The authors report the results of 322 assays of plasma fibronectin. The object of this study was to specify its diagnostic contribution. The assays have been effected from February 1987 to October 1988 by the laser nephelometric method in all the hospitalized patients without exclusion. We have defined two populations on clinical diagnostic criteria into 193 control cases and 129 patients suffering from inflammatory rheumatisms or bone neoplasias. The statistical processing has been effected in the Department of Medical Computer Science. The results are as follows: The rate of plasma fibronectin is inclined to increase with age (p = 1 X 10(-4) a little higher among men than among women (p = 0.05). With regard to the other biological markers of inflammation, there was no obvious correlation with the rate of fibrin, the sedimentation rate at the first hour, or the presence of C-reactive protein. This study does not confirm the specificity of this assay in the inflammatory rheumatisms although our two populations are strictly comparable with respect to age and sex, and the clinical and biological parameters (sedimentation rate and fibrin) would allow us to discern the difference, if any, between the two populations. Plasma fibronectin assay has no diagnostic or discriminating value on a large scale.