Somewhere south of Newcastle, amid the wide-open prairie and rolling hills, rests a mass grave. A femur here. A tooth there. A tip of a tail barely poking through the ground somewhere else.

The cause of death is unknown. It could have been a lightning strike, disease or an attack by a band of marauding T. rexes.

The victims: At least four U-Haul-sized, plant-eating triceratopses.

Paleontologists worked for two months this summer and found 250 bones. Only 950 more to go.

On a hot day in mid-August, one paleontologist held up a pterygoid for inspection. A pterygoid is a portion of a triceratops palette in its skull. It’s roughly the size of a loaf of bread, and had never previously been found complete and alone.

Some portions measure only a single millimeter thick. Removing it from the earth was a painstaking task. The ground was hard and the bone weak.

“There are maybe 10 people in the world who care about this bone,” said Matt Larson, a paleontologist for the Black Hills Institute of Geologic Research.

“And four are here.”

What it represents is entirely different. That pterygoid could belong to the most complete triceratops skeleton ever found — something many more people care about.

The institute’s research team is unearthing what is, at minimum, four triceratops skeletons. Scientists believe the collection could be the key to answering how one of the prehistoric world’s unique vegetarians lived and died.

Experts always thought the triceratops was a loner. Skeletons were never found grouped together like some other horned dinosaurs, said Peter Larson, founder of the Black Hills institute.

Remains were most often limited to a skull in one place or a femur in another. They must have lived alone, because they all seemed to die alone.

This new find, hidden beneath layers of sand, silt and lignite, could tell a very different story of the life of the world’s best-known three-horned dinosaur.

Triceratopses roamed the wetlands of western North America 67 million years ago. It was the end of the Cretaceous period and shortly before the extinction of dinosaurs. Water prevented their movement west to other continents, and an inland seaway separated them from the east.

Only three skeletons have been found with more than 50 percent of their bones. Two were in Wyoming, one in North Dakota.

The most complete skeleton, a dinosaur named Lane, only has 75 percent of its original bones. Kelsey and Raymond, the other two triceratopses, are about 50 percent complete. The ones you gawk at in museums are actually collages of bones from many animals, some of which may not even belong to a triceratops, Larson said.

Why have so few bones been found of a dinosaur that measured 20 feet long and 8 feet high at the shoulders? Blame Tyrannosaurus rex.

“T. rex would not just eat the flesh from triceratops, it would eat a good share of the carcass as well,” Larson said. “It would ingest bones and everything else in some instances.”

Other creatures would likely scavenge the parts the T. rexes didn’t eat, acting like prehistoric coyotes and vultures.

The Newcastle bones may have been fed upon after the triceratopses died. But, until the scientists find teeth marks or actual T. rex teeth at the scene, they won’t know for sure.

And they may never know what killed the beasts.

Late-August 2012, an amateur paleontologist approached rancher Donley Darnell. He’d found dinosaur bones on Darnell’s land while looking for them on a nearby ranch.

Darnell hadn’t given him permission to be there, and he wasn’t going to let the man take the bones. What was buried was more than a lone collector could handle, he said. Instead, he called Larson.

The rancher is no stranger to fossils, museums or collectors. He has collections of invertebrate skeletons — mostly shellfish — at natural history museums in Denver and New York.

Records show dinosaur bones were cherry-picked from the area as early as the 1910s.

The way the dirt settled in the area over millions of years created an environment perfect for dinosaur preservation. Almost like a delta, the land sank, filled in with sediment and then sank again, perhaps many more times. At one point, as much as a mile of earth covered the triceratops bones, Larson said.

Land eroded away as the Black Hills rose and left some bones exposed and others covered by only feet of soil.

“More rapid sedimentation would be able to preserve moments in time,” he said. “They’re snapshots in history.”

Larson would know. He started the Black Hills Institute of Geologic Research in 1974 in neighboring Hill City, S.D. Since then, he’s helped uncover Sue, a famous T. rex, and two of the most complete triceratopses.

But, it wasn’t the triceratops site that first interested the Black Hills Institute. Darnell showed them a few T. rex bones he’d found in another location on his land. The paleontologists jumped at the chance of a T. rex, calling a museum in the Netherlands, the Naturalis Biodiversity Center, that was looking for a T. rex skeleton.

The two institutions partnered on the T. rex dig. When the scientists couldn’t find more of the meat-eater’s bones, they switched to the triceratops site.

Larson and his crew wrapped up digging for the year in mid-August.

On one of the last days of the dig, the paleontologists exposed two frills, the iconic shields behind the triceratopses’ heads, a few ribs, the pterygoid and a tooth.

Each solid-looking bone is actually fractured into thousands of tiny pieces from the compression of tons of earth. The scientists clean them with small knives and paintbrushes and squeeze glue into the cracks. Then they cover the entire bone with another type of glue, flip it over and do the same to the other side.

Some bones are so intertwined the team takes them out in large blocks.

When they started digging in early May, it looked like they had three triceratopses: two adults and one youth.

They soon realized they wouldn’t be done in June as planned. Perhaps the end of August, Larson speculated.

They just kept finding bones, including another two femurs. The site now has at least three adults and one juvenile — a gangly teenager, all legs and no real body size.

“We have this big mass of bones we just can’t separate,” Larson said. “We will finish it next summer or spring.”

If the bones keep creeping into the hillside, it may take even longer.

The real work begins when the bones are all removed and in a lab.

Each triceratops has about 300 bones. To bring one animal from field to display takes about 20,000 hours, said Matt Larson, one of two of Peter Larson’s sons who work for the institute.

The skeletons haven’t been sold, yet. They will likely go to Naturalis, a partner in the dig.

“Naturalis will expand its dinosaur hall, and a triceratops skeleton — or maybe even a little herd — would certainly be an interesting addition,” wrote Anne Schlup, a paleontologist for Naturalis, in an email.

Darnell, the rancher, doesn’t care as much where the skeletons end up, as long as they’re someplace public where people can see them.

“And then maybe we will have some answers,” he said.