Adding prebiotics or probiotics to infant formula to improve the intestinal flora of formula-fed infants is considered to be a major innovation. Several companies have brought relevant formulations onto the market. However, comparative data on the effects of pre- and probiotics on the intestinal microflora of infants are not available. The present study aimed to compare the effects of infant formula containing a mixture of galacto- and fructo-oligosaccharides or viable Bifidobacterium animalis on the composition and metabolic activity of the intestinal microflora. Before birth, infants were randomised and double blindly allocated to one of three formulas. The prebiotic (GOS/FOS) group (n 19) received regular infant formula supplemented with a mixture of galacto-oligosaccharides and fructo-oligosaccharides (6g/l). The probiotic (Bb-12) group (n 19) received the same formula supplemented with 6·0×1010 viable cells of B. animalis per litre. The standard group (n 19) received non-supplemented regular formula. A group of sixty-three breast-fed infants was included as a reference group. Faecal samples were taken at postnatal day 5 and 10, and week 4, 8, 12 and 16. Compared with the groups fed Bb-12 and standard formula, the GOS/FOS formula group showed higher faecal acetate ratio (69·7% (sem 2·7), 69·9% (sem 3·9) and 82·2% (sem 5·3); P<0·05) and lactate concentration (11·3 (sem 7·9), 3·1 (sem 2·3) and 34·7 (sem 10·7) mmol/kg faeces) and lower pH (6·6 (sem 0·2), 7·1 (sem 0·2) and 5·6 (sem 0·2); P<0·05) at 16 weeks. Differences in percentage of bifidobacteria between the GOS/FOS (59·2% (sem 7·7)), Bb-12 (52·7% (sem 8·0)) and the standard (51·8% (sem 6·4)) groups were not statistically significant at 16 weeks. Feeding infants GOS/FOS formula resulted in a similar effect on metabolic activity of the flora as in breast-fed infants. In the Bb-12 group, composition and metabolic activity of the flora were more similar to those of the standard group.