In summary, HIV vaccine studies described have generally not been designed to measure the effect of the adjuvant or to make comparisons between adjuvants. In only one study was a head-to-head comparison made between HIV antigen alone and antigen formulated with different adjuvants. We hope that future experiments with HIV/SIV vaccine candidates will be designed to determine the relative potency and safety of different adjuvants. Unfortunately, such experiments tend to be tedious and expensive. The design of these studies will need to address a number of variables which influence the response to the vaccine, including route and schedule of immunization, genotype and species of the vaccinated subject, and intrinsic characteristics of the antigen. In addition, the immunologic endpoints should include measurement of both B and T cell function. The carrier/adjuvant/antigen formulation should be hand-tailored and then standardized so that it is manufactured reproducibly without producing different biological effects between lots, and the vaccine formulation should be stable on storage and shipping. Finally, we obviously need to identify and test the protective antigen or antigens. The best adjuvant will never correct the choice of the wrong epitope.