Mathematical modelling of the course of the immune response is undoubtedly one of the most progressive and most promising areas of modern immunology. Mathematical models (along with computer programs) can be taken as "the only means of thoroughly testing and examining a large and intricate theory" (Partridge et al. 1984). The first phase of construction of mathematical models is the formulation of assumptions based on the knowledge of the facts to be modelled (manifested usually in a scheme of the presumed course of the modelled process). The first mathematical models of immune response were based on the hypothesis of a two-stage differentiation of cells participating in the humoral response, published in Prague 23 years ago (Sercarz and Coons 1962; Sterzl 1962) and illustrated by the X----Y----Z scheme. Many contemporary mathematical models still stem from this scheme which undoubtedly fits the fundamental data concerning the immune system.