The studies described in this thesis examine the effects of neural and mechanical factors on production of active and passive force in human muscles. The first series of studies investigated the the level of neural drive human subjects can exert during maximal voluntary contractions. A second series of studies examined mechanical behaviours of human muscles and tendons at rest and during contraction.

The first study used the method of twitch interpolation to determine if the level of neural drive subjects could produce during maximal voluntary contractions was reduced during concurrent contraction of muscles from the contralateral limb. Subjects performed maximal voluntary contractions of the thumb adductors with or without concurrent contraction of the contralateral elbow flexors or thumb adductors. All subjects were capable of attaining near-maximal, but not usually maximal, levels of neural drive. Maximal voluntary force and neural drive were similar in unilateral and bilateral contractions. Under these conditions, concurrent contralateral contraction had little or no effect on voluntary force or neural drive.

In a second study, we extended the use of the twitch interpolation method to measure neural drive to elbow flexor muscles during maximal concentric contractions. Subjects attained even higher levels of neural drive in maximal concentric contractions than in maximal isometric contractions. To determine if failure of neural drive contributed to fatigue, subjects repeatedly lifted and lowered a weight and continued doing so, with minimal levels of assistance, until the muscle was fatigued. Under these conditions, subjects were able to maintain high levels of neural drive. This suggests that the fatigue associated with our dynamic fatigue protocol is primarily of peripheral origin, and does not result from a failure of neural drive.

It is widely believed that neural drive to muscles can be enhanced with training. Indeed, this is thought by many to be the primary mechanism producing increases in strength associated with the first few weeks of high resistance exercise (the "neural training hypothesis"). We used twitch interpolation to show that eight weeks of isometric strength training did not increase maximal voluntary neural drive to the elbow flexors in isometric contractions. In addition, contrary to an earlier report, imagined training did not increase strength or neural drive. These findings challenge the neural training hypothesis.

The studies described above used twitch interpolation to measure neural drive. This involves delivery of an electrical stimulus to contracting muscles and measurement of the resulting "interpolated twitch". To determine what factors determine the amplitude of the interpolated twitch, we implemented a computer model of twitch interpolation. Results of simulations suggest that the amplitude of the twitch is slightly attenuated by antidromic and reflex effects. It was shown that the muscle force and the amplitude of the interpolated twitch become insensitive to changes in neural drive at high forces. Thus real changes in the amplitude of the interpolated twitch may be indicative of large changes in neural drive.

Two final studies explored properties of muscles and tendons, the muscle force actuators. In the first, ultrasonography was used to systematically map changes in pennation of human brachialis muscle fibres in vivo. It was found that the brachialis muscle undergoes large increases in pennation with even weak contractions, particularly at short lengths. This influences the capacity of muscle fibres to generate torques. There was no change in pennation of the relaxed muscle with changes in_joint angle, suggesting that the muscle may be slack at rest. This would limit the capacity of brachialis muscle afferents to sense joint displacements when the muscle is relaxed.

In the final study, ultrasonography was used to measure changes in the length of muscle fascicles in two human lower limb muscles in vivo. Length changes in muscle fascicles were much smaller than the change in origin-to-insertion distance. This was interpreted to mean that much of the change in length is taken up by a change in the length of the tendons. Tendon compliance, like slack in muscles, reduces the capacity of resting muscles to "sense" joint displacement.