We compared capture rates and exposure to SGARs of wood mice (Apodemus sylvaticus) and house mice (Mus domesticus) in autumn/winter on farms that currently used, had previously used, and never used SGARs. 6-10 weeks after baiting programmes began, 15 % of 55 wood mice and 33 % of 12 house mice had detectable liver SGAR residues. Wood mice with residues occurred on farms not using rodenticides, reflecting the high mobility of these animals, and four had multiple liver residues, possibly due to cross-contamination of baits. The winter decline in wood mouse numbers was similar on farms that did and did not use SGARs, suggesting little long-term impact of SGARs on populations on farms. Our results indicate residual levels of rodenticides will be ever present in small mammal prey across the agricultural landscape unless all farms in a locality cease application. The implications for secondary exposure and poisoning of predators are discussed.