Assessment of inherent fluctuations of mitotic and labelling indices of human tumours

Summary.-A method is presented to evaluate the influence of statistical errors and inherent variation on the determination of mitotic and labelling indices of human tumours. In most of the experiments reported here, sufficient cells were counted to yield a statistical error which is small in comparison to the inherent differences in the proliferative indices, both between different sites in the same tumour and between different tumours of the same histological type. These inherent fluctuations are, therefore, a critical factor in cell kinetic studies of human tumours. THE study of cell population kinetics ideally requires tissue samples which are large, or multiple, or both. Generally, this does not pose a serious problem when the subject of research is an experimental tumour population. But the goal of most tumour cell population studies is a better understanding of human cancer, an