Abstract: Response surface methodology (RSM) had been employed to optimize the clarifying starch medium for citric acid production by Aspergillus niger. Firstly, according to the results of amono-factor experiment, a Plackett-Burman design was used to investigate the effects of different factors in the ferment medium, with the three statistically significant factors being identified as:corn steep liquor, urea and p H. Secondly, the steepest ascent procedure wasemployed to define the optimal response region for these three factors. Finally, the BoxBehnken method was employed to design the reaction system. The results of the experiments were analyzed by RSM with Minitab15.0 software. The optimal fermentation medium for production of citric acid consisted of the following (g/L) :corn steep liquor 5.15g/L, urea 1.83g/L, p H4.15, Mg SO40.2g/L, K2HPO40.4g/L. The average yield and conversion rate of citric acid, compared with unrefined starch culture, improved about 10%, meanwhile, the total residual sugar reduced about 88%.