In the first of two passive transfer experiments, three groups of four macaques were injected intraperitoneally with a normal serum pool, an immune serum pool (pool 1) collected 132-172 weeks postinfection with the 11/88 pool of SIVmac251, or with a pool of four neutralizing monoclonal antibodies (KK9, 17, 54, and 56) raised against gp120 of the 11/88 pool. Sera were given at a dose of 13 ml/kg whereas the MAb pool was given at 30 ml/kg. In a second experiment, a further four macaques were injected with an immune serum pool (pool 2) collected 12 weeks postinfection with simian-grown SIVmac251 at a dose of 19 ml/kg. Animals in both experiments were challenged with SIVmac251 grown in simian peripheral blood lymphocytes. Despite high levels of circulating antibodies in the serum of animals that received either the immune serum pools or the MAbs, all macaques became infected following challenge. The results described are in contrast to a previous report in which passive transfer of sera from animals infected with SIVsm successfully protected against challenge with the homologous virus grown in human PBMCs. Challenge with SIVmac251 grown in simian PBMCs may be the reason for these conflicting results. Nevertheless, the results suggest that in this model the presence of circulating neutralizing antibodies alone does not necessarily confer protection against challenge with SIVmac251 grown in simian cells.