The mechanical behavior of muscle during locomotion is often predicted by its anatomy, kinematics, activation pattern and contractile properties. The neuromuscular design of the cockroach leg provides a model system to examine these assumptions, because a single motor neuron innervates two extensor muscles operating at a single joint. Comparisons of the in situ measurements under in vivo running conditions of muscle 178 to a previously examined muscle (179) demonstrate that the same inputs (e.g. neural signal and kinematics) can result in different mechanical outputs. The same neural signal and kinematics, as determined during running, can result in different mechanical functions, even when the two anatomically similar muscles possess the same contraction kinetics, force-velocity properties and tetanic force-length properties. Although active shortening greatly depressed force under in vivo-like strain and stimulation conditions, force depression was similarly proportional to strain, similarly inversely proportional to stimulation level, and similarly independent of initial length and shortening velocity between the two muscles. Lastly, passive prestretch enhanced force similarly between the two muscles. The forces generated by the two muscles when stimulated with their in vivo pattern at lengths equal to or shorter than rest length differed, however. Overall, differences between the two muscles in their submaximal force-length relationships can account for up to 75% of the difference between the two muscles in peak force generated at short lengths observed during oscillatory contractions. Despite the fact that these muscles act at the same joint, are stimulated by the same motor neuron with an identical pattern, and possess many of the same in vitro mechanical properties, the mechanical outputs of two leg extensor muscles can be vastly different.