IE 11 is not supported. For an optimal experience visit our site on another browser.

Why did feathers evolve? A pterosaur fossil offers new hints

Pterosaurs, ancient reptiles and close cousins of dinosaurs, appear to have had branched feathers, similar to those of modern birds, a new study found.
Paleontologists found evidence of two types of feathers when examining the crest of a type of pterosaur, called Tupandactylus imperator. The simple feathers were likely dark, while the branched feathers were likely lighter-colored.
Paleontologists found evidence of two types of feathers when examining the crest of a type of pterosaur, called Tupandactylus imperator. The simple feathers were likely dark, while the branched feathers were likely lighter-colored.Julio Lacerda; Maria McNamara

A few years ago, Maria McNamara was invited to Brussels by fellow paleontologist Pascal Godefroit and presented with an intriguing opportunity. 

At the time, they were collaborating on a study of an amazingly well-preserved — and feather-covered — dinosaur. But Godefroit was also approached by a private collector keen to have scientists examine the partial skull of a Tupandactylus imperator, a type of pterosaur known for an enormous sail-like crest that extended from its head. Godefroit asked McNamara if she wanted to take a look.

Pterosaurs were ancient reptiles and close cousins of dinosaurs. Pterosaurs, dinosaurs and birds all belong to the group Avemetatarsalia and share a common evolutionary ancestor. 

“I remember looking at it and thinking, ‘God, we could do something very interesting with this,’” McNamara, a professor of paleontology at University College Cork in Ireland, said. 

In a study published Wednesday in the journal Nature, McNamara, Godefroit and their colleagues report that this pterosaur possessed two types of feathers, including branched feathers similar to those of modern birds. 

Close up of branched feather.
Close up of branched feather.Maria McNamara

The pterosaur suggests feathers emerged around 250 million years ago through the common ancestor of dinosaurs, birds and pterosaurs -- and shifts the origin of feathers to 100 million years earlier than previously thought. 

“To me, these fossils seal the deal — pterosaurs really had feathers,” said Steve Brussate, a professor of paleontology at the University of Edinburgh who peer-reviewed the study.

“Seeing photos of this fossil blew me away,” he said. “Feathers aren’t just a bird thing, or even just a dinosaur thing, but evolved deeper in time.” 

The 113-million-year-old fossil is preserved within four limestone slabs. The partial skull, more precisely, is the pterosaur’s crest. A bird’s crest is a tuft of feathers on its head — think of the red crown of a cardinal. A pterosaur’s crest consists of skin stretched over slender bones. 

In the feathers of modern birds, there is a correlation between the shape of melanosomes -- granules of the pigment melanin -- and the color of the feather. 

Through the use of a high-powered electron microscope, the team found preserved melanosomes within the feathers and in the skin. This study is the first time different melanosome shapes have been found in a pterosaur, suggesting both a genetic link to birds and that the feathers on the pterosaur were multicolored. 

A micrograph of preserved melanosomes in the two feather types associated with the Brazilian pterosaur. On the left are the elongated melanosomes found in the monofilaments, and on the right are the stubby, short melanosomes found in the branched feathers.
A micrograph of preserved melanosomes in the two feather types associated with the Brazilian pterosaur. On the left are the elongated melanosomes found in the monofilaments, and on the right are the stubby, short melanosomes found in the branched feathers.Maria McNamara

While the small size of the feathers and their lack of secondary branching means they played no role in helping pterosaurs fly, it’s previously been suggested that pterosaurs had fluff to maintain their internal temperature. This study finetunes this idea, claiming that instead of mere fluff, the ancient reptiles had feathers that helped regulate body temperature and with visual communication. 

“One of the big questions is: Why did feathers evolve?” McNamara said. “What was the function? We think we have really good evidence here that visual communication was an important driving factor in feather evolution.” 

It’s also believed that the unbranched feathers observed in theropod dinosaurs served a similar purpose. It’s been thought that the crests of some pterosaurs may have conveyed some type of communication — they may have been used to attract mates or intimidate rivals. 

The presence of feathers reinforces this idea, and suggests communication was the secondary reason they emerged, following thermoregulation. 

Entire fossil specimen of a cranial crest with the back of the head to the right.
Entire fossil specimen of a cranial crest with the back of the head to the right.University College Cork

When the scientists examined the well-preserved specimen, they observed two critical details: small, whisker-like monofilaments and larger branching structures. 

“At this point, we started to get very excited because it’s been known for a long time that pterosaurs had some sort of fluffy covering, but it was thought this was some kind of hair-like structure that wasn’t related to feathers,” McNamara said. 

An earlier study co-written by McNamara suggested that the fluff associated with pterosaurs — called pycnofibres — may actually be feathers. This finding was challenged by some scientists, including David Martill, a professor of palaeobiology at the University of Portsmouth in the U.K. In 2020 he co-wrote a rebuttal to the 2018 paper, skeptical of its claims. 

He still has his doubts. While he finds the melanosome details within the new study exciting, Martil does not think the new research changes the debate on whether pterosaurs had feathers. 

“Although the authors seem convinced that they [the features on the crest] are some sort of feather, the evidence is far from convincing,” Martill said. “I have no objection to the notion that pterosaurs might have had a feather-like pelt. But to suggest that the frayed fibrous structures seen on some pterosaurs are protofeathers and share a homology with feathers requires better evidence than this.” 

University College Cork professor Maria McNamara.
University College Cork professor Maria McNamara. Courtesy University College Cork

McNamara is undeterred by this criticism. 

“Maybe it just boils down to semantics,” she said. “For me, if something has the same structure as a feather, contains melanosomes, and shows chemical signatures for keraton — these are all defining characteristics of feathers. There’s no need to invent a new name for it.” 

Now, Ph.D. students working with McNamara are attempting to determine the true color of the pterosaur through further examination of the melanosomes. For now, it’s thought the bottom of the crest displayed dark, short strands and lighter branched feathers.

These findings also suggest scientists should re-examine the pycnofibres on other specimens, McNamara said. This specimen represents just the crest of the creature: It’s possible a hunt for melanosomes on other pterosaurs could reveal that the pycnofibres are actually feathers. 

The Tupandactylus imperator is no longer in Brussels. McNamara and her colleagues made efforts to repatriate it to Brazil, where it was originally discovered. Today it is found at the Earth Sciences Museum in Rio de Janeiro. 

“Paleontology is a fabulous gateway science,” McNamara said. “People deserve to have their own fossils back in their own country and to feel that sense of pride in their heritage.”