In recent years research into the formation of nitrogenous disinfection by-products (N-DBPs) in drinking water - including N-nitrosodimethylamine (NDMA), the haloacetonitriles (HANs), haloacetamides (HAcAms), cyanogen halides (CNX) and halonitromethanes (HNMs) - has proliferated. This is partly due to their high reported toxicity of N-DBPs. In this review paper information about the formation yields of N-DBPs from model precursors, and about environmental precursor occurrence, has been employed to assess the amount of N-DBP formation that is attributable to known precursors. It was calculated that for HANs and HAcAms, the concentrations of known precursors - mainly free amino acids are insufficient to account for the observed concentrations of these N-DBP groups. However, at least in some waters, a significant proportion of CNX and NDMA formation can be explained by known precursors. Identified N-DBP precursors tend to be of low molecular weight and low electrostatic charge relative to bulk natural organic matter (NOM). This makes them recalcitrant to removal by water treatment processes, notably coagulation, as confirmed by a number of bench-scale studies. However, amino acids have been found to be easier to remove during water treatment than would be suggested by the known molecular properties of the individual free amino acids.