The antimicrobial properties of florfenicol were investigated for the bovine respiratory tract pathogens, Mannheimia haemolytica and Pasteurella multocida . Three in vitro indices of efficacy and potency were determined; minimum inhibitory concentration (MIC), minimum bactericidal concentration (MBC) and in vitro time‐kill curves for six pathogenic strains of each organism. Each was monitored in two matrices, Mueller Hinton broth (MHB) and calf serum. MBC:MIC ratios were low, 1.8 : 1 for M haemolytica in both MHB and serum and 2.4 : 1 and 2.1 : 1 for P multocida in MHB and serum, respectively. The killing action of florfenicol had the characteristics of concentration dependency against M haemolytica and codependency (on time and concentration) against P multocida . Modelling of the time‐kill data after 24 hours exposure was undertaken to quantify three levels of activity for the ratio, area under concentration‐time curve over 24 hours (AUC 24h )/MIC; bacteriostatic action (no change in bacterial count), 3log 10 reduction and 4log 10 reduction in bacterial count. Mean AUC 24h /MIC values for P multocida in MHB (and serum) were 22.0 (23.3) hour, 34.5 (39.9) hour and 45.8 (50.4) hour, respectively. Similar numerical values were obtained for M haemolytica. For both bacterial species, interstrain variability was low; coefficients of variation ( per cent) in serum for 3log 10 and 4log 10 reductions in count were, respectively, 14.3 and 24.1 for P multocida and 7.8 and 11.4 for M haemolytica . These data form a rational basis for dosage selection for treatment of calf pneumonia caused by M haemolytica or P multocida .