Field experiments were conducted to identify the impact of post-anthesis rainfall on the concentration of deoxynivalenol (DON) and zearalenone (ZON) in harvested wheat grain. Winter wheat plots were inoculated with Fusarium graminearum at stem extension (GS31) and prothioconazole was applied at mid-anthesis (GS65) to split plots and plots were subsequently mist irrigated for 5 days. Plots were either covered by polytunnels, irrigated by sprinklers or left as non-irrigated uncovered control plots after medium-milk (GS75). Plots were harvested either when ripe (GS92; early harvest) or three weeks later (late harvest). Fusarium head blight (FHB) was assessed each week from inoculation. At harvest, yield and grain quality was measured and grains were analysed for DON and ZON. Differences in rainfall resulted in contrasting disease pressure in the two experiments, with low FHB in the first experiment and high FHB in the second. Difference in FHB resulted in large differences in grain yield, quality and mycotoxin content. DON concentration was significantly (P < 0.05) higher in irrigated compared to covered and control plots in the first experiment, whereas in the second experiment, DON was significantly (P < 0.05) higher in the covered plots compared to the control and irrigated plots. ZON concentration was significantly (P < 0.05) higher in irrigated plots in both experiments. Later harvesting resulted in an approximate fivefold increase in ZON in the first experiment, but was not significantly different in the second experiment. Prothioconazole significantly (P < 0.05) reduced DON in both experiments, but gave inconsistent reductions to ZON. This is the first report to show that the post-anthesis rainfall can significantly increase ZON in wheat, which can increase further with a delayed harvest but may be significantly reduced with the application of prothioconazole. Importantly, in the absence of moisture late season, ZON remains at very low concentrations even when wheat is severely affected by FHB.