Abstract: Due to the high growth rate of Chinese economy, environmental pollution has become a social focus. SO2 is a typical traditional pollutant, which has attracted the attention of Chinese government since the early 1990s. This article employed China's provincial panel data between 1992 and 2012 to investigate the absolute and relative convergence in per capita SO2 emissions across provinces. We utilized static and dynamic panel data estimators, and analyzed the main impact factors which may affect the convergence speed. The empirical results indicate that in the chosen sample period, there were absolute and conditional convergences in per capita SO2 emissions. Compared with the period after the year of 2001, the convergence speed was faster during the pre-2001 period. According to the factors that might influence convergence speed, the ratio of secondary industry to GDP has a significant positive impact on SO2 emissions. Therefore, a higher ratio of secondary industry to GDP would reduce the convergence speed. The real GDP per capita and population density are both estimated to be negatively related with SO2 emissions. Hence higher income per capita or higher population density would accelerate the convergence progress.