This paper reports the results of a preliminary investigation into the existence of a Representative Elementary Area (REA) in the context of hydrologic modeling at the catchment scale. The investigation was carried out for an actual catchment topography as represented by Coweeta River experimental basin with synthetic realizations for rainfall and soils. The hydrologic response of this catchment was modeled by a modified version of topmodel^{{black star}} {black star} Beven and Kirkby (1979) which is capable of modeling both infiltration excess and saturation excess runoff and incorporating the spatial variability of soils, topography, and rainfall. The effect of scale was analyzed by first dividing the catchment into smaller subcatchments and determining the average water fluxes for each subcatchment. The preliminary results lead to the following conclusions: (1) a Representative Elementary Area (REA) exists in the context of catchment hydrologic responses; (2) the REA is strongly influenced by the topography; and (3) based on our initial results, the length scale of rainfall seems to have only a secondary role in determining the size of the REA; however, increases in the variability of rainfall and soils between subcatchments increase the variability of runoff generation between subcatchments.