Analysis of pathological data in the 10th year of follow-up of a multicentre trial of the management of operable breast cancer has confirmed the correlation of prognosis with tumour grade, tumour size and lymph-node status. For each factor examined there was no difference in survival between the 2 treatment groups ("watch policy" and radiotherapy) but patients in the WP group whose tumours were of Grade II or III or greater than 2 cm, or with lymph-node metastases, had a greater chance of local recurrence. Cellular reaction had no relationship with prognosis, except in patients with Grade III tumours. The clinical relevance and application of these results are discussed.