seriously affect results of acute or long-term tests, regardless of which 

 bioassay endpoint has been selected. Usually holding fish for an acclima- 

 tion period in the laboratory before testing will insure a reliable 

 response if fish remain disease free and accept food normally. It also may 

 be desirable to subject samples of the test fish to the effects of a 

 reference toxicant for which response has been well documented with the 

 species. 



Another factor that is important in verifying the reliability of 

 endpoints is the degree of crowding of test organisms in the test chamber. 

 When small fish are used, 1-2 g of fish per liter of water in the test 

 chamber will usually allow sufficient space to permit free movement and pre- 

 vent secondary effects from too many test specimens. If large fish are 

 used, especially adults of some species, antogonism between individuals can 

 seriously affect final results by causing fish to reach the designated end- 

 point at lower toxicant concentrations than fish not stressed by behavioral 

 patterns. Test conditions which overstimulate the fish to activity or de- 

 press activity unnaturally will affect the validity of results at the 

 selected endpoint. 



Three factors influencing the validity of a selected endpoint are 

 temperature, oxygen concentration, and pH. Temperature has a marked effect 

 on the sensitivity of test organisms and consequently on the calculated 

 LC50 or other endpoints selected. The effect of temperature cannot be pre- 

 dicted with certainty. For example, the tolerance of fathead minnows and 

 goldfish to hydrogen sulfide is greatly increased by a 15 C lowering of 

 temperature (Figure 2). In contrast, tolerance of bluegills is decreased 

 by a lowering of temperature (Figure 3). It is therefore important that 

 when an endpoint for a bioassay is selected, the test temperature is 

 related to the objective of the test. A standard test temperature of 25 C 

 does not necessarily relate to the ambient ecological condition in nature 

 where the test results are expected to be applied. 



Oxygen concentrations below 4 or 5 mg/liter will increase the sensitiv- 

 ity of test species to most toxicants. At low dissolved 2 concentrations 

 (below 4 mg/liter) a new stress is added that increases the adverse 

 response of the organism to the toxicant. Similarly, extreme variations in 

 the hydrogen ion concentration (pH) of test water can alter response and 

 affect the validity of the endpoint chosen. This influence may be exer- 

 cised through the effect of pH on ionization of the material being tested 

 or on the physiological conditions imposed upon the fish which make absorp- 

 tion or blood changes more or less responsive to changes in concentration 

 of toxicant. 



A factor frequently overlooked in choosing bioassay endpoints for 

 various fish species has been the difference in tolerance of eggs, larvae, 

 juveniles, and adult fish. Frequently fish in the early life-history 

 stages are much more sensitive than older fish and in consequence a satis- 

 factory endpoint for one life-history stage will not necessarily 

 demonstrate the sensitivity of the species through its entire life cycle. 

 Examples of differences can be drawn from H 2 S studies where frey or larvae 

 are the most sensitive form and may vary markedly from juveniles. In con- 



40 



