Sieve deposits were once considered to be one of the building blocks in alluvial fan stratigraphy. Later reinterpretation of sieve lobes as debris-flow deposits, favored because no visual records of active sieve deposition had been reported, undermined their significance, divided opinions, and left this issue unresolved. Here I document active deposition of sieve lobes in natural settings, in support of the original model. Sieve deposition can easily occur in natural settings such as proglacial outwash fans, small arid alluvial fans, or perennial streams when there is a scarcity of fine material, significant bedload, high slope, permeable ground, and discharges moderate enough to allow infiltration. The only hydrodynamic requirement for sieve deposition is a high rate of water loss promoted by permeable bed sediments. Under some circumstances alluvial fans can be built almost entirely of sieve deposits, as shown here. One effect of the rapid extraction of water is the creation of sigmoidal fan profiles. A gradation from sieve deposition to sheetflood occurs if sediment becomes progressively less permeable or if water flow increases, overcoming bed permeability. Sieve deposition is a universal depositional process based simply on infiltration, and it explains matrix-poor clast-supported gravels, while alternative hypotheses, such as matrix winnowing of debris flows used to dispute the sieve model, still need to be proven by observations in nature.