As scientific puzzles go, this one is a bit of a monster.

"There hasn't even been agreement which way is up," said Lauren Sallan, an assistant professor of earth and environmental science at the University of Pennsylvania.

She is talking about the 300-million-year-old Tullimonstrum gregarium, commonly called the Tully monster, a soft-bodied marine creature with a pincers mouth and eyes on the ends of stalks.

Scientists have debated where the weird beast fits into the tree of life ever since its partly fossilized remains were discovered in 1958. Last year, a pair of research teams made waves by announcing that Tully was a vertebrate, with a primitive sort of backbone.

Not so fast, countered Sallan and a team of colleagues from England and France, writing in February in the journal Palaeontology. There is much we do not know about Tully, but the best evidence suggests it is more akin to certain prehistoric mollusks or perhaps early arthropods, whose modern cousins are insects and lobsters, she said. No backbone.

Spirited disagreement in paleontology is nothing new, given that its practitioners must work with the limited offerings nature has decided to preserve. When prehistoric animals are preserved as fossils, the remains generally include just a fraction of the complete animal. Untold numbers of long-gone species were never preserved at all.

The point is to determine how the various branches of life are connected, allowing scientists to understand evolution, extinction, biomechanics, and natural processes such as climate change.

The two teams of scientists arguing that Tully monster was a vertebrate based their arguments on several lines of evidence. One study focused on the animal's eye-like appendages, which the authors argued were characteristic of other kinds of creatures that had a backbone. In the other study, the authors contended that Tully's remains contained a central rod-like structure called a notochord – a primitive version of a backbone.

Sallan and her coauthors dismissed this feature as a light-colored stain. They say it is not a notochord for a variety of reasons, among them that internal structures are not preserved in other fossils found at the site, the Mazon Creek in Illinois. The fossils are preserved as  "outlines and flattened forms in nodules of ironstone from several areas," according to the Illinois State Museum.

As for the eye-like structures, Sallan noted that light-sensitive organs have evolved multiple times in the animal kingdom. The eyes on the Tully monster are more akin to "pigment cups," which are not characteristic of vertebrates, she and her coauthors wrote.

Victoria McCoy, the lead author of the 2016 paper that called the rod-like structure a notochord, made that call after eliminating other anatomical features that could account for the light-colored line running down the center of the Tully fossils.

Then at Yale University, and now a postdoctoral researcher at the University of Leicester in England, she is not persuaded by Sallan's new analysis. But she welcomed the debate.

"You can never say something is absolutely certain," McCoy said. "Eventually, over time, hopefully there will be a scientific consensus."