Mesoscale internal combustion engines for a variety of new combustion system applications have dimensions that are far smaller than conventional macroscale engines, yet unlike true microscale engines allow significant mean flow and turbulence to be created in the combustion chamber by the injection process. The resulting flow allows minimization of the combustion time by augmenting flame propagation across the combustion chamber to provide maximum power and efficiency. However injection and ignition locations that give minimum combustion times in traditional macroscale engines are of limited relevance in such small-scale engines. This study has therefore examined premixed flame propagation in a generic mesoscale combustion chamber for various injection and ignition configurations to develop guidelines for minimizing combustion times in mesoscale engines. Numerical simulations based on the G-equation, coupled with mean flow and turbulence Fields for various injection configurations, reveal key features of flame propagation and combustion times for combinations of injection and ignition sites. Mean flow velocities and turbulent flame speeds are found to be comparable in such mesoscale engines, indicating that both the mean flow and turbulence must be matched to the combustion chamber geometry to minimize the overall combustion time τB . Results show that τB in such mesoscale engines can be 8-20 times faster than the laminar combustion time scale Lmax/SL0 that governs true microscale engines, and can be 1-2 times faster than the turbulent combustion time scale Lmax / SL0. Minimum τBes are found to result from injection across the smallest chamber dimension, with ignition on the opposite side of the chamber providing rapid initial flame growth and a secondary flow that assists in flame propagation. Moving-boundary simulations permit analysis of flame propagation in realistic engine configurations, and assessments of resulting combustion times.