Previous studies relating unit discharge to elevation indicated that large floods in the Rocky Mountains may be limited by elevation. However, high-elevation data are sparse in Montana and the indications may not be entirely correct. Based on data at 19 sites in Montana, a strong log-linear relation exists between large-flood discharge and drainage area. The use of unit discharge (peak discharge divided by drainage area) to compare flood magnitude from site to site may thus be biased and tend to overstate flood magnitude for small basins. Removal of the bias by use of a revised unit discharge (peak discharge divided by drainage area raised to the 0.16 power) results in no apparent relation between revised unit discharge and elevation in two areas of Montana. However, because of a paucity of data, the magnitude of revised unit discharge at elevations greater than 1,650 m is largely unknown. Additional data and research are needed to resolve questions about mountain flood hydrology.