The effectiveness of bromadiolone for controlling breeding house mouse populations in irrigated soybean crops was tested when mouse densities were around 200-300 per hectare. A single application of wheat with 0.005% bromadiolone, placed in bait stations (200 g per station) 20 m apart throughout the crop, effectively reduced mouse numbers at one of two sites. Where few other irrigated summer crops were close by (Farm 1), bromadiolone reduced mouse numbers relative to the unbaited control by between 43% and 62% within 22 days, depending upon the index of mouse abundance used. Because of a general decline in mouse abundance after this time, the effect of baiting lasted through days 22 to 42. However, the overall benefit from baiting was marginal, because mouse numbers just before harvest (42 days after baiting) were still around 43% of pre-baiting levels. Where other irrigated summer crops were in close proximity (Farm 2), the single application of bromadiolone did not significantly reduce mouse numbers, and there were no effects which could be attributed to baiting, either at day 22 or 42. Although few marked individuals survived baiting, mice from neighbouring crops apparently replaced those killed. When the baiting trials commenced, growers had just become aware that mice were causing noticeable damage to soybean crops. By this time, mice were evenly distributed from 0.5 to 90 m into the crop and, as the mean movement of resident mice was 10.9 m, perimeter baiting only would have been ineffective. These data also indicate that the distance between bait stations in irrigated soybean crops should not exceed 20 m.