It is shown that the extent of miscibility of blends of polycarbonate (PC) and poly(methyl methacrylate) (PMMA) described in recent papers has been over estimated. The solvent preparation methods used cause the two polymers to become kinetically trapped in a homogeneous, but non-equilibrium, mixture below the glass transition. The phase separation that occurs on heating above the glass transition is not a result of a lower critical solution temperature (LCST) just above T(g) but stems from an extremely slow rate of phase separation from the trapped non-equilibrium state. Based on this work, we believe that none of the LCST boundary may lie above the T(g) line when the components have molecular weights in the commercially useful range. There is some uncertainty at the extremes of the composition scale because of the excessively long times required to reach equilibrium. The effects of polycarbonate molecular weight in this situation are considered. Even though the PC-PMMA interaction is not as favourable as originally thought, it is clear that this interaction is only weakly unfavourable for mixing.