One hundred and thirty-nine patients with pulmonary tuberculosis were evaluated by means of gallium-67 and thallium-201 scintigraphy. The disease was clinically active in 83 and inactive in 56. The uptake ratio between the lesion and the contralateral normal lung field was calculated. The ratio determined by 67Ga scintigraphy was expressed as GR, and that determined by 201T1 scintigraphy (early or delayed) as ER or DR. The 201T1 retention index (RI) was calculated using the following equation: RI = DR - ER/ER x 100. The sensitivity, specificity and accuracy of 201T1 scintigraphy with respect to the activity of pulmonary tuberculosis were better than those of 67Ga scintigraphy (the figures for 201T1 scintigraphy were 88.0%, 82.1% and 85.6%, respectively, and those for 67Ga scintigraphy, 83.1%, 60.7% and 74.1%). We found a significant correlation between GR and ER, but there was no significant correlation between RI and ER. When the relationships between ER, C-reactive protein and 1-h erythrocyte sedimentation rate were examined among patients with abnormal uptake, no significant relationships were noted. In the 27 patients who could be followed up, GR and ER decreased with duration of the therapy, indicating a decrease in disease activity which was consistent with clinical findings. The RI was significantly higher in the early stages of therapy than in the later stages (P < 0.01), suggesting a temporary delay in 201T1 washout in the early stages. The washout seemed to be promoted by the effects of the therapy. Overall, it is concluded that 201T1 scintigraphy is more useful and more suitable than 67Ga scintigraphy for the evaluation of disease activity and therapeutic effects in patients with pulmonary tuberculosis.