Experiments have been made on quasi two-dimensional sails of small camber and at small incidence. Four excess-length ratios have been tested at a Reynolds number of 1.2 x 105. The results for lift, tension, centre of lift, maximum camber and its position, and leading- and trailing-edge membrane angles have been compared with existing inviscid theories and show poor agreement in general. This is attributed to leading- and trailing-edge flow separations as indicated by supplementary flow-visualization experiments. The optimum incidences in particular are much greater than the theoretical value of 0°. Luffing occurs at slightly negative incidences and appears to be a dynamic instability. The highest lift-to-drag ratio obtained was 16.5 on a membrane with an excess-length ratio of 0.03.

In a recent paper Hooper & Boyd (1983) have shown that the unbounded stratified Couette flow of two viscous fluids of equal density and with no interfacial tension is always unstable. They found that the instability arises at the interface between the two fluids and occurs at short wavelengths where viscosity is more important than inertia. The purpose of this note is to provide a simple physical explanation of the mechanics of the instability.