1. Measurements were made of the energy produced as heat and work (h + w) and the chemical changes which occurred between the beginning and end of each of two periods of stimulation. The muscles contracted tetanically under isometric conditions. Each period of stimulation (tetanus) lasted 5 sec and there was an interval of 3 sec between them. The tension developed in the second tetanus was 91% of that in the first.2. The observed (h + w) was greater in each tetanus than the quantity expected from the measured chemical changes. This excess was 230 +/- 43 mJ/g dry wt. (mean +/- S.E. of mean) in the first tetanus, but only 89 +/- 46 mJ/g in the second tetanus (mean +/- S.E. of mean). The result for the first tetanus agrees with previous findings.3. The observed (h + w) was divided into two parts, labile and stable, which were defined by Aubert (1956). This division was made on the basis of the time course of the (h + w) production, without reference to the chemical changes. The labile part of the (h + w) has an exponentially declining rate, and the stable part has a constant rate.4. The observed (h + w) was less in the second tetanus than in the first. This was due largely to the decrease in the labile part in the second tetanus; it was only ca. 0.35 of that in the first tetanus. However, the stable part remained relatively constant, 0.83 of that in the first tetanus. Aubert & Maréchal (1963) and Aubert (1968) have reported similar results.5. Having divided the (h + w) into the stable and labile parts, we found that the stable part could be accounted for by the chemical change, but the labile part could not. Thus, for both tetani, the stable part of the energy is about equal to the explained energy, and the labile part is about equal to the unexplained energy. The possible interpretations of these equalities are discussed.