Limb lengthening of the left femur was performed in 12 sheep, 7-8 months of age. After weakening of the cortex by drill holes, a controlled fracture was created in the diaphysis. The bone segments were fixed in reduced position using a unilateral external fixation device. The fixator was left in neutral position for 10 days postoperatively when lengthening started. By means of daily distraction at a rate of 1.0 mm once a day for 3 weeks, an average lengthening of 1.9 cm (11.9%) was obtained. Two weeks after distraction had been discontinued, distinct callus bridging of the lengthening zone was confirmed radiographically in all animals. At this time the rigid state of fixation was converted to a system allowing axial compression by loading in half the animals. The other half of the animals served as controls with the fixator remaining unchanged. The external device was left on until 5 weeks after the end of distraction. At this time the animals were killed. Both femora of all animals were tested mechanically in torsion. The average ultimate torsional strength of the elongated femur compared with control was 71% in the group with dynamic axial fixation and 65% in the control group. No significant difference in relative strength of the tested bones in the two groups was found. We conclude that intermittent axial compression by loading during consolidation had no significant effect on the bone-healing process evaluated by radiography and mechanical testing.