Objective. The recent identification of antigens or nucleic acids of infectious agents in the joints of patients with reactive arthritis has raised questions about whether chlamydial or other infectious agent nucleic acids are also present in normal joints. We had the opportunity to study synovium from 30 asymptomatic volunteer subjects by use of polymerase chain reaction (PCR) for attempted identification of Chlamydia and other infectious agents. Methods. All subjects had blind needle synovial biopsies with the Parker-Pearson needle. DNA was extracted and PCR performed using primers for Chla-mydia trachomatis, Chlamydia pneumoniae, Borrelia burg-dorferi, and pan bacterial 16S ribosomal RNA (rRNA). Results. Two subjects were identified with nucleic acid for the 16S rRNA gene of C trachomatis. All other PCR reactions were negative except for the pan bacterial 16S rRNA in the C trachomatis-positive subjects. Both subjects, although symptom free, had some evidence of synovial reaction. Conclusion. C trachomatis appears to occasionally be disseminated to joints without producing overt disease. The identification over the past few years of antigens or nucleic acids from Chlamydia (1-7) and several other organisms (see later) in the joints of patients with reactive arthritis has led to a reevaluation of the pathogenesis and treatment of these patients. Our studies and others have shown that evidence of bacterial components (most often Chlamydia trachomatis) in joints, although most common in reactive arthritis, can also occur in patients with other diagnoses (5,8). Since it is known that many other organisms, such as mycobacteria, can be widely disseminated without producing symptoms, we thought that it was important to examine normal control subjects to begin to get some impressions about whether and how often C trachomatis and other bacteria associated with arthritis might be disseminated to joints without producing symptoms. This report describes the first study of apparently normal subjects, a survey of 30 asymptomatic volunteers who consented to blind needle biopsies of knee syno-vium, for evidence of chlamydial or other infectious agents in these joints. SUBJECTS AND METHODS Thirty normal volunteers gave their informed consent and were entered into a protocol approved by the Institutional Review Board at the clinical center at the National Institute of Arthritis and Musculoskeletal and Skin Diseases to study a number of aspects of the features of normal synovium. All subjects were required to have been totally free of any knee or other joint symptoms at any time and to have no objective signs of arthritis. All were also questioned about any antecedent infections that might have later been associated with arthritis, but subjects were not excluded on this basis. All subjects had normal findings on knee radiographs. Results of physical examinations were normal except for mild, painless, retropatellar crepitus noted in 4 subjects. Subjects were age and sex matched to reflect the composition of an ongoing study our g...