Background. Each year millions of liters of fire-retardant chemicals are applied to wildfires across the nation. Recent laboratory studies with long-term fire-retardant chemicals indicate a significant photoenhanced toxicity of products containing sodium ferrocyanide corrosion inhibitors. Our objective of this study was to determine the toxicity of fire-retardant chemicals to fathead minnows during exposure in experimental outdoor streams. Methods. Stream tests were conducted to determine the potential toxicity of a pulse of exposure as might occur when fire retardant chemical is rinsed from the watershed by rainfall. Two artificial 55-meter experimental streams were dosed with different concentrations of Fire-Trol?? GTS-R, or uncontaminated for a control. Replicate groups of fathead minnows were added to screened containers (10 fish per container) and exposed to retardant chemicals in the recirculating flow of the stream for up to 6 hours. Results and Discussion. Under field conditions toxicity of GTS-R only occurred in the presence of sunlight. When GTS-R was tested on sunny days, 100% mortality occurred. However, when tested during heavily overcast conditions, no mortality occurred. Conclusions. Lethal concentrations of cyanide were measured when GTS-R with YPS exposures were conducted under sunny conditions, but not under cloudy conditions, indicating that a minimum UV level is necessary to induce toxicity as well as the release of cyanide from YPS. The toxicity observed with GTS-R was likely associated with lethal concentrations of cyanide. Rainwater runoff following applications of this fire-retardant at the recommended rate could result in lethal concentrations in small ponds and streams receiving limited water flow under sunny conditions. Recommendations and Outlook. In addition to avoiding application to aquatic habitats, it is important to consider characteristics of the treated site including soil binding affinity and erosive properties.