2IO BULLETIN OF THE BUREAU OF FISHERIES. 



Even if this illumination is only slightly reduced the effect becomes quite evident, 

 as seen in the following observations: 



On August 15, 1 1 a. m., the side wall of the crystallizing dish was covered with black 

 paper, thus forming a black cylinder 6 cm. high in order to slightly reduce the illumina- 

 tion from above. The fish at this time was fully adapted to the spotted background 

 fully illuminated from above and below. It had numerous dark brownish spots and 

 the three ocelli also had a brownish cast. At 5 p. m. the fish in general had become 

 much lighter. The three ocelli were at this time nearly black, but the dark regions 

 around them and the dark spots were much smaller and much less distinct. The light 

 from the mirror was now again intercepted and at 6.30 p. m. the surface was again 

 about the same as it was in the beginning of the experiment. These tests were re- 

 peated several times the following day and in general it was found that the fish assumed 

 a lighter shade when the background was illuminated from below in addition to the 

 illumination from above, and in general the pattern in the skin appeared to be less 

 conspicuous, although the spots in the background, as seen from above, appeared much 

 more conspicuous. The ocelli and spots in the skin did not appear quite as large or as 

 dense as they did when adapted to the background illuminated only from above, but 

 the light regions appeared somewhat whiter. 



It is consequently evident that a sHght reduction of light from above has a marked 

 effect on the reactions in the skin; but the following observations indicate that there 

 is no strict proportionality between these reactions and the amount of light received 

 from above compared with that received from below. 



On August 12, 6 a. m., the cylinder was removed, admitting light from above. The 

 skin of the fish immediately became speckled. The three ocelli became black, a row 

 of five conspicuous black spots appeared near the base of the dorsal fin and another 

 row of the same number near that of the ventral; and some spots also appeared in the 

 fins and the tail. When the cylinder was put in place and closed these spots rapidly 

 disappeared again and the ocelli became lighter. This change required less than two 

 minutes. When the cylinder was removed the spots appeared again but not quite so 

 rapidly. With the cylinder open on top, admitting some light from above, the results 

 appeared to be the same as they were when it was closed, admitting no light from above. 

 This was tried several times between 6 and 10 a. m., after which the cylinder was put in 

 place, closed and left. At 1 1 .45 a. m. there was no change ; the ocelli were still clearly 

 visible, the rest of the surface was very light gray with a slight brownish cast. There 

 was much reddish sediment on the mirror. This was removed and the entire aquarium 

 cleaned. At 1.15 p. m. the three areas apparently had become somewhat darker and the 

 rest of the surface lighter, the brownish cast having entirely disappeared. After the 

 cylinder had been removed a row of dark spots appeared almost at once along the edge 

 of the body near the fins, and the three ocelli became much darker. The light from the 

 mirror was now intercepted so that the background was illuminated only from above. 

 At 1.22 p. m. the dark spots had become darker and many more had appeared and each 

 of the three ocelli was surrounded by a dark patch fully i cm. in diameter. Observations 

 were made from time to time during the afternoon, but no further changes were seen. 

 At 5 p. ra. the light from the mirror was again turned on and left. At 6.10 p. m., how- 

 ever, no reduction in the number or the size of the dark spots was detected. This test 



