Sap flow measurements based on the heat balance method offers the opportunity to evaluate directly and quite easily the mass flow rate of water in plants. However, extrapolation of measurements of water use by individual stems to that for a canopy is tricky. In the present study, 14 sugarcane stems, out of a canopy of nearly 200 000, were equipped with Dynamax sap flow gauge. We extrapolated these individual measurements to determine the transpiration of the canopy and compare this transpiration to the crop evapotranspiration calculated on the basis of the Penman-Monteith method. The method used for the extrapolation assumes that the transpiration of a sugarcane plant is proportional to its leaf area. Transpiration of the canopy determined by this method was overestimated by more 35% as compared to the reference evapotranspiration results. Different sources of possible errors were examined and lead to suppose that it is very difficult to determine the transpiration of a heterogeneous canopy in growth by using the sap flow measurement technique.