From 1982 to 1985 two pressure gauges were deployed across the continental shelf northwest of Spitsbergen to determine the seasonal variations in the barotropic component of the West Spitsbergen Current. The variation in current velocity was calculated after analysis and removal of the tidal signals and correction of the pressure difference across the 47-km section for bottom buoyancy effects. The pressure-derived velocity fluctuates about 10 cm s-1 peak to peak with the maximum northward flow occurring in the winter. The maximum appears to be correlated with the seasonal cycle of wind stress curl in the northern Greenland Sea. In fact, the magnitude of the velocity variation corresponds to the Sverdrup transport due to the fluctuation in wind stress curl at 79¿N, 7.5¿E. The idea that the observed variation might also be due to the spin-up of the basin by wind stress curl is explored with two simple analytical models. Basin spin-up does appear to be a possible cause. However, East Greenland Current velocity data from Muench et al. (1986) indicate that the fluctuation about the mean in the deep currents on the western side of the basin is northward in winter, while basin spin-up would produce a southward fluctuation. This supports the idea that Sverdrup transport dominates basin spin-up, at least at the margins of the basin. The fluctuations in transport and heat flux though the pressure gauge section are estimated to be 2 Sv and 30 to 60 TW peak to peak. These values are major fractions of historical estimates of the mean transport and heat flux values in the West Spitsbergen Current, and they are significant fractions of the total oceanic fluxes into the Arctic Ocean. This suggests that long time series are required to make meaningful estimates of the heat budget of the Arctic. ¿ American Geophysical Union 1991