Whole milk, skim milk and an emulsion of milk fat in water, inoculated with approx. 10(5) cfu/ml of Listeria innocua, were treated at 30 degrees C with 100 IU/ml of nisin, homogenization at 200 bar or both procedures. Nisin activity and survival of L. innocua after treatments were determined. Recovery of nisin activity from non-homogenized whole milk treated with 100 IU/ml of nisin was complete, whereas a loss of 18 to 28% of activity was detected in non-homogenized fat-in-water emulsion. Loss in nisin activity due to homogenization represented up to 64% in whole milk and 62% in fat-in-water emulsion. Nisin addition by itself achieved a reduction in L. innocua counts of 3.7-3.8 log units in whole milk and 3.6 log units in fat-in-water emulsion compared to numbers in untreated samples. When nisin-containing whole milk and fat-in-water emulsion were homogenized, L. innocua counts were only reduced by 2.6-2.9 log units and 2.5 log units, respectively, compared to numbers in untreated samples. Homogenization of nisin-containing skim milk resulted in a loss of nisin activity of 20% but achieved a reduction of 3.0 log units in L. innocua counts.