Increased precipitation storage as soil water is crucial to dryland production of wheat (Triticum aestivum L.) and grain sorghum [Sorghum bicolor (L.) Moench] on the semiarid southern Great Plains. At the USDA-ARS Conservation and Production Research Laboratory at Bushland, TX, surface runoff from a Pullman clay loam (fine, mixed, superactive, thermic Torrertic Paleustoll) is typically greater with no-till (NT) than stubble-mulch (SM) tillage under the 3-yr wheat-sorghum-fallow (WSF) dryland crop rotation. Our objective was to quantify the effects of NT or SM tillage with bare or retained residue on infiltration of simulated rain, sediment transport, and related aggregate stability for continuous wheat (CW) and WSF rotation plots established in 1983. Compared with bare soil, the retained wheat residue cover increased mean 60-min cumulative infiltration across all tillage and rotation combinations by >25 mm and decreased soil loss. Cumulative rain infiltration and total soil loss did not vary significantly (P < 0.05) with tillage, but the mean infiltration rate at 60 min was 15.0 mm h(-1) less for NT than SM. The CW rotation typically increased infiltration regardless of residue cover or tillage compared with the WSF rotation, which we attributed to greater aggregate stability. The greater aggregate stability for CW may have decreased soil loss compared with WSF, which was greatest for bare SM tilled sites. We conclude that residue cover significantly increases rain infiltration over bare soil conditions independent of any tillage or rotation treatment effects. Nevertheless, increased rain infiltration due to residue cover was not sustained in the absence of soil disturbance for NT.