Acne vulgaris is a common disease of the pilosebaceous unit. Antibiotics and non-antibiotic antimicrobial drugs which reduce the growth of Propionibacterium acnes (P genes) are of therapeutic value. The .study- aimed to compare the activity of benzoyl peroxide 5% with or without topical erythromycin 3% against P. acnes and to determine the development of antimicrobial resistance. Methods: A double-blind, randomized .study was performed on 75 patients with mild to moderate acne vulgaris. The effect of treatment was assessed by lesion counting system. Samples were obtained from the forehead at 0, 4, N, 12 and 10 weeks of treatment. Brain heart infusion agar with supplements was used for isolation of P aches in an anaerobic chamber Agar dilution method was performed for minimal' inhibitory, concentration MIC of erythromycin. Results: Both treatment groups produced significant clinical improvements with no better results in the combination therapy group. Reduction in the total lesion counts was .similar in both groups at the end of treatment. The $MIC_90$ value of P. aches was 0.125 l$mu$g-ml for erythromycin prior to therapy Isolation rate of P. aches was similar in both treatment groups. The $MIC_90$ values of erythromycin were higher in P aches isolated from the combination therapy group than benzoyl peroxide group (p, 0.001). Conclusion: these data suggest that the combination erythromycin and benzoyl peroxide is not synergestic against P aches and resistance to erythromycin develops with the combination therapy