In this study, five mixed substrates – SSFGF, FSSGF, FSSFF, FSSFG, and FSFGF – consisting of fly ash, sludge, soil, fine cinder, gravels, and fine sand, were made in order to test the removal effect of lead. Multiple comparisons and analysis of variance were used to analyze physicochemical properties and the removal efficiency of lead. The test results showed that for five substrates, the effluent concentrations were not significantly different among 10 mg/L and 40 mg/L initial concentrations. For the other three initial concentrations (20 mg/L, 80 mg/L and 160 mg/L) from five substrates, it was significantly different. Moreover, when decomposition time was 0.25 h~48 h, the removal efficiency of Pb from the solution was better for substrate SSFGF-based wetland than those of the other substrate-based wetlands at low and medium initial concentrations (10 mg/L, 20 mg/L, 40 mg/L, and 80 mg/L), and then it was slightly better for substrate FSSFF-based wetland. At relatively high initial concentrations (160 mg/L), the removal efficiency was better for substrate FSSFF-based wetland than those of the other substrate-based wetlands, and then it was slightly better for substrate SSGFF-based wetland. Our study further suggested that SSFGF and FSSFF were the two relatively ideal substrate materials suitable for removal of Pb from a constructed wetland system.