Résumé : Some models describing the kinetics of cell proliferation present the major advantage of requiring a small set of parameters for the complete description of the kinetic behaviour of cell populations. Among such models, that proposed by Burns & Tannock (1970) offers the further advantage of allowing a generalized description of cell cycle that covers the ‘Go’ concept. This model assumes that the cell life cycle consists of two phases: G‐phase and C‐phase. After division, the cells enter G‐phase and remain there for any length of time. They leave this phase at a constant rate (kG) to enter C‐phase where they divide. C‐phase has a constant duration: θ. Only the two parameters θ and kG are required for a complete kinetic description in the steady‐state case. This is also true for an exponentially growing population, provided cell loss is null or negligible. We started from the equations relating kG, θ and NG/N (where NG is the number of cells in G‐phase and N the total cell numbers) associated with the following models: (1) Burns & Tannock (1970) model based on the hypothesis that the cells may differentiate from any phase of the cell cycle. (2) A variant of this model where only the cells in G‐phase may differentiate. We describe an original method allowing the determination of the parameters of the model from published ‘labelled mitosis’ curves. We observe that the experimentally obtained values of those parameters constitute a non‐randomly limited part of the set of values that one may theoretically have. This suggests the necessity for further assumption(s) restricting the model. Copyright © 1975, Wiley Blackwell. All rights reserved