The cooling system of a nuclear power plant was tested after a period of shutdown, during which noncirculating sea water was in contact with a copper-nickel tube in the system. The first water to be discharged had Cu 1800 mu g/litre and after 30 days there was 20 mu g/litre; inlet water had 1 mu g/litre. After that test about 1500 Haliotis rufescens and H. cracherodii were found dead in the discharge area [original population not stated]. Wild Haliotis of a range of weights and lengths were collected and held for 2 weeks without feed then exposed for 96 h to standing water without extra Cu or with 10, 32, 56, 100 or 180/109 added as sulphate. Larvae of H. rufescens about 48 h old were exposed for 48 h without extra Cu or to 7 concentrations from 10 to 640/109. The 96-h LD50 was 50/109 for H. cracherodii and 65/109 for H. rufescens adults; the 48-h LD50 was 114/109 for the larvae. All adults survived up to 32/109 and the increase in mortality among larvae was significant at 80/109 or higher. Mean Cu concentration in tissues of adults generally increased with concentration of Cu in water, except in digestive glands at higher concentrations in water. Gills accumulated Cu significantly in water with as little Cu as 56/109 and digestive glands did so with as little as 32/109 for H. cracherodii, and 56/109 for H. rufescens. Cu content of those that died during trials was significantly greater than that of survivors. Lowest concentration in gills of those that died was 61.9/106 for H. rufescens and 92/106 for H. cracherodii, the "estimated detrimental accumulation levels (EDAL)". Wild H. rufescens had about the EDAL but can be acclimatized to Cu. There was little doubt that Cu released during testing the cooling system had killed the molluscs and the copper tube has been replaced with Ti.