Evaluation of an Automated Fish Ventilatory Monitoring System in a Short-Term Screening Test for Chronic Toxicity

Abstract

A continuous automated fish ventilatory monitoring system was evaluated for use in short-term screening tests to determine the chronic toxicity of materials to fish. The multi-parameter ventilatory system monitors bluegills (Lepomis macrochirus) for four parameters: ventilatory rate, average ventilatory depth, cough rate, and percent movement. The system was evaluated in tests with technical chlordane, 1,3,5-trinitrobenzene (TNB) and a solvent mixture (triton X-100 and acetone) that was used as a carrier solvent in the chlordane test. The lowest concentrations of chlordane and TNB causing changes in the ventilatory parameters during a 6-day exposure period were compared to reported chronic toxicity values. Linear regression analysis of ventilatory signals monitored both visually and by the computer indicated high computer accuracy for ventilatory rate (R² = 0.976, slope = 0.972). The accuracy of the ventilatory rate led to the assumption that the computer was accurate for average depth. Cough rate accuracy was poor in the solvent and chlordane tests (R² <0.300) but was better in the TNB test (R² = 0.766, slope = 1.177). Percent movement accuracy was fair for all three tests (R² = 0.586, slope = 0.764). The solvent mixture ventilatory test revealed no significant differences between control and solvent-exposed fish for any of the four parameters monitored. In the chlordane ventilatory test, no significant responses in any of the four parameters monitored were found at concentrations up to seven times the chlordane concentration reported to cause major chronic toxicity. The sensitivity of the ventilatory system to chlordane-induced effects may have been reduced by either poor cough accuracy, (possibly related to the small size of the bluegills tested) or by an insufficiently long period of exposure. TNB (using larger bluegills) caused significant ventilatory responses at concentrations as low as 0.128 mg/L, which compares favorably to the lowest concentrations of TNB that caused toxicant effects in early life stage tests with fathead minnows (0.12 mg/L) and rainbow trout (0.17 mg/L).