There is a large and growing base of knowledge about the protective immune response to encapsulated bacterial pathogens. This information is a clear advantage to the development of glycoconjugate vaccines. It is well understood for such pathogens as Haemophilus b, pneumococcus, meningococcus and group b streptococcus that the induction of a functional anti-saccharide antibody response is likely to be protective. The most important question for glycoconjugate vaccines is the magnitude of the response required for protection. Even after extensive field trials with Haemophilus b vaccines, there is not a precise antibody correlate of immunity. In fact, cited values still rely on older studies with polysaccharide vaccine [11]. For that reason, vaccine such as the multivalent pneumococcal formulations will require field trials to measure their efficacy and hopefully determine the appropriate correlates of immunity. However, the addition of new serotypes into a multivalent pneumococcal formulation after efficacy trials are completed will probably have to be based on antibody correlates since it will not be possible to do efficacy studies on individual serotypes especially since they will have low prevalence. A similar dilemma may be faced with a vaccine for the group C meningococcus where the low disease incidence will make efficacy studies difficult. As was the case with mixing DTP and Hib vaccines, in the future combinations of effective antigens will continue to rely on comparative immunogenicity studies to ensure efficacy. It is clear that the immune system is capable of responding to a diversity of antigens simultaneously. However, it is also clear that vaccine components can interfere with each other and this interference is not necessarily predictable [12]. The ability to measure the stability of each of the components of a combined product reliably will become even more critical as the complexity of the mixtures increases.