1. To study the means whereby ankle biomechanics are represented in the interneuronal circuitry of the spinal cord we examined stretch-evoked reflex interactions between the physiological extensors flexor hallucis longus (FHL) and flexor digitorum longus (FDL) as well as their interactions with gastrocnemius (G), soleus (S), and the quadriceps group (Q) in 34 unanesthetized decerebrate cats. To evoke stretch, DC motors provided ramp-hold-release length changes to tendons detached from their bony insertions. Semiconductor myographs measured resultant muscle force response. Reflexes were examined under both quiescent (no active force generation) and activated conditions; muscle activation was achieved through either crossed-extension or flexion reflexes. 2. FHL and FDL share mutual excitatory stretch-evoked interactions under most conditions examined. These interactions depended on muscle length, were asymmetric (with FHL contributing a larger magnitude of reflex excitation onto FDL), and occurred at a latency of 16 ms. Mutual Ia synergism previously described for these two muscles provides a basis for all of the above findings. Our data demonstrate that for this muscle pair, reflex connectivities revealed at the intracellular level can be extrapolated to cover the entire motoneuron pool; further, our data directly demonstrate the net mechanical result of ensemble synaptic events. 3. FHL was found to share strong, mutually inhibitory stretch-evoked interactions with G, S, and Q. Stepwise regression statistical analyses determined that these interactions depended on recipient muscle force and donor muscle force. These reflex interactions all occurred at a latency of 28 +/- 4 (SE) ms. Further, the heterogenic inhibition between FHL/G and FHL/S was attenuated by strychnine infusion (intravenous) but unaffected by either mecamylamine, picrotoxin, or baclofen infusion (intravenous, intrathecal). Disynaptic Ib inhibition previously described among hindlimb extensors provides a basis for the above findings; our data demonstrate that under certain conditions the ensemble activity of this system can cause a dramatic decline in whole muscle force output. 4. By contrast, FDL was found to share mutually inhibitory, stretch-evoked reflex interactions with G, S, and Q that were much weaker than those observed between FHL and these same muscles. The small magnitude of inhibition observed in these interactions made it difficult to assess reflex latency or to determine the factor(s) that best predicted the heterogenic inhibition. 5. This study provides further evidence of intrinsic differences in interneuronal organization between muscles whose activity occurs in a periodic manner during locomotion ("stereotypical") and a muscle whose locomotor activity is characterized by both periodic and nonperiodic components ("facultative").(ABSTRACT TRUNCATED AT 400 WORDS)