The study examines the impact of pre-monsoon rainfall deviation on maize yield in Manipur. Secondary data on rainfall was extracted from high resolution 0.50×0.50 daily gridded data obtained from India Meteorological Department for the period of twenty-eight years (1980-2007). The annual yield of maize was regressed on time trend and rainfall (January, February, March, April, May). Different regression models were worked out using different explanatory variables and functional form and the log-lin model turns out to be the best model based on logistic (sign and coefficients), statistical (p-value) and econometric (value of R-square) criteria. The regression model shows that time trend has positive and significant (1% level of significance) impact on yield of maize and March rainfall deviation has negative and significant (5% level of significance) impact on yield of maize. This is because, in Manipur, the tasseling and silking stage (critical stage of maize) falls during March. Hence, 1 per cent increase in March deviation leads to 5.6 unit decrease in yield of maize. The study recommended planting of drought-tolerant maize variety in the study area.