Carbohydrates in the form of capsular polysaccharides and/or lipopolysaccharides are the major components on the surface of bacteria. These molecules are important virulence factors in many bacteria isolated from infected persons. Immunity against these components confers protection against the disease. However, developing vaccines based on polysaccharides is difficult and several problems have to be solved. First of all, most of the bacterial polysaccharides are T-lymphocyte independent antigens. Anti-polysaccharide immune response is characterised by lack of T-lymphocyte memory, isotype restriction and delayed ontogeny. Children below 2 years of age and elderly respond poorly to polysaccharide antigens. Secondly, the wide structural heterogeneity among the polysaccharides within and between species is also a problem. Thirdly, some bacterial polysaccharides are poor immunogens in humans due to their structural similarities with glycolipids and glycoproteins present in man. The T-lymphocyte independent nature of a polysaccharide may be overcome by conjugating the native or depolymerised polysaccharide to a protein carrier. Such neoglycoconjugates have been proven to be efficient in inducing T-lymphocyte dependent immunity and to protect both infants as well as elderly from disease. Another approach to circumvent the T-lymphocyte independent property of polysaccharides is to select peptides mimicking the immunodominant structures. Several examples of such peptides have been described.