At the Cancer Institute we are using RF capacitive hyperthermia as an adjuvant to radiotherapy and/or chemotherapy in the local control of soft tissue sarcomas. We have studied the influence of bolus conductivity, electrode and phantom sizes on the rate of heating of agar phantoms. We have varied the bolus conductivity by varying the saline concentration in the bolus bags from zero to 2.0 per cent, during heating. We found that the rate of heating of phantoms increases and that of the bolus decreases with the increase in the saline concentration of bolus up to 1 per cent, irrespective of phantom and electrode sizes. However, for a given size of electrodes the rate of heating decreased with the increase in the phantom size. When the diameter and height of the phantom were equal to the diameters of electrodes the rate of heating of the phantom was nearly uniform. However, when the diameter of the phantom was larger than that of electrodes the rate of heating in the radial axis decreased with the increase in the radial distance. On the basis of this data we suggest the use of electrodes larger in size by 1.0-3.0 cm than the size of the tumour, where the size of the anatomical site to be heated is larger than the electrode size to be used. Phantom and clinical data have indicated that the presence of bone in the field of heating can lead to hot spots. Preliminary clinical results have shown that the response of sarcomas to thermo-chemo-radiotherapy was superior to that of either thermo-radiotherapy or radiotherapy alone.