'Saturation analysis' is the term originally proposed by one of us (Barakat & Ekins 1961) to describe the general analytical method relying on progressive saturation by the test compound of a specific reagent. It has, in the last ten years, been applied to widely differing groups of physiologically important compounds using a variety of classes of reactive material. This review is intended to summarise the fundamental requirements of the method, with particular reference to the aim of achieving an 'accurate' result. Some of the mathematical principles underlying optimal assay design are discussed.