Abstract : The classical Prandtl eddy viscosity model, which was formulated for an incompressible jet into still air, is modified to account for a density difference between the jet and a non-zero velocity coflowing stream. The constant in the potential core of the Prandtl model is freed and empirically determined as a function of the local mass flux ratio of the secondary to primary flows. The resulting predictions are compared to available axisymmetric jet data. The model is expected to be of use in the computation of thrust augmenting ejector problems where the primary jets are heated by the use of a combination of engine and bypass air. (Author)