We evaluated analytic performances of an automated treponemal test and compared this test with the Venereal Disease Research Laboratory test (VDRL) and fluorescent treponemal antibody absorption test (FTA-ABS). Precision performance of the Architect Syphilis TP assay (TP; Abbott Japan, Tokyo, Japan) was assessed, and 150 serum samples were assayed with the TP before and after heat inactivation to estimate the effect of heat inactivation. A total of 616 specimens were tested with the FTA-ABS and TP, and 400 were examined with the VDRL. The TP showed good precision performance with total imprecision of less than a 10% coefficient of variation. An excellent linear relationship between results before and after heat inactivation was observed (R(2) = 0.9961). The FTA-ABS and TP agreed well with a κ coefficient of 0.981. The concordance rate between the FTA-ABS and TP was the highest (99.0%), followed by the rates between FTA-ABS and VDRL (85.0%) and between TP and VDRL (83.8%). The automated TP assay may be adequate for screening for syphilis in a large volume of samples and can be an alternative to FTA-ABS.