In interstitial heating one of the main requirements for achieving a certain elevated temperature in a tumour is that the effective power per applicator (Peff), i.e. the power which is actually deposited in the tissue, is sufficiently high. In this paper this requirement is discussed for the applicators of the 27 MHz multielectrode current source (MECS) interstitial hyperthermia (IHT) system. To minimize power reflection, the applicator impedance was matched with the generator impedance by adjusting the length of the coaxial cable in between. Transmission line losses, applicator efficiency and subsequently Peff were computed for several applicator types. The actual Peff per electrode was obtained from calorimetric measurements. Experiments with RC loads, which can be seen as perfect applicators, were performed to investigate the effect of mismatching on Peff. Applicator losses were measured for clinically used applicators, both single- and dual-electrode, utilizing saline phantoms. A simple spherical tumour model, using the effective heat conductivity (keff) to account for heat transport, was used to estimate Peff for a given tumour size, implant size and applicator density. Computations of Peff of various MECS-IHT electrodes were in close agreement with the phantom measurements. Most of the initial generator power was absorbed in the transmission line (60-65%). The efficiency of the applicators was about 65%. For both single- and dual-electrode applicators the effective electrode power was found to be about 1 W. Model calculations show that Peff of 1 W is sufficient to reach a minimum tumour temperature of 43 degrees C in well perfused tumours (keff = 3 W m-1 degree C-1), using a typical implant with 2 cm electrodes and 1.5 cm spacing. Mismatching can considerably affect Peff. Both a reduction to almost zero and a two-fold increase are possible. However, because the matching theory is well understood, mismatching is not a serious problem in clinical practice and can even be used to increase Peff if necessary. We conclude that the applicator design and the impedance matching method chosen in the MECS system allow heating to temperatures in the therapeutic range with implants used in clinical practice.