748 children with severe pneumonia in three hospitals in Papua New Guinea were randomised to receive intramuscular injections of either chloramphenicol alone or chloramphenicol plus penicillin. Sequential analysis showed no difference between the two treatments. 48 (13%) of the 377 children in the chloramphenicol alone group died, and 3 (0.8%) were changed to different treatment. 62 (17%) of the 371 children in the chloramphenicol-plus-penicillin group died, and 6 (1.6%) were changed to different treatment. The difference in failure rates (death or withdrawal for change of treatment) was 4.8% +/- 5.2% (+/- 95% confidence limits). In children with severe pneumonia, treatment with chloramphenicol alone is as effective as treatment with chloramphenicol plus penicillin.