Two experiments were conducted in Argentina's semiarid pampa (36°46'S, 64°17'W, altitude 210 m). The soil was an Entic Haplustoll. Long (LC), intermediate (IC) and short (SC) cycle wheat varieties, with a plant density of 180, 250 and 320 pl m^-2 , respectively, five cultivars on every cycle. They were sown May 24, June 23 and August 2, respectively, and kept under dryland conditions or were irrigated. Plots were 6 m long and 7 rows, 0,20 m apart. Crops received 50 kg ha^-1 DAP at sowing time and 100 kg ha^-1 at Zadoks' 14 stage. Volumetric soil water content was measured at 20-cm-depth intervals and only to the depth of the petrocalcic layer (approximately 1 m). Water was computed by adding precipitation, the change in the stored water within the soil profile along the life cycle, and the water added. IC cultivars had the highest yield in the irrigated treatments as much as in the dryland treatments with 3617 and 3126 kg ha^-1, respectively. Yield of dryland treatments were from 65.9 to 86.4% of the irrigated ones for each cycle. Yield increase of the irrigated treatments was mainly due to an increase in the number of grains per m² in SC cultivars and to a lesser extent to an increase in grain size. Yield increase in LC and IC cultivars was due to an increase in seed size. Total dry matter was highest in LC cultivars especially in plants under dryland conditions and was lowest in SC cultivars and were the ones that gave the highest response to irrigation (20.5%). Harvest index was highest in IC cultivars, and in the irrigated plants. Water consumption was similar in all dryland treatments and also within the irrigated treatments. The water use efficiency was highest in SC cultivars.