Juvenile chinook salmon Oncorhynchus tshawytscha (40-50 mm total length, TL) and striped bass Morone saxatilis (30-40 mm TL) were exposed to serial dilutions (100, 50, 25, and 12.5%) of agricultural subsurface drainwater (WWD), reconstituted drainwater (RWWD), and reconstituted seawater (IO). Agricultural subsurface drainwater contained naturally elevated concentrations of major ions (such as sodium and sulfate) and trace elements (especially boron and selenium), RWWD contained concentrations of major ions that mimicked those in WWD but trace elements were not elevated, and IO contained concentrations of total dissolved salt that were similar to those in WWD and RWWD but chloride replaced sulfate as the dominant anion. After 28 d of static exposure, over 75% of the chinook salmon in 100% WWD had died, whereas none had died in other dilutions and water types. Growth of chinook salmon in WWD and RWWD, but not in IO, exhibited dilution responses. All striped bass died in 100% WWD within 23 d, whereas 19 of 20 striped bass had died in 100% RWWD after 28 d. In contrast, none died in 100% IO. Growth of striped bass was impaired only in WWD. Fish in WWD accumulated as much as 200 μg/g (dry-weight basis) of boron, whereas fish in control water accumulated less than 3.1 μg/g. Although potentially toxic concentrations of selenium occurred in WWD (geometric means, 158-218 μg/L), chinook salmon and striped bass exposed to this water type accumulated 5.7 μg Se/g or less. These findings indicate that WWD was toxic to chinook salmon and striped bass. Judging from available data, the toxicity of WWD was due primarily to high concentrations of major ions present in atypical ratios, to high concentrations of sulfate, or to both. High concentrations of boron and selenium also may have contributed to the toxicity of WWD, but their effects were not clearly delineated.