Multiple field experiments and a controlled-environment temperature study were conducted to investigate soil and weather conditions responsible for herbicide phytotoxicity in lentil (Lens culinaris Medik.) from post-emergent application of metribuzin. A linear relationship was observed between plant injury (% necrosis) and metribuzin rate in all 12 environments, but in only 11 environments for anthesis dry weight and nine environments for both plant density and grain yield. Grain-yield reduction from label metribuzin rates of 135 g a.i. ha^–1 for sand and 285 g a.i. ha^–1 for clay ranged from 0% to 32% and 0% to 67%, respectively, across all environments. Principal component analysis of soil and weather factors around the time of herbicide application suggested that metribuzin-induced plant damage in lentil was due to a combination of multiple soil and weather factors. However, heavy rainfall within 10 days of herbicide application, particularly on light-textured soils or where soil moisture was low, was most strongly linked to plant damage. Experiments targeting the impact of reductions in temperature post-metribuzin application showed no effect, and of light intensities pre- and post-metribuzin application showed low effects on plant-damage measures. Because rainfall in the 10 days after application is a major determinant of metribuzin damage in winter-grown lentil in southern Australia, a higher level of selective tolerance to metribuzin than that present in commercial cultivars is needed for its safe post-emergent use. Early and late measures of plant damage will be required to assess accurately plant tolerance to post-emergent metribuzin application in lentil.