It has been suggested that during a skeletal muscle contraction the metabolic energy cost at the onset may be greater than the energy cost related to holding steady-state force. The purpose of the present study was to investigate the effect of contraction duration on the metabolic energy cost and fatigue process in fully perfused contracting muscle in situ. Canine gastrocnemius muscle (n = 6) was isolated, and two contractile periods (3 min of isometric, tetanic contractions with 45-min rest between) were conducted by each muscle in a balanced order design. The two contractile periods had stimulation patterns that resulted in a 1:3 contraction-to-rest ratio, with the difference in the two contractile periods being in the duration of each contraction: short duration 0.25-s stimulation/0.75-s rest vs. long duration 1-s stimulation/3-s rest. These stimulation patterns resulted in the same total time of stimulation, number of stimulation pulses, and total time in contraction for each 3-min period. Muscle O2 uptake, the fall in developed force (fatigue), the O2 cost of developed force, and the estimated total energy cost (ATP utilization) of developed force were significantly greater (P < 0.05) with contractions of short duration. Lactate efflux from the working muscle and muscle lactate concentration were significantly greater with contractions of short duration, such that the calculated energy derived from glycolysis was three times greater in this condition. These results demonstrate that contraction duration can significantly affect both the aerobic and anaerobic metabolic energy cost and fatigue in contracting muscle. In addition, it is likely that the greater rate of fatigue with more rapid contractions was a result of elevated glycolytic production of lactic acid.