The aim of this study was to determine the anatomical restraints to dislocation of the second metatarsophalangeal joint and to assess the biomechanical efficacy of a technique that is commonly used to stabilize this joint. Cadaveric feet were disarticulated at the ankle, with preservation of the long flexor tendons at the medial malleolus. The hindfoot was transfixed to an aluminum jig, and a contoured nylon block was secured to the dorsum of the second metatarsal. A Kirschner wire was passed transversely through the proximal phalanx and was attached to a wire loop through which a constant vertical displacement was applied with a universal testing machine. A preload of five newtons was applied, followed by a constant displacement of two millimeters per minute, and the load-displacement curves were measured. The volar plate and the collateral ligaments were divided in five feet each. In another ten feet, both of these structures were divided simultaneously. Each load cycle was repeated four times. The force required to dislocate the joint, in the position in which testing was performed, was reduced by a mean of 30 per cent when the volar plate was divided and by a mean of 46 per cent when the collateral ligaments were divided. Division of both of these structures created an unstable joint, which dislocated at an applied load of five to ten newtons. The metatarsophalangeal joint was then repaired with use of a flexor tendon transfer in all twenty feet. This repair technique restored the load-displacement curves to that of the normal toe.