Arthrodesis of the subtalar joint, triple arthrodesis (involving the subtalar, talonavicular, and calcaneocuboid joints), double arthrodesis (involving the talonavicular and calcaneocuboid joints), arthrodesis of the talonavicular joint, and arthrodesis of the calcaneocuboid joint were simulated in a cadaver model, and the range of motion of each joint not involved in the simulated arthrodesis was measured with a three-dimensional magnetic space tracking system. The excursion of the posterior tibial tendon was also measured under all of these conditions. We found that any combination of simulated arthrodeses that included the talonavicular joint severely limited the motion of the remaining joints to about 2 degrees and limited the excursion of the posterior tibial tendon to 25 per cent of the preoperative value. Simulated arthrodesis of the calcaneocuboid joint had little effect on the range of motion of the subtalar joint, and it reduced the range of motion of the talonavicular joint to a mean of 67 per cent of the preoperative value; a mean of 73 per cent of the excursion of the posterior tibial tendon was retained. After simulated arthrodesis of the subtalar joint, a mean of 26 per cent of the motion of the talonavicular joint, 56 per cent of the motion of the calcaneocuboid joint, and 46 per cent of the excursion of the posterior tibial tendon was retained. The talonavicular joint is the key joint of the triple joint complex. The talonavicular joint had the greatest range of motion, and simulated arthrodesis of this joint essentially eliminated motion of the other joints of the complex.