The formation and fate of trihalomethanes (THM) during the third injection, storage, and recovery test at Lancaster, Antelope Valley, California, were investigated as part of a program to assess the long-term feasibility of using injection, storage, and recovery as a water-supply method and as a way to reduce water-level declines and land-subsidence in the Antelope Valley. The program was conducted by the U.S. Geological Survey in cooperation with the Los Angeles County Department of Public Works and the Antelope Valley-East Kern Water Agency. The water used for injection, storage, and recovery must be disinfected before injection and thus contains THMs and other disinfection by-products. THMs (chloroform, CHCl3, bromodichloromethane, CHCl2Br, dibromochloromethane, CHClBr2, and bromoform, CHBr3) are formed by reaction between natural dissolved organic carbon that is present in water and chlorine that is added during the disinfection step of the drinking water treatment process. THMs are carcinogenic compounds, and their concentrations in drinking water are regulated by the U.S. Environmental Protection Agency. During previous cycles of the Lancaster program, extracted water still contained measurable concentrations of THMs long after continuous pumping had extracted a greater volume of water than had been injected. This raised concerns about the potential long-term effect of injection, storage, and recovery cycles on ground-water quality in Antelope Valley aquifers. The primary objectives of this investigation were to determine (1) what controlled continued THM formation in the aquifer after injection, (2) what caused of the persistence of THMs in the extracted water, even after long periods of pumping, (3) what controlled the decrease of THM concentrations during the extraction period, and (4) the potential for natural attenuation of THMs in the aquifer. Laboratory experiments on biodegradation of THMs in microcosms of aquifer materials indicate that aquifer bacteria did not degrade CHCl3 or CHBr3 under aerobic conditions, but did degrade CHBr3 under anaerobic conditions. However, the aquifer is naturally aerobic and CHCl3 is the dominant THM species; therefore, biodegradation is not considered an important attenuation mechanism for THMs in this aquifer. The alluvial-fan sediments comprising the aquifer have very low contents of organic matter; therefore, sorption is not considered to be an important attenuation mechanism for THMs in this aquifer. Laboratory experiments on formation of THMs in the injection water indicate that continued THM formation in the injection water after injection into the aquifer was limited by the amount of residual chlorine in the injection water at the time of injection. After accounting for THMs formed by reaction of this residual chlorine, THMs behaved as conservative constituents in the aquifer, and the only process affecting the concentration of THMs was mixing of the injection water and the ground water. The mixing process was quantified using mass balances of injected constituents, the sulfur hexafluoride (SF6) tracer that was added to the injected water, and a simple descriptive mathematical mixing model. Mass balance calculations show that only 67 percent of the injected THMs and chloride were recovered by the time that a volume of water equivalent to 132 percent of the injection water volume was extracted. Pumping 250 percent of the injection water volume only increased recovery of injected THMs to 80 percent. THM and SF6 concentrations in the extracted water decreased concomitantly during the extraction period, and THM concentrations predicted from SF6 concentrations closely matched the measured THM concentrations. Because SF6 is a conservative tracer that was initially only present in the injection water, parallel decreases in SF6 and THM concentrations in the extracted water must be due to dilution of injection water with ground water. The simple descriptive mixing mode