Commercialization of UV-C treatment of horticultural produce in order to induce beneficial responses in the produce following treatment requires both accurate dose delivery and a method of treating large quantities of produce efficiently. Furthermore, it has long been assumed that such effects require the entire surface of the horticultural commodities - typically fruit - to be exposed to UV-C. This has invariably been achieved by manually rotating the fruit in a UV-C field whilst reducing the dose delivered at each rotation in direct proportion to the number of rotations. However, the resulting UV-C dose distributions achieved under these circumstances are generally not reported in the literature. In the work described here a polystyrene sphere (dia. 70 mm) was used to simulate fruits such as tomatoes, apples, peaches, etc., that have an approximately spherical form in order to provide a means of measuring the total doses of UV-C accumulated during treatment and comparing such estimates to theoretically-derived ones. This was achieved using dosimetry based on spores of Bacillus subtilis in which spore-impregnated membranes were attached to the surface of the sphere. The fraction of spores surviving exposure was used to estimate dose from a dose-response curve for the spores. Under irradiation conditions leading to a theoretically calculated dose of 10.6J, spore dosimetry yielded estimates of 9.1, 10.7 and 6.1 J for UV-C delivered in, respectively, one, two or four exposures. In the case of exposure of the sphere during continuous mechanical rotation for the same length of time (80 s) a value of only 3.5J was obtained. Irradiation conditions resulting in the spores being subject to intermittent exposure to UV-C led to dose estimates below the theoretically derived ones. The circumstances under which spore dosimetry can be used to obtain surface dose distributions are discussed. (C) 2010 Elsevier Ltd. All rights reserved.