Time should be considered in developmental ecotoxicity test

Developmental toxicity tests are often used for the hazard assessment of chemicals and environmental media. One of the most widely used is the oyster embryo larval test (OEL), in which the development of oyster larvae is arrested at a single fixed time (e.g. 24 or 48 h) of toxic exposure, and the proportion of normal larvae measured. However, a major problem with this conventional approach is the lack of information on temporal trends in development. In this study, Pacific oyster Crassostrea gigas embryos were exposed to nominal concentrations of copper (CuSO₄) of <0.001 (control), 0.60, 1.25, 2.5, 5.0, 10.0 and 20.0 μgl^-1 (at 20°C, salinity 35‰ and pH 8.1). Three replicates from each group were arrested and examined every 8 h during 24-72 h of exposure, and the number of viable larvae developed to D-shape was determined. The results revealed that the number of viable D-shape larvae in the control increased rapidly and reached an optimum at 32 h, before declining gradually due to starvation. Analysis of covariance (ANCOVA) showed that larval developmental rates during 0-32 h were significantly inhibited by Cu at all concentrations. This paper demonstrates that arrest and measurement at different time periods are important and should be incorporated into the OEL test. This would maximise the sensitivity of the test in detecting developmental effects in spiked or environmental samples. © 2002 Elsevier Science Ltd. All rights reserved.