In a first step the effects of 10 components of a culture medium designed for L-lysine production were evaluated with a 2(10-6) factorial design. Among them, glucose, fish silage, and ammonium sulphate showed a significant effect. In a second step, an orthogonal-central composite experimental design and response surface methodology was performed with five from the 10 initial compounds. The determination coefficient (R-2) of the fitted second-order model was 0.990. L-lysine production with the optimised medium increased 2.6 times as compared with the original medium.