Temperature distribution is an important factor in thermo-radiotherapy and it is greatly dependent on the applied heating technique. Consistency of the heating method is therefore important in translating in vivo experimental data to the clinical situation. To further evaluate the combination of interstitial hyperthermia and interstitial radiotherapy, an experimental interstitial hyperthermia system has been developed for small (500-2000 mm3) tumours growing in the flank of a rat. The system used reproduces the properties of our clinical current source interstitial hyperthermia system. The heating system consists of four applicators, each with independent tuning and power control. The applicators are situated inside plastic afterloading catheters and are capacitively coupled with the surrounding tissue. The tumour is heated through dissipation of a 27 MHz current flowing to an external ground plane. An effective RF-filter allows reliable thermocouple temperature measurements when the power is switched on. The tumour temperature is easily controlled by means of a continuous temperature read-out and a clear temperature display. A minimum temperature up to 46 degrees C can be reached within 4-10 min and maintained (+/-0.5 degrees C) throughout the treatment period. Modelling calculations performed for this heating system indicate that the applicator temperatures should be kept equal in order to minimize the difference between maximum and minimum temperature. Significantly higher applicator currents are needed at larger distances from the ground plane. In addition, the homogeneity of the temperature distribution is improved when either the tumour is isolated or when the environmental temperature is increased. The calculations also show that temperature distribution is strongly dependent on effective heat conductivity. A description of the system and its performance is presented.