Abstract: OBJECTIVE: To confirm that the midshaft dorsal cortex of the third metacarpal bone experienced higher compressive strains during fast exercise than the medial or lateral cortices, and that the strain peak occurred earlier in the hoof-down phase of the stride on the dorsal cortex than the medial or lateral cortices. DESIGN: Observations of a single horse. PROCEDURE: Strains were collected from a single, sound, 3-year-old Thoroughbred mare during treadmill exercise from rosette strain gauges implanted onto the medial, lateral and dorsal surfaces of the midshaft of the right cannon bone, simultaneously with data from a hoof switch that showed when the hoof was in the stance phase. RESULTS: Peak compressive strains on the dorsal surface of the third metacarpal bone were proportional to exercise speed and occurred at about 30% of stance. Peak compressive strains on the medial surface of the non-lead limb reached a maximum at a speed around 10 m/s and occurred at mid-stance. Peak compressive strains on the lateral surface varied in timing and size between strides at all exercise speeds, but remained less than -2000 microstrains. CONCLUSIONS: The timing of peak compressive strains on the dorsal cortex suggests a relationship to deceleration of the limb following hoof impact, so the main determinants of their size would be exercise speed and turning (as shown in previous experiments). This experiment confirms data from other laboratories that were published but not discussed, that peak compressive strains on the medial surface occur at mid-stance. This suggests that they are related to the support of body weight. The strains on the lateral cortex occurred at variable times so may be associated with the maintenance of balance as well as the support of body weight. Understanding the loading of the third metacarpal bone will help to determine causes of damage to it and ways in which the bone might be conditioned to prevent such damage.