A field experiment was conducted in a well-puddled paddy field developed on the Tama River alluvial soil in the Farm of Tokyo University of Agriculture and Technology, Japan, to study the effect of intermittent irrigation on water use efficiency of paddy rice. Four treatments were arranged with 2 replicates: continuous flooding irrigation treatments (CFI), and three intermittent irrigation treatments Ⅱ-0, Ⅱ-1 and Ⅱ-2, in which plants were re-irrigated when the soil water potential fell below 0, -10, and -20 kPa, respectively, at soil depth of about 5 cm. Water consumption was lower in treatment Ⅱ-0 than in treatment CFI because the percolation rate was reduced by the reduction in the hydraulic head of ponded water. Intermittent irrigation led to soil repeated shrinking and swelling in Ⅱ-1 and Ⅱ-2 plots and, therefore, soil cracks developed rapidly. Since they became the major routes of water percolation, the soil cracks increased water consumption in treatments Ⅱ-1 and Ⅱ-2. There were no significant differences in dry matter production and grain yields between treatment Ⅱ-0 and treatment CFI, but the dry matter production and grain yields in treatments Ⅱ-0 and CFI were significantly higher than those in treatments Ⅱ-1 and Ⅱ-2. Therefore, the water use efficiency in the treatments was in the order of Ⅱ-0 > CFI > Ⅱ- 2 > Ⅱ- 1. A field experiment was conducted in a well-puddled paddy field developed on the Tama River alluvial soil in the Farm of Tokyo University of Agriculture and Technology, Japan, to study the effect of intermittent irrigation on water use efficiency of paddy rice. Four treatments were arranged with 2 replicates: continuous flooding irrigation treatments (CFI), and three intermittent irrigation treatments II-0, II-1 and II-2, in which plants were re-irrigated when the soil water potential fell below 0, -10, and -20 kPa, respectively, at soil depth of about 5 cm. Water consumption was lower in treatment Ⅱ-0 than in treatment CFI because the percolation rate was reduced by the reduction in the hydraulic head of ponded water. Intermittent irrigation led to soil repeated Since they became the major routes of water percolation, the soil cracks increased water consumption in treatments Ⅱ-1 and -2. There were no significant differences in dry matter production and grain yields between treatment Ⅱ-0 and treatment CFI, but the dry matter production and grain yields in treatments Ⅱ-0 and CFI were significantly higher than those in treatments Ⅱ-1 and II-2. Thus, the water use efficiency in the treatments was in the order of II-0> CFI> II-2> II-1.