The anti-SSB antibodies were measured using two enzyme immunoassays (ELISA). The difference between the both results from the preparation of the SSB antigenic extract. The first method, developed in our laboratory, uses a purified SSB antigen extracted from calf thymus, while the other uses an antigen cloned by genetic engineering. We have realized an analytic investigation about the repetability, the reproducibility and the detection limit of our ELISA method and we have refined its evaluation in using a clinic study carried out on 203 subjects (55 had a Sjögren syndrome, 47 had a systemic lupus erythematosus, 17 had a rheumatoid arthritis, 13 a progressive systemic sclerosis, 10 a polymyositis and 61 healthy subjects (blood donors)). These measures were worked out in order to compare them with the Ouchterlony method of reference. The results we have obtained are totally similar to the ELISA methods, with a global correlation factor of 0.96 in spite of the difference on the preparation of SSB antigenic extract. The enzyme immunoassay is a lot more sensitive than the Ouchterlony method since, for a Sjögren sample, we obtain a sensitivity of 0.69 while the sensitivity is only of 0.51 for the immunoprecipitation. In the lupus sample, the sensitivity is respectively of 0.42 and 0.25.