The focus of this dissertation is on the preparation, performance, and long term stability of SOFC composite cathodes prepared by infiltration methods. The majority of the work that follows aims to improve the understanding of the processes contributing to cathode deactivation and to propose strategies to lessen the extent of this deactivation. Through this understanding of the factors governing cathode performance, improvements can be made in overall cathode performance which can in turn lead to lower operating temperatures. The fuel cells used in this work were prepared by tapecasting and infiltration methods. Composite YSZ-perovskite electrodes were prepared by infiltration of stoichiometric ratios of perovskite precursor nitrate salts into a porous YSZ scaffold. First, the influence of ionic conductivity on the performance of solid oxide fuel cell cathodes was studied for electrodes prepared by infiltration of 40-wt% La0.8Ca0.2FeO3 (LCF), La0.8Sr0.2FeO3 (LSF), and La0.8Ba0.2FeO3 (LBF) into porous YSZ scaffolds. Although ionic conductivity varied by over an order of magnitude, no significant difference was observed in the performance of each material, suggesting that oxygen ion diffusion through perovskite film is not a rate limiting step for the oxygen reduction process within the cathode. Next, the effect of various infiltrated dopants on the performance of SOFC cathodes was examined. The addition of dopants had little influence on the 1123-K composite electrodes but all dopants tested improved the performance of the 1373-K, suggesting that the improved performance is related to structural changes in the electrode, rather than to improved catalytic properties or ionic conductivity. Based on these results, a model was developed to understand the performance of these electrodes. Two rate-limiting cases are considered for oxygen transfer into the YSZ fins: diffusion through the perovskite film or reactive adsorption of O2 at the perovskite surface. In agreement with the experimental results, an important implication from the model is that ionic conductivity of the pervoskite phase does not limit performance, for most commonly used perovskites, and that surface adsorption is likely limiting. Finally, strategies for improving cathode performance and stability are discussed.