The decomposition rate of fire ant baits and the consequent leaching of the pesticide mirex from the bait were investigated for one year in three types of environments (i.e., on the ground, in the soil, and in the pond) by means of the nylon litter bag method. There was no significant difference in the decomposition rates between toxic and non-toxic baits. The bait proved to be stable and slow to decay. After one year of exposure in the field, about 75% of bait material remained in the bags that were deployed on the ground, 60% in the soil, and 25% in the pond. The removal of mirex from the bait was due to leaching from, and natural decay of the bait. The loss of mirex from the bait to the environment varied significantly among the three habitats. At the end of 12 months, mirex disappearance from baits exposed on the ground was 41.1%, in the soil 56.3%, and in the pond 83.9%. The rate of mirex loss was slower during fall and winter. The toxicities of the bait decomposed in situ for different lengths of time (12, 9, 6, 3, and 0 months) to the land isopod Armadillidium vulgare and the soil millipede oxidus gracilis were higher than the undecomposed baits. This indicated that weathered mirex baits left in the field have greater biological impact than freshly applied baits, at least on the two soil animals tested.