and ecological considerations, for efficient experimental investigation of 

 undesirable effects on fish growth of water pollutants, especially toxic 

 ones. The proposed studies are directed toward reasonably reliable 

 estimation of limits of water quality alterations having virtually no such 

 harmful effects in nature. Because of the unavoidable complexity and 

 difficulty of sufficiently thorough investigations directed toward the 

 stated goal, it is clearly desirable to minimize the number of the most 

 difficult and costly experiments to be performed. In the plan of 

 investigation suggested here, the sequence of different experiments will 

 provide, in my opinion, for maximal efficiency of the studies of a 

 practical nature, which should not be expected to lead to complete 

 understanding of observed effects of pollutants. 



A reduction of growth rates of fish can be a consequence of any one or 

 more of the following effects of degradation of water quality: (1) reduc- 

 tion of the available food supply; (2) impairment of the appetite of the 

 fish for food; (3) reduction of the feeding activity of the fish and their 

 ability to find and capture their prey; and (4) impairment of the effi- 

 ciency of metabolic utilization of food by the fish and its conversion into 

 body tissues. We must look for each of these possible effects. 



Maximum concentrations of toxic water pollutants having no material 

 effect on over-all food resources are very difficult to determine experi- 

 mentally, because most fish depend, directly or indirectly, on a large 

 variety of aquatic organisms for their food supply. Demonstrable adverse 

 effects on the reproduction and growth of some food organisms may not be 

 assumed to bring about a lower availability of food, since the production 

 of other, more tolerant species that can be utilized by fish may increase 

 as competition for space and food by the less tolerant species declines. 

 In waters polluted with energy-rich organic materials, such as pulp mill 

 wastes, the abundance of fish foods in the aggregate sometimes even 

 increases, while some disappear, because the growth of some microorganisms 

 that serve as a primary food resource for invertebrates is stimulated. 

 Often, therefore, reliable prediction of effects of pollutants on the 

 availability of fish foods in natural habitats can be achieved only through 

 experiments performed under nearly natural conditions, such as experiments 

 with artificial or modified natural streams in which complex plant and 

 animal communities can be maintained. The results of such costly experi- 

 ments often are difficult to interpret if there are no good reasons to 

 believe that any observed impairment of the growth of fish could have been 

 due only to a reduction of availability of foods and not to one of the 

 other possible causes mentioned above. Obviously, it is not with difficult 

 experiments of this kind that one should begin in seeking to reduce the 

 number or range of concentrations of a pollutant that need to be tested in 

 other additional experiments. 



On the other hand, the maximum concentration of a toxic substance 

 having no pronounced adverse effect on the appetite of juvenile fish, or on 

 the highest rate of food consumption of which they are capable, or on their 

 efficiency of utilization of food resources if their activity is not 

 materially depressed by the poison at that level, can be quite easily deter- 

 mined through laboratory experiments. To find the level that does not 



56 



