ABSTRACT Seven groups of rats were fed during 1 mo using 1 infant formula containing Bifidobacterium bifidum and Bifidobacterium longum, 3 infant formulas containing 4-galactosyllactose at 1.2%, 5.0%, and 10.0%, and 3 infant formulas containing both ingredients. During 3 periods, corresponding to day 8 to 10, 18 to 20, and 28 to 30, fecal samples were collected for total aerobes, total anaerobes, and bifidobacteria counts. Results showed that bifidobacteria represented an important proportion out of the anaerobe group in the 1st period. However, in the 2nd period bifidobacteria decreased significantly, and in the 3rd period bifidobacteria counts increased, especially in the group fed diet containing galactooligosaccharides 1.2%. In a 2nd study, 12 human infants were fed with the infant formula containing B. bifidum and B. longum, whereas the other 12 were fed using a control infant formula. Fecal samples were collected at the age of 1, 3, 5, 7, 9, and 12 mo for total anaerobes, bacteroides, bifidobacteria, clostridia counts, as well as for fecal pH determination. Infants fed infant formula containing bifidobacteria in samples corresponding to 7th and 9th mo of age had significantly (P< 0.05) higher bifidobacteria counts and a lower fecal pH than those fed control infant formula.