A simple, reliable method for predicting evapotranspiration (ET) for crops is needed. The purpose of this study was to formulate and test empirical relationships far estimating soil evaporation and transpiration based on evaporative demand and crop development with limiting and non‐limiting soil water regimes. The relationships were compared with evaporation and transpiration data for irrigated cotton (Gossypium hirsutum L.).Evaporation from the soil surface was determined by measuring diurnal changes in the surface soil‐water content and weight changes in embedded containers. Evapotranspiration was calculated from the quantity of irrigation water applied, rainfall, drainage, and changes in soil‐water content measured by the neutron scattering technique. Evaporation from the soil surface on the 1st day after irrigation (ESo) was related to potential evaporation (Eo) and leaf area index (LAI) as: ESo = Eo exp(−0.623 LAI). On successive days after irrigation, surface evaporation (ES) was computed as: ES = 5,8t0.6 − 5.8(t − l)0.6, where t is time in days. For transpiration on the day after irrigation (EPo), the following relationship was found: EPo = Eo(l − exp(−0.623 LAI)). l'or limiting soil‐water conditions, transpiration (EP) Mas computed with: EP = EPo/ (1 + 6.2 exp − 15.2 FAW), where FA W is the fraction of available water in soil. The results obtained with these equations agreed well with values for ES and EP measured in this study and with data presented by Ritchie (1972).