In this research, daily meteorological data for 15 years (1999-2013) were used in 153 synoptic stations throughout Iran to determine the best method among pan evaporation methods to determine the reference evapotranspiration for different climate conditions. For all synoptic stations of the country, evapotranspiration was calculated from 12 different evaporation pan methods, and the FAO-Penman-Monteith method was used as the standard method for evaluation. Different statistics tests as RMSE, MBE, R2 and t tests were used to evaluate the accuracy of different methods. The final results showed that in two arid and semi-arid climates, Mohammad et al. (2008), FAO-24 (in dry conditions) (Doorenbos and Pruitt, 1997) and Christensen methods were the most suitable methods and Schneider (1992), Allen and Prout (1991), Raghuwanshi and Wallender (1998) were the most inappropriate methods. For the Mediterranean climate condition, Mohammed et al., Konica (1989) and Schneider were the most suitable and Pereira's, FAO-24 (in dry conditions), Raghuwanshi and Wallender were identified as the most unsuitable methods. In the semi-humid climate, the modified Schneider method, Allen and Pruitt (Allen & Pruitt, 1991), and Pereira (Pereira et al., 1995) were the most suitable methods and Mohamad et al., FAO-24 (in dry condition), Ragvanshi and Waldner and were the most unsuitable methods. In the wet and humid climates conditions, modified Schneider method, (Allen et al., 1998) were the best and Pryira, FAO-24 (in dry conditions) and Mohammed et al. (2008), Raghuwanshi and Wallender methods were the worst methods. Moreover, the results which obtained and reported in this study did not match with the other results reported by the other researchers. Finally, comparing the results of lysimetric studies in semi-arid climate confirmed the selected methods.