More than half a century after amateur fossil collectors first cracked open iron-stone nodules along Illinois strip-mine spoil heaps and found something inside that looked like nothing else on Earth, Tullimonstrum gregarium still defies classification. The creature, roughly 300 million years old, has been called a vertebrate, an invertebrate, a mollusk, an arthropod relative, and a thing unto itself. Two high-profile studies published in Nature in 2016 seemed to settle the matter by declaring it a vertebrate related to lampreys, but subsequent research has pushed back hard, and the debate is arguably more open now than it was before those papers appeared.
What the Tully Monster Actually Looks Like
The animal is genuinely strange, even by the standards of Carboniferous sea life. Its body was elongate and soft, segmented like a worm but ending in a two-lobed tail fin. A long, narrow proboscis projected forward from its head, tipped with a pincer-like set of tiny teeth. Most striking of all, its eyes sat on the ends of a rigid transverse bar that stuck out sideways from the head, giving the creature a hammerhead-shark silhouette in miniature. The combination of features is what makes the animal so difficult to place: no living group of animals has that exact package, and no other fossil does either.
The formal scientific name honors Francis Tully, the amateur collector who found the first specimen in 1958 in the Mazon Creek fossil beds southwest of Chicago. The species name, gregarium, reflects how common these fossils are. Thousands of specimens exist in museum collections, which is unusual for a soft-bodied animal and a big part of why researchers keep returning to it with new techniques.
How a Soft-Bodied Animal Survived 307 Million Years
Almost everything we know about the Tully Monster comes from one place: the Mazon Creek Lagerstätte in northeastern Illinois. A Lagerstätte is a fossil deposit with unusually good preservation, and Mazon Creek is one of the best in the world for late Paleozoic life. It preserves not just bones and shells but soft tissues, pigments, and delicate structures that vanish in nearly every other geological setting.
The secret is siderite, an iron carbonate mineral. When organisms died and sank into the seafloor mud roughly 307 million years ago, bacteria broke down their tissues. That microbial activity triggered a chemical chain reaction in the surrounding sediment: sulfate-reducing bacteria consumed available sulfate quickly, coating tissues in a thin layer of pyrite, and then methane-producing microbes generated bicarbonate that cemented siderite around the remains before the sediment could compact and crush them flat. The result is a three-dimensional mold locked inside a hard, rounded nodule, sometimes preserving details down to individual cell layers.
This preservation pathway is critical to the Tully Monster debate because it means researchers are not looking at the animal itself but at mineral replacements and impressions of its tissues. Interpreting what a particular dark smear or raised ridge on a split concretion actually represents in terms of original anatomy is where much of the disagreement begins.
The Case for a Vertebrate
In 2016, two papers published in Nature made the strongest case yet that the Tully Monster belonged among the vertebrates. One study, led by Victoria McCoy, examined over a thousand specimens and used synchrotron X-ray imaging to reinterpret the animal’s internal anatomy. The team identified a structure running along the body as a notochord (the flexible rod that defines chordates), spotted what they interpreted as vertebra-like elements called arcualia surrounding it, and found what looked like a single nostril and tectal cartilages in the head region. Their phylogenetic analysis placed Tullimonstrum as a stem lamprey, a primitive relative of modern jawless fish.
The companion study focused on the eyes. Using scanning electron microscopy, researchers found that the dark structures at the tips of the transverse bar preserved a retina made up of two distinct types of melanosomes, tiny pigment-containing bodies, arranged in layers. Among living animals, having two melanosome shapes arranged in layered retinal pigment epithelium is a trait found exclusively in vertebrates. The authors argued this was strong evidence that the Tully Monster had vertebrate-grade eyes, and therefore was itself a vertebrate.
A 2020 chemical study added another layer of support. Researchers used Raman spectroscopy to analyze the molecular signatures of preserved soft tissues across a range of Mazon Creek fossils, both known invertebrates and known vertebrates. Invertebrate tissues showed chemistry consistent with fossilized chitin, a polysaccharide found in arthropod exoskeletons and similar structures. Vertebrate tissues showed chemistry consistent with fossilized proteins. When the Tully Monster’s tissues were analyzed, they grouped cleanly with the vertebrates in the resulting chemical space, not with the invertebrates.
The Case Against
The vertebrate interpretation did not go unchallenged for long. A 2017 paper examined the same anatomical features from a different angle and concluded that several key identifications were problematic. The structures interpreted as myomeres (the segmented muscle blocks characteristic of fish) did not match the expected pattern for vertebrates. The supposed tri-lobed brain, tectal cartilages, and fin rays were also reinterpreted as not comparable to vertebrate equivalents. The authors argued that the phylogenetic analysis placing Tullimonstrum among vertebrates worked only because the dataset was restricted to chordates with limited outgroups. When you only offer the computer chordate options to choose from, it will pick the closest chordate match, even if the real answer lies somewhere else entirely.
The same critique noted that certain structures you would expect to find preserved in a Mazon Creek vertebrate were missing. Otic capsules (the bony or cartilaginous structures of the inner ear), for instance, preserve well in other fossil vertebrates from the same deposit but are absent in the Tully Monster. Body pigment patterns typical of Mazon Creek fish are also absent. These gaps are hard to explain if the animal really was a lamprey relative.
A more recent study used synchrotron tomography to build three-dimensional models of Tully Monster anatomy, getting past the limitations of looking at flattened specimens in split concretions. The resulting 3D dataset cast further doubt: the structures previously identified as vertebrate hallmarks looked less convincing when viewed in full spatial context. The authors concluded that Tullimonstrum might be a non-vertebrate chordate (something like a lancelet) or could even belong among the protostomes, the vast branch of animal life that includes insects, mollusks, and worms.
The Menagerie of Alternative Identities
If the Tully Monster is not a vertebrate, what is it? Over the decades, researchers have proposed a remarkable range of possibilities. It has been compared to arthropods, including distant relatives of anomalocarids, those bizarre Cambrian predators. It has been likened to various types of worms. It has been placed tentatively among tunicates (sea squirts), lancelets, and vetulicolians, an obscure group of early deuterostomes.
One of the more creative proposals links it to heteropod gastropods, a group of pelagic sea snails. Modern heteropods like Pterotrachea coronata, sometimes called the “sea elephant,” are translucent, elongate, free-swimming predators with a proboscis-like snout and large eyes. The body plan has a superficial resemblance to the Tully Monster’s layout, though the similarities may be coincidental rather than genealogical. The heteropod hypothesis remains a minority view, but it illustrates how wide open the field of candidates is.
The deeper problem is that the Tully Monster’s combination of features does not fit neatly into any existing group. A pincer-mouthed proboscis, stalked eyes on a rigid bar, a segmented soft body, and a vertebrate-style tail fin: pick any two of these and you can find an animal group that has them, but no single group has all four. This is what makes Tullimonstrum a genuine problematicum, the technical term for an organism that resists classification.
Why Taphonomy Makes Everything Harder
A major thread running through the entire debate is how much of what we see in a Tully Monster fossil actually reflects the living animal’s anatomy, and how much is an artifact of the preservation process. Decay experiments on modern chordates and invertebrates have shown that soft tissues undergo complex transformations after death: structures collapse, shift position, and merge into one another in ways that can be deeply misleading. Eyes lose their layered structure. Cartilage warps. Muscles disintegrate at different rates depending on their position in the body.
These taphonomic filters mean that even when researchers agree on what they are looking at in a specimen, they can disagree on what it originally was. A dark band might be a compressed notochord, or it might be a gut tube that settled into a similar position after the body deflated. A layered structure in the eye region might preserve genuine retinal architecture, or it might be a mineralization artifact that happens to mimic one. The fact that siderite concretions preserve soft tissues in remarkable detail does not mean they preserve them without distortion.
Researchers studying decay sequences have built atlases documenting exactly how different anatomical features transform and disappear during decomposition of modern lampreys, lancelets, and other relevant animals. These atlases help set expectations for what a fossil vertebrate “should” look like after millions of years, but applying them to the Tully Monster requires first assuming which group it belongs to, which is the very question in dispute. The reasoning risks becoming circular.
What the Tully Monster Can Tell Us About Swimming
While the identity debate continues, some researchers have taken a different approach: instead of asking what the animal was, they have asked what it could do. A computational study used both 3D and 2D fluid dynamics simulations to model how the Tully Monster would have moved through water. The results suggested that the eyebar and proboscis were not just sensory and feeding structures but also played a role in hydrodynamics. The tail fin complex could generate pressure differences consistent with propulsion, but the overall picture was of a slow swimmer with a tendency to sink. The animal probably spent much of its time in the lower water column rather than cruising near the surface.
This matters because it constrains the ecological role the creature could have played. A slow, descending swimmer with a pincer-tipped proboscis was probably not chasing fast-moving prey. It may have been a bottom-feeder or a slow-pursuit predator picking off small, less mobile organisms. The Mazon Creek ecosystem was rich with potential prey, from small crustaceans to soft-bodied worms, and the Tully Monster’s abundance in the fossil record suggests it was successful at whatever it ate. While the broader Mazon Creek food web included serious predators capable of crunching horseshoe-crab relatives, the Tully Monster’s build suggests it occupied a humbler niche.
Why Thousands of Specimens Have Not Settled the Question
One of the most counterintuitive aspects of the Tully Monster story is that the fossil is not rare. Thousands of specimens sit in museum drawers and private collections, far more than exist for most soft-bodied Paleozoic animals. You might expect that with so much material, the anatomy would be settled by now. But abundance has not translated into clarity, for a few reasons.
First, the fossils are all from a single deposit. Every known Tully Monster comes from Mazon Creek, which means every specimen has been subjected to the same preservation pathway with the same potential biases. There is no independent locality to provide a different taphonomic window on the same animal. If the siderite concretion process systematically distorts a particular structure, every single specimen will show that same distortion, and researchers have no way to detect it by comparison.
Second, the quality of preservation varies enormously from specimen to specimen. Some show clear detail in the proboscis but nothing useful in the body. Others preserve the tail beautifully but the head is a smear. Building a complete picture of the animal requires compositing information across many specimens, and reasonable people can disagree about which specimens are showing real anatomy versus artifacts.
Third, the animal has no close living or fossil relatives that everyone agrees on. In most classification disputes, researchers can point to a “nearest neighbor” in the fossil record and say, this animal clearly belongs in the same group as that one. The Tully Monster has no such anchor. Every proposed relative is itself a hypothesis, which means the usual shortcuts for resolving ambiguity do not work.
Chemistry Versus Morphology
The chemical evidence from Raman spectroscopy is worth dwelling on because it represents a fundamentally different approach to the problem. Rather than trying to interpret shapes and structures in fossils, which are subject to all the taphonomic complications described above, the chemical approach asks: what were these tissues made of when the animal was alive? If the fossilized soft tissues retain molecular signatures of their original composition, that composition can point toward vertebrate or invertebrate identity regardless of what the shapes look like.
The 2020 study found a clean separation between known invertebrates and known vertebrates in the Mazon Creek collection based on their tissue chemistry, with invertebrate tissues retaining polysaccharide (chitin-derived) signatures and vertebrate tissues retaining protein-derived signatures. The Tully Monster fell on the vertebrate side of that divide.
Critics have noted that chemistry alone cannot settle the question. Some invertebrates have protein-rich tissues (muscle, for instance), and the chemical preservation pathway itself could vary depending on the organism’s burial environment, not just its biology. Still, the chemical data remain one of the strongest independent lines of evidence favoring the vertebrate hypothesis, precisely because they sidestep the morphological ambiguities that fuel most of the disagreement.
Illinois and Its State Fossil
The Tully Monster was designated the official state fossil of Illinois in 1989, making it one of the few state fossils that scientists cannot confidently assign to a phylum. It appears on the logo of the Field Museum in Chicago and is a point of pride for fossil collectors across the region. The Mazon Creek fossil beds, exposed primarily through coal mining activity, have been a collecting destination for amateurs and professionals alike since the mid-twentieth century, and the Tully Monster is the site’s most iconic find.
The cultural attachment matters for the science, too. Because so many specimens have passed through private collections over the decades, the total number of known Tully Monsters is difficult to pin down, and specimens vary widely in how well they have been prepared and documented. Museum collections represent only a fraction of what has been pulled from the ground. Researchers working on the animal often depend on amateur collectors willing to loan or donate specimens, and the quality of available material depends heavily on which collections they can access.
Where the Debate Stands Now
As of the mid-2020s, there is no consensus. The vertebrate camp can point to the melanosome evidence from the eyes, the chemical grouping with vertebrate tissues, and the original anatomical reinterpretations from 2016. The skeptics can point to the 3D studies showing that supposed vertebrate structures do not hold up under closer scrutiny, the absence of expected vertebrate features like otic capsules, and the argument that the phylogenetic analyses were biased by restricted datasets. Both sides have published in top-tier journals, and both sides have legitimate methodological points.
Some researchers have suggested the Tully Monster may represent something genuinely new: not a vertebrate, not a conventional invertebrate, but an animal from a lineage that has no surviving members and no other fossil representatives. If that is the case, trying to force it into an existing group may be the wrong approach entirely. The Carboniferous period was a time of extraordinary evolutionary experimentation, and not every body plan that evolved during that era left descendants. The Tully Monster may simply be a dead end that looks a little bit like several things but is not any of them.
New analytical techniques continue to emerge. Synchrotron tomography is improving in resolution. Chemical mapping methods are becoming more sophisticated. Machine learning approaches to morphological comparison are being developed. Each new tool offers a fresh angle on the same old nodules. Whether any of them will finally pin down what the Tully Monster was, or whether it will remain paleontology’s most famous unresolved question, is something only future research can answer.