Using the Raji cell radioimmunoassay (RCRIA) reported by Theofilopoulos et al., 26 serum samples from systemic lupus erythematosus (SLE) patients were evaluated. All of those patients had definite granular immune deposits in renal biopsy specimens, and the presence of circulating immune complex in these patients was also suspected. RCRIA values of these patients were generally higher than those of normal human sera. RCRIA values were compared with other serological parameters of SLE including serum complement (CH(50)), serum gammaglobulin, anti-double stranded DNA antibody, anti-nuclear ribonucleoprotein (RNP) and anti-lymphocyte antibody. There was no correlation between the RCRIA values and anti-lymhocyte antibody. The correlation with CH(50) was also rather negative. There was positive correlations between the RCRIA values and serum gammaglobulin and anti-ds DNA antibody. However there were four serum samples which were negative to anti-ds DNA antibody but high RCRIA values. Three samples had high anti-RNP antibody titiers. The data suggest that not only the ds-DNA antigen-antibody system, but also other kinds of antigen-antibody systems may compose the circulating immune complex. The RCRIA values of sera from SLE patients using Raji cells killed by 100 % acetone instead of living cells were generally higher. The increases in the RCRIA values were regarded as being due to the participation of anti-nuclear antibodies present in the patients' sera. Thus, in evaluating SLE sera, there is the possibility that the RCRIA may give false positive results due to the presence of antibodies, including antinuclear antibodies, which react with Raji cells. As long as this possibility is kept in mind, the RCRIA value may become a valuable serological index of SLE.