Macromolecular interactions observed using surface plasmon resonance technology (BIAcore, Pharmacia) often display kinetic behavior which deviates from the pseudo-first-order time dependence that has been predicted for 1:1 interactions of ligand and ligate. In the present study we review the major reasons for such deviations, and present results which suggest that the most common source of deviations from the pseudo-first-order kinetic approximation of BIAcore kinetic data is likely to be heterogeneity of the immobilized ligand sites. A simplified analysis of the adsorption stage of BIAcore data is presented in terms of thenet observedpseudo-first-order rate constant,k_obs, rather than in terms of the association and dissociation rate constants,k_aandk_d. The analysis is then extended to the determination of the dissociation equilibrium constant for the interaction of ligand and ligate in the solution phase from sensorgrams reflecting competition between soluble and immobilized forms of ligand for ligate.