Evidently a direct relation exists between the duration of work and the increase of threshold. For instance, the threshold is higher after a muscle is fatigued for two hours than it is at the end of the first hour. The relation between the work done and the threshold is not so clear. In some animals the thresholds were higher after 120 grams had been lifted 120 times a minute for 30 minutes than they were in others in which 200 grams had been lifted 240 times a minute for the same period. The muscle in the latter instances did almost four times as much work, yet the threshold was lower. The difference may be due to the general condition of the animal. A few experiments were performed on animals in which the nerve supplying the muscle was cut seven to fourteen days previous to the experiment. The muscle, therefore, had within it no living nerve fibres. The average normal threshold for the denervated muscle in 6 animals was 61.28 units. As in the normal muscle, the percentage increase due to fatigue was large. The Slow Restoration of Fatigued Muscle to Normal Irritability by Rest That rest decreases the fatigue threshold of both nerve-muscle and muscle can be seen in Fig. 18. The time taken for total recovery, however, is dependent upon the amount of work done, but this change, like that of fatigue, varies widely with different individuals. In some animals the threshold returned to normal in 15 minutes; in others, in which the same amount of work was done, it was still above normal even after 2 hours of rest. This may be due to the condition of the animals--in some the metabolites are probably eliminated more rapidly than in others. There were also variations in the rate of restoration of the normal threshold when tested on the nerve and when tested on the muscle in the same animal. In Fig. 18 (at 3) the nerve-muscle returned to normal in 30 minutes, whereas the muscle (at 4) after an hour’s rest had not returned to normal by a few β units. This, however, is not typical of all nerve-muscles and muscles. The opposite condition--that in which the muscle returned to normal before the nerve-muscle--occurred in as many cases as did the condition just cited. The failure of the two tissues to alter uniformly in the same direction may be explained as due to variations in the location of the electrodes when thrust into the muscle at different times (e. g., whether near nerve filaments or not). The results from observations made on the nerve are more likely to be uniform and reliable than are those from the muscle. The time required for the restoration of the threshold from fatigue to normal, in denervated muscles, is approximately the same as that for the normal muscle. The Quick Restoration of Fatigued Muscle to Normal Irritability by Adrenin