Did Lucy live in the woods?

Humans may have walked first in the forest, not on the savannah, says Jerome Burne

There are plenty of mysteries about human evolution. Are we all really descended from one pre-human woman? How did we develop the power of speech? Until very recently, though, there was pretty wide agreement about our first step. This was standing upright and walking on two legs, or bipedalism as it's technically known. Look in any relevant textbook and you'll see the secular, triumphalist version of the Garden of Eden story.

Once, the old theory goes, we were like all the other apes, living in trees in the lush forest, picking fruit from the branches whenever we wanted it. Then came a change in the weather, the forests thinned out and we had to come down from the trees and survive in the harsher, more arid world of the savannah. But being thrown out of Eden was the making of us. We stood upright, we learnt to hunt, our brains grew and we went on to dominate the planet.

However, unlike religious myths that are impervious to facts, scientific stories constantly have to earn their keep - and the evidence that once underpinned the savannah story is looking increasingly shaky. "The savannah paradigm has been overthrown," says Phillip Tobias, senior palaeoanthropologist at the University of Witwatersrand, Johannesburg. "We have to look for some other explanation."

Only three years ago, palaeontologist Elizabeth Vrba of Yale University was able to confidently declare: "The lineage of upright primates was one of the founding groups of the great African savannah biota [the life of a region or period]." She had found that between about 2.5 and two million years ago - when hominids were starting to use tools - conditions got harsher and forest-living animals disappeared, to be replaced by ones suited to the savannah. Other researchers found similar evidence for global cooling around five million years ago - about the time humans were supposed to be going bipedal.

This fitted with the theory that walking upright happened before we started growing our big brains. Key evidence for this came from Lucy, the fossil found 20 years ago, and dated to 3.2 million years ago, who had a chimp- sized brain but already walked upright. It also fitted in with the famous footprints found frozen in volcanic ash at Laetoli, Tanzania - showing an early hominid family walking on two legs 3.5 million years ago.

But the theory that this change was climate-driven was looking increasingly shaky. Andrew Hill and John Kingston, also of Yale, looked at carbon isotopes in geological strata in an area of Kenya where many hominid fossils had been uncovered. Plants from arid periods leave different ratios of isotopes from those flourishing in wet soils. They found no evidence for a dramatic shift to grassland. Other new finds also undermined the savannah theory, such as early hominid remains along with pigs whose legs were adapted for life in heavy woods. Ocean-based research found evidence for a cooling in Africa only 2.8 million years ago - far too late to explain Lucy's new gait. Yet more evidence was suggesting that the ground Lucy was covering may not have been that dry after all.

The possibility that hominids were walking around upright in a forested area at least a million years before Lucy came dramatically alive last year when two new sets of fossils were found by Lake Turkana in Kenya. One, known as Australopithecus anamensis, was dated to 4.2 million years while the other, Ardipithecus ramidus, was estimated at 4.4 million years old.

Anamensis was walking upright, but according to a co-discoverer, Alan Walker of Penn State University, not exclusively on the savannah. The lake, he says, was much bigger than it is now and was fed by massive rivers that would have supported forests a mile or two wide on either bank.

The clincher should be ramidus. These hominids definitely lived in dense woods along with monkeys and antelope adapted for forests. But there is an academic cliff-hanger here. The team that found it are refusing to say whether it is bipedal until they have finished a full analysis and that won't be until 1998. If it is bipedal, the savannah story is certainly dead.

The great attraction of the savannah theory, apart from the way it seemed to fit the facts as far as they were known, was that it was a tale of plucky little hominids surviving against the odds. But if our distant ancestors weren't pushed into walking upright we are left with the considerable mystery of why they did do it. It's not as if it's a particularly good way of getting about, and osteopaths say our upright gait is the reason why humans generally are so prone to back problems.

There is no shortage of theories but they all have problems. One says that it was a development of what some chimps do when they are feeding on the ground - they reach up and pull down branches. Another suggestion is that it comes from chimps' aggression display, when they pull themselves up to full height to scare off an attacker. But that still doesn't answer the question of why we took these occasional behaviours to such an extreme.

If the drive wasn't feeding or fighting then perhaps it could be sex, or rather parenting. That's what Owen Lovejoy of Kent State University in Ohio suggests. He believes that what distinguished Lucy and her upright forebears was that dad helped out more with the kids, which would have improved their chances of survival. The parents stayed together and worked as a team and in return he got more sex.

Instead of just fending for himself the male would go foraging and bring back a selection of fruit and vegetables. But to do this he had to have his hands free - which walking upright gave him. What this scenario doesn't deal with is that the whole point of the savannah theory is that life was too easy in the forest.

The facts, then, are turning against the savannah scenario. But we also know that scientific theories have a cultural dimension. Perhaps in the face of the looming threat of global warming, we have suddenly become less keen on a theory which is so heavily based on the bracing effect of climatic change.