AIMS--To determine the sensitivity of commercially available diagnostic assays for Chlamydia trachomatis using a simple method. METHODS--Nine commercial assays and an "in-house" polymerase chain reaction (PCR) were evaluated using serial dilutions of a laboratory grown H serovar--four of them using a laboratory grown E serovar. Seven of the assays were further tested using dilutions of several cervical samples known to contain chlamydiae. RESULTS--The most sensitive assays were the MicroTrak direct fluorescent antibody (DFA) test (Syva) and the PCR which detected C trachomatis at a 10(-8) dilution of the H serovar, while the two least sensitive, Clearview (Unipath) and TestPack (Abbott), were positive only at 10(-4) and-3 dilutions, respectively. A range of enzyme immunoassays (EIAs) and a nucleic acid hybridisation test were of intermediate sensitivity. The results with serovar E were consistent with these. When clinical samples were examined, the DFA test detected C trachomatis in dilutions at least 10-fold greater than any other assay. CONCLUSIONS--The range of sensitivity of diagnostic assays determined by the laboratory dilution procedure is very wide. Sensitivity assessed in this way, however, reflects the ability of the assays to detect C trachomatis in large scale clinical trials. The dilution procedure, which is simple to undertake, could therefore be applied by any laboratory before a new diagnostic method is considered for routine use.