IntroductionOur biomechanical link segment models of human gait assume rigid segments. The primary confirmation of that assumption has been a power balance within each segment; Robertson and Winter (1980) demonstrated in a planar analysis that all but the foot segment during stance satisfied a power balance. The purpose of this paper is to reinvestigate the power imbalance within the foot in a 3D analysis and using a more precise measurement system than has been used previously. The interpretation of this imbalance relates to the absorption of energy by the soft tissue under the foot and the muscle function at the metatarsal-phalangeal (M-P) joints (which are not part of the rigid foot segment).Theory and MethodologyThe rigid segment is defined as the segments between the ankle joint and the M-P joints. The power balance for the foot segment equates the rate of change of energy of the foot with the sum of the translational and rotational powers at the ankle joint:dE f = F . V + M . ω + P i where: F is the ankle reaction force, V is the ankle translational velocity M is the ankle moment, ω is the angular velocity of the foot P i is the power imbalance (residual)A common ankle joint is defined in conjunction with the leg segment markers by a calibration protocol while the subject stands in the anatomical position. This common ankle joint differs during walking, thus the joint is defined as the mid position between the independently defined ankle joints. Similar protocols were used to define the knee and hip joints. Two subjects were studied while walking barefoot and shod at slow, natural and fast cadences. Both limbs were analysed.ResultsThe power in balances in both left and right leg and thigh segments averaged 0.3 W over both stance and swing for all three walking speeds; the foot segment during swing also had almost perfect power balances. Thus our assumptions of rigid thigh and leg segments and foot segment during swing are justified. However, the 3D power imbalance during stance had three distinct phases. Figure 1 plots dE f , F . V and M . ω for one of the shod trials. Phase A was negative and was associated with the compression of the fat pad under the heel and ball of the foot during weight acceptance. Phase B was negative and began at heel off and was associated with extension of the M-P joints and compression of the fat pads under the phalanges. This negative power is both active (due to lengthening of M-P flexors) and passive (due to compression of the phalanges fat pads). Phase C, just before toe-off was associated with the final stage of push-off when the M-P joint is flexing and the M-P flexors shorten (flexor digitorum longus and hallucis longus). The work done during each of these phases were compared with the negative work done by the ankle dorsiflexors as the foot is lowered to the ground after heel contact.Discussion and ConclusionsThe major finding is the large energy absorption by compression of the fat pads during weight acceptance (Phase A) compared with the minor active absorption by the ankle dorsiflexors/invertors. The tibialis anterior is traditionally credited with the major shock absorption during weight acceptance; this no longer is seen to be true. Also, the role of the M-P. flexors is not insignificant as an energy absorber and generator during the latter half of stance. The push-off work by these flexors (Phase C) is about 35% of the large work by the plantarflexors during push-off (Winter, 1983).