Field studies were conducted to determine if laboratory protocols accurately predict shrimp mortality under field conditions. Fenthion, a mosquitocide, was applied to coastal marshes in several truck-mounted ultra-low volume (ULV) adulticide operations, and by direct application at the larvicide rate. Pink shrimp (Penaeus duorarum) were deployed in floating cages and observed for mortality. Water samples characterized the exposure regime and fate of fenthion at the field sites. The acute flow-through 96-h LC50 of 0.11 micrograms/l was used as a conservative estimate of the expected toxicity in field exposures. An exposure profile based on measured field concentrations was used for laboratory pulse-exposures: fenthion was metered for 2 h to specified maximum concentrations, then flushed with seawater to cause a 6 to 8 h exposure, yielding a no-observed-effect concentration (NOEC) of 0.84 micrograms/l. Four ULV sprays produced water concentrations less than NOEC and no fenthion-induced shrimp mortality. However, the direct application's water concentrations ranged from 15 to 20 micrograms/l (less than NOEC) and caused extensive mortality (90 to 100%). Field observations confirmed our hypothesis that if peak fenthion concentrations were lower than the laboratory NOEC, then no mortality would occur, whereas if maximum concentrations exceeded the NOEC, mortality would occur. These laboratory toxicity tests can predict the range of lethal and non-lethal acute field exposures to fenthion for pink shrimp when exposure regimes are similar.