A previously described model of solar radiation interception was applied to a spatially discontinuous canopy: that of a traditional vineyard in which the classical terms of the radiative balance and the spatial distribution of the radiation transmitted to the soil were measured. Comparison of measured and simulated data gave satisfactory agreement for reflected radiation (fig 4), but major discrepancies appeared for mean transmitted radiation (fig 5). The use of small stationary sensors for measuring the transmitted radiation explains the latter observation, since most of the time they measured radiation received on the ground in the sunflecks or in the shaded area rather than mean radiation. This was verified by comparing the measured and simulated spatial distribution of transmitted radiation (figs 7, 8). Finally, the influence of the woody parts which were not taken into consideration in the model was clearly identified : it significantly reduced the transmission of incident radiation (fig 9), and to a greater degres the closer the sensor was to the vegetation row.