In this study, the Frequency and the spell of rainy days was analyzed in Lake Uremia Basin using Markov chain model. For this purpose, the daily precipitation data of 7 synoptic stations in Lake Uremia basin were used for the period 1995- 2014. The daily precipitation data at each station were classified into the wet and dry state and the fitness of first order Markov chain on data series was examined using Chi-square test at a significance level of 0.01 and was approved. After computing transition probability matrix, the persistent probability, average spell of dry days and rainy days and weather cycle was calculated. By calculating the frequency of 1-10 rainy, the spell of this periods and 2-5-days return period were calculated. The results show that in this study period the average of rainy days is 25% and the probability of P_dd is more than other states (P_ww ، P_dw و P_wd). The average spell of rainy days in the study area was estimated at about two days. Generally, in all stations the persistent probability of wet state is more than rainy state. Estimation of frequency and spell of rainy days and 2-5-days return period show that with increasing duration, the frequency of rainy days decreases. Also with increasing duration of rainy days, their spell is reduces and return period increases.