Prognostic factors predicting for breast recurrence following breast conserving surgery and radiotherapy have been identified in a prospective randomised trial comparing two different radiotherapy techniques. The first used megavoltage radiation to encompass the whole breast and regional nodes (WF group). The second used an electron field directed to the tumour bed only (LF group). With a median follow up of 8 years survival in both groups was the same (72%). There was an increased rate of breast recurrence in the LF group: 25%, compared to the WF group: 13% (P = 0.00008), expressed in actuarial terms at 8 years. In a multivariate analysis of factors predicting for breast recurrence, once the difference in treatment technique is allowed for, only two factors were found to be significant: histological grade (P = 0.013) and lymphovascular invasion in the histological specimen (P = 0.037).