A double-blind placebo controlled investigation was carried out to study the effect of peroral colonization. Human volunteers were given mixtures of bifidobacteria and lactic acid bacteria. Measurements were made over a 1 week treatment period and for another week after the end of the treatment. Two different bacteriological preparations were used, one consisted of Enterococcus faecium and Bifidobacterium longum (a total of 6.4 x 10(8) cfu d-1); the other consisted of Lactobacillus acidophilus, Bif. bifidum, Lact. delbrueckii ssp. bulgaricus, and Streptococcus thermophilus (a total of 9 x 10(9) cfu d-1). Together with a placebo preparation, they were given to 24 healthy controls (eight in each group). Microbiological examinations of jejunal aspirates showed that viable counts of most species were below the detection limit. However, the test preparation containing Ent. faecium and Bif. longum significantly reduced the anaerobe: aerobe ratio in faeces by a factor of three during treatment (P = 0.03), and increased it by a factor of 30 during the following week (P < 0.02). This study shows that peroral administration of certain bacterial cultures may affect the distal intestinal microflora.