Our aims were to establish the incidence and clinical characteristics of early and late onset Group B Streptococcal (GBS) septicaemia in neonates in our hospital over a period of 1 year. Routine screening for maternal GBS was not standard practice in the hospital. GBS was isolated from high vaginal swabs (HVS) obtained antenatally or postnatally for risk factors as determined by the obstetrician or neonatologist in charge. Data obtained were analyzed separately and these did not form part of the study. By a system of clinical case review and follow-up, mail, telephone and home visits, the outcome of all 15,062 livebirths in the hospital over a 1-year period were verified and reported. Our results show a low incidence of GBS infection in neonates in the hospital: early onset disease was 0.265 per 1,000 livebirths and late onset a quarter of that. The majority of our cases of early onset GBS disease were in premature infants. Because of our low incidence, prophylaxis schedules would have to ensure an acceptably smaller number of mothers exposed to antibiotics over and above the current level and the cooperation of our obstetricians. We have devised a schedule incorporating a current PROM (prelabour premature rupture of membranes) protocol which would result in only an additional 2.2% of mothers requiring prophylactic antibiotics.