Search

Sublethal toxicity tests successfully measured the improved quality of pulp mill effluents from the first cycle of environmental effects monitoring (1992–1996) to the second cycle (1997–2000). Test endpoints showed notable shifts to higher concentrations (less toxic). During the second cycle of monitoring, significantly more tests showed no effect in full-strength effluent. Five case studies were considered as part of this exercise. Most of the improvement came with installation of secondary treatment. Twelve Ontario mills with secondary treatment showed reduced toxicity, compared to results with primary treatment. All 29 sets of sublethal data showed higher IC25s during the second cycle, and 23 of these differences were statistically significant. Any other changes between the two cycles of study caused only marginal overall improvement in toxicity, judging by 12 freshwater mills in British Columbia which had secondary treatment during both cycles. Sublethal tests successfully predicted the zone of potential effect in receiving water, agreeing with effects observed in biological surveys. Overlapping zones from multiple discharges could also be demonstrated. In a situation near Niagara Falls, sublethal tests estimated the proportions of toxic loading that four mills contributed to one water body. The prediction was realistic; the actual toxicity found for a mixed effluent was 57% of that predicted from separate toxicities. The conservative prediction agrees with the usual less-than-additive sublethal action of combined toxicants.

Three groups of Atlantic salmon (Salmo salar) parr were acclimated to water temperatures of 3, 11, and 19 C. When exposed at their acclimation temperature to lethal concentrations of zinc, cold-acclimated salmon survived longer than warm-acclimated salmon. However, the threshold lethal concentration (LC50) of zinc was highest for salmon at 19 C, i.e. there was less toxicity at the high temperature. At test temperatures lower than prior acclimation temperatures, survival times in zinc were longer but the threshold LC50 was lower, i.e. fish were less tolerant of zinc. The opposite results were seen at temperatures above prior acclimation temperatures.The threshold LC50 may be predicted by the equation LC50 = 1.9611 − 0.1873 Ta + 0.8699 Te − 0.0414 Te2 + 0.00004 Ta Te3, where Ta = temperature of acclimation and Te = temperature of exposure.In polluted rivers, zinc concentrations may be most damaging to fish populations during the winter months. Measured concentrations of zinc and copper in the Miramichi River, New Brunswick were highest in winter when estimated toxicity of metal was also greatest.

Bleached kraft mill effluent (BKME) seems to contain at least two different toxic mechanisms or materials which affect aquatic animals. BKME stored under clean and quiescent conditions, lost much of its toxicity for salmon parr ( Salmo salar L.) in 1 week, and all in 2 weeks. By contrast, it lost little toxicity to lobster larvae ( Homarus americanus Milne-Edwards) in 2 weeks. Bio-oxidation of BKME for 1 week eliminated toxicity for salmon parr, but did not greatly change toxicity for lobster larvae. Biooxidation for 2 weeks reduced but did not eliminate toxicity to lobster larvae. A waste treatment process which reduces toxicity of BKME for one species may not protect other species.

It is now possible to predict toxicity of mixtures of two or more pollutants on the basis of chemical measurements. Toxicities of individual components are added up as fractions of the incipientlc50 (lethal threshold concentration). Total toxicity of the mixture is expressed as a single number. A standard terminology is given for the effects of two or more toxicants acting simultaneously. Modifying conditions (temperature, water hardness, etc.) greatly affect toxicity. Much fragmentary information should be brought together and examined for underlying patterns. Concise summaries for estimating modifying effects have been published for rainbow trout. Computerized multivariable analysis is a promising tool, especially for studies of sublethal toxicity. Effects of fluctuating concentrations of toxicants may be evaluated by four approximate techniques; the theoretical basis for understanding effects still seems unproven. Acclimation to toxicants should be studied in detail for representative pollutants. There is a great deal of information on relative resistance of different species of fish but it needs summarizing. A trend towards use of ‘standard’ fish species in research is good. With invertebrates and algae, as much as with fish, tests of sublethal effects such as growth and development are often more meaningful than tests of acute toxicity. Chemical autopsy methods seem promising, more so than histopathological approaches, which suffer from practical difficulties in field situations.

The lethal threshold concentration of neutralized bleached kraft pulp mill effluent for juvenile Atlantic salmon (Salmo salar L.) is about 12–15% in fresh water. This agrees in general with work on other salmonid fish. Larval lobsters (Homarus americanus Milne-Edwards) are reasonably resistant to this waste, and concentrations below 10% did not greatly affect their survival in sea water. Adult lobsters in sea water are more resistant than lobster larvae or salmon parr.

The review describes profitable methods for measuring lethal levels of pollutants for aquatic organisms. Methods for research in the laboratory are emphasized but the same principles could be applied in field work. Greater use of standard toxicological methods and terminology is urged. For 211 out of 375 toxicity tests reviewed, acute lethal action apparently ceased within 4 days, although this tabulation may have been biased towards short times by a large number of static tests. The incipient LC50 (lethal concentration for 50 per cent of individuals on long exposure) is recommended as the most useful single criterion of toxicity. If this cannot be estimated, the 4-day LC50 is a useful substitute, and often its equivalent. A desirable first step in toxicity tests is to estimate median lethal time for each of a series of concentrations. A toxicity curve should be drawn by plotting median survival times against concentrations on logarithmic paper. The curve helps to reveal any unusual features of toxicity. Whenever possible, tests should be prolonged until the toxicity curve becomes parallel to the time axis, indicating a lethal threshold concentration. The incipient LC50 is then estimated by selecting an exposure time from the asymptotic part of the toxicity curve; for this exposure time, observed mortality is plotted against concentration on log-probit paper, and the LC50 is read from an eye-fitted line. Confidence limits of the LC50 may also be estimated by simplified methods. These should be given in published work along with a value for slope of the probit line. Alternative approaches use reciprocal transformations or estimate LC50's for a series of exposure times. A graph is given for estimating partial replacement times of water in tanks of continuous-flow tests. Rate of flow should give a short partial replacement time, such as a flow equal to volume in 3–5 hr, and also adequate water for respiration of fish, usually 2 or 3 l/g of fish/day, or more.

COPPER and zinc are surprisingly widespread pollutants of water. In wilderness areas they may endanger fish in otherwise clean rivers as the result of mining operations1. In British industrial areas, they commonly contribute an average of 23 per cent of total toxicity of mixed pollution for fish, and in some rivers a much higher proportion2. Furthermore, trace concentrations, one-tenth to one-twentieth of accepted standards for drinking water, can be lethal for fish in regions where surface water is very soft1. In such circumstances, year-round prevention of “spills” of metal pollution can be difficult. The trisodium salt of nitrilotriacetic acid (NTA) seems to be a promising “anti-pollutant”, to prevent fish-kills in case of short-term breakdown of normal pollution controls.