Autoimmune mechanisms are thought to be involved in the pathogenesis of the chorioretinal changes in ocular onchocerciasis. The humoral autoimmune response was determined by measuring serum levels of autoantibodies, directed against human S-antigen and interphotoreceptor retinoid binding protein (IRBP) using an enzyme immunoassay. The cell-mediated immune response to these antigens and a crude retinal extract was investigated by means of a two-step migration inhibition factor assay. Patients with onchocerciasis (n = 50) were subdivided into three groups: 1. without ocular involvement (n = 10), 2. with ocular onchocerciasis limited to the anterior segment (n = 19), 3. with onchocercal chorioretinopathy (n = 21). A group of endemic controls from Sierra Leone, West Africa were also studied. The cellular immune response to Concanavalin A was measured to assess the general capacity of lymphocytes to respond to a mitogen. High levels of anti-human S-antigen and IRBP antibodies were detected in patients with onchocerciasis and endemic controls. The levels of both anti-human S-antigen and IRBP antibodies were significantly higher in onchocerciasis patients compared to endemic controls (Mann-Whitney ranksum test; p less than 0.001 respectively 0.002). No relationship could be demonstrated between the anti-retinal antibody level and the occurrence of chorioretinitis in ocular onchocerciasis. The occurrence of the anti-retinal antibodies as a result of crossreactivity of anti-retinal antibodies with parasitic antigens or of induction of polyclonal B-cell activation due to parasitic infection is discussed, since high antibody levels were also found in patients with Bancroftian filariasis from Papua New Guinea and Surinam. The migration inhibition factor assay, in which the cell-mediated immune response to human S-antigen, IRBP and retinal extract was tested, showed that four out of 50 (8%) patients with onchocerciasis and four out of 25 (16%) endemic controls reacted with at least one retinal antigen. From the patients with onchocercal chorioretinopathy two out of 21 (10%) showed a positive cellular response. The general mitogen response tested with Con A was positive in all these individuals. In conclusion, circulating antibodies against human S-antigen or human IRBP are thus nor specific for onchocerciasis and in themselves not sufficient to cause chorioretinopathy in onchocerciasis, although their pathogenic role in an ongoing chorioretinitis cannot be excluded. Furthermore a role for a cell-mediated anti-retinal autoimmune mechanism in the pathogenesis of chorioretinitis in onchocerciasis as studied with human S-antigen, IRBP or crude retinal extract could not be demonstrated.