203 Biliary atresia in most countries occurs in 1 in 10-20000 live births. At Baragwanath Hospital (a referral centre in Soweto serving a mainly black population), 20-24 new cases pa were documented from 1988; it was unclear if this was due to an increased incidence in the community. We hypothesized that biliary atresia occurs more frequently in the Soweto/Gauteng area than reported in world literature. Aims. The purpose of this study was to determine the local incidence, age at presentation, operability and short term operative outcome. Methods. The records of cases of biliary atresia seen at Baragwanath Hospital from 1993-1996 were reviewed. Place of residence, age at presentation, gender, race, management, age at operation and short-term operative outcome were reviewed. Inoperability was assessed climically (poor liver synthetic function despite replacement), histologically (established cirrhosis) or because of advanced age (age > 4-6 months); laparotomy was performed if there was uncertainty regarding operability. Operative outcome was considered successful- jaundice resolved; partial- stools were pigmented but jaundice persisted; failed- stools remained acholic. Results. Seventy-one infants with biliary atresia were diagnosed - 23 in 1993, 23 in 1994, 9 in 1995 and 16 in 1996. Male:female ratio was 0.58:1, with a racial distribution of 67 (94.4%) black, 2 (2.8%) white and 2 (2.8%) of mixed race. Information regarding place of residence was available in 63/71; 31 were living in Soweto - 9 seen in 1993, 11 in 1994, 3 in 1995 and 8 in 1996 - giving an incidence of approximately 1 in 2500 live births for 1993/4/6 and 1 in 8000 in 1995, based on annual Soweto birth rates. For the areas outside of Soweto, a lack of accurate data on the population denominator precluded further analyses. Median age at presentation was 95.5 days (range 9-677); 24/71 (34%) were less than 60 days of age. Management was as follows: 1 died preoperatively of gram-negative septicaemia, operation was refused in 1, 1 unknown, 23 (32%) were inoperable clinically, 6 (8%) were inoperable at laparotomy, and 39(55%) underwent portoenterostomy. Surgery was successful in 10/39 (25.6%), partially successful in 14 (35.9%) and failed in 15 (38.5%). Of the total of 71 infants, only 10 (14%) were jaundice free. These infants tended to be younger, with a trend towards poorer outcome with increasing age at surgery. Conclusion. There is a high incidence of biliary atresia in black infants from Soweto, which remained constant from 1988 to 1994 and 1996, with a decrease in 1995. Operative management yielded poor results. These findings appear to confirm both the high incidence and poorer outcome of biliary atresia in black infants from Soweto. Identification of cases and early diagnosis are of paramount importance particularly in South Africa where liver transplantation is not an option for the majority of patients.