OBJECTIVE: A minimal cancer incidence data for Karachi, the largest city of Pakistan, is being presented here, for the years 1998-1999. The city has a population of 9,802,134; males 5,261,712 (52.6%) and females 4,540,422 (47.4%); census 1998. METHODOLOGY: A predominantly mixed (passive and active) registration system has evolved in Karachi, the data sources being the hospitals within the Karachi Division. The reported/retrieved cancer data sets at the Karachi Cancer Registry are checked, coded, computerised in an analytical format and analysed. RESULTS: The incident cancer cases registered in Karachi, during the 2-year period, 1st January 1998 to 31st December 1999 were analysed. The age-standardised incidence rate (ASR) of cancer, all sites was 132.4/100,000 for the males. Cancer of the lung 10.8%; ASR 17.3 was the most frequently recorded malignancy, followed by oral cavity 10.5%; ASR 13.2 and larynx 5.0%; ASR 7.4. The age-standardised incidence rate (ASR) of cancer, all sites was 133.0/100,000 in the females. Cancer of the breast, 32.0%; ASR 40.7 was the most frequently recorded malignancy, followed by oral cavity 8.1%; ASR 11.7 and gall bladder 3.6%; ASR 5.5. CONCLUSION: The present data has been calculated with an estimated 15-20% probable under ascertainment. Tobacco-associated cancers in Karachi were responsible for 38.3% of the tumours diagnosed amongst the males. Two principal cancers, breast and oral cavity were responsible for 40.1% of the cancers in females. A rare finding was the high incidence of gall bladder cancer in the females. At present it is difficult to determine whether this indicates a genuine high risk or a selection bias. A continuous process of cancer registration to study the trends in the incidence and an adequate cancer control program are possible and essential for Pakistan and can be based on the pattern being practiced in Karachi.