“…From the limited data for coniferous vegetation approximately 6.4% (Thiry et al, 1990cited in Myttenaere et al, 1993 to 10.1% (Yamagata et al, 1969) of the total 137Cs inventory is stored in the above-ground biomass, whereas the remainder is in the litter and mineral soil. This can be compared with approximately 3.3% for pre-Chernobyl 137Cs stored in (upland) grasses in Cumbria, UK (Coughtrey et al, 1989). Additionally, wet and dry radionuclide deposition in forested areas is spatially more heterogeneous than grassed areas for the following reasons: (i) stand composition may be a complex mix of various species with different aerodynamic surface areas (Wiman et al, 1990), surface to volume ratios (Whicker, 1983) and thus different interception factors for wet and dry fallout (Winteringham, 1989) -this variability is increased if there is a mix of coniferous and deciduous vegetation; (ii) soil to plant 137Cs transfer coefficients are species-specific and this, combined with variations in leaf-fall, increases the spatial variation in the 137Cs inventory at a given site; (iii) stemflow generally causes the greatest 137Cs activity to be concentrated around tree trunks, with a radial decrease towards the outer canopy (Franklin et af., 1967;Adriano et al, 1981;Bunzl et al, 1989;Forster and Schimmack, 1992), and stemflow 137Cs varies with bark type and canopy architecture (Franklin et al, 1967;Gersper, 1970); and (iv) common sampling designs are extremely difficult to implement in forested areas, and complications arise when attempting to define a representative sample site (i.e.…”