The initial variations in the rates of NO dissociation and CO + NO reaction on Rh(110) and their relation to the changes in the surface structure and composition have been studied by means of mass spectrometry, thermal-programmed desorption (TPD) and low-energy electron diffraction (LEED). The changes in the partial pressures of the reaction products, N2 and CO2, have been used as a measure of the reaction rates for three catalyst temperatures: 460, 510 and 670 K. The composition of the surface layer at various stages of the reaction was determined from the TPD spectra of N2 and O2. Various LEED patterns were observed in the course of the reactions which pass through several stages depending on the reaction temperature and CO/NO partial pressure ratios. In the case of NO dissociation the variations in the rate of N2 production at 460 and 510 K correlate well with structural changes. The LEED shows N-related 3 x 1 and 2 x 1 structures in the early stages of the reaction and an oxygen-induced 1 x n reconstruction of the surface in the later stages. The latter leads to a successive formation of structures, from p(2 x 2) and c(2 x 4), related to mixed O + N layers, to (2 x 2)pg and c(2 x 6) related to oxygen alone. At 460 and 5 1 0 K, when the steady state is reached, the catalyst surface contains oxygen in a c(2 x 6) structure with some nitrogen possibly buried subsurface. Similar stages in the rate of N2 production and related surface structures were observed during the CO + NO reaction. Because of the ''cleaning'' effect of CO at steady state the catalyst surface contains oxygen and nitrogen in a c(2 x 4) structure. At 670 K the variations in the reaction rates and surface structures are governed only by oxygen. The observed anti-phase variations in the rates of N2 and CO2 production, when the CO + NO reaction is carried out at 460 and 510 K, are correlated to the changes in the structure and composition of the catalyst surface. Models for some of the structures observed in the course of the reactions are suggested.