The effects of combined drugs were compared uith the effects of single drugs in vitro against Mycobacterium kansasii. The single drugs isoniazid 1.0 microgram/ml, streptomycin 2.0 microgram/ml, and ethambutol 5.0 microgram/ml and the combinations of 1.0 microgram/ml isoniazid and 2.0 microgram/ml streptomycin, 1.0 microgram/ml isoniazid and 5.0 ethambutal and 1.0 microgram/ml isoniazid, 2.0 microgram/ml streptomycin and 5.0 microgram/ml ethambutol were evaluated as to their effects on M. kansasii organisms from 22 patients with pulmonary mycobacterial disease. These drugs were incorporated into Middlebrook 7H11 medium. Colonial growth was then observed and assigned numerical values for statistical analysis. The results showed that isoniazid was not significantly better than ethambutol. Streptomycin, however, was significantly better than both isoniazid and ethambutol alone. The double combinations were better than the single drugs, with isoniazid--streptomycin being better than isoniazid--ethambutol. The combination isoniazid--ethambutol was not significantly better than streptomycin alone. However, the triple combination of isoniazid--streptomycin--ethambutol was markedly better than all single drugs, better than isoniazid--ethambutol, but not better than isoniazid--streptomycin. Combined-drug testing thus showed a pattern of susceptibility not demonstrated by single-drug testing.