Acute oral toxicity of zinc phosphide: an assessment for wild house mice (Mus musculus)

Irregular plagues of house mice, Mus musculus, incur major economic impacts on agricultural production in Australia. The efficacy of zinc phosphide (ZnP), the only registered broadacre control agent for mice, is reported as increasingly variable. Have mice become less sensitive over time or are they taking a sub‐lethal dose and developing aversion? In this laboratory study, the sensitivity of mice (wild caught; outbred laboratory strain) was assessed using oral gavage of a range of ZnP concentrations. The estimated LD50 values (72–79 mg ZnP/kg body weight) were similar for each mouse group but are significantly higher than previously reported. The willingness of mice to consume ZnP‐coated grains was determined. ZnP‐coated grains (50 g ZnP/kg grain) presented in the absence of alternative food were consumed and 94% of wild mice died. Mice provided with alternative food and ZnP‐coated wheat grains (either 25 or 50 g ZnP/kg grain) consumed toxic and non‐toxic grains, and mortality was lower (33–55%). If a sublethal amount of ZnP‐coated grain was consumed, aversion occurred, mostly when alternative food was present. The sensitivity of wild house mice to ZnP in Australia is significantly lower than previously assumed. Under laboratory conditions, ZnP‐coated grains coated with a new higher dose (50 g ZnP/kg grain) were readily consumed. Consumption of toxic grain occurred when alternative food was available but was decreased. Our unambiguous findings for house mice indicate a re‐assessment of the ZnP loading for baits used for control of many rodents around the world may be warranted. The LD50 for Australian wild house mice is more than twofold higher than previously reported. Mice readily consumed grains coated with 5% ZnP. The mortality rate was highest in the absence of alternative food. These results have direct implications for mouse management in cropping systems.