the spots are slightly larger and more 

 dense. This statement and all others of a similar nattu'e are based upon a comparative study of a 

 series of prints made from each of the different negatives, and upon descriptions of the effect of the 

 different backgrounds ^vritten during the process of the experiments. 



Fig. ss- Individual (B), on black and white background (dots 5 mm. in diameter) from August 

 2, 12 m., to August 3, I p. m.; no observable change after 8 a. m. The black spots are practically the 

 same as those produced on dots 2 mm. in diameter, but only half as far apart (fig. 54). 



Putting the matter in subjective terms this work shows that these animals do not recognize dots 

 0.5 ram. in diameter as individuals, but that they do recognize those i mm. in diameter and that they 

 distinguish dots 2 mm. in diameter from those 3 mm. in diameter and dots 3 mm. in diameter from those 

 S mm. in diameter. 



Plates X X X I II-X X X V . — Simulation of colors in the background. 



Pl.we XXXIII. 



For a fuller account of the environment of the specimens autochromed see text. 



To compare the color of the different individuals eliminate the background by covering the auto- 

 chromes with tlie accompanying gray sheet containing holes. These autochromes prove conclusively 

 that the color in Paralichihys and Ancylopseila changes within wide limits so as to harmonize fairly 

 accurately with that of the background. 



Fig. 56. P. albiguitus, 17 cm. long, on chrome yellow from September i to 11. The color is only 

 fairly accurately reproduced. 



97867°— vol 34—16 IG 



