To gain insight into the relative importance of force feedback to ongoing ankle extensor activity during walking in the conscious cat, we isolated the medial gastrocnemius muscle (MG) by denervating the other ankle extensors and measured the magnitude of its activity at different muscle lengths, velocities and forces accomplished by having the animals walk up and down a sloped pegway. Mathematical models of proprioceptor dynamics predicted afferent activity and revealed that the changes in muscle activity under our experimental conditions were strongly correlated with Ib activity and not consistently associated with changes in Ia or group II activity. This allowed us to determine the gains within the force feedback pathway using a simple model of the neuromuscular system and the measured relationship between MG activity and force. Loop gain increased with muscle length due to the intrinsic force-length property of muscle. The gain of the pathway that converts muscle force to motoneuron depolarization was independent of length. To better test for a causal relationship between modulation of force feedback and changes in muscle activity, a second set of experiments were performed in which the MG muscle was perturbed during ground contact of the hind foot by dropping or lifting the peg underfoot. Collectively, these investigations support a causal role for force feedback and indicate that approximately 30% of the total muscle activity is due to force feedback during level walking. Force feedback's role increases during upslope walking and decreases during downslope walking providing a simple mechanism for compensating for changes in terrain.