The applicability of classical liquid electrochemical concepts to porous cathode structures used in high temperature fuel cells is discussed. It does appear that the performance of molten carbonate fuel cell cathodes can be satisfactorily interpreted in terms of conventional electrochemical engineering invoking thin films of molten electrolyte distributed in the partially flooded porous cathode structure. In contrast, the behaviour of porous oxide cathodes deposited on ceramic oxide electrolytes is dominated by interfacial oxygen surface exchange reactions and diffusive processes. The importance of the characteristic length term, L-c (D double dagger/k), which indicates the change-over from surface to bulk control is emphasised. It is suggested that for porous oxide cathode structures, in which the grain size is usually around 1 mu m, the oxygen reduction process will invariably be dominated by the interfacial oxygen exchange process. For mixed conducting cathodes, chemical processes such as changes in the oxide stoichiometry, can contribute significantly to the overpotential within the electrode. This can lead to erroneous interpretations if the I-V-characteristics are analysed in terms of conventional electrochemical expressions. Finally the design of optimised solid state cathodes is discussed and the need for oxygen surface diffusion data is emphasised.