In a randomized, double-blind, placebo-controlled study in January and February 1992, the safety and immunogenicity of two doses of a new oral cholera vaccine was assessed. The vaccine was prepared from the recombinant B subunit of the toxin and from killed whole cells (rBS/WC) in 1165 individuals between the ages of 12 months and 64 years in Barranquilla, Colombia. Participants received two doses of either the vaccine or a placebo (killed Escherichia coli K12) over a 2-week interval. Few symptoms were detected during the 3 days following administration of the initial dose and even fewer following the second one. Sera obtained upon administration of the first dose and 2 weeks after administration of the second dose were tested for Vibrio cholera 01 Inaba vibriocidal antibodies and antitoxins. Geometric mean titers (GMTs) of vibriocidal antibodies were found to increase two-fold in subjects receiving the vaccine. In the paired samples taken from vaccinated subjects, two-fold or greater increases were observed in 44% and four-fold or greater increases were observed in 34%. In comparison, similar increases were found only in 9.2% and 2.2% of the sera taken from those receiving placebo (p .05). The GMTs of IgG and IgA antitoxins, as determined by ELISA, increased by factors of 4 and 3.2, respectively, in those receiving the vaccine as compared with factors of 1.1 and 1.1, respectively, in those given the placebo (p .001 for IgG and p .01 for IgA). Approximately 80% of the paired samples from the vaccinated group showed an increase of both IgG and IgA antitoxins or= 1.5 as compared with only about 20% of those in the placebo group (p .000001). Belonging to the O blood group did not significantly affect the immune response. Children under the age of 4 years tended to show a weaker vibriocidal antibody response and stronger antitoxin response than did older subjects. The two doses of oral vaccine were found to be safe and without attributable side effects.