Some 3.6 million years ago the now extinct Sadiman volcano erupted in Laetoli, Tanzania. It released a plume of ash into the atmosphere. This was the rainy season and the rains changed ash into mud. Elephants, antelopes, hares, giraffes, pigs, rhinos, as well as some bird species walked over the muddied terrain. Among the footprints were those from a pair of (and perhaps even three) hominins, walking side-by-side. A second eruption released more ash into the air covering over the footprints, preserving them as a layer of tuff.
And so the it remained for more than three-and-a-half million years.
Mary Leakey sent an expedition to investigate Laetoli in 1974. One afternoon in 1976, a group of paleontologists were passing the time by throwing elephant dung at each other. Admidst the mud flinging, palaeontologist Andrew Hill found himself standing atop the now eroded ash layer. Archaeologists set about painstakingly excavating the footprints. The layer was friable and crumbled easily. After years of meticulous excavation, the footprints were exposed in all their glory; the grand prize being the fifty metre trail left by the hominins.
They are perhaps the clearest evidence for the early adoption of bipedal walking in our lineage. The footprints are thought to belong to Australopithecus afarensis, the species which included the famous fossil Lucy. However, there has been some debate as to whether these tracks represent fully bipedal locomotion or were more similar to the bent-knee, bent-hip gait seen when modern chimpanzees adopt a bipedal locomotion.
In a study that recently appeared in the journal PLoS ONE, human subjects were asked to walk over a specially constructed walkway. The surface of the track was covered with a damp sand, to mimic the soft underfoot condition that existed at Laetoli when the footprints were laid-down. The subjects walked twice across the trackway and then a further two times assuming a bent-knee, bent-hip gait. Walking with a normal modern human gait produced foot impressions with nearly equal heel and toe depths. In contrast, the bent knee gait resulted in footprints with deeper toe impressions than heel impressions. When non-human apes walk bipedally, weight is transmitted from the heel, along the outside of the foot, with toe-off occurring around the middle of the foot. We on the other transmit weight along the heel to the ball of the foot, finally toeing-off with the big toe. This is the more efficient way to walk bipedally. The impressions from Laetoli best match the pattern made by modern humans.
However I would be cautious about drawing too many conclusions from this study. One major drawback of this study is that walking with a bent-knee, bent-hip gait is not a natural gait for us. The impressions left by modern humans walking with this posture are probably not exactly the same as the footprints that a chimpanzee would leave when walking upright. While this study suggests that these hominins walked with a gait similar to our own, there is still room for debate as to exactly how similar the footprints are to our own. Regardless of these drawback, this study is a step in the right direction (no pun intended).
Raichlen, D., Gordon, A., Harcourt-Smith, W., Foster, A., & Haas, W. (2010). Laetoli Footprints Preserve Earliest Direct Evidence of Human-Like Bipedal Biomechanics PLoS ONE, 5 (3) DOI: 10.1371/journal.pone.0009769