There exists a good deal of indirect evidence, from several locations around the world, that there is a substantial eddy field over continental shelves. These eddies appear to have typical swirl velocities of a few centimeters per second and have horizontal scales of perhaps 5–10 km. These eddies are weak compared to typical, wind-driven, alongshore flows but often seem to dominate middepth cross-shelf flows. The idea that motivates the present contribution is that the alongshore wind stress ultimately energizes these eddies by means of baroclinic instabilities, even in cases where obvious intense fronts do not exist. The proposed sequence is that alongshore winds over a stratified ocean cause upwelling or downwelling, and the resulting horizontal density gradients are strong enough to fuel baroclinic instabilities of the requisite energy levels. This idea is explored here by means of a sequence of idealized primitive equation numerical model studies, each driven by a modest, nearly steady, alongshore wind stress applied for about 5–10 days. Different runs vary wind forcing, stratification, bottom slope, bottom friction, and Coriolis parameter. All runs, both upwelling and downwelling, are found to be baroclinically unstable and to have scales compatible with the underlying hypothesis. The model results, combined with physically based scalings, show that eddy kinetic energy generally increases with bottom slope, stratification, wind impulse (time integral of the wind stress), and inverse Coriolis parameter. The dominant length scale of the eddies is found to increase with increasing eddy kinetic energy and to decrease with Coriolis parameter.