AbstractIn the Laboratory of Experimental Geomorphology experiments were carried out on loess from the Pleniglacial A of the Weichselian. Most of the experiments had as their objective determination of the effect of slope angle, rainfall intensity, raindrop impact, added overland flow and concentrations of suspended material, on the erosion and redeposition of loess.At various stages during the experiments, as well as at the beginning and end, undisturbed samples of loess were collected for micromorphological investigation.—Pluvial runoff (overland flow with raindrop impact) supplied with or without terminal runoff, gave rise to laminated deposits, tightly packed and moderately differentiated in size and mineral species.—After flow (the flow that occurs in the field during a short period after rainfall has ceased) and meltwater flow were simulated by terminal runoff. If the discharge and suspended load were moderate (respectively 0.81/min/dm and 20,000 mg/l), the sediments produced were very well laminated, well sorted and minerally differentiated, and loosely packed. In an exceptional case, with a very high discharge (81/min/dm) and suspended solid load (150,000 mg/l), the sediments were virtually unlaminated, and consisted of very loosely packed aggregates, which must have been formed during transport.—Splash (material transported exclusively under raindrop impact) was applied successively to: calcium carbonate rich loess, decalcified loess and Ap‐material from a loess soil profile. In these three sediments no lamination, sorting and practically no mineral differentiation were produced that could be observed in the thin sections. Comparing the laboratory results with the situation in two loess quarries, one near Leuven and one near Tongrinne (both in Belgium), the micromorphological interpretation of the sediments in the field ranged from the very simple to the very complex.—The small quarry near Leuven contained a cross‐section through a gully filled with laminated calcareous loess. This sediment could be interpreted as having probably been formed by pluvial runoff alternating with after flow.—The large quarry near Tongrinne was more complicated. The several deposits here could have been formed under conditions ranging from pluvial runoff alternating with after flow, to splash and/or meltwater flow with high wash‐load concentrations.The micromorphological research suggested that the loess was not only reworked by meltwater but also by pluvial runoff; the latter being probably the most important process, at least during Early We