Abstract: To predict the gas content of coal reservoirs more accurately under situation of lacking measured coring data in field, by adopting the conventional loggi ng curves, influencing features and mechanisms of limestone's thickness, the shale content, effective porosity, water content, buried depth and the contact relationship between limestone and coal seam on the gas content of coal reservoirs were analyzed, and multivariate regression model for gas content of coal reservoir was establis hed based on the characteristics of limestone. The results showed that the gas content of coal reservoir was found to be inversely associated with the effective porosity and water content, and positively correlated with buried depth and shale content; effective porosity was the main controlling factor of gas content of coal reservoir. Ther e was no obvious relationship between limestone's thickness and gas content of coal reservoir. Limestone and coal seam were indirect contacted in local area, the cont act relationship could not affect the gas content of coal reservoir; regression model was verified to be good, but on the edge of the basin and fault developing zone, the error was larger.