We have compared the total microbial biomass and the fungal/bacterial ratio estimated using substrate-induced respiration (SIR) in combination with the selective inhibition technique and using the phospholipid fatty acid (PLFA) technique in a pH gradient (3.0-7.2) consisting of 53 mature broad-leaved forest soils. A fungal/bacterial biomass index using the PLFA technique was calculated using the PLFA 18:2w6,9 as an indicator of fungal biomass and the sum of 13 bacterial specific PLFAs as indicator of the bacterial biomass. Good linear correlation (p < 0.001) was found between the total microbial biomass estimated with SIR and total PLFAs (totPLFA), indicating that 1 mg biomass-C was equivalent to 130 nmol totPLFA. Both biomass estimates were positively correlated to soil pH. The fungal/bacterial ratio measured using the selective inhibition technique decreased significantly with increasing pH from about 9 at pH 3 to approximately 2 at pH 7, while the fungal/bacterial biomass index using PLFA measurements tended to increase slightly with increasing soil pH. Good correlation between the soil content of ergosterol and of the PLFA 18:2w6,9 indicated that the lack of congruency between the two methods in estimating fungal/bacterial ratios was not due to PLFA 18:2w6,9-related non-fungal structures to any significant degree. Several PLFAs were strongly correlated to soil pH (R-2 values > 0.8); for example the PLFAs 16:1w5 and 16:1w7c increased with increasing soil pH, while i16:0 and cy19:0 decreased. A principal component analysis of the total PLFA pattern gave a first component that was strongly correlated to soil pH (R-2 = 0.85, p < 0.001) indicating that the microbial community composition in these beech/beech-oak forest soils was to a large extent determined by soil pH.