In 1922, Hartree and Hill (1) described a delayed heat-production after a short tetanic stimulus in a muscle deprived of oxygen, amounting to about one-third of the recovery beat in oxygen. In 1923 the same authors (2) re-examined the matter, with greater precautions to exclude oxygen, and found a "most probable " value for this delayed anaerobic heat of about one-quarter of the initial, or one-sixth of the recovery, heat. The existence of this heat has remained an obscure phenomenon, a complication in an otherwise comparatively simple scheme, and a further attempt was made by Furusawa and Hartree (3) in 1926 to trace its source. In spite of all precaution to exclude oxygen, and to obviate physical effects (lack of uniformity in the muscle, etc.) which might produce the same apparent result, the delayed heat persisted, and Furusawa and Hartree concluded that its minimum value was about 12 per cent, of the initial heat ; in many cases. It was more, sometimes much more. It should be noted that all the investigations referred to dealt with a short tetanus, not with a single twitch. The increment produced by stimulation in the resting heat-rate of a muscle under anaerobic condition, described in a previous paper of the present series, is one of the factors responsible for the effect discussed. Unless the galvanometer-zero, and the temperature of the thermopile-chamber, be extremely constant, it is easy to misinterpret the permanent shift of position of the galvanometer alter stimulation in nitrogen, and to deduce the existence of a long-continued slow production of beat gradually diminishing to zero. The reactions underlying the increment in the resting beat-rate are not part of the process of activity itself, although induced by it, and it is incorrect to attribute the energy the liberate to the preceding contraction; this is obviously the case, since the increment in beat-rate we know now to be permanent, so that the energy liberated is not constant but proportional to the time during which one choose to follow the galvanometer defection. It will be shown, indeed, by Hartree and Hill in a later paper of this series that such anaerobic delayed heat as really exists, in the case of a tetanus, occupies only a minute or two alter the stimulus; the long-continued part of the delayed heat, described in each of the three papers cited above, is an error due to a misinterpretation of the facts, which could not then be observed with the same accuracy as is possible now.