Abstract Existing literature lends credence to the idea that channel sinuosity and channel gradient are independent morphologic elements responding to characteristics of the water and sediment regime of a stream. At the same time, sinuosity and gradient are viewed as closely related variables, since sinuosity is often defined as a ratio of valley gradient to channel gradient over a stream reach. These conceptions of the sinuosity variable are essentially contradictory if viewed from the standpoint of process, since in the latter case, water and sediment characteristics have no direct bearing on the determination of sinuosity, but only the potential of indirect influence operating through the channel gradient. This paper presents an empirical analysis aimed at evaluating the validity of these two conceptions of the sinuosity variable. The results of the analysis strongly support the gradient ratio concept and suggest that the traditional sinuosity variable is not reasonably viewed as an independent element of stream channel morphology, since it is essentially determined by the same process as that which determines the channel gradient. Thus, channel sinuosity is insignificantly influenced by the direct effects of water and sediment discharge. Certain indirect effects of these factors do exist, however, given the role of channel gradient in the determination of sinuosity. A summary of these results is given in the form of a two-equation recursive model.