Flame spread over solid materials is commonly described in the literature by a two-dimensional reactive boundary layer solution first formulated by Emmons (1956). In the recent past, experimental measurements associated with material flammability testing (e.g. NASA-STD-6001) are compared with the Emmons solution, as modified by Pagni and Shih (1979), and found in disagreement. In the classical solution, the B-number (Spalding mass transfer number) appears as one of the boundary conditions, and has been traditionally assumed a constant for mathematical simplicity. However, experimental results show that the B-number is not a constant, but varies with time and space. A simple modification to the standard measurement procedure is described, which allows calculation of the B-number from the flame stand-off distance. Measurements of flame standoff distance are performed to determine the B-number as a function of time. The classical combustion problem is revisited to model flame spread over a combustible surface. An experimentally-obtained B-number is used to model flame propagation, showing good agreement with current and previous experimental data obtained from the literature. Implications to microgravity flame spread are discussed.