Abstract
An experiment has been designed to test the predictions of nongeostrophic baroclinic stability theory. The apparatus is similar to the conventional rotating annulus experiments, except that the vertical temperature difference can be controlled as well as the horizontal temperature difference. Therefore, the Richardson number can be decreased by heating the bottom of the annulus relative to the top. The first qualitative observations derived from the experiment are described and are found to agree well with the theory. With no vertical temperature difference applied, the motion consists of a conventional baroclinic instability superimposed on the basic thermal wind. As the fluid is destabilized symmetric instabilities first appear superimposed on the baroclinic instability. As further destabilization occurs the symmetric instabilities completely replace the baroclinic instability, and are themselves subsequently replaced by small-scale, nonsymmetric instabilities.
