In treating brain abscess we have three primary concerns: saving life, preserving or restoring function, and preventing sequels such as epilepsy and recurrence of infection. Our success with the first of these is accurately expressed by the immediate mortality rate, but information about the last two can only be obtained by long-term follow-up studies. And for statistical assessment larger figures are required than any one clinic can supply. Furthermore, the methods of treatment tend to differ in different clinics. In Oxford, for instance, we have records of about 120 cases of brain abscess treated since 1938, but most of these have been dealt with by excision after successful aspiration, and we have little useful information about aspiration or other forms of drainage as the sole procedure. Accordingly we approached Professors Norman Dott and Sir Geoffrey Jefferson who kindly placed at our disposal the records of their cases treated in Edinburgh and Manchester over the last 25 years or so. From the three centres there were altogether 295 cases. Of these patients, 118 died soon after operation (40%). The remaining 177 patients survived for periods up to 26 years, and of these it has been possible to obtain adequate follow-up information in 173 cases. As we were particularly interested in the long-term results, we have not included any cases treated within the last 12 months. We found that in general three methods of treatment had been used (Table I): aspiration (with or without a decompression); continuous drainage by a tube, or by marsupializing the abscess; and excision (with or without previous aspiration and/or decompression). No clinic had applied any of these methods to the exclusion of others, but there was a strong preference for excision in one clinic, and for continuous drainage in another. In the third clinic there was less rigidity and the various methods all had a fair trial.