Understanding fall precipitation effects on rangelands could improve forage production forecasting and inform predictions of potential climate change effects. We used a rainout shelter and water addition to test effects of seasonal precipitation on soil water and annual net primary production of C-3 perennial grass, C-4 perennial grass, annual grasses, forbs, and all plants combined. Treatments were 1) drought during September-October and April-May (DD); 2) drought plus irrigation during September-October and drought during April-May (WD); 3) year-long ambient conditions (WW); and 4) ambient plus irrigation during September-October (W + W). Treatments created conditions ranking among the driest and wettest September-October periods since 1937. Fall water effects on soil water were not detectable by May at 15 cm and 30 cm. Effects persisted into July at 60 cm and 90 cm, depths below the primary root zone. With spring drought, annual net primary production was 344 kg ha(-1) greater when the previous fall was wet rather than dry. No differences were detected between fall water treatments when spring was wet and fall was about 184% (1 938 +/- 117 kg ha(-1)) or 391% of the median (1 903 +/- 117 kg ha(-1)). Fall water increased C-3 perennial grass when spring was also wet and had no effect under spring drought, when forage production concerns are greatest. Fall water did not affect C-4 perennial grass, and extremely wet fall conditions reduced forb production about 50%. The greatest effect of fall water was increased annual grass production. Even record high levels of fall water had minor effects on biomass, functional group composition, and soil water that were short-lived and overwhelmed by the influence of spring precipitation. Movement of fall water to deep soil by the growing season suggests plants that would most benefit from fall precipitation are those that could use it during fall (winter annuals), or deep-rooted species (shrubs). Published by Elsevier Inc. on behalf of The Society for Range Management.