The degradation of chlorsulfuron and triasulfuron was investigated in alkaline soils (pH 7.1±9.4) spiked at 40 lg a.i. kg ±1 under laboratory conditions at 25°C and a moisture content corresponding to 70% ®eld capacity (±33 kPa), using high-performance liquid chromatography. Degradation data for the two herbicides did not follow ®rst-order kinetics, and observed DT 50 values in surface soils ranged from 19 to 42 days and from 3 to 24 days for chlorsulfuron and triasulfuron respectively. Disappearance of both chlorsulfuron and triasulfuron was faster in non-sterile than in sterile soil, demonstrating the importance of microbes in the breakdown process. The persistence of chlorsulfuron increased with increasing depth, which can be attributed to the decline in the microbial populations down the pro®le. The DT 50 value for chlorsulfuron at 30±40 cm depth was nearly four times higher than that in the top-soil. The results obtained show that persistence of these herbicides in alkaline surface soils at 25°C and at a moisture content of 70% ®eld capacity is similar to those reported in other European and North American soils. The study shows that if these herbicides are contained in surface soil layers, the risk of residue carry-over under southern Australian conditions is small. However, the rate of their degradation in alkaline subsoils is very slow, and under conditions conducive to leaching their prolonged persistence in the soil pro®le is possible. Materials and methods Soils and herbicidesThis study entailed three experiments. Experiment 1 consisted of chlorsulfuron and triasulfuron degradation studies in freshly collected surface soils (0±10 cm) of Avon (calcareous sandy loam), 84 A K Sarmah et al.

The ability of bioassay, enzyme-linked immunosorbent assay (ELISA) and high-performance liquid chromatography (hplc) methods to detect sulfonylurea herbicides in soil was evaluated as part of a project studying the leaching and persistence of these herbicides in the alkaline soils of south-eastern Australia. Soil samples with known concentrations between 0.1 and 10 lg a.i. kg A1 chlorsulfuron, metsulfuron-methyl or triasulfuron were prepared by an independent laboratory and supplied in coded bags to separate laboratories for testing. The accuracy of the results was analysed, and the merits of each method are discussed. Bioassay was suitable for measuring biologically active residues from 0.1 to 1.0 lg a.i. kg A1 . ELISA accurately measured residues in the range of 0.1±10 lg a.i. kg A1 , making it the most widely adaptable assay tested. It will be useful for measuring residues in sodic subsoils where bioassay plants grow poorly. There was good reproducibility between the bioassay and ELISA. The hplc technique used in this study was not as accurate as bioassay or ELISA at quantifying residues of 3.0±10 lg a.i. kg A1 and could not detect residues at or below 1.0 lg a.i. kg A1 .