The fish are exposed to the test substance preferably for a period of 96 hours. Mortalities are recorded at 24, 48, 72 and 96 hours and the concentrations which kill 50 per cent of the fish (LC50) are determined where possible. One or more species may be used, the choice being at the discretion of the testing laboratory. At least seven fishes must be used at each test concentration and in the controls. The test substance should be administered to, at least, five concentrations in a geometric series with a factor preferably not exceeding 2.2. The limit test corresponds to one dose level of 100 mg/L. This study includes the observations of fish at least after 24, 48, 72 and 96 hours. The cumulative percentage mortality for each exposure period is plotted against concentration on logarithmic probability paper.