Transpiration from a hawthorn (Crataegus monogyna L.) dominated hedgerow in southern England was measured continuously over two growing seasons by the sap flow technique. Accompanying measurements of structural parameters, microclimate and leaf stomatal and boundary layer conductances were used to establish the driving factors of hedgerow transpiration. Observed transpiration rates, reaching peak values of around 8 mm day(-1) and a seasonal mean of about 3.5 mm day(-1), were higher than those reported for most other temperate deciduous woodlands, except short-rotation coppice and wet woodlands. The high rates were caused by the structural and physiological characteristics of hawthorn leaves, which exhibited much higher stomatal and boundary-layer conductances than those of the second-most abundant woody species in the hedgerow, field maple (Acer campestre L.). Only in the hot summer of 2003 did stomatal conductance, and thus transpiration, decrease substantially. The hedgerow canopy was always closely coupled to the atmosphere. Hedgerow transpiration equaled potential evaporation (calculated by the Priestley-Taylor formula) in 2003 and exceeded it in 2004, which meant that a substantial fraction of the energy (21% in 2003 and more than 37% in 2004) came from advection. Hedgerow canopy conductance (g(c)), as inferred from the sap flow data by inverting the Penman-Monteith equation, responded to solar radiation (R(G)) and vapor pressure deficit (D). Although the response to R(G) showed no systematic temporal variation, the response to D, described as g(c)(D) = g(cref) - mln(D), changed seasonally. The reference g(c) depended on leaf area index and the ratio of -m/g(cref) on long-term mean daytime D. A model is proposed based on these observations that predicts canopy conductance for the hawthorn hedge from standard weather data.