Thirty feet beneath the murky waters of the Aucilla River, about 40 miles southeast of Tallahassee, archaeologists found evidence of some of the earliest known humans in the Americas.
In submerged sediments dating back 14,550 years, a team of scuba-diving researchers uncovered six stone artifacts -- including knives and flaked pieces of rock -- at the underwater site known as Page-Ladson. They also pulled up a mastodon tusk with cut marks on it experts say were made when these ancient people butchered its carcass alongside a lake bed.
The findings, published Friday in Science Advances, provide the first indication communities of hunter-gatherers were living in the southeast United States 1,500 years earlier than many scientists previously believed, suggesting a new story line for when and how people first came to the Americas.
"We are getting enough data now, particularly on the East Coast, to know that people have been around here for a very long time," said Dennis Stanford, an anthropologist at the Smithsonian National Museum of Natural History who was not involved in the work.
From the 1930s through the early 2000s, the dominant narrative among archaeologists held that the first Americans were members of the so-called Clovis culture. These pioneers left their distinctive spearheads scattered across a region that now covers the United States, Mexico and northern South America, almost like archaeological bread crumbs.
Most researchers believed members of this group were the descendants of big-game hunters who followed their prey across the Bering land bridge from Siberia into Alaska around 13,000 or 14,000 years ago. From there, they migrated down into Canada and quickly spread across the enormous land mass to the south.
"If you disagreed with that you were relegated to being on the periphery of the science community," said Dennis Jenkins, senior research archaeologist at the Museum of Natural and Cultural History at the University of Oregon, who was not involved in the Florida discovery. "You were seen as a lunatic, or way out there on the fringe."
Over the past few decades, persuasive challenges to the Clovis-First narrative have begun to emerge.
In 1997, for instance, researchers confirmed an archaeological site in Chile held evidence of human activity dating back to 14,500 years ago -- a full millennium before the Clovis-First theory would allow. In 2002, Jenkins led an excavation that unearthed 14,300-year-old fossilized human feces in Oregon's Paisley caves. A handful of other pre-Clovis sites popped up, too, including one in Wisconsin and another in Texas.
The 14,500-year-old butchered mastodon at Page-Ladson is now among the two or three oldest archaeological sites in the New World, and the oldest in the American southeast.
"It's important because it adds another site to the very small list of well-dated pre-Clovis sites," Jenkins said. "We don't find much evidence of these people because there weren't very many of them. We're looking for a very, very few people in a huge, huge haystack."
The finding also adds a new wrinkle to another commonly held theory human hunters caused the rapid extinction of late Ice Age mammals such as mastodons, mammoths and giant ground sloths. All of these megafauna disappeared around 12,500 years ago.
"However humans and mastodons interacted, it took at least 2,000 years for the process of extinction to run to completion," said Daniel Fisher, a paleontologist at the University of Michigan who worked on the study.
Fisher was responsible for interpreting the suite of cut marks on the nearly 7-foot-long mastodon tusk found buried in sediment in one whole piece. His conclusion is the cuts were made as these ancient people tried to rip the tusk from its base, perhaps to gain access to edible tissue inside.
"A tusk this size would have held over 15 pounds of tender, nutritious tissue in its pulp cavity and that certainly would have been of great value," he said.
The excavations at Page-Ladson are not new. In the 1980s and '90s, another research team lead by archaeologist James Dunbar and paleontologist David Webb spent several seasons at the site. Using radiocarbon dating, they determined the site was more than 14,000 years old. But their findings, published in a book in 2006, were disregarded by the archaeology community at large.
"It was just impossible for them to accept at that time that there was anything older than Clovis," said Jessi Halligan, a geoarchaeologist at Florida State University who co-led the new work. "People said, 'Oh, it's an underwater site? We can't assess that. It's impossible to validate.'"
In 2012, John Ladson, who owns the Page-Ladson site, reached out to Michael Waters, director of the Center for the Study of the First Americans at Texas A&M University. Ladson asked Waters if he might want to pick up where Webb and Dunbar left off.
Waters brought in Halligan, who specializes in underwater archaeology, and together they led excavations over three seasons, from 2012 to 2014.
Field work was challenging. The Aucilla River is cold, and the water is so dark and murky divers can't see without using lights mounted on caver helmets.
They used trowels to dig -- five to 10 centimeters at a time -- and marked the site off with string, just like they would on land.
Wworking underwater also has some benefits. "It's almost a zero-gravity environment, so you can dig upside down and sideways -- whatever is the most convenient way to get in there," Halligan said.
Over three years, the team uncovered six artifacts in a layer of sand, fine gravel and mastodon dung as much as 15 feet below the riverbed. The artifacts were deposited next to what was once the western edge of a pond at the bottom of a sinkhole.
The radiocarbon dates came from bits of dung deposited in the sediment layers where the artifacts were found. The researchers also dated the sediments above and below the strata containing the pre-Clovis artifacts, just to be sure the layers hadn't mixed over time.
"The radiocarbon dating is impeccable," Jenkins said. "All things considered, I think the study puts the question of the validity of the site away."
Not everyone agrees, however. Don Grayson, an anthropologist at the University of Washington, said the team that first started excavating back in the '80s raised concerns that the organic material they dated might be contaminated by ancient carbon coming out of the Florida aquifer.
"If so, that could mean the actual age of the dated material could be substantially younger than the radiocarbon ages obtained from the site," Grayson said. "Until they address this, I see no reason to accept their radiocarbon dates as accurate."
Halligan responded that the team dated the remains of plants that acquired their carbon during photosynthesis, not from any later exposure to groundwater. She added that their pretreatment methods would have scrubbed any contamination resulting from exposure to groundwater.
Waters said his group's work meets the high standard necessary to demonstrate that a site predates Clovis.
"One, you need clear evidence of human activity, usually in the form of stone tools. Second, these tools must occur in a solid geological context. And third, these artifacts must be dated using a reliable dating technique," he said. "At Page-Ladson, we meet all three criteria."
Now the researchers hope to turn to other questions.
"I want to see if we can find other sites with more information and features to give us a more complete picture of how these people were living," Halligan said.