The aim of this study was to quantify how elite high jumpers used their free limbs in a competitive high jump and to estimate the contribution that these made to vertical take-off velocity. This was achieved by analysing the competitive performances of six elite male high jumpers using 3D motion analysis and assessing limb function using the relative momentum method. The mean peak relative momentum of the arm nearest to the bar at take-off was 9.4 kg m s(-1), while that of the arm furthest away from the bar was 11.3 kg m s(-1) and these did not differ significantly. The free (lead) leg reached a mean peak relative momentum of 20.9 kg m s(-1). At touch-down the free leg had a large positive relative momentum that was offset by the negative relative momentum of the arms, although their combined value still remained positive. The mean combined free limbs' relative momentum at touch-down was 13.8 kg m s(-1) and reached a peak of 37.6 kg m s(-1). The difference between these two values amounted to 7.1 % of whole-body momentum, which was judged to be the amount by which the free limbs contributed to performance. The arms had a greater influence on performance than had the lead leg. This was because the lead leg increased its relative momentum little during the contact period while the arms had an initial negative value that increased markedly after touch-down. The compressive force exerted by the motion of the free limbs, estimated by the change in the combined free limbs' relative momentum, reached a mean peak of 366 N and was greatest at 37% of the contact period. It was concluded that to maximize the contribution the free limbs can make to performance, given the restraints imposed on technique by other performance requirements, the arms should have a vigorous downward motion at touch-down to make the most use of the high (but little changing) relative momentum of the lead leg.