Just beyond a shallow, narrow sea lay an open topography of grassy savanna, populated by plentiful game and few predators. This watery barrier—likely not more than five kilometers wide—would have been but a small obstacle for a group of modern humans accustomed to navigating African lakes with boats and rafts. But this short crossing, enabled by coincidental climate change, might have led the species—possibly for the first time—out of Africa and into Arabia, and eventually deeper into Asia, Europe and the rest of the globe.

After finding a trove of Paleolithic stone tools in what is today the United Arab Emirates (U.A.E.), a team of researchers now proposes that just such a pivotal journey across what is now the Red Sea occurred at least 125,000 years ago—about 75,000 years after Homo sapiens are thought to have evolved and tens of thousands of years earlier than they were thought to have left the African continent. And although small watercraft certainly helped, it was a trick of climatic shifts—a window of plentiful rains on the heels of a glacial period—that made the trip possible.

Direct human fossil evidence for such an early—and southeastward—migration is still lacking, however, the sand deposits around the stone tools suggest they have been buried 100,000 to 120,000 years. A middle Stone Age residence in this area would suggest that humans reached the Arabian Peninsula not from the more-northern Nile Valley 119,000 to 81,000 years ago or from the Mediterranean Sea's shores 65,000 to 40,000 years ago—as previous evidence has suggested—but rather directly from the Horn of Africa, and much earlier.

Even with "the confounding lack of diagnostic fossil evidence," says Chris Stringer, a professor of paleontology at the Natural History Museum in London and who was not involved in the research, the new archaeological work "provides important clues that early modern humans might have dispersed from Africa across Arabia, as far as the Strait of Hormuz, by 120,000 years ago." The new findings will be published in the January 28 issue of Science, and researchers think that the results could have broad implications for thinking not just about when and where humans first decamped from Africa, but also why and how.

"The mechanisms of getting out of Africa should be understood in a different way," Hans-Peter Uerpmann of the University of Tübingen in Germany and a co-author of the new study said in a telephone conference call with reporters on Wednesday. "Up until now we thought of cultural developments leading to the opportunity of people to move out of Africa. Now we see, I think, that it was the environment that was the key."

Flakes of Jebel Faya
The site where the tools were found, Jebel Faya, is about 65 kilometers from the coast of both the Persian Gulf and the Indian Ocean. Some 125,000 years ago, it would have been a grassy home to gazelles, ibex and wild asses. But before March 2006, when the first piece of a hand ax was identified, the site was known to local research teams as simply "a nice shady picnic place for a weekend without archaeology," Uerpmann said. But once striking stone tools started turning up, researchers realized they were on to something.

At the site, which is a collapsed rock shelter, excavators found three distinct layers of ancient tools (labeled A, B and C, with C being the oldest). "Assemblages A and B were similar, but assemblage C, the lowest, was radically different," said Anthony Marks, of Southern Methodist University in Dallas and research team member. The tools from the C group, which were dated to approximately 120,000 years ago, included denticulates, end-scrapers, foliates, hand axes and side-scrapers.

"We looked at what was in southeastern Arabia at that time, there was literally nothing,", Marks said during the Wednesday call. And, as Stringer points out, "the fact that artifacts in assemblage C at Jebel Faya do not resemble those associated with contemporaneous Homo sapiens [east of Egypt] signals yet more complexity in the exodus of modern humans from Africa."

Further analysis of the earliest group of tools—and the techniques used to make them—showed a strong similarity to known tools from that time period in East Africa. The presence of two techniques, Levallois flaking and bifacial reduction—which in the outer chunks of a rock are flaked off to separate a useable core—underscore the makers' African origins. "In East Africa and northeast[ern] Africa bifacial reduction is a constant part of the technological repertoire of people," Marks said. "In Africa they produce hand axes, but also what are called foliates—leaf-shaped bifacial pieces," he added, noting that "either you have a connection to East Africa or you have an independent invention of this technique from a group that has no history in the area."

And the later layers of tools, groups A and B, did not bear evidence of the use of these techniques. "These marked differences led us to conclude that assemblages A and B developed locally and in basic isolation," Marks noted.

Even though the tools were not accompanied by any direct H. sapiens fossil evidence, the research team is confident that they were not made by another more primitive human relative, such as Neandertals. "There's no evidence for any Neandertals south of that [Russian-Asian] temperate zone to the east," Marks said. Certainly another group cannot be ruled out but, he noted, that would have meant that an errant group of Neandertals "took a turn south, went several thousand kilometers into what at the time was desert," he said. "It seems to me a very difficult explanation and one that doesn't follow any reasonable logic."

Not everyone in the archeology field, however, is convinced that the stone tools in the C grouping have an undeniable connection to Africa. "I think the verdict here is ambiguous," says John Shea of Stony Brook University's Department of Anthropology, who was not involved in the research but is familiar with the paper. Given that all H. sapiens originated in Africa, he adds, "it is likely that either [the tools'] makers or their makers' ancestors came from Africa"—but not necessarily directly. "When the Earth offers up what you have long been seeking," he adds, "a good archaeologist needs to be on guard."

Marking the time of sands
Because the evidence for human occupation at the Jebel Faya is not biologically based—from bone or other animal or plant material—the samples could not be radiocarbon dated (aside from some shells that turned up in an overlaying layer, which were dated at some 9,700 to 10,400 years ago). The research team thus had to find a different means to date the tools. "Archaeology without ages is like a jigsaw with the interlocking edges removed," Simon Armitage, of Royal Holloway, University of London, and co-author of the paper, said on Wednesday. "You have lots of individual interesting pieces of archaeological information, but you can't fit them together to produce the big picture."

He used single-grain optically stimulated luminescence (OSL) dating to determine how long the surrounding sand grains had been buried. "When a sand grain is buried it's exposed to very low levels of naturally occurring radiation from the surrounding sediments," Armitage explained. The radiation separates electrons from their host atoms and entraps them in a different part of the crystal. Once exposed to sunlight again, the electrons are freed, "and in doing so, they emit a tiny amount of light called luminescence," Armitage said, likening the process to discharging a rechargeable battery. If collected without being exposed to sunlight, the sand grains' luminescence can be measured with OSL dating, revealing the time that has passed since it was last buried.

Many of the recent brushstrokes of human migration have been painted not with sand grains, but rather with the help of DNA analysis. Sampling of genetic material from people around the world has put the first major human exodus out of Africa at some 60,000 years ago. But as defenders of rock-hard stone evidence, researchers behind the new paper stand behind the accuracy of their findings. "Genetics can only tell us which event happened before another or after another, and the estimates of time are very rough," Uerpmann said. He said he was not surprised that their results contradicted DNA estimates and suggested that it will "stimulate thinking" about the genetic approach.

Nevertheless, the researchers acknowledge that fragmentary artifact-based evidence is not perfect for following early human dispersal around the globe. "To track them archaeologically would be extremely difficult," Marks said. "What we have been able to do here is a first step."

The findings in the U.A.E. do not preclude the Nile route out of Africa from being taken as well. "Our findings open a second way," Uerpmann said, adding that in his opinion the Arabia route "is more plausible for massive movements than the northern route."

A favorable climate
If early modern humans were on the verge of leaving Africa today, the Arabian Peninsula would not likely be the most enticing route. The current global climate is currently vastly different than it was 200,000 to 130,000 years ago. As a global ice age locked up much of the Earth's water in frigid poles, sea levels receded, narrowing straights practically down to large lakes. By the end of the glacial period, water levels in the Red Sea would have been more than 100 meters lower than they are currently.

As the glaciers began to thaw, weather patterns would have shifted, bringing the Indian monsoon system to drench Arabia with rains, bringing with them greenery and freshwater lakes and rivers. This transitional time, Adrian Parker, also of Royal Holloway and a member of the research group, explained on the Wednesday conference call, was the chance climatic opening during which it made sense for H. sapiens to endeavor crossing the then-small sea.

"This led to a brief window of time when sea levels were still low, and Arabia experienced a wetter climate," Parker said.

After the makers of the C group of tools arrived in Arabia, continued glacial melting would have raised sea levels again, making more trans–Red Sea trips less likely. But, Uerpmann noted, "another window opened shortly after this event, increasing monsoon during the full interglacial enabled hunter–gatherers to cross the desert separating south[ern] Arabia from southeast[ern] Arabia. A drop in sea level during the subsequent cooler period provided an opportunity to reach parts of Mesopotamia, extending much farther into what is now the Persian Gulf and the Iranian coast—and the Indian subcontinent could be reached [from there]."

The climatic conditions so favorable to human expansion did not last indefinitely. The Jebel Faya archaeological site has yielded evidence from the Neolithic, Bronze and Iron ages as well, but the trail of artifacts has a gaping hole between 38,000 and 10,000 years ago when the area would have once again become desertlike.

The Paleolithic period climatic cycles' influence on early human dispersal likely have many more discoveries to divulge. The lowering and rising of seas and the shifting of precipitation patterns "happened many times during the Quaternary [period], and this leaves a lot of possibilities for human migrations," Uerpmann said. "And keeping this in mind might change our view completely."