Using murine monoclonal antibodies (MAbs) raised against the common antigen of group A rotavirus (RV), two single-sandwich ELISA systems were developed for detection of RV in stools: one using polyclonal antibody (PAb) as capture and a MAb as detector antibody (referred to as PAb-MAb assay); and the other based on the use of two different MAbs as capture and detector antibodies (referred to as MAb-MAb assay). In each single-sandwich ELISA system, samples and peroxidase-labeled MAb were incubated sequentially (two-step method) or simultaneously (one-step method). Using the two-step procedure on purified RV, 50 pg of protein was detected in the PAb-MAb as well as in the MAb-MAb assay, whereas the one-step method detected 0.4 ng and a conventional double-sandwich ELISA detected 3.2 ng of viral protein. Titration of RV samples from stools and cell cultures showed that single-sandwich ELISA titers were, on the average, 10-100-fold higher than those obtained by electron microscopy (EM), but 10-100-fold lower than those obtained by solid-phase immune EM (SPIEM). However, when 200 stool samples previously examined by EM or SPIEM were tested by the single-sandwich ELISA systems, specificity and sensitivity of these assays were 100%, and comparable to SPIEM. No false positive results were obtained when 54 samples of meconium and 91 stools from newborns in the first five days of life were tested. The two-step procedure appeared to be somewhat preferable over the one-step method, which, although faster, gave a marked prozone with a few samples in the MAb-MAb assay. The use of MAbs in rapid single-sandwich ELISA systems for RV detection in stools appears highly convenient, due to reliable results and short test performance times.