Dinosaur hemoglobin may have been discovered in fossils

Scientists have announced the discovery of damaged hemoglobin in the hollow, blood vessel–like structures of dinosaur fossils. The researchers used a novel technique they hope may put an enduring controversy to rest.

Twenty years ago, Mary Schweitzer, a paleontologist at North Carolina State University (NCSU), and colleagues reported in the journal Science that they’d found elastic, vessel-like structures in dissolved bits of a 68-million-year-old Tyrannosaurus rex femur unearthed in Montana.

“The vessels and contents are similar in all respects to blood vessels recovered from extant ostrich bone,” they wrote in the paper. (Like all birds, ostriches descended from dinosaurs.) But many of Schweitzer’s colleagues never fully embraced this discovery, even after she discovered similar structures in an 80-million-year-old Brachylophosaurus canadensis specimen, also from Montana.

There is a long-standing debate about whether dinosaur soft tissue persists over millions of years, with some paleontologists doubting this is even possible. Schweitzer believes these tissues could provide crucial details on dinosaur physiology, appearance, and genetic relatedness.

“I think the bias in the community is that these things can’t preserve,” she tells C&EN. “I mean, we’ve used over two dozen methods now to demonstrate the presence of organic molecules in deep-time fossils. And it still is not well accepted.”

In her new study, published recently in the Proceedings of the Royal Society A (DOI: 10.1098/rspa.2025.0175), Schweitzer took a different tack: using a form of Raman spectroscopy to find evidence of hemoglobin.

Resonance to the rescue

Schweitzer first reached out to physicist Hans Hallen, also at NCSU, who specializes in Raman spectroscopy, which uses a laser to illuminate a substance and then analyze the light that scatters off its surface.

But dealing with ancient remains is not simple. “Molecularly, fossils are a mess,” Hallen says, with different compounds at various levels of degradation that make the results difficult to interpret. So Hallen, Schweitzer, and their collaborators decided to use resonance Raman instead, which aims for specific molecules.

“If you use the precise wavelength of light that is absorbed by the molecule, it has the effect of making the Raman signal much bigger,” he explains. “This way, you can look for one specific molecular type.” To verify if the vessels found in the T. rex and B. canadensis fossils contain traces of blood, they decided to look for hemoglobin, the molecule that carries oxygen in the blood of most vertebrate animals.

Hemoglobin consists of four sets of a ring-shaped molecule surrounding an iron atom, the heme—which is where the oxygen binds—connected to a large, rounded protein, the globin. If these two parts were still connected, Hallen and colleagues determined, a green (532 nm) laser should be ideal to amplify their signal. That was exactly what happened in both fossils.

To double-check they weren’t just looking at heme-like structures left behind by bacteria, the researchers also tried a blue (473 nm) laser more likely to resonate with heme molecules not bound to proteins. This wavelength didn’t resonate much if at all, strengthening the earlier results.

The wavelength shifts in the scattered light also suggest that the hemoglobin isn’t entirely intact. Specific outer parts of the ring appear to be broken, which would be expected in such old material.

Probing preservation in color patterns

The study also unveils new information about the dark-light patterns detected in the ancient vessels. One of the detected-wavelength peaks suggests that the lighter areas may represent ferric oxyhydroxide (FeOOH) crystals of a mineral known as goethite, caused by the oxygenation of the central iron atom in the heme.

This may have absorbed enough oxygen to create alternating areas rich in oxygen, where crystals were formed, and poor in oxygen, where the iron atom may instead have driven a cycle of reduction and oxidation that created crosslinks between the proteins, resulting in stiffer, darker material.

“In this way, hemoglobin might actually have helped the preservation of this material,” says Schweitzer, who has long suspected this might be the case. “We’ve seen the beginnings of very similar processes in ostrich bones that we tried to artificially age, sometimes in the presence of additional hemoglobin, to study its impact.”

Interestingly, the T. rex samples have better-preserved heme and better-formed goethite than those of the much older Brachylophosaurus, which may indicate a change with age or be due to the different conditions in which the fossils—both found in sandstone—were buried.

Dinosaur debate persists

“I don’t think this work settles the debate on the preservation potential of blood in dinosaur blood vessels, but it does provide another piece of the puzzle,” says Valentina Rossi, a paleobiologist at University College Cork who studies the preservation of soft tissue in fossils and was not involved in this study. “More work will be required to validate this application in other fossils.”

“Resonance Raman spectroscopy seems to be very good at identifying hemoglobin in biological tissues and in forensic settings,” she adds. “But whether iron-rich organic matter not derived from blood can give false positive results needs to be tested further. I would have loved to see the resonance Raman signals of other organic matter found in association with the fossil bones.”

Such material was not collected in this case, as the bones were museum items found many years ago. Even so, she says, “I hope that researchers excavating new fossils will make sure to collect it for comparison.”

That’s the plan, Hallen and Schweitzer say.

“We did two dinosaurs in this work,” Hallen says, “but there’s at least half a dozen more of different types of dinosaurs of different ages buried in different environments to look at. It’s time to buckle down.”