A newly described feathered dinosaur from China’s Changma fossil bed, a close relative of Velociraptor, appears to have glided on four wings while hunting early birds. The specimen, recovered from a site rich in avian fossils, carries exceptionally long feathers on both its forelimbs and hindlimbs. Senior author Jingmai O’Connor and corresponding author Matthew Lamanna led the study, which presents the animal as the only non-avian dinosaur found among hundreds of bird remains and crushed-bone clusters at the site, raising pointed questions about whether four-winged flight in small raptors evolved as a predatory tool rather than a means of escape or display.
Why a four-winged raptor changes the debate over dinosaur flight
The discovery matters because it directly challenges the long-standing assumption that feathered limbs on small dromaeosaurids served primarily for locomotion, thermal regulation, or courtship. The ecological context of the Changma fossil bed tells a different story. According to the Field Museum release, this animal lived in an environment dominated by birds, and the crushed-bone clusters at the site resemble the pellets that modern raptorial birds regurgitate after consuming prey. If those clusters originated from the new dinosaur, it would mean a non-avian predator was routinely catching and eating birds in their own habitat.
That scenario supports a specific ecological hypothesis: four-winged gliding in small dromaeosaurids provided a selective advantage for ambush-style attacks on flying prey in dense, low-canopy environments. Rather than covering long distances or launching from the ground into sustained flight, these animals may have used their hindwing feathers to execute short, controlled glides from elevated perches, dropping onto birds below. The Changma setting, packed with avian species, would have been an ideal hunting ground for exactly that kind of predator.
Separate fossil evidence strengthens the case. Specimens of the related genus Microraptor preserve avian remains in the gut, confirming that at least one four-winged dromaeosaurid ate birds. That gut-content evidence, combined with the new Changma taxon’s ecological setting, builds a pattern: these animals were not occasional scavengers of avian prey but active hunters whose anatomy was shaped by that diet.
If four-winged raptors were already exploiting aerial or arboreal prey, it reframes the evolutionary pathway to powered flight. Instead of feathers and partial wings evolving first for insulation or display and only later being co-opted for locomotion, predation-focused gliding could have been an early and persistent driver. In this view, selection would favor individuals capable of precise aerial maneuvers-steep dives, controlled braking, and quick directional changes-because those traits directly increased hunting success in bird-rich habitats.
The Changma animal may therefore represent an intermediate stage where gliding performance was good enough for repeated predatory strikes but not yet refined into the full flapping flight seen in true birds. Its anatomy potentially captures a snapshot along a gradient from tree-leaping predators to active fliers, with predation pressure acting as the main evolutionary engine.
Fossil anatomy and aerodynamic modeling behind the gliding hypothesis
The physical evidence starts with feather length. The new specimen’s limb feathers are described as exceptionally long relative to body size, a trait shared with other four-winged dromaeosaurids analyzed in prior peer-reviewed work. A separate study in Nature Communications documented another small raptorial dinosaur with similarly extensive hindlimb plumage, providing comparative data on how widespread and pronounced this trait was among close relatives of birds. The consistency of long hindlimb feathers across multiple species suggests the feature was not an anomaly but a functional adaptation under active selection pressure.
Long feathers on the legs and tail would have increased total lifting surface, turning the hindlimbs into secondary wings. For a lightweight predator moving among trees, this extra surface could stabilize leaps, slow descents, and allow fine-tuned adjustments during a glide. The Changma fossil’s feather proportions, as described in institutional materials, fit this pattern of expanded aerodynamic surfaces rather than the short, contour-like feathers associated purely with insulation.
Aerodynamic performance has been tested more directly in related animals. A modeling study published in the Proceedings of the National Academy of Sciences examined gliding performance in Microraptor gui under different hindwing configurations. Researchers used physical models and wind-tunnel tests to show that when the hindwings were fully extended, the animal achieved stable gliding postures, maintaining controlled descent without tumbling or stalling. Varying the angle and spread of the rear feathers produced different aerodynamic profiles, but a biplane-like arrangement of fore and hind wings consistently allowed effective gliding at angles consistent with short-range, steep dives.
These modeling results align well with the ambush-hunting hypothesis. A short, stable glide from a tree branch or rock outcrop would give a small predator enough speed and control to intercept a bird in flight or to pounce on one perched or foraging below. The hindwings would not need to generate sustained lift over long distances; they would need to provide drag control, pitch stability, and maneuverability for a brief, directed attack. The Microraptor data show that such maneuvers were aerodynamically plausible in at least one four-winged dromaeosaurid.
By analogy, the Changma dinosaur’s long hindlimb feathers likely contributed to similar performance. If its mass, wing loading, and feather arrangement were broadly comparable, then short-range gliding ambushes become a reasonable default scenario. The presence of abundant bird fossils and pellet-like bone accumulations at the same site makes a predatory interpretation even more compelling, tying the anatomical potential for gliding directly to a plausible ecological role.
Gaps in the fossil record and what to watch next
Several questions remain open. The primary osteological measurements and feather-length data for the new Changma taxon have not yet appeared in a peer-reviewed journal. Outside the institutional announcement, detailed metrics on limb proportions, feather attachment points, and body mass estimates are unavailable. Without those numbers, independent researchers cannot rigorously test how closely this animal’s flight capabilities matched those of Microraptor or other modeled taxa.
The crushed-bone clusters at the fossil bed also require closer scrutiny. Their resemblance to modern raptor pellets is suggestive but not definitive. To confirm that they are regurgitated remains, researchers will need to examine breakage patterns, acid etching, and the taxonomic composition of the fragments. If the pellets can be confidently linked to the new dinosaur-for example, by tight spatial association with multiple individuals or by consistent size and content-they would offer rare, direct evidence of diet and feeding behavior in a non-avian theropod.
The aerodynamic case has limits as well. The best-studied four-winged dinosaur, Microraptor gui, differs from the Changma species in several important ways, including overall body size and the exact arrangement of feathers on the legs and tail. Small changes in these variables can significantly alter glide angle, stall speed, and maneuverability. Until researchers build species-specific models for the Changma dinosaur, the gliding hypothesis for this particular animal rests on comparative inference rather than direct testing.
Future work will likely focus on three fronts. First, a formal anatomical description will need to quantify feather lengths, wing areas, and skeletal features relevant to climbing and perching, such as claw curvature and limb joint mobility. Second, taphonomic studies of the bone clusters and surrounding sediments could clarify whether the pellets truly represent regurgitated bird remains and, if so, whether they can be tied to this taxon rather than to early birds or other predators. Third, physical or computational aerodynamic models tailored to the Changma specimen could test how effectively it glided, how steeply it could dive, and whether it was capable of any powered flapping.
Even with these uncertainties, the Changma four-winged raptor adds a crucial piece to the puzzle of how flight evolved. It situates a gliding, bird-eating dinosaur squarely within a bird-dominated ecosystem, implying that aerial predation may have been an important selective pressure shaping the earliest experiments in winged locomotion. As more fossils emerge from Changma and other feather-rich deposits, paleontologists will be better positioned to judge whether four-winged gliding was a brief evolutionary side branch or a key stepping stone on the path to true avian flight.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
