Reports results of clinical trials of Ca-Al-PAS on 20 tuberculosis patients in the conditions when (1) there is no radiological improvement of lesions (bacteriology remaining) consistently +ve), (2) radiological regression is slight (bacteriological results +ve) (3) radiological regression is very definite but limited (bacteriology - ve), (4) radiological cleaning is considerable with some doubtful anomalies persisting with -ve bacteriology, and (5) when radiological cleaning is very considerable and bacteriology -ve.