The relationship between K-agglutination antibody titres and protection against experimental challenge with Dichelobacter nodosus, the effect of increasing the number of D. nodosus fimbrial antigens, and the importance of the nature of additional antigens in multivalent vaccines on antibody response and protection against experimental challenge with D. nodosus were examined in Merino sheep. A total of 204 Merino sheep were allocated to one of 12 groups, and vaccinated with preparations containing a variable number of rDNA D. nodosus fimbrial antigens. The most complex vaccine contained ten fimbrial antigens from all major D. nodosus serogroups, while the least complex contained a single fimbrial antigen. In addition to D. nodosus fimbrial antigens, other bacterial rDNA fimbrial antigens (Moraxella bovis Da12d and Escherichia coli K99), and bovine serum albumin (BSA) were used in some vaccines. Antibody titres to fimbrial antigens and BSA were measured by agglutination and ELISA tests, respectively. Antibody titres were determined on five occasions (Weeks 0, 3, 6, 8, and 11 after primary vaccination). All sheep were exposed to an experimental challenge with virulent isolates of D. nodosus from either serogroup A or B, 8 weeks after primary vaccination. For D. nodosus K-agglutinating antibody titres, a strong negative correlation between antibody titre and footrot lesion score was observed. This relationship was influenced by the virulence of the challenge strain. Increasing the number of fimbrial antigens in experimental rDNA D. nodosus fimbrial vaccines resulted in a linear decrease in K-agglutinating antibody titres to individual D. nodosus serogroups. Similarly, a linear decrease in protection to challenge with homologous serogroups was observed as the number of D. nodosus fimbrial antigens represented in the vaccine increased. The reduction in antibody titres in multicomponent vaccines is thought to be due to antigenic competition. The level of competition between individual antigens is not constant and appears to be related to the immunodominance (nature) of the competing antigens. Both BSA ELISA, and M. bovis K-agglutinating antibody titres were adversely affected by the presence of two D. nodosus fimbrial preparations, whereas the antigenicity of E. coli K99 was unchanged by the presence of two additional D. nodosus antigens. Further studies are required to determine the step(s) in the immune response which are influenced by antigenic competition. Our results suggest that antigen presentation, particularly following primary vaccination, is the step most strongly influenced by antigenic competition.