Forces in the myocardial wall can be measured in several ways or calculated using certain simplifying assumptions. In this study we investigated the reliability of two measurement methods, one of which was introduced by Feigl et al (1967), whereas the other method was developed in our laboratory. Both methods were tested in actively contracting skeletal muscle and beating hearts of open-chest dogs by comparing the force transferred from the muscle to the transducer under various circumstances. It appeared that changes in muscle length, be it through initial length changes or through shortening during contractions, had a great influence on the transfer of force to the transducer, for both methods, in both preparations. In the heart a decrease in internal left ventricular diameter of 15% resulted in a 50% reduction of force transferred to the transducer, independent of whether the length took place as a change in filling or as a change in ejection volume. In skeletal muscle the length-dependent effects during shortening were larger and those resulting from initial length changes were more variable than in beating hearts. That the effects of muscle length changes are of such magnitude means that, if no other errors exist, they alone would invalidate that until principally different methods of measuring wall stress in the myocardium are discovered, attempts at accurate calculation of myocardial wall stress are a better approach than wall stress measurements.