for some time in the presence or in the absence of a poison. The number of 

 food organisms introduced into each aquarium should be such as to make it 

 impossible for the fish to become satiated. It can be less than the number 

 that can be consumed by the fish at once when the food organisms are very 

 abundant and easily found and caught, or it can be greater than that number 

 if the initial density of the food organisms or the cover provided for them 

 are such that these organisms are not too vulnerable to predation. The 

 food organisms should be as uniform in size as possible. The food 

 organisms remaining in the aquaria with and without the toxicant being 

 tested can be counted when only a few remain in the control aquaria. If 

 the foraging activity and efficiency of the fish are unaffected at a tested 

 concentration of poison, about as many of the food organisms, on the 

 average, should remain in the aquaria with the contaminated water as in the 

 control aquaria at the end of the test period. Because of the progressive 

 decline of the numbers of food organisms in the aquaria during a test, the 

 experimental and control fish will be confronted with a desirable variety 

 of foodorganism densities in such tests. 



For experiments with large fish, small, artificial ponds like those 

 that have been used in the already mentioned experiments on the influence 

 of dissolved oxygen on the feeding and growth of largemouth bass can be 

 used. However, such ponds are costly and require much space, and the 

 maintenance in them of constant concentrations of toxic pollutants, by 

 sufficiently rapid replacement of the water or otherwise, can be 

 difficult. Therefore, experiments with laboratory models of more modest 

 size have been undertaken recently. In these exploratory tests long, rec- 

 tangular aquaria are being used, with a shelf made of fine-mesh wire or 

 plastic netting suspended at each end a short distance below the water sur- 

 face. Mosquitofish ( Gambusia ) with which these tanks are stocked soon 

 learn to use the area over each shelf as a refuge, remaining there most of 

 the time and escaping to one of these sanctuaries if they can when they are 

 pursued by a largemouth bass also placed in the aquarium. The bass catch 

 some of the mosquitofish that spontaneously leave the protection of the 

 cover from time to time, or that the bass are able to flush from the cover 

 by some maneuver, but they are unable to follow the prey in the shallow 

 water above the shelves and capture it there. Consequently, they cannot 

 fully satisfy their appetite, and their food consumption and growth are 

 dependent on the density of the prey, just as were those of the bass in the 

 ponds. Other things being constant, any reduction of their foraging vigor 

 and agility must result in a reduction of food intake and slower growth. 



Because of differences of the foods and feeding habits of fish of 

 different kinds in various waters, experimental methods suitable for the 

 study of effects of water pollution on the foraging activity and effi- 

 ciency of some species are unsuited to other forms. The contriving of the 

 most appropriate methods sometimes may not be easy, challenging the most 

 imaginative and inventive biologist's ingenuity. But experiments that are 

 very easily designed or standardized soon cease to be interesting. Artifi- 

 cial streams with circulated water can be used for tests with fishes that 

 normally inhabit rapidly flowing waters. 



63 



