The central bleed jet in an axisymmetric, compressible base flow with a Mach 2.46 free stream was visualized using an acetone planar laser-induced fluorescence (PLIF) technique. Three bleed flow rates, a low-bleed case (I=0.0038) and two others that bracket the optimal bleed case (I=0.0113 and I=0.0226), were examined. The injection parameter, I, is defined here as the bleed mass flow rate normalized by the product of the base area and freestream mass flux. This study shows that the bleed jet in the low-bleed case is almost instantly deflected outward toward the outer shear layer after it exits from the bleed orifice. When I=0.0113, the bleed fluid carries enough momentum that the bleed jet remains coherent and generally aligned along the symmetry axis for approximately one base radius before it is deflected outward by the primary recirculation region. For the highest bleed rate examined (I=0.0226), the bleed jet remains coherent and aligned along the symmetry axis for approximately 2.5 base radii downstream, where the impingement of freestream fluid causes the bleed jet to symmetrically eject mass downstream. Because the behavior of the bleed jet is much different for pre- and post-optimal bleed rates, some general conclusions can be made about the effect of bleed rate on base drag. For the two lower-bleed cases, there is a base pressure increase because the bleed fluid partially isolates the primary recirculation region from the outer flow, and provides part of the mass necessary for entrainment in the outer shear layer. The bleed jet in the high-bleed case is not as effective at increasing the base pressure, because the jet is shielded from the outer shear layer by the secondary recirculation region, which exists in the annulus between the bleed orifice and the base corner.