A formalin-inactivated Rift Valley fever vaccine, originally produced in primary monkey kidney cells, has been used to protect laboratory workers. A trial of a modified vaccine, newly formulated in well-characterized diploid fetal rhesus lung cells, was conducted with 114 men aged 19--24 years. Of the 107 subjects who received up to three injections of 0.1 to 1 ml vaccine (an additional seven received a placebo) one had a local hypersensitivity-type reaction and another a generalized urticarial syndrome. Both cases had a prior history of hypersensitivity states. No pyrogenicity was detected and only insignificant systemic reactions were recorded. Mild and transient local reactions ranged from 5% at the lowest dose level to 43% at the highest. Serologic response, as assessed by plaque reduction neutralizing antibody titers, was dose dependent. Within a single vaccine lot tested at multiple dose levels, peak (day 42) geometric mean titers ranged from 48 (at 0.1 ml x 3) to 436 (at 1.0 ml x 3). Reciprocal titers of greater than or equal to 40 are considered to be protective. Comparison of three lots at the 0.5 ml level indicated between lot variability, though this was not statistically significant. A sharp decline in antibody titers was observed in all vaccination groups by day 84; six months after vaccination apparently protective antibody titers were present only in groups that received 1 ml x 3 and 0.5 ml x 3 of the most antigenic lot of vaccine. These results suggest that 1) the vaccine is generally nonreactogenic, but individuals with a prior history of hypersensitivity states should be observed for allergic side effects; 2) existing vaccine supplies cannot be extended by using lower dose levels without a lower and less sustained serologic response; 3) a booster dose is necessary six months or more following the primary series; 4) although the current TSI-GSD-200 vaccine is immunogenic, a more potent vaccine is needed.