Abstract.A model of water flow from the soil into the plant, and from the plant to the atmosphere is described. There are three state variables in the model: the soil, root and shoot water contents. The flow rate of water from the soil to the root is calculated by dividing the gradient in water potential by a resistance, comprising the resistance from the bulk soil to the root surface, and that from the root surface to the root interior. The resistance in the soil depends on the soil hydraulic conductivity, which in turn depends on the soil water potential. The flow rate from the root to the shoot is given by the gradient in water potential divided by a resistance, which depends on the structural dry mass of the plant. Transpiration is described by the Penman‐Monteith equation. The plant water characteristics can be modified to take account of osmotic and cell wall rigidity parameters. The model incorporates the concept of shoot/root ‘messages’ of water stress, which influence stomatal conductance. The message works through the generation of a hormone as the pressure potential in the shoot (mesophyll) or root falls. This hormone induces a shift of osmoticum from the guard cells to the surrounding mesophyll cells, which causes an increase (i.e. closer to zero) in the osmotic potential in these cells. This, in turn, causes a decrease in their pressure potential, and so reduces stomatal conductance. The model is used as a framework to address some of the issues that have recently been raised concerning the role of water potential in describing water flow through plants. We conclude that, with the hormone present, there is unlikely to be a unique relationship between stomatal conductance and shoot total water potential, since stomatal conductance depends on the pressure potential in the guard cells, which may differ from that in other cells. Nevertheless, this does not imply that water potential is not an important, and indeed fundamental, component for describing water flow through plants. Other aspects of water flow through plants are also considered, such as diurnal patterns of shoot, root and soil water potential components. It is seen that these may differ from the commonly held view that, as the soil dries down, they all attain the same values during the dark period, and which, as we show, is largely unsubstantiated either theoretically or experimen