The application of a hospital-based presurgical autologous blood deposit programme to support of radical hysterectomy is described over 4 years (1991-95), during which 48 patients participated in the autologous blood programme, and 63 did not to do so. All but one of the autologous donors received autologous blood but only two received allogeneic blood. Forty-three of the autologous nondonors received no transfusion and 20 received allogeneic blood. These differences are highly significant. For the 48 autologous donors, 91% of the blood requested was collected, and of that 91% was used, for a "wastage' rate of 9%. The mean blood use was significantly greater in the autologous donors. Blood loss was not significantly different between the two groups. The data confirm the validity of the autologous blood order schedule of 2 units for radical hysterectomy.