New Fossil Species Suggests Dinosaur Flight Evolved Twice, Independently

Paleontologists have described a new species of small, feathered dinosaur from northeastern China whose skeleton and bone structure indicate that the capacity for flight did not arise once and pass down to both birds and their closest dinosaur relatives, but evolved separately in each group under different evolutionary pressures. The fossil, named Norellraptor barsboldi, is described in a paper published Tuesday in Nature Communications by a team led by researchers at the Chinese Academy of Geological Sciences in Beijing, with collaborators in Italy and Slovakia.

The animal belonged to Microraptorinae, a group of small, four-winged dromaeosaurid dinosaurs closely related to birds that lived alongside early avians during the Early Cretaceous. Its remains, a nearly complete skeleton with patches of feather imprints still visible in the rock, were recovered near Lamadong Town in Jianchang County, western Liaoning Province, from the Jiufotang Formation, a fossil-rich layer of the Jehol Biota roughly 120 million years old. The specimen, cataloged as 130108-MHGU-F4281, is now held at the Museum of Hebei GEO University.

What the Bones and Feathers Show

The skeleton measures 57 centimeters, or about 22 inches. Growth-ring analysis of the radius bone, based on lines of arrested growth laid down each year, indicates the animal was at least three years old when it died, a middle-aged individual by the standards of its relatives: two other small microraptorines, Wulong bohaiensis and a Sinornithosaurus specimen, died before age two, while the largest known microraptorine, Changyuraptor yangi, lived to at least five.

The researchers placed Norellraptor among the more derived, or "late-diverging," members of Microraptorinae, indicating that flight-related anatomical features accumulated gradually within that lineage rather than appearing all at once. Comparing 57 traits that arose independently in both microraptorines and true birds, or Avialae, the team found that the order in which those features appeared differed consistently between the two lineages: a statistical test of the sequences turned up no significant correlation between them (Spearman's rank correlation coefficient of -0.08, with related tests on forelimb and shoulder traits alone also showing no significant relationship). Combined with the bone-growth evidence pointing to a distinct limb-development pattern in microraptorines, the authors write that the findings "dismiss a shared developmental pattern driving the evolution of the flying dinosaurs" and instead support "multiple and independent selective regimes" behind the origin of winged, flight-capable animals in this part of the dinosaur family tree.

A Long-Running Argument About How Flight Began

Whether flight, or at least gliding ability, was present in the last common ancestor of birds and their closest dinosaur relatives, or evolved separately on different branches afterward, has occupied paleontologists since the first four-winged dromaeosaurids were unearthed in China roughly two decades ago. Those animals, with long feathers on both their arms and legs, raised the possibility that early fliers experimented with body plans quite unlike those of modern birds. Changyuraptor yangi, described in a 2014 Nature Communications paper, had tail feathers nearly 30 centimeters long, close to a third of its body length, which researchers at the time argued would have helped control its descent during a glide or shallow flight.

A broader 2020 analysis incorporating the largest dataset of feathered dinosaurs and early birds assembled up to that point reached a similar conclusion to the one in the new paper. That study, whose authors included a McGill University paleontologist, found that the potential for powered flight evolved independently at least three times among feathered theropods, once in birds and twice among dromaeosaurids. The new fossil adds anatomical and developmental detail to that picture, tracing within a single microraptorine lineage how aerial features appear to have been assembled piece by piece, and on a different schedule than in birds.

The genus name honors "the exceptional contributions" of paleontologists Mark Allen Norell and Rinchen Barsbold "to the study of the bird-like dinosaurs," the authors write in the paper.

That tribute carries particular weight this year: both scientists, whose decades of fieldwork and research helped establish the dinosaur-bird connection, are listed in the paper as having died in 2025. Norell, an American paleontologist, and Barsbold, a Mongolian paleontologist, are honored respectively in the genus name Norellraptor and the species name barsboldi.

Why a Single Fossil Matters

The finding is a narrow but telling data point in a larger question in evolutionary biology: how complex, multi-part adaptations such as flight can arise more than once along different branches of the same family tree. Flight is widely regarded as one of the most consequential innovations in vertebrate history, underlying the later diversification of birds into more than 10,000 living species. For working paleontologists, establishing whether flight-related traits had one deep origin or several shallower, independent ones changes how evolutionary trees of Paraves, the group containing birds and their nearest dinosaur relatives, are built and interpreted, and which anatomical changes get read as inherited versus separately, convergently acquired.

There are limits to how far conclusions can be drawn from a single specimen. Fossils that preserve both feather impressions and the internal bone microstructure needed for growth-rate analysis, the two lines of evidence combined in this study, remain rare, and most known examples come from a handful of exceptionally fine-grained deposits, chief among them the Jehol Biota of northeastern China, which has produced the majority of feathered dinosaur specimens described over the past three decades. Early coverage of the study noted that testing how widespread this pattern of independently assembled flight features was among other bird-like dinosaur groups will require further fossil and bone-growth evidence. Any such evidence will likely depend on new finds of similar quality emerging from those same Liaoning beds.

The paper, "Independent assembly of the flight apparatus in a non-avian dinosaur clade," was published in the peer-reviewed, open-access journal Nature Communications. Its authors are affiliated with the Chinese Academy of Geological Sciences, Hebei GEO University and China University of Geosciences in China, the OPHIS Museo Paleontologico e Centro Erpetologico in Italy, and the University of Pavol Jozef Šafárik in Slovakia.