Organoclays were assayed as matrices in which to associate herbicides, with the aim of decreasing product losses that could give rise to water contamination from agricultural activities. Fenuron was selected as model of a very mobile and highly water-soluble herbicide. Two different organoclays of high (A-HDT) and low (H-C18) reversible fenuron sorption were selected. Herbicide-organoclay complexes were prepared from the two organoclays and with two different fenuron contents (20 and 40 g AI kg-1) and two different mixing times, so as to form a series of weak and strong complexes. The release of fenuron from those complexes into water and water/soil suspensions gave values of T50 (time to release 50% of the fenuron content) ranging from 0.3 min to 2400 h. The total fenuron released in these closed systems ranged from 48 to 80% of the fenuron in the complex. The organoclay type (high or low sorptivity) had the greatest influence on fenuron release, followed by the strong or weak complex, suggesting that herbicide-organoclay interactions are the main factors controlling release. Soil column leaching experiments showed fenuron-organoclay complexes to be effective in reducing the peak herbicide concentration in the leachate to a half (6 microns) or a quarter (3 microns) of that obtained from the free technical compound (12 microns). Herbicide lost through leaching was reduced from 78% for the free technical fenuron to 50-30%, depending on the organoclay used as carrier and the strength of the complex. Bioassay with ryegrass showed that the weak fenuron/H-C18 complex (40 g AI kg-1) gave the same herbicidal activity as technical fenuron. The potential suitability of low-sorptive organoclays for conferring slow-release properties on the fenuron complex has been demonstrated.