In this article special emphasis has been placed on the problem of reproducibility of the description of tumour spread. Tumour size is an important factor for the prognosis of patients with cervical carcinoma. This article has described several methods for determining tumour extent. I have also presented a simple method applicable to routine measurements, and based on stereological principles, and have compared it to more sophisticated tumour-size models. The main intention of this work was to define objective and comparable parameters for the classification of tumour spread.