In order to gain insight into the mechanisms of the force-sharing between cat medial gastrocnemius (MG) and soleus (SOL), direct measurements of MG and SOL tendon forces were performed for a variety of movement tasks, while simultaneously recording the kinematics and kinetics of the target movements. The purpose of this thesis was to investigate changes in MG and SOL forces for a variety of unrestrained movements, such as different speeds and intensities of walking, jumping, and paw shaking, and to develop a theoretical framework to gain insight into the control of MG and SOL forces, specifically, and voluntary movements in general. MG actiViities systematically increased with increasing mechanical requirements for all movement tasks. SOL activation during locomotion appeared to be regulated so that decreases in force potential, associated with changes in the contractile conditions, were offset to maintain SOL forces approximately constant. During ju1 ping, SOL was deactivated prematurely, and the corresponding forces fell to zero prior to the end of the take off phase. SOL shortening speed during jumping reached and exceeded the maximal speed of shortening, and it appears that the muscle was deactivated so as to preserve energy in a situation where it could not contribute effectively to force production. Deactivation always occurred at or near the instant when SOL speed reached its maximal shortening value, therefore it seems that deactivation might have been triggered by a speed sensitive mechanism. Finally, a theoretical model was developed to quantify the proximity of the component of the ground reaction force (GRF) produced by an individual muscle to the resultant GRF produced by the entire system. MG forces were tightly related to proximity, suggesting that MG might be preferentially activated to control the direction of the GRF. This result is in agreement with a conceptual model of muscle control proposed by the group of van lngen Schenau several years ago, but it provides first validation of these ideas through the newly developed theoretical model of the cat hindlimb, and the corresponding direct muscle force measurements.