1. The contractile properties of fast‐twitch (extensor digitorum longus or EDL) and slow‐twitch (soleus) muscles in the rat were followed for periods of between 4 and 10 months after denervation. The effects of chronic electrical stimulation during the last 3‐8 weeks of denervation were investigated. 2. The fall in tetanic tension that follows axotomy ended after about 4 months' denervation. The equilibrium tension was about 0.75% of control tension in EDL and 0.2‐0.3% in soleus. 3. The low tension in soleus was due partly to the small diameter of the muscle fibres (atrophy) and partly to their necrosis that resulted in an 8‐fold fall in specific tension (the force per unit cross‐sectional area). Similar but less extreme changes occurred in EDL. 4. It is speculated that the final level of tension reached by unstimulated denervated muscles is an equilibrium between decrease in force due to atrophy and necrosis and increase due to regeneration. Differences between the final tension levels in soleus and EDL cannot be accounted for quantitatively by known differences in atrophy alone. Therefore, the rate of necrosis in soleus and of regeneration in EDL may be higher. 5. Chronic stimulation of long‐term denervated muscle increased force generation by about 7‐fold in EDL and between 20 and 55 times in soleus. The final tension reached was between 4 and 5% of normal in both muscles. Specific tension of fibres was almost completely restored by stimulation and the number of fibres was normal. The failure to recover full tension was largely due to failure to reverse denervation atrophy completely. 6. Twitch contraction and relaxation times were identical in denervated‐stimulated soleus and EDL. There was no evidence for dependence on duration of stimulation or tension of the muscle. The normalized maximum rate of rise of tetanic tension remained higher in EDL than soleus.

1. Rat soleus muscles were denervated for 6‐10 months; some of these were chronically stimulated for the last 3‐8 weeks before recording. The muscles were fixed at physiological lengths and embedded in epoxy resin. 2. Sections for light microscopy were stained with p‐phenylenediamine. Denervation reduced the mean cross‐sectional area of fibres to 3% of controls (peak at 20 microns2). The cross‐sectional areas of the stimulated fibres had a peak at 70 microns 2. In light micrographs of denervated muscles, the total number of fibres appeared to be reduced; however not all fibres could be identified (see paragraph (3)). 3. In the electron microscope, many fibres showed disarrayed myofilaments. Few fibres had more or less normal cross‐striations. Muscle fibres as small as 1 micron in diameter were seen. The smallest fibres did not contain myofilaments. Some unequivocally necrotic fibres were seen. 4. Most stimulated denervated fibres showed an almost normal sarcomere pattern. Scattered throughout the muscle were single fibres as small as 2 microns in diameter, but these were well organized and could be recognized in the light microscope. Few fibres were necrotic. Often fibres were serially arranged. Satellite cells were prominent. 5. It is concluded that in long‐term denervated rat soleus the original fibres are lost and those seen are the result of repeated cycles of regeneration and necrosis. Stimulation maintains the sarcomeres and probably prevents secondary degeneration and necrosis.

1. A fast (extensor digitorum longus) and slow (soleus) twitch muscle were denervated in rats and guinea‐pigs and isometric and isotonic contractions were followed for periods of up to 6 months after. 2. The decay of tetanic tension with time could be described as exponential. The rate of decay of tension was greatest in rat soleus and least in guinea‐pig soleus by a factor of more than three. The fast muscles atrophied at intermediate rates. 3. The contraction and relaxation times of soleus and extensor digitorum longus of rat, initially prolonged by denervation, became shorter after 3 weeks. There was no such reversal in either guinea‐pig muscle, indeed extensor digitorum longus twitch became even more prolonged. Guinea‐pig muscles often showed signs of repetitive response to a single stimulus, resulting in distortion of relaxation of the twitch. 4. There was a slowing of isotonic shortening velocity in the late stage of denervation of guinea‐pig extensor digitorum longus, accompanied by a fall in the rate of development of isometric tetanic tension. There was a just‐significant (P less than 0.1) increase in the shortening velocity of rat soleus. None of the other muscles showed any change in either rate characteristic. 5. In guinea‐pig extensor digitorum longus the type I fibres atrophied less than type II fibres; in all other muscles the atrophy was more uniform, possibly faster in type II. Guinea‐pig soleus remained pure type I contrasting with an increase in the numbers of type II fibres in rat soleus. There was a possible increase in the number of type I fibres in guinea‐pig fast muscle and no change in the rat.

1. Rat soleus muscle was denervated by sciatic transection and electrically stimulated for periods of between 3 and 9 weeks with intermittent 1 s bursts of pulses. Most of the bursts were either repeated every 90 s and pulses within them had frequencies between 10 and 100 Hz, or had a frequency of 50 Hz and were repeated at intervals between 60 and 600 s. Comparisons were made with continuous stimulation at 10 Hz. 2. At the end of the period of stimulation, isometric twitches and tetani were measured and, in a proportion, also isotonic shortening velocity. 3. Isometric twitch duration (contraction and relaxation) decreased with time of stimulation. Very similar effects were seen in all animals in which intermittent stimulation had been used. There was a significant relationship between the change in twitch duration and the frequency used within the bursts of chronic stimulation, with slightly larger effects at frequencies of 40 and 60 Hz. The lowest burst repetition rate produced the largest effects. 4. It was confirmed that similar changes were found in denervated muscles that were not stimulated, although these changes were smaller and developed more slowly. 5. The extreme loss of tetanic tension induced in the muscle by denervation was reduced by chronic stimulation, with no significant difference between different regimes, although there were small differences which showed the same patterns of effectiveness described for twitch durations. 6. Continuous stimulation at 10 Hz maintained the twitch contraction and relaxation phases at the values found 3 weeks after denervation, that is it prevented secondary shortening of the twitch. Continuous stimulation reduced tension loss but was, perhaps, less effective than intermittent stimulation. 7. Twitch‐tetanus ratio increased with denervation with little spontaneous reversal later. Stimulation at all frequencies reduced the ratio, but it did not reach normal values. 8. Isotonic shortening velocity was measured in many of the muscles. Maximum velocity was estimated and normalized by muscle length.(ABSTRACT TRUNCATED AT 400 WORDS)

In recording the contractile (or dynamic) properties of muscle. two standard sets of conditions are commonly used. In one, the isometric set, the muscle is held at a constant length and the tension that develops is measured. The other, namely the isotonic set, involves allowing the muscle to shorten under constant load and recording changes of length. In the body, of course, muscle activity is more complex, sometimes involving periods approximating to isometric or isotonic, but more often shortening occurs under varying load. When lifting a weight, for example, arm muscles must first contract isometrically, developing sufficient tension to overcome the force of gravity on the weight (and on the arm). This will be followed by shortening but with the load on the muscles greater than in the isometric phase because additional force must be generated to accelerate the mass of the load (and of the arm). Only if a constant speed of movement is reached, will the shortening be truly isotonic. As the movement is ended and the arm is slowed, the inertia of the mass will exert a force in the opposite direction, which has to be opposed by action of antagonists. Finally, isometric contraction will hold the weight up. Natural contractions are also more complex than in experimental recordings because, as in the example described above, the activation of the muscle is not constant. A varying number of motor units are recruited, with changing discharge rates at different stages of the activity, according to the command signal from the brain and feedback from many receptors.

Rat soleus muscles were denervated and stimulated in vivo for periods of up to 104 days. Stimuli used were trains of 1 ms pulses at 100 Hz delivered for periods of 1 s; trains were repeated every 10-100 s. In a majority of animals the tension of the muscles was maintained at about 10% of normal, equivalent to muscles denervated but unstimulated for 20 days. At the longest periods the stimulated muscles developed ten times more tension than ones that were denervated but not stimulated. In denervated and denervated-stimulated muscles twitch contraction and relaxation times were prolonged, compared with controls, for up to 3 weeks. Thereafter both sets showed a speeding of the isometric twitch that was greater in the stimulated muscles. At the longest periods the twitch was as short as that of a denervated fast muscle. Stimulation did not affect contralateral denervated muscles. Twitch: tetanus ratios remained high despite stimulation, and muscles showed little post-tetanic potentiation. Tension developed more rapidly in the tetani of the stimulated muscles, even allowing for larger final values. Maximum velocity of shortening was increased in many of the stimulated muscles, and there was a proportional flattening of the force-velocity curve, i.e. a/P 0 increased. Maximum velocity and a/P 0 increased reciprocally with twitch time to peak, so that those muscles that had twitches most changed by stimulation also had their isotonic properties modified to the greatest extent. Even at the longest period of stimulation, twitch time course and tetanic tension were not converted to those of normal fast muscle.

1. Cat soleus (slow twitch) was cross‐reinnervated with nerve to flexor hallucis or digitorum longus muscle (fast twitch). More than 3 years later motor unit isometric contractions and muscle immunohistochemistry and histochemistry were investigated. All muscles differed from normal fast or slow muscle. 2. The motor units could be divided into two groups: one with fast twitches and low tetanic tension, the other with slow twitches and high tension. This is the reverse of the relation between motor unit twitch time and tetanic tension in normal muscle (fast or slow). 3. Motor unit twitch time to peak decreased with axonal conduction velocity, as in normal muscle, but so did tetanic tension, which is abnormal. 4. Twitch—tetanus ratio increased with twitch time to peak in the group of slow units but not in the fast group (although the range of ratios was as great). 5. A tetanus depressed the twitch tension of slow motor units and potentiated fast ones as in normal muscle but the potentiation was often accompanied by an abnormal prolongation of the twitch. 6. The mean conduction velocity of axons was slightly higher than at 6 months' reinnervation but below the normal value for fast muscle. 7. Antibody to slow myosin was bound strongly to Type I fibres but not to Type II fibres, confirming the histochemical division of fibres into Types I and II. 8. More than 95% of the fibres were oxidative, with Type I predominating over Type II a in the ratio of about 2:1. 9. The higher tension of the slow motor units was the result of three factors: the number of fibres per motor unit was at least three times that in the fast; Type I fibres had cross‐sectional areas little less than Type II (a and b together) and were estimated to develop more tension per unit area. All three findings were different from those in normal fast muscle. 10. One flexor hallucis longus muscle was self‐reinnervated and examined histochemically. This muscle was abnormal in that a large majority of the fibres were Type I.

1. Isometric contractile properties of motor units were measured in cross-reinnervated fast (flexor digitorum longus) and slow (soleus) twitch muscles of the cat. All but one cross was at least 95% pure. 2. There was a reduction in the number of motor units in all muscles, but totals remained about equal in cross-reinnervated soleus and flexor digitorum longus. 3. Motor unit tensions (mean and maximum values) were higher in cross-reinnervated soleus than in cross-reinnervated flexor digitorum longus, reversing the differences between normal muscles. This was due to increases in muscle mass and in the tension developed per unit cross-sectional area. There were motor unit tensions larger and smaller than those seen in normal muscle, but the range was comparable with that seen in self-reinnervated muscle. 4. The changes in twitch time to peak of whole muscle following cross-reinnervations resulted from a change over the whole range of motor units. The conversion of soleus was less complete than that of flexor digitorum longus, and the time to peak of its fastest motor unit was twice as long as any seen in normal flexor digitorum longus. 5. In neither of the cross-reinnervated muscles were the fast contracting motor units larger than the slow contracting ones, and in cross-reinnervated soleus they were smaller. 6. Axonal conduction velocity was correlated with motor unit tension in both muscles and with twitch time to peak in cross-reinnervated flexor digitorum longus, but in all cases less clearly than in normal muscles. 7. The ratio of twitch to tetanic tension increased with increasing twitch time to peak, as in normal muscles.