Gerald Johnson displays the fossilized remains of a mastodon that were found at a creek near Yorktown, Va., in 1983. (Timothy C. Wright for The Washington Post)

At least 12,000 years ago, a mastodon with a toothache died on top of a pile of seashells beside a creek near what is now Yorktown, Va. The flesh decayed, leaving teeth and bones that were gradually covered by sediment from the creek. Over the years, the bones of other mastodons in Virginia dissolved. But this one was in just the right place to survive.

“This guy took me there,” Johnson said, speaking admiringly of a man who had a fourth-grade education and a wealth of curiosity. “He saw this piece of wood that shone in the water, that turned out to be the tooth.”

In 1983, a brick mason named Lawnell Hart was out hunting when he spotted an enormous tooth in the creek. Hart quickly showed the site to Gerald Johnson, a geologist at the College of William and Mary.

There was more than a tooth.

Virginia's first full mastodon?

Together, the bricklayer and the geologist found a molar, several rib fragments, half of a lower jaw and what Johnson described as “fully articulated bones of a foot.”

“I looked at this stuff and said, ‘Oh my God, ’ ” Johnson recalled.

“You find a tooth here, maybe a tusk there,” he said. “This is the first time we’ve ever found multiple parts east of the Blue Ridge.”

Johnson and Hart found some of the bones in the water; they spotted others exposed along the bank of the stream. They gathered the remains in a single day, confident that a formal excavation would yield even more. Johnson expected to begin digging right away for the rest of the skeleton.

“So I said, ‘Okay, let’s go to the owners,’ ” he said. “The owners of the land wouldn’t let us dig it up without paying money. So for 32 years it’s been dormant.”

A fossilized piece of a Chesapecten jeffersonius, a 4-to-5 million year old scallop, was found laying on the bottom of a small creek near Yorktown moments before this photograph was taken. Just a few feet from where the scallop was found, the first fossil bones of a mastodon found east of the Blue Ridge were recovered. (Timothy C. Wright for The Washington Post)

The site was a closely held secret for decades. Mastodon bones are valuable, and the two men worried about theft. Johnson, Hart and a few confidants waited and hoped for an opportunity to return.

Hart died in 2011. Johnson held on to the bones, occasionally showing them to his students and eventually taking the collection home when he retired from teaching geology. Finally, the land was sold and the new owners invited Johnson to return and search for the rest of the mastodon.

On a recent afternoon, Johnson trekked through the woods toward the site with a bucket in one hand, a battered U.S. Army entrenching tool in the other. He has used the collapsible spade for excavation since bringing it back from his deployment to occupied Germany in the 1950s.

As he walked, he lectured on fossils with measured cadence. (He began teaching geology during the Kennedy administration.) Arriving at the site, he poked a stick down a narrow hole beside a fast-flowing stream. The stick came up wet, and Johnson sighed. An unusually wet spring and summer resulted in a very high water table. The water level was only 12 inches beneath the surface. Four feet of soil lay between Johnson’s feet and the remaining bones. Until the groundwater receded, any attempt to remove the overlaying soil would result in a mud pit that would be impossible to work in.

Little digging could be done that day.

The dig is taking place on a shoestring budget. A local excavation company has offered to improve an access road to the site and remove the first layers of soil. Volunteers will assist with the more delicate work that follows. Free use of a ground-penetrating radar system was provided by the nearby city of Newport News. (The city uses this equipment to investigate historical sites.) Johnson is personally absorbing other costs associated with the project.

The American mastodon was probably once common across the North American continent. It was a distant relative of modern elephants and mammoths, having diverged about 27 million years ago. Mammoths specialized in chewing grasses with their broad, flat teeth and tended to be more common in the northern steppe regions of the continent while mastodons were probably forest-dwelling browsers of spruce trees and other conifers. They lacked the mammoth’s sloping back and high-domed head.

Johnson, a retired geologist from William and Mary, sits in a future exacavation site in a forest where the fossilized remains of a mastodon have been discovered near Yorktown along a small creek. (Timothy C. Wright for The Washington Post)

Johnson hunts to find the remaining bones of the first mastodon found east of the Blue Ridge. The sticks in the ground are bore holes to determine ground water levels. (Timothy C. Wright for The Washington Post)

About 15,000 years ago, mastodons and mammoths in continental North America began a gradual decline that resulted in their disappearance from the fossil record by about 12,000 years ago. (A small number of mammoths may have persisted on Russia’s Wrangel Island in the Arctic Ocean until as recently as 2,000 years ago.)

Camels, giant ground sloths, saber-toothed cats and cow-size beavers all faded away at the same time in North America. The causes of these extinctions are unclear. There is little evidence of widespread hunting of mastodons and mammoths by humans. Climate change may have been a factor. The discovery of these bones in eastern Virginia may help scientists understand how the mastodons in the region died out.

“This is an extremely important problem for early human colonization of the Americas, for understanding the subsistence practices of early humans and for understanding the nature and dynamics of terrestrial ecosystems of that time,” Daniel Fisher, a mastodon expert at the University of Michigan, said in an e-mail. “Understanding what happened back then is relevant to the general and more current problem of human impacts on landscapes and biota. On a practical level, much of our understanding of how the climate system works over longer spans of time comes from studying episodes of climate change in the past, and the Pleistocene/Holocene transition [11,700 years ago] is an especially important case to investigate.”

This is only the second site in Virginia where a group of mastodon bones has been found. Another site was excavated from 1992 to 1997 in Saltville, 300 miles away in the mountainous region near the state’s border with Tennessee. Aptly named, Saltville is the home of large deposits of salt that attracted wildlife, resulting in a concentration of fossils. Other than the Saltville find, Virginia has revealed only scattered mastodon bones, which don’t illuminate an animal’s life.

Patterns of light and sand on the bottom of a small creek near Yorktown. The bottom of this creek is littered with countless fossils as old as 5 million years old. (Timothy C. Wright for The Washington Post)

Beside the creek near Yorktown, a few dozen square yards have been carefully cleared, staked out and probed with narrow bore holes. Scans produced by ground-penetrating radar have shown that bone-size objects are waiting about five feet down. Four feet of soil will be dug out using heavy equipment, then the last foot will be carefully removed by hand.

Rain is normally mildly acidic and will eventually dissolve bones in the ground, but this mastodon met its end in an auspicious spot.

“This animal died on a streambed that had eroding shells,” Johnson said. “This produces an alkaline environment where the bones are preserved.”

The shells of oysters, scallops and other bivalves neutralized the acid of the rain,protecting the bones and teeth of the mastodon.

The animal’s bed of shells was created about 5 million years ago, when sea levels were higher and the area was underwater. The age of the bones themselves has not yet been determined. Carbon dating was attempted but proved inconclusive.

“We took one of the toe bones I collected and sent it in,” Johnson said. “And the collagen was not datable. Too much gone.”

Johnson holds open a small bag filled with the broken fossils from the toe of a mastadon found in the woods near Yorktown. (Timothy C. Wright for The Washington Post)

As more bones are found, it may be possible to date them. Cause of death is still a mystery, but Johnson has a theory.

“We do know that the animal was probably ill, because it had an impacted tooth,” Johnson said. “Maybe it got the infection and it died of the infection?”

An enormous cavity in the molar appears to line up with a crater in a piece of lower jaw. The tooth is badly worn down from years of chewing, suggesting that this was not a very young mastodon.

The animal’s exact age — and possibly the cause of its death — may be revealed if a tusk is found. Information about a mastodon’s diet and physical condition was embedded in its ever-growing tusks, as with the growth rings of a tree. If the Yorktown mastodon was unable to eat because of pain from the infection in its jaw, the final layers should show poor nutrition.

Much remains to be learned about the bones. Johnson, primarily an expert on geology and rock formations, has been operating on the theory that he is digging up a single mastodon. When he found the foot bones, they were positioned right next to each other in a way that seemed to be lifelike positions. Bones that remain arranged as they were in life are called “articulated” by scientists. But after viewing photographs of the bones, Fisher, a paleontologist, suggested that they may have come from two different mastodons.

Johnson holds the fossilized remains of a mastodon's humerus bone close to his body for scale. (Timothy C. Wright for The Washington Post)

The large bone closest to him is part of a jaw. Closer to the camera are the broken remains of a tooth. The dark object to the upper right is a toe bone. (Timothy C. Wright for The Washington Post)

“I am puzzled by an apparent size discrepancy between the [two toe bones],” Fisher said by e-mail. “I could have made an error in identifying them, or there could be more than one individual.”

“This might seem like something that could be judged visually without too much trouble — and in some cases, it can — but especially with foot bones, you need to know the anatomy to discern whether things are, or are not, in articulation. Most people see bones next to one another in the sediment and will report them as articulated. Of course, it’s important to get this straight for understanding the history of a site.”

Johnson accepted Fisher’s analysis.

“I thought a few years ago that there might be bones from more than one mastodon here,” Johnson said. “I have been very conservative in my conclusions. I think that Fisher is correct. I would cite him as the expert in this realm.”

Excavation is expected to continue throughout the summer and fall. Work may continue into 2016, but after waiting 32 years, Johnson hopes to be finished sooner than that.

“I’m 79 and three-quarters years old,” he said. “I want to see this thing dug up and finished before I’m finished.”

Landers is an author and freelance writer in Charlottesville.