One of the most unusual fossils ever to be found are strange tall structures recovered across Nebraska, primarily in the state’s northwestern badlands and in neighboring parts of Wyoming. Known locally as Devil’s Corkscrews, each structure is the infilling of a left- or right-handed spiral or helix that can extend up to seven feet into the ground. At the deep end of the spiral, a tunnel extends sideways and up at an angle. These structures became exposed by weathering of the soft rock enclosing them on the sides of bluffs or ravines. They mainly occur in the fine-grained sandstones of the Harrison Formation, which dates from the Miocene epoch and are about 20 to 23 million years old.

It was paleontologist Erwin H. Barbour who first discovered them. “Their forms are magnificent; their symmetry perfect; their organization beyond my comprehension,” he wrote .

Barbour assembled a marvelous fossil collection at the University of Nebraska in Lincoln in the late 19th century. Ably assisted by his wife Margaret and with financial support from one of the university’s trustees, he built a foundational collection of fossil mammals from Nebraska, dating mostly from the Neogene, about 23 to 2.58 million years ago. Today, the University of Nebraska State Museum of Natural History is famous for its fossil treasures, which document the diversity of mammals large and small living when the grasslands of the mid-continent developed. Its most spectacular exhibition is a parade of the many extinct species of elephants that once roamed across what is today the midwestern United States.

While exploring the western part of Nebraska, Barbour collected dozens of examples of the giant spiral structures, reporting on them in 1892 and naming them Daimonelix (Greek for “devil’s screw,” often spelled Daemonelix ). Their origin was a mystery and there was nothing else like them in the fossil record. After first considering them as possible remains of giant freshwater sponges, Barbour surmised that the fossils of Daimonelix were the remains of plants, possibly root systems, because he had discovered plant tissues inside the helices.

A year later, the legendary American vertebrate paleontologist Edward Drinker Cope rejected Barbour’s interpretation of the fossils, noting that “the most probable explanation of these objects seems to be that they are the casts of the burrows of some large rodent.”

In the same year, the Austrian paleontologist Theodor Fuchs, an authority on trace fossils, independently arrived at the same conclusion. He noted “thus we are justified in viewing these strange fossils as really nothing more than the underground homes of Miocene rodents, probably related to Geomys [pocket gophers].”

But Professor Barbour would have none of this and published a critique of Fuchs’s analysis in 1894. Assuming that the rocks of the Harrison Formation were lake deposits, Barbour commented that “Dr. Fuchs’ gopher is left to burrow and build its nest of dry hay in one or two hundred fathoms of Miocene water.” (Fuchs had doubted that the surrounding rocks were lake deposits and interpreted the plant remains found by Barbour as hay stored by the burrow-maker.)

Another American paleontologist, Olaf Peterson, collected specimens of the Devil’s Corkscrews for the Carnegie Museum in Pittsburgh. He observed that they often contained skeletons of an ancient beaver, Palaeocastor, which was slightly larger than today’s black-tailed prairie dog. And so, Peterson supported Cope’s reinterpretation.

But Barbour vehemently defended his identification of the Devil’s Corkscrews as a kind of plant fossil. He responded to supporters of the rodent-burrow hypothesis, “If this is in truth the work of a gopher then it must stand as a lasting monument to the genius of that creature which laid the lines of his complex abode with such invariable precision and constancy.”

Fuchs and others interpreted strange grooves on the infillings of the burrows as claw marks left by the digging animal. In time most researchers, including Barbour’s former student and successor at the State Museum, C. Bertrand Schultz, considered the structures fossil rodent burrows.

For many years, no further research was undertaken on the identity of Daimonelix and the issue remained in a stalemate.

Enter Larry Martin, an expert on fossil mammals at the University of Kansas. In the early 1970s, Martin and his student Deb Bennett studied many of the Devil’s Corkscrews in the field and in the lab. Their research on Daimonelix , published in 1977, painted a completely new picture of these strange spiral structures and their origin.

By the time the Kansas researchers started their work, geologists had long rejected the lake deposit theory of the Harrison Formation and established that its fine-grained sediments were instead accumulated by wind under seasonally dry conditions quite similar to the prevailing conditions in western Nebraska today. These deposits not only preserved the Devil’s Corkscrews, but also abundant fossil plant roots and burrows made by insects and small mammals.

Martin and Bennett found that the incisor teeth of the extinct beaver Palaeocastor were a perfect match for the grooves on the infillings of the Devil’s Corkscrews. These tooth marks affirmed that they were, in fact, burrows, spiraling tunnels that the beaver Palaeocastor built mainly by excavating the soil with left- and right-handed strokes of its large, flat incisors. The animal also left claw marks, but they tended to be confined to the sides and bottom of the burrows. The initial burrow extended down as a tightly coiled spiral. At the bottom, the beaver started digging upwards at an angle of up to 30 degrees to create a chamber for itself. This portion of the burrow sometimes extended up to 15 feet.

The Daimonelix -building Palaeocastor sported large, flat incisors . It lived and, based on finds of bones of young beavers, raised its litters at the end of this straight chamber. The tall, tightly coiled spiral entrance forming the top portion of the burrow is now thought to be an ingenious method for helping to retain moisture and control temperature in the animal’s burrow.

Scattered clusters of the burrows of Palaeocastor are often found in great numbers. These clusters probably resembled the “towns” of present-day prairie dogs. Interestingly, other animals occasionally visited the burrows—including an extinct relative of martens and weasels, probably looking to make a meal of the burrow’s maker.

But what of the plant tissues that Barbour had discovered inside the burrows? To solve that mystery, Martin and Bennett noted that the rocks containing the Daimonelix burrows were laid down in a seasonally dry environment. Under such conditions, plants would have difficulty finding enough moisture to survive. But inside the Daimonelix tunnels there was much more humidity and moisture-seeking plants quickly grew their roots into the walls of the burrows. In fact, the growth was so abundant, the interior of the burrow walls would have to be cropped back by the beavers from time to time in order to maintain access. Since the rocks of the Harrison Formation contain a lot of ash from nearby volcanoes, rainwater flowing through the soil would become saturated with silica. Plant roots readily absorbed silica. Gradually, the root-lined walls became mineralized and eventually the entire burrow was filled in with silicified roots.

Mystery solved. What started out with the finding of curious fossils from the badlands of Nebraska led to a detailed reconstruction of an ancient ecosystem and the lives of some of its inhabitants. Every fossil carries this potential—to clue researchers into discoveries about the ancient environment and the plants and organisms that once thrived in it. As for Barbour, he apparently went to his grave denying that Daimonelix was a rodent burrow.

The Daimonelix burrow with a skeleton of its maker, the extinct beaver called Palaeocaster is on view in the new fossil hall "Deep Time" at the Smithsonian's National Museum of Natural History in Washington, D.C.