Healthy young adult volunteers, 778 in number, without HBV markers were randomly distributed into groups and administered different lots of a yeast-derived hepatitis B vaccine (YDV) at different dose levels or a commercial plasma-derived vaccine (PDV), according to a 0, 1, 2, 12-month vaccination schedule. The YDV proved to be safe and well tolerated, even when partly purified lots were given. Reactions were mild and transient, comparable to those observed after PDV. One month after three YDV doses, 0-7% of subjects overall had failed to seroconvert; all those evaluated one month after the booster dose had seroconverted. No significant difference was found between the two vaccine types as far as seroconversion rates were concerned. Geometric mean anti-HBs levels following three vaccine doses were higher in seroconverters of the PDV groups. However, a booster dose of YDV resulted in high anti-HBs levels in all groups varying from 11,474 to 51,404 IU l-1 (purified YDV lot), 4915 to 18,832 IU l-1 (partly purified YDV lots) and 11,008 to 15,805 IU l-1 (PDV lots). Of seroconverters to the purified lots of YDV 93% attained 1000 IU l-1 after the booster dose, thus ensuring protection for a number of years. Dose-response studies provided a basis for the selection of 20 micrograms of highly purified YDV as the standard dose.