Abstract
Previous studies of jumping in frogs have found power outputs in excess of what is possible from direct application of muscle power and concluded that jumping requires the storage and release of elastic strain energy. Of course, the muscles must produce the work required and their power output should be consistent with known muscle properties if the total duration of muscle activity is known. Using the Cuban tree frog, Osteopilus septentrionalis, I measured jumping performance from kinematics and used EMG measurements of three major jumping muscles to determine the duration of muscle activity. Using the total mass of all the hindlimb muscles, muscle mass-specific work output up to 60 J kg-1 was recorded. Distributed over the duration of the jump, both average and peak muscle mass-specific power output increased approximately linearly with the work done, reaching values of over 750 and 2000 W kg-1, respectively. However, the muscles were activated before the jump started. Both preactivation duration and EMG amplitude increased with increasing amounts of work performed. Assuming the muscles could produce work from EMG onset until toe-off, the average muscle mass-specific power over this longer interval also increased with work done, but only up to a work output of 36 J kg-1. The mean power above this value of work was 281 W kg-1, which is approximately 65% of the estimated maximum isotonic power. Several reasons are put forward for suggesting this power output, although within the known properties of the muscles, is nevertheless an impressive achievement.