Adjuvants have traditionally been appreciated for their immunoenhancing effects, whereas their impact on immunological memory has largely been neglected. In this paper, we have compared three mechanistically distinct adjuvants: aluminum salts (Alum), Ribi (monophosphoryl lipid A), and the cholera toxin A1 fusion protein CTA1-DD. Their influence on long-term memory development was dramatically different. Whereas a single immunization i.p. with 4-hydroxy-3-nitrophenyl acetyl (NP)-chicken γ-globulin and adjuvant stimulated serum anti-NP IgG titers that were comparable at 5 wk, CTA1-DD-adjuvanted responses were maintained for >16 mo with a half-life of anti-NP IgG ∼36 wk, but <15 wk after Ribi or Alum. A CTA1-DD dose-dependent increase in germinal center (GC) size and numbers was found, with >60% of splenic B cell follicles hosting GC at an optimal CTA1-DD dose. Roughly 7% of these GC were NP specific. This GC-promoting effect correlated well with the persistence of long-term plasma cells in the bone marrow and memory B cells in the spleen. CTA1-DD also facilitated increased somatic hypermutation and affinity maturation of NP-specific IgG Abs in a dose-dependent fashion, hence arguing that large GC not only promotes higher Ab titers but also high-quality Ab production. Adoptive transfer of splenic CD80(+), but not CD80(-), B cells, at 1 y after immunization demonstrated functional long-term anti-NP IgG and IgM memory cells. To our knowledge, this is the first report to specifically compare and document that adjuvants can differ considerably in their support of long-term immune responses. Differential effects on the GC reaction appear to be the basis for these differences.