The herbicide alachlor was added to Neosho River water samples and subjected to a typical treatment sequence used for treatment of public water supplies in Kansas. No significant alachlor removal was caused by chlorination, lime-soda softening, alum flocculation, or sand filtration. Partial removal was achieved when anthracite was used in filtration. Activated carbon in batch treatment for taste and odor removed 70 percent-80 percent of the alachlor. Granular activated carbon used in filtration removed alachlor to below the detection limit. These results indicate the need for routine activated carbon treatment of public water supplies to prevent public exposure to alachlor. Alachlor, the active ingredient in Lasso herbicide, has been identified as a possible cancer threat to humans (Schatzow, 1984). Although restrictions and cautions have been placed on its application, alachlor is still in use. With a lifetime in soil that may range from two months (Wauchope, 1978) to nine months (Ashton, 1982), alachlor persists long enough that there is little doubt that some will be transported to surface waters by runoff from treated cropland. The extent of loss is highly variable, depending on the method of application and the timing and intensity of rainfall. Under normal conditions, it is estimated that 0.5 percent of the applied alachlor will be lost (Wauchope, 1978), but as much as 22 percent may be lost in a worst-case scenario with intense rainfall occurring soon after broadcast application (Baker and Laflen, 1979). Although alachlor has been used extensively in Kansas, no reliable data are available on amounts or methods of application. However, alachlor is found frequently in surface waters of eastern Kansas; during the period 19751984, the Kansas Department of Health and Environment found alachlor in 28 of 80 lake samples (Snethen, 1985). In those positive samples, the mean alachlor concentration was 0.86 #g/liter, and the maximum was 3.1 This content downloaded from 207.46.13.128 on Tue, 06 Sep 2016 06:08:35 UTC All use subject to http://about.jstor.org/terms 42 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE gg/liter. Fifty-two stream samples were found to contain alachlor at 0.18 to 4.4 Ag/liter. Since several cities in Kansas draw public water supplies from surface waters fed in part by runoff from cropland, there is little doubt that those water supplies will at times contain traces of alachlor. No safe level of human exposure to alachlor has been determined, so it is of vital concern that finished water supplies be kept free of such contaminants. Little is known of the fate of alachlor in water treatment plants. The purpose of this project is to describe the fate of alachlor during a typical treatment sequence involving chlorination, lime-soda softening, flocculation with alum, activated carbon adsorption, and filtration. With such information, water treatment plant operators can better protect public health by optimizing alachlor removal. EXPERIMENTAL PROCEDURE Water samples were collected on three dates in May and June 1985 from the Neosho River at the intake for the Emporia city water treatment plant. The flow conditions varied from normal to near-flood. Alachlor was added at concentrations less than 0.5 mg/liter, and the samples were allowed to equilibrate at room temperature under gentle aeration for 8-10 days before