Otosclerosis is a localized bone dystrophy of unknown etiology mainly involving the stapes. The hypothesis of a persistent infection by the measles virus was based on the inconstant detection of the virus by various methods, including reverse transcription-PCR (RT-PCR) of patients' stapes samples. The aim of this work was to investigate the presence of the measles virus in stapedial otosclerosis foci by different sensitive methods. Pathologic stapes samples were obtained from 35 patients suffering from otosclerosis. Measles virus detection was performed by (i) cocultures of Vero cells and primary cell cultures of bone samples (n = 7), (ii) immunofluorescence study of these cocultures (n = 3), and (iii) RT-PCR on RNA directly obtained from fresh frozen samples (n = 28) and on RNA extracted from the primary cell cultures (n = 2). Viral genomic regions coding for N (nucleoprotein) and M (matrix) proteins were separately amplified. PCR sensitivity was optimized on the measles virus Edmonston strain. Glyceraldehyde-3-phosphate dehydrogenase mRNA was used as a marker of total RNA recovery. PCR products were tested by Southern blot hybridization technique to improve sensitivity and specificity. PCRs amplifying the M and the N protein genes were able to detect the control measles virus RNA at titers as low as 0.1 and 0.01 50% tissue culture infective dose, respectively. With these highly sensitive methods, we could not evidence the presence of the measles virus in any of our bone samples or primary bone cell cultures. Our results do not confirm the hypothesis of persistent measles virus infection in otosclerosis.