Ten water-soluble ions (F–, Cl–, NO3–, PO43–, SO42–, Na+, NH4+, K+, Ca2+, Mg2+), distributed among four fractions of particulate matter, PM, were investigated in an urban background site. The PM was sampled twice a week during a two-year sampling period. Mass distribution among the PM fractions and ambient concentrations of the ten PM-bound ions in the heating and non-heating periods, the seasonal effects in the PM fraction-bound ion concentrations (generalized regression model), and the principal components of all the resulting sets of ambient concentrations (principal component analysis) were determined, discussed, and interpreted in the terms of source apportionment of PM emissions. The formation of secondary inorganic aerosol in transformations of gaseous precursors (e.g., SOx, NOx, NH3) appeared to be most probable and significant source of PM2.5, especially of its sub-fraction PM1–2.5, in the non-heating period. In the heating period, PM and PM2.5 bound water-soluble ions originated mainly from combustion of coal and other solid fuels, or waste. Coarse particles (PM2.5–10 and PM10–40) and some PM2.5–40-bound ions (e.g. Na+, K+, Mg2+) may come from re-suspension of mineral matter and road dust. In some part, coarse PM may consist of mineral and salt particles containing gaseous and semi-volatile compounds.