DEPARTMENT OF MARINE BIOLOGY. 161 



In studying the stomach contents of snappers I have been struck by 

 the comparative rarity of specimens of the food-fishes not well advanced in 

 the process of digestion. My attention being thus called to the rate of 

 digestion in these fishes, a few experiments were made bearing upon the point. 

 Fishes of various genera, Iridio, Scarus, Hmmulon, Eupomacentrus, Siphostoma, 

 Monacanthus, etc., as well as crabs and shrimps, were marked with thread, fed 

 to the snappers, and in two instances recovered from their stomachs after 

 having been subjected to the process of digestion for 2% and Z}4 hours, 

 respectively. At the end of the shorter period most of the fishes were 

 readily recognizable by one who knows the fishes of the region. Skin and 

 pattern were usually intact upon any face not directly in contact with the 

 lining of the stomach. Crabs such as Mithrax hispidus and various small 

 Portunidse were practically unchanged in color or texture. After 3>^ hours 

 specimens of M. hispidus were still green, firm, and intact, though a shrimp of 

 larger size, perhaps bitten in two before being swallowed, was far advanced in 

 digestion. Many of the fishes still showed bits of their original color pat- 

 tern. Most of them were identifiable by jaws, pharyngeal teeth, or other 

 pecuHarities. Some, however, had passed the point where their species could 

 have been more than guessed at, if they had not borne my marks of identi- 

 fication. From comparison of marked fishes and Crustacea recovered, with 

 other specimens eaten under natural conditions, enough was learned to prove 

 that snappers in general feed more or less continuously through the night. 



In reports of earlier years much has been written regarding the obliterative 

 coloration of tropical reef fishes. To pass from a study of such color to a 

 consideration of the agencies that have possibly called it into being is no long 

 step. Among these agencies must be included natural selection through 

 attacks of predaceous fishes. But if these have worked in any large way in 

 directing racial development of color and pattern toward obliterative effects, 

 it may well be anticipated that they have decided powers of discrimination 

 in respect to colors and patterns. 



Since Professor Reighard ^ has already investigated the ability of the gray 

 snapper to discriminate colors, I undertook only a study of their power to 

 distinguish two simple patterns and to form associations between such pat- 

 terns and the quality of distastefulness. Minnows of the species Atherina 

 laticeps provided the test material. These fishes are so lightly pigmented as 

 to be almost uniformly pale in coloration". They were given either a longitu- 

 dinal dark stripe through the eye to the base of the tail, or two transverse 

 dark bands upon the body, somewhat narrower than the interspaces, by paint- 

 ing them with silver nitrate. 



In the first test, which was several times repeated on different groups of 

 snappers, 25 unmarked atherinas were first fed, after which 25 striped and 25 

 banded fish came in alternation with one another, to be followed themselves 

 at last by 25 plain fish like those first offered. The results of these tests were 

 uniform. The unmarked fishes were accepted with least hesitation, and the 

 striped more readily than the cross-banded. 



In order to determine whether the banded fishes were taken least readily 

 merely because they were darkest, or for another reason, a second series of 

 tests was made, in which the striped fish were as before, while the banded 



^Papers frem Department of Marine Biology, Carnegie Inst. Wash. Pub. No. 103. 1908 



