A new fossil from western Liaoning, Norellraptor barsboldi, shows microraptorines and birds assembled flight apparatus independently
Synopsis
A team reports in Nature Communications a 57-cm, exquisitely preserved new feathered dinosaur, Norellraptor barsboldi, from the Lower Cretaceous Jiufotang Formation at Lamadong Town, Jianchang County, Liaoning, placing it in Microraptorinae as sister to Zhongjianosaurus; optimizing 194 microraptorine apomorphies along the tree shows about 30% (57) were convergently acquired in the bird lineage (Avialae), yet the order in which the two lineages acquired them differs, and radial osteohistology hints at a peculiar limb growth model in Microraptorinae, leading the authors to argue that the two flight apparatuses did not arise from a single shared developmental regime but evolved multiple times independently.
Interpretation
The study names and describes a new microraptorine, Norellraptor barsboldi, based on holotype 130108-MHGU-F4281, a 57-cm skeleton that is complete with only moderate disarticulation of the appendicular girdles and preserves plumage imprints near the forelimb, hindlimb and tail end; the forelimb probably bore asymmetrical pennaceous feathers, and the distal tail feathers have a rachis-dominated morphology. Anatomical and ecological information for Microraptorinae had come mainly from Microraptor, Sinornithosaurus, Wulong and Zhongjianosaurus; the new specimen adds skeletal and plumage data for a late-diverging member of the clade and is distinguished from known taxa by features including extensive fenestration of the rostral antorbital fossa and a gently curved pubis with a small foot. The basis is a nearly complete, largely articulated skeleton whose integrity and original locality the authors verified, with microraptorine synapomorphies present in all anatomical regions; osteohistology, complete fusion of tibiotarsi and tarsometatarsi, and smooth long-bone surfaces support a post-juvenile stage at death.
Osteohistology indicates the individual was at least about three years old when it died: the mid-diaphysis of the radius preserves two lines of arrested growth dividing the periosteal bone into three zones, with zone 3 only beginning to be deposited and never circumferentially complete, a periosteal-to-endosteal bone index of 5.8 to 11.7, and no trabecular structures in the medullary cavity, suggesting extensive pneumatization. The authors place this age between the less-than-two-year-old Wulong bohaiensis holotype and Sinornithosaurus sp. DNHM-2140 and the at-least-five-year-old Changyuraptor yangi holotype, and note that the radial growth pattern differs from the uninterrupted femoral pattern and may indicate an inverse forelimb-hindlimb regime relative to more mature Microraptor individuals. Evidence comes from successive physical transverse sections of a distal projection of the left radius midshaft, examined in transmitted, elliptically polarized and fluorescent light at 52 micrometres thickness; the authors note the mid-diaphysis was crushed post-mortem by overlying sediment but that microstructure remained readable in the sampled region, and that zone 3 measures only 15.7 micrometres at its thickest.
Phylogenetic analysis places Norellraptor among late-diverging microraptorines as sister to Zhongjianosaurus, with which it shares a bulbous distal process of the ulna, extensive fusion of the proximal metatarsal shafts and an arctometatarsalian metatarsus; the analysis found 50,000 shortest trees of score 8735, with a consistency index of 0.31 and a retention index of 0.54. On this basis the authors redefine Microraptorini and report Halszkaraptorinae and Unenlagiinae as closer to birds than dromaeosaurids and troodontids, independently replicating an alternative phylogenetic scenario that emphasizes the numerous avian-like features of unenlagiids. The result rests on an updated version of the Cau (2024) matrix analysed under equally weighted parsimony in TNT 1.6 (maxtree = 50,000, 1,000 traditional search runs) with an implied-weighting analysis at weighting strength 1000 as the reference topology; the authors also report that enforcing microraptorines as members of Avialae yields shortest trees 23 steps longer than the unenforced topologies.
Within the reconstructed framework, about 30% of 194 microraptorine apomorphies (57 features) were convergently acquired along the avialan lineage, including wing feathers, hook-shaped toe bones and a forearm bone longer than the upper arm bone; however, the order in which the two lineages acquired these features differs, and rank correlation is not significant (Spearman coefficient -0.08, p = 0.50 for all features; 0.15, p = 0.47 for forelimb and pectoral features alone). The authors conclude that the bird-like traits of microraptorines cannot be considered the expression of a developmental homology shared with birds, quantitatively challenging the hypothesis that the flight apparatus is an incomplete synapomorphy inherited from the paravian common ancestor and supporting multiple independent origins of flight among Paraves. The evidence comes from character-state optimization on the strict consensus of the shortest trees using fast optimization to avoid oversampling microraptorine internodal apomorphies due to preservation artifacts, plus correlation tests in PAST 4.04; the authors note that manual phalanx shortening occurred at the root of Microraptorinae whereas carpometacarpal fusion, a fenestra in the deltopectoral crest and relative sternum elongation occurred later in Microraptorini, with the opposite order in Avialae.
Perspective
This work speaks to readers interested in bird origins, paravian phylogeny and dinosaur osteohistology, and applies to the specific question of whether the flight apparatuses of Microraptorinae and birds share a single origin. It provides reusable material and data: the holotype 130108-MHGU-F4281 is housed at the Museum of Hebei GEO University, the phylogenetic matrix and character list are deposited in FigShare, and the new names are registered in ZooBank. Next steps include sampling forelimb and hindlimb osteohistology in more microraptorine individuals to test whether the proposed inverse forelimb-hindlimb growth pattern is general, and adding more paravian taxa to the matrix to test whether the topology placing Halszkaraptorinae and Unenlagiinae closer to birds is stable.
The authors themselves call the histological evidence preliminary and note that the radial midshaft was crushed post-mortem and that zone 3 is not circumferentially complete, so growth-model inferences still need testing on more individuals. The exact length and detailed morphology of the plumage are difficult to ascertain because of poor preservation, and the asymmetrical pennaceous forelimb feathers are preserved separated from the rest of the plumage. On phylogeny, the authors report that enforcing microraptorines within Avialae yields shortest trees 23 steps longer, and note that the positions of Graciliraptor and Sinornithosaurus within Microraptorinae vary between analyses, so topological conclusions are sensitive to matrix and weighting settings. In addition, this evidence bundle is a summary-level reading that does not include the supplementary figures or the full character list, so questions about specific character numbers and their one-to-one correspondence require checking the original article and supplementary materials.
