The consequences of the relatively short lower limbs characteristic of AL 288-1 have been widely discussed, as have the causes and consequences of the short limbs of Neanderthals. Previous studies of the effect of limb length on the energetic cost of locomotion have reported no relationship; however, limb length could have accounted for as much as 19% of the variation in cost and gone undetected (Steudel and Beattie, 1995; Steudel, 1994, 1996). Kramer (1999) and Kramer and Eck (2000) have recently used a theoretical model to predict the effect of the shorter limbs of early hominids, concluding that the shorter limbs may actually have been energetically advantageous. Here, we took an experimental approach. Twenty-one human subjects, of varying limb lengths, walked on a treadmill at 2.6, 2.8, 3.0 and 3.2m.p.h., while their expired gases were analyzed. The subjects walked for 12 minutes at each speed and their rates of oxygen consumption (VO2) over four minutes were averaged to estimate VO2. We also measured each subject's height, weight and lower limb length. Lean body mass and % fat were determined using dual-energy x-ray absorptiometry. ANCOVA with total VO2 at either speed as the dependent variable and total lean mass, % fat and lower limb length as covariates resulted in all three covariates having a significant positive effect on VO2 at p<0.01. Subjects with relatively longer lower limbs had lower locomotor costs. Thus the short lower limbs characteristic of some hominid taxa would have resulted in more costly locomotion, barring some physiological anomaly. The magnitude of this effect is substantial; Neanderthals are estimated to have had locomotor costs 30% greater than those of contemporary anatomically modern humans. By contrast the increase in lower limb length seen in H. erectus would have mitigated the increase in locomotor costs produced by the increase in body size.