Alachlor [2-chloro-2′,6′-diethyl-N-(methoxymethyl)-acetanilide] mixed with plaster-of-paris in concentrations of 0.5, 1, 2, and 4% (w/w) was cast into tablets of three sizes. The controlled release of the herbicide was tested by the following methods: (a) in the laboratory with eduction by soaking tablets in water, (b) in the greenhouse with azalea (Rhododendron obtuseum Planch. ‘Richardii’) stock in container, and (c) in the lathe house with azalea stock in containers. In the laboratory experiment, the quantity of herbicide educed from tablets of equivalent volumes, weights, and concentrations were similar regardless of surface area of the three tablet sizes. Successively decreasing quantitites of alachlor were educed with each of the first 10 soakings. Following these soakings, the quantity of alachlor educed per soaking stabilized for the next 30 soakings. In both greenhouse and lathe house experiments, azalea plants treated with as much as 45 kg/ha alachlor formulated in tablet(s) were not damaged. In the best greenhouse treatments, two small tablets per container, effective weed control was obtained for up to 7.5 months from a single application of alachlor at rates of 11, 22, and 45 kg/ha.