Synovial fluid (SF) from patients with osteoarthritis (OA), rheumatoid arthritis (RA), and various other arthridites was analyzed to assess the prevalence of interleukin-1 (IL-1) using both radioimmunoassay competitive inhibition specific for the beta form of IL-1 and the D10.G4.1 cell line bioassay which measures both alpha and beta forms of IL-1. Using radioimmunoassay competitive inhibition, IL-1 beta was found in 45% and 60% of individual samples from patients with OA and RA respectively. When RA and OA SF were examined in sequentially obtained samples, IL-1 beta occurred in one or more samples from 8 of 10 patients studied, suggesting the probability that it can be produced at some time in SF by all patients with these conditions. No correlation between SF leukocyte counts and the occurrence of IL-1 beta was noted and no effect of antiinflammatory drug treatment on the prevalence of IL-1 beta was found. When tested for the presence of IL-1 by the D10.G4-1 cell line, 66% and 50% of RA and OA patients respectively were found to contain IL-1. These were not in total concordance with results obtained by RIA. Of all SF tested, seven were negative by RIA but positive by D10.G4.1 and these are considered to contain IL-1 alpha. Seven samples which were RIA positive and D10.G4.1 negative were tested for their ability to inhibit IL-1 responses in the bioassay. Five of these contained inhibitor and one markedly enhanced the proliferative response of D10.G4.1 to a known amount of IL-1.(ABSTRACT TRUNCATED AT 250 WORDS)