Abstract: In the present study, L. casei, L. farciminis and L. delbrueckii were inoculated together at a ratio of 1:1:1 to produce cured grass carp. The effects of total inoculum amount, salt amount, curing temperature and curing time on the quality of cured fish were studied. The optimal levels of the parameters were optimized using one-factor-at-a-time and orthogonal array design methods to be 106 CFU/g, 5%, 10 ℃ and 4 d, respectively. Products with close texture, low salinity and the flavor of conventional cured fish were obtained under these conditions. The TVB-N and peroxide value were 18.72 mg/100 g and 0.18 g/kg, which decreased by 34.9% and 51.6%, respectively, when compared with naturally fermented cured fish.

Key words: Lactobacillus, cured fish, one-factor-at-a-time method, orthogonal array design