The purpose of this article is to outline the interaction between ophthalmologists and internists in the management of uveitis. Two issues will be addressed: 1) which strategies should the internist follow when asked to investigate a case of uveitis; and 2) in which systemic diseases should the internist order an ocular examination to rule out intraocular inflammation. The modern approach to the diagnosis of uveitis is based on the naming-meshing system popularized by Smith and Nozik. After a short history (ocular complaints, general health) an ophthalmic examination is carried out to determine the anatomic structures involved. Based on the results a uveitis is classified as anterior uveitis, intermediate uveitis, posterior uveitis, or panuveitis. Associated factors (eg, unilateral versus bilateral, acute versus chronic, granulomatous versus nongranulomatous, etc.) are also assessed. Based on this information the type of uveitis will be named (eg, acute, nongranulomatous, unilateral, anterior uveitis) and matched (meshing) to a potential list of etiologies (eg, viral iritis, HLA-B 27 associated iritis). Targeted questioning and selected medical and laboratory investigations based on the shortlist will then identify a possible cause for a particular patient's uveitis. In other words the ophthalmologist should never ask the internist to run the full battery of tests in a patient with uveitis. He rather should indicate which type of uveitis is present and what are the most likely diagnoses to be excluded. Many systemic diseases cause diffuse inflammation and are associated with uveitis. These include tuberculosis, spirochaetal diseases such as Lyme disease and syphilis, sarcoidosis, Behçet syndrome, juvenile idiopathic arthritis, and HIV infection amongst many others. Routine ophthalmic examination in patients with systemic disease may be indicated for diverse reasons: to prevent profound damage due to asymptomatic uveitis in JIA; to detect diagnostic clues in patients with febris e causa ignota; or to rule out opportunistic infections in HIV positive patients. It is clear that the information gained from routine examination in systemic disease will be greatly dependent on the prevalence of ocular involvement in a particular disease.