Rain splash soil erosion in the presence of rock fragments and different initial conditions was tested in laboratory flume experiments under controlled conditions. The aim of the experiments was to ascertain whether cumulative soil erosion is proportional to the area of soil exposed to raindrop detachment under the condition of constant and uniform precipitation. The surface area exposed to rain splash erosion was adjusted by placing rock fragments onto the surface of identically prepared soil in laboratory flumes. The laboratory results showed that the eroded cumulative mass depended on the cumulative runoff, and that soil erosion was proportional to the soil surface area exposed to raindrops, in situations where an initially dry, plowed and smoothed soil surface was ensured. The results showed that this relationship was controlled to a smaller extent by the soil's initial moisture content, bulk density and soil surface characteristics. When the initial conditions were more complex, soil erosion was proportional to the area exposed only at steady state. Then, sediment concentrations during the first part of the erosion event were instead more sensitive to the initial state of the soil surface, whereas at steady state it was observed that the concentrations of eroded sediments were controlled mainly by the effective rainfall and area exposed to raindrops. Previously published field data on rain splash soil erosion were analyzed to ascertain whether the same behavior was evident under field conditions. For this case it was found that rain splash erosion is in general not proportional to the area exposed. In contrast to the controlled laboratory experiments, the field experiments were characterized by non-uniform initial surface roughness, surface soil aging and heterogeneous rock fragment size and spatial distribution. However, the presented laboratory results showed clearly that, for soils with negligible surface roughness, erosion depends on (i) the area of soil exposed to rainfall and (ii) the cumulative runoff, and that it is only slightly dependent on other soil variables.