Background: Surveys have been conducted to measure prevalence of eye disease in Africa, but not of incidence, which is needed to forecast trends. The incidence of visual loss is reported in southwest Uganda. Methods: A rural population residing in 15 neighbouring villages was followed between 1994-5 (R1) and 1997-8 (R2). Survey staff screened adult residents (13 years or older) for visual acuity using laminated Snellen's E optotype cards at each survey. Those who failed (VA >6/18) were evaluated by an ophthalmic clinical officer and an ophthalmologist. Incidence of visual loss (per 1000 person years (PY)) was calculated among those who had normal vision at R1. Results: 2124 people were studied at both survey rounds (60.9% of those screened at R1); 48% were male. Participants in R1 were older (34.7 versus 31.5 years at R2, p<0.001). Visual loss in R2 occurred in 56 (2.8%) of 1997, yielding a crude incidence rate of 9.9, and an age standardised incidence rate of 13.2, per 1000 PY. Incidence of visual loss increased with age from 1.21 per 1000 PY among people aged 13-34 to 64.2 per 1000 PY in those aged 65 years or older (p for trend >0.001). The six commonest causes of visual loss were: cataract, refractive error, macular degeneration, chorioretinitis, glaucoma, and corneal opacity. If similar rates are assumed for the whole of Uganda, it is estimated that 30 348 people would develop bilateral blindness or bilateral visual impairment, per year. Conclusions: Cataract and refractive error were the major causes of incident visual loss in south west Uganda. These data are valuable for forecasting and planning eye services.