Studies using a rainfall simulator were carried out to test the stochastic hypothesis embedded in the original formulation of the stochastic interception model and to determine the vegetation-dependent stochastic model parameters. These studies showed that for all the species investigated: (a) the rate of wetting was greatest for small volume drops--this is in agreement with the stochastic wetting hypothesis; (b) the maximum canopy storage that was achieved is not, as is conventionally assumed, constant, but is dependent on drop size. The reduction in maximum canopy storage with increasing drop size is attributed to a dependence on drop kinetic energy.