Our goal was to investigate whether post-stenotic dilatation (PSD) enhances collateral blood flow. In vitro experiments and computer modeling analysis were used to study the flow through stenotic segments and through collateral channels in the presence and absence of PSD. Pulsatile blood flow was provided by a left heart simulator primed with glycerol or normal saline. Pressure gradients across the stenosis were measured at different "cardiac" outputs. Computer models were constructed to simulate the experiments. Flow patterns and pressure drop across the stenosis were determined for a steady flow of 3 L/min. We observed that PSD was associated with a larger pressure drop across the stenosis than the absence of PSD when the flow was occurring through the stenosis only. There was, however, no difference in the pressure drop between the two geometries when the flow was occurring through both the stenotic orifice and the collateral channels when saline solution was used, but a small pressure difference prevailed for glycerol. At all different geometries there was considerable turbulence at PSD, and PSD geometry was found to be either at a disadvantage or at no advantage when compared to the tapered geometry for the total flow past the stenosis. The PSD geometry, however, enhanced the flow through the collateral while the flow through the orifice decreased concomitantly, resulting in no net increase in the total flow. This was true for any proportion of the total flow going through the collateral channels. For the total flow past the stenosis, PSD does not offer a benefit over tapered geometry.