Abstract: Based on the 50-year meteorological data series of Beijing, the reference crop evapotranspiration calculated with two commonly used methods; Priestley-Taylor and Penman methods were compared. Comparison of the yearly results showed that Priestley-Taylor method had significant lower value than that of Penman method, with the former being 15%～31% lower than that of the later. The maximum, minimum and multi-year average difference is 378.3 mm，150.9 mm, and 255.9 mm, respectively. Comparison of monthly value indicated that the two methods agreed very closely in July and August, in which there was no significant difference. In other months, the difference is significant. The difference is caused by both precipitation and the aerodynamic term in the Penman method, with the later being the main contributor to it. The ratio of aerodynamic term to net radiation is inversely related to the agreement of the two methods. A higher ratio will result in a worse agreement. In contrast, A lower ratio will result in a better agreement.