Scientists find a feathery dinosaur tail trapped in amber

Researchers examine the feathers of a rare 99-million-year-old dinosaur tail preserved in amber. 

It sounds like science fiction: a 99-million-year-old, feathery dinosaur tail encased in amber. But the specimen is real, and it is helping scientists envision how feathers evolved.

"A lot of what we know about the evolution of feathers we've pieced together from two-dimensional data," Alan Turner, a paleontologist at Stony Brook University, tells The Christian Science Monitor. Most ancient feathers are preserved in compression fossils, formed in rock as preserved material is squished and flattened in the sediment. But feathers are not flat. So "having something three-dimensional like this is nice because the three-dimensional geometry is preserved."

The amber specimen is described in a paper published Thursday in the journal Current Biology.

A young, sparrow-sized dinosaur's tail likely brushed up against a tree sticky with resin some 99 million years ago. The middle section of the feathery tail was stuck. Over time, that resin hardened and turned into amber, perfectly preserving the dinosaur's tail and the feathers on it.

Building a feather

Today, birds have feathers with a three-tiered branching structure. At the center of a long feather is the rachis, or central shaft. Smaller barbs stick out of the rachis and even smaller barbules emerge off of those barbs. The barbules interlock like velcro to hold each feather together, making it more useful for flight, among other things. 

But when a team of researchers scanned the chunk of amber containing the juvenile dinosaur's tail, they saw something different.

"In this specimen, we're seeing barbs that are looking like they're branching off of each other" rather than a rachis, says study co-author Ryan McKellar, curator of invertebrate paleontology at the Royal Saskatchewan Museum, in an interview with the Monitor. And the barbules don't seem to zip together like they do in the flight feathers of modern birds. 

This would situate this dinosaur's feathers in a unique spot in feather development.

In the 1990s Richard Prum, an evolutionary ornithologist at Yale University who was not involved in the new study, came up with a model of feather development by studying modern birds. He found that feathers start as a simple, hollow filament, called stage 1. Stage 2 is a tuft of several filaments, or barbs. 

Stage 3 is broken up into stage 3a and stage 3b, separate versions which later merge to become stage 3a+b. In stage 3a, a central rachis has formed with barbs coming off it in a series, but no barbules. Stage 3b sees barbules, but no rachis. Stage 3a+b sees feathers with a rachis, barbs, and barbules, but the barbules don't interlace. Stages 4 and 5 see the complexity of the barbules increasingly becoming like the feathers seen on birds today.

"The morphology that the authors are presenting provides a glimpse into something that fits overall into the developmentally based models for feather evolution," says Stony Brook's Professor Turner, who was not involved in the research. It is "a new data point that will, over time, help us refine the model."

Dr. McKellar and his colleagues assert that the amber-encased dinosaur feathers are a sort of hybrid between stage 3a and stage 3b, but not quite stage 3a+b because the central shaft has barbules coming off it more like a barb than a fully developed rachis.

But Dr. Prum disagrees. "To me, this is absolutely stage 3a+b," he tells the Monitor. "Maybe the fact that it does have barbules on its rachis may indicate that its rachis identity has not proceeded to that strong stage, but I don't see it as separate."

As such, Prum says the dinosaur's feathers are consistent with his model of feather evolution.

That the dinosaur was a juvenile "introduces a little bit of uncertainty as to whether or not it might have had a different feather structure as an adult," McKellar admits. Birds today undergo molts as they grow, losing their baby feathers and replacing them with differently structured adult feathers.

But, he says, "the feathers that we're seeing in this particular specimen match up better with adult contour feathers in modern birds than they do with the downy, first coat of feathers that most birds produce. So we think we're dealing with adult-like feathers."

Dinosaur or bird?

At one time, a specimen like this one would have been controversial, says Mark Norell, a paleontologist at the American Museum of Natural History who was not involved with this study. Scientists didn't always realize that birds are dinosaurs, so finding a dinosaur with feathers would have been a big surprise. 

But now it's an understood relationship among the vast majority of scientists that "birds are a kind of dinosaur and dinosaurs other than birds have feathers," Dr. Norell tells the Monitor. It's so unsurprising to find a dinosaur with feathers now, he says, "It's like predicting that a fossil mammal would have hair."

Still, McKellar and his colleagues checked to be sure that this specimen was not actually an early bird. "Modern birds and their close relatives have something called a pygostyle, which is a fused series of tail vertebrae where they form a rod that supports the tail plumage," he explains. "And we're not seeing this in this particular specimen. We've got a flexible, whip-like tail instead of one of these big, sort of plowshare-shaped tails that we see in modern birds."

The dinosaur's tail was probably used only for signaling or display of some sort, McKellar says, and wouldn't have been useful for flight or much else. The team's scans of the specimen revealed that the feathers on the topside of the tail were a chestnut brown color, while the underside of the tail was paler or perhaps white in color, somewhat like a white-tailed deer, he says.

With just a handful of vertebrae and feathers preserved, the scientists know little about the dinosaur itself. But they were able to determine that the little beast was a coelurosaur theropod, a type of two-legged dinosaur.

Dinosaur-era feathers have been found encased in amber before. Some were attached to entire bird wings, but most of these preserved feathers were not associated with the flesh or bones of an animal, so scientists couldn't determine much about the beast that wore them.

"It's stunning to see these beautifully formed feathers from a representative that is clearly a coelurosaur," Prum says.

Bringing science fiction to life?

One other tantalizing clue the team found in the amber was the chemical signature of ferrous iron in the thin carbon film where the animal's soft tissues would have been. That form of iron comes from blood proteins, like hemoglobin and ferritin.

"It's thought, based on other specimens in the vertebrate fossil record, that traces of this particular form of iron might be linked to decay products from things like blood or tissue left behind," McKellar says. So this could allow researchers to study the soft tissues that almost never survive a hundred million years.

Any Jurassic Park fan's interest might be piqued by the mention of dinosaur blood in amber, as the premise of the tale hinges on replicating the beasts' DNA from just that. In the fictional story, scientists clone dinosaurs using DNA extracted from blood that was preserved for millions of years in the digestive system of mosquitos trapped in amber. And, with bits of an actual dinosaur itself preserved in amber, this non-fictitious specimen seems to get closer to that possibility than even that science-fiction scenario.

But don't expect live dinosaurs anytime soon, McKellar says. Research has shown that DNA cannot last in the fossil record for nearly long enough to survive from the age of the dinosaurs. With a half-life of about 521 years, it would take less than 7 million years for an organisms' DNA to be completely destroyed. The non-avian dinosaurs died about 65 million years ago, so their DNA is long gone.

And, McKellar points out, with all kinds of debris and insects also stuck in the tree resin, the DNA sampled from an amber specimen may be from another organism, or even the tree itself. So, as exciting as it would be to clone 99-million-year-old organisms, a tree doesn't quite have the same thrill-factor as a dinosaur.