A simple extraction procedure has been developed to assess rapidly the probable extent of the transfer of radiocaesium into ruminant food products soon after a nuclear accident. The in vitro extractions were validated against true absorption measurements of different forms of radiocaesium in the sheep gut. Extractions were performed on a range of different radiocaesium sources. Some of these sources were artificial (ionic radiocaesium absorbed onto bentonite, silica spheres and filter-papers) and others were environmentally contaminated [silt from the Ravenglass Estuary contaminated by effluent from British Nuclear Fuels Limited (BNNFL) Sellafield, and upland grass and heather contaminated by Chernobyl fallout]. Laboratory experiments concentrated primarily on the use of simple inorganic extractants in competitive ion-exchange processes. Of the reagents used, 0.1 mol dm^–3 stable caesium chloride solution was the most effective extractant. The proportion of radiocaesium extracted by 0.1 mol dm^–3 ceasium chloride correlated well with measurements of true absorption. Extracting radiocaesium using 0.1 mol dm^–3 caesium chloride proved to be an inexpensive and rapid method of predicting the availability of radiocaesium for absorption in the ruminant gut, giving results within 24 h. Further extractions were carried out using cellulase/pepsin simulated digestions and ovine rumen fluid. Results suggested that the availability of radiocaesium from some inorganic sources may be underestimated using such techniques.