A New Enantiornithine Bird with Unusual Pedal Proportions Found in Amber

Report

Highlights Authors d New fossil is ﬁrst avian species recognized from amber Lida Xing, Jingmai K. O’Connor, Luis M. Chiappe, ..., Han Hu, Ming Bai, d Elektorornis is distinct from all other birds based on the Fumin Lei proportions of the foot Correspondence d Scutellae scale ﬁlaments on foot suggest probing function for elongated third toe [email protected] (L.X.), [email protected] (J.K.O.)

In Brief Xing et al. describe an avian hindlimb preserved in 99 million-year-old Burmese amber. Its unusual pedal proportions support the erection of a new species, Elektorornis chenguangi gen. et sp. nov. The elongated third toe may have been a feeding adaptation, aided by scutellae scale ﬁlaments with possible tactile function.

Xing et al., 2019, Current Biology 29, 2396–2401 July 22, 2019 ª 2019 Elsevier Ltd. https://doi.org/10.1016/j.cub.2019.05.077 Current Biology Report

A New Enantiornithine Bird with Unusual Pedal Proportions Found in Amber

Lida Xing,1,2,11,* Jingmai K. O’Connor,3,4,11,12,* Luis M. Chiappe,5 Ryan C. McKellar,6,7,8,11 Nathan Carroll,5 Han Hu,9 Ming Bai,10 and Fumin Lei10 1State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Beijing 100083, China 2School of the Earth Sciences and Resources, China University of Geosciences, Beijing 100083, China 3Key Laboratory of Vertebrate Evolution and Human Origins of the Chinese Academy of Sciences, Institute of Vertebrate Paleontology and Paleoanthropology, Beijing 100044, China 4CAS Center for Excellence in Life and Paleoenvironment, Beijing 10010, China 5Dinosaur Institute, Natural History Museum of Los Angeles County, Los Angeles, CA 90007, USA 6Royal Saskatchewan Museum, Regina, Saskatchewan S4P 4W7, Canada 7Biology Department, University of Regina, Regina, Saskatchewan S4S 0A2, Canada 8Department of Ecology & Evolutionary Biology, University of Kansas, Lawrence, KS 66045, USA 9Zoology Division, School of Environmental and Rural Sciences, University of New England, Armidale, NSW 2351, Australia 10Key Laboratory of Zoological Systematics and Evolution, Institute of Zoology, Chinese Academy of Sciences, Beijing, 100101, China 11These authors contributed equally 12Lead Contact *Correspondence: [email protected] (L.X.), [email protected] (J.K.O.) https://doi.org/10.1016/j.cub.2019.05.077

SUMMARY Holotype HPG-15-2 (Hupoge Amber Museum, Tengchong City Amber As- Recent discoveries of vertebrate remains trapped in sociation, China) is an incomplete avian right hindlimb with middle Cretaceous amber from northern Myanmar plumage from the left wingtip preserved encased in amber, [1, 2] have provided insights into the morphology of measuring 34.8 mm 3 34.4 mm 3 8.2 mm and weighing soft-tissue structures in extinct animals [3–7], in 5.51 g. The bird appears to have undergone signiﬁcant decay particular, into the evolution and paleobiology of prior to resin polymerization. In many places, the skin from the early birds [4, 8, 9]. So far, ﬁve bird specimens have foot has sloughed off the bones and drifted short distances through the amber. been described from Burmese amber: two isolated wings, an isolated foot with wing fragment, and two Etymology partial skeletons [4, 8–10]. Most of these specimens Elektorornis,‘‘Elektor,’’ the word for amber; ‘‘-ornis,’’ Greek, contain the remains of juvenile enantiornithine birds meaning bird. The species name ‘‘chenguangi’’ is in honor of [4]. Here, we describe a new specimen of enantiorni- Chen Guang, a curator at the Hupoge Amber Museum. thine bird in amber, collected at the Angbamo locality in the Hukawng Valley. The new specimen includes a Diagnosis partial right hindlimb and remiges from an adult or Small enantiornithine (distal condyles of the tibiotarsus contact- subadult bird. Its foot, of which the third digit is ing medially, J-shaped metatarsal I, metatarsal IV mediolaterally much longer than the second and fourth digits, is reduced relative to metatarsals III and IV, metatarsal IV trochlea distinct from those of all other currently recognized reduced to a single condyle, and recurved pedal unguals) with Mesozoic and extant birds. Based on the autapo- the unique combination of following traits: pedal digit III, 20% morphic foot morphology, we erect a new taxon, longer than tarsometatarsus; hallux, more than 86% length of Elektorornis chenguangi gen. et sp. nov. We suggest pedal digit II ( 80% in pengornithids); and pedal digit II, 59% that the elongated third digit was employed in a length of digit III, proportionately shorter than that of any other Mesozoic bird (autapomorphic within the Enantiornithes; unique foraging strategy, highlighting the bizarre Table S1). morphospace in which early birds operated. Locality and Horizon RESULTS Late Albian-Cenomanian 98.8 ± 0.6 Ma [11] Angbamo locality, Hukawng Valley, Kachin Province (Tanai Township, Myitkyina Systematic Paleontology District), northern Myanmar. Aves Linnaeus 1758 Ornithothoraces Chiappe 1995 Ontogenetic Assessment Enantiornithes Walker 1981 The specimen is considered to be subadult to adult, based on Elektorornis gen. nov. the complete fusion of the proximal tarsals to the tibia and the Elektorornis chenguangi sp. nov. (Figures 1A–1C) distal tarsals to the metatarsals [12].

2396 Current Biology 29, 2396–2401, July 22, 2019 ª 2019 Elsevier Ltd. Figure 1. Tarsal Structure and Integumentary Structures Preserved in Elektorornis chenguangi HPG-15-2 (A) HPG-15-2 overview, with inset providing greater detail on foot, arrowheads marking different apices of unguals and ungual sheathes where visible, and red arrow marking base of mt III 4 shared with (D). (B and C) Osteological details. (D) Tuft of elongated SSFs near apex of mt III ph 3, with horizontal arrowhead marking edge of reticulae from digital pad, inclined arrowhead marking edge of scute, white arrow marking sloughed reticulae, and red arrow marking base of ungual in (A). (E) Detail of lowermost SSFs in (D), showing hollow cores (arrowheads) and mottled outer walls, presumably due to feather oils. Fe, femur; ﬁ, ﬁbula; lc, lateral condyle; mc, medial condyle; mt, metatarsal and corresponding digit; ph, phalanx; tb, tibia. Scale bars, 5 mm in (A); 1 mm in (A) inset; 0.5 mm in (D); and 0.25 mm in (E). See also Figures S1, S2, and S4.

Description enantiornithines [13, 19]. This trochlea is also slightly plantarly Osteology displaced relative to the coplanar trochleae III and IV. In plantar The distal end of the right femur is jointed with a fully articulated view, the medial condyle of the trochlea has greater plantar pro- tibiotarsus, fibula, and pes (Figures 1B and 1C). The femur is jection, and metatarsal III displays a slight depression just prox- obscured by dense soft tissue. A separate, poorly preserved imal to the trochlea. The trochlea of metatarsal IV appears element contacts the lateral surface of the knee (possibly the distal reduced to a single condyle as in other enantiornithines [20]. ends of the ulna-radius). The tibiotarsus seems to bear a circular Metatarsal I is approximately 20% of the length of metatarsal proximal articular surface that is expanded relative to the shaft, II, proximally tapered, and mediolaterally compressed. Similar as in many enantiornithines [13]. The tibiotarsal shaft has an oval to HPG-15-1 [4], the shaft of metatarsal I articulates with the cross-section with the long axis oriented mediolaterally. The distal medial surface of metatarsal II, somewhat extending onto the end of the tibiotarsus is expanded so that the articular surface plantar surface such that the lateral surface of the shaft of meta- formed by the condyles is much wider than the shaft. The medial tarsal I is deeply concave. The distal articular surface of meta- condyle is wider than the lateral condyle, a condition typical of tarsal I projects plantarly, nearly perpendicular to the longitudinal enantiornithines and other basal pygostylians [14–17]. The medial axis of the shaft (J-shaped in medial view), a condition similar to surface of the medial condyle appears somewhat excavated (Fig- some other enantiornithines [14, 21]. The hallux was fully ure 1B). The articular surface extends onto the caudal surface of reversed in the anisodactyl condition. the tibiotarsus. The fibula is mediolaterally compressed and The non-ungual pedal phalanges are gracile (Figures 1B and approximately three-quarters of the length of the tibiotarsus. 1C). The third digit is the longest, followed by the fourth, the sec- The proximal tarsometatarsus appears fully fused (Figures 1B ond, and the hallux, which is 86% of the length of the second and 1C). The proximal ends of the metatarsals are expanded digit. The foot is unusual in that the third digit is proportionately relative to their shafts, which remain unfused to one another— much longer than the other digits and in that the first digit ap- a condition typical even in adult enantiornithines. The dorsal proaches the length of the second digit (typically, the second surface of metatarsal III is convex, similar to the condition in avi- digit is proportionately longer) (Table S1). The penultimate pha- saurid enantiornithines [18, 19]. Typical of enantiornithines, lanx is the longest in each digit, as in most perching birds [22]. metatarsal IV is narrower than metatarsals II and III [13]. Meta- The first phalanx of digit I is very long, shorter only than the penul- tarsal III is the longest; metatarsals II and IV are subequal, ex- timate phalanx of the third digit. The ungual phalanx of digit I is tending to just above the trochlea of metatarsal III. The trochlea slightly more curved than that of the other digits. The proximal of metatarsal II is the widest, a proportion common among phalanx of digit II is short, approximately half the length of the

Current Biology 29, 2396–2401, July 22, 2019 2397 Figure 2. Wing Fragment Plumage (A) Overview of primary and secondary feather exposure at polished edge of amber piece, with inclined arrows marking primary rachises (P1 and P2 weakly distinguished from secondaries and marked in red); vertical arrows mark secondaries; horizontal arrows mark pale areas in wing; and lettered circles mark positions of (B) and (C). (B) Weakly pigmented reduced barbules from primary barbs in leading edge of wing. (C) Dark brown barbules from primary barbs in base of posterior vane of primary. Scale bars, 2 mm in (A); 0.25 mm in (B) and (C). See also Figures S1 and S2.

interacting with the surrounding resin. The crural tract contains contour feathers with very short rachises and loosely vaned barbs composed of elongated, blade- shaped barbules with no visible hooklets or clear pigmentation patterns (Figures S1H and S1I). HPG-15-2 contains plumage from the tip of the left wing. Either ten primaries and four secondaries, or eleven primaries and penultimate phalanx. The hypertrophied third digit bears three three secondaries, are exposed at the polished surface of the elongated non-ungual phalanges; the first two are subequal amber piece. The distinction between primaries and secondaries and roughly two-thirds the length of the penultimate phalanx. is based upon narrower rachises and a slightly different orientation The claw of this digit is slightly longer than that of the other digits. in the latter feathers, which cannot be traced back to their inser- The proximal three phalanges of the fourth digit are short and tions (Figures 1AandS2D). These feathers are clearly remiges subequal in length; they are followed by a more delicate phalanx based on their size and morphology. The exposed rachises that is 30% longer and a claw that is slightly shorter than those of have subcylindrical cross sections, and their pith cavities are filled the other pedal digits. with milky amber and bubbles. Asymmetry between the primary Soft Tissue vanes is strong, with barbs in the leading vanes less than half of Plumage and soft tissue are visible on much of the lower leg, but the length of those in the trailing vanes. Barbules within both vanes visibility is restricted by particulates in the surrounding amber are reduced and blade-shaped, with no clear indication of nodes, (Figure 1). The keratinous ungual sheaths are visible as translu- and weak hooklets. Proximal barbules closer to the base of the cent masses around the bones, extending the functional length feather diverge from the ramus at 30 degrees and curve gently of each ungual by approximately one-third (e.g., Figure 1A). adapically; distal barbules diverge at 47 degrees and angle Patches of scutellae and reticulae are visible in sheets of skin adapically in a weakly developed pennulum (Figures 2Band that have detached from the bone (Figure S1); scutes appear 2C). Diffuse traces of pigments within the barbules give the wing to be restricted to a narrow band on the dorsal margin of each an overall dark brown color, with paler barb rami and rachises, digit. The scutellae are round and small (0.12 mm in diameter), and with two large-scale pale wing spots visible anteriorly (Fig- and the reticulae are approximately one-quarter of this diameter ure 2A). Entire barb apices within the posterior vanes of the pri- (making them difficult to differentiate from particulates in the sur- maries are either pale or white in color, which would have made rounding amber). Scutellae scale filaments (SSFs) [4] are sparse the underside of the wing pale or striped. upon the dorsal and lateral surfaces of the phalanges. The SSFs in HPG-15-2 originate from the distal margins of scutes and scu- DISCUSSION tellae. A cluster of longer and more robust SSFs is present near the base of the ungual on digit III; a less prominent cluster is Based on the presence of a proportionately hyper-elongated third visible near the apex of phalanx 3 on digit IV (Figures 1D, 1E, pedal digit (Tables S1 and S2; Figures 4 and S3), HPG-15-2 can be S1B, and S1G). SSFs are also present on the dorsal surface of distinguished from all other known Mesozoic birds, justifying the the metatarsals (Figures S1E and S1D) and in greater density erection of a new taxon, Elektorornis chenguangi gen. et sp. nov. on the ventral surface, surrounding the scutellate heel pad (Fig- Cladistic analysis confirms that Elektorornis is an enantiornithine ures S1J and S1K). Shed SSFs clearly indicate a flared basal (Figure 3). Most previously described Hukawng birds represented cavity (Figure S1E) that tapers into a narrow hollow core running very young juveniles, whereas the tarsals appear fully incorporated the length of the filament (Figures 1E, S1E, and S1G). Their outer into the tibiotarsus and tarsometatarsus, indicating that HPG-15-2 walls have a mottled brown surface that may reflect either partial represents a subadult or adult [12]. Despite these apparent ontoge- preservation of keratin or (more likely) surface oils on the feather netic differences, HPG-15-2 is approximately the same size as

2398 Current Biology 29, 2396–2401, July 22, 2019 was predominantly arboreal [24]. Compared to other enantiornithines, the hallux is proportionately even longer in Elektorornis relative to pedal digit II (Table S1). This may suggest increased grasping abilities in the foot of Elektorornis chenguangi,aspro- posed for the relatively elongated hallux of pengornithid enantiornithines [23]. In the absence of any extant analogs, the function of the elongated third toe is uncertain. It could represent a feeding specialization, similar to the Malagasy Aye-aye, or it may have facilitated greater arboreal function, or both. Similarly elongated toes are also observed in some tree-climbing lizards [25], supporting an arboreal function for this unusual feature in Elektorornis. The soft tissue of the podotheca also lacks modern analogs. Elongated feathers and scutate scales of varying degrees can be found in modern birds as adaptations to speciﬁc substrate interactions [26], although no modern examples display the morphology of the SSFs present in HPG-15-2. SSFs are consistently present in Hukawng birds [4, 10],whichinthe future may shed light on the development of protofeathers or the production of avian pedal scutes. The curious distribution and relative size of SSFs on the podotheca in HPG-15-2 may play a role in substrate interaction by increasing traction in wet tropical forests. Another possibility is that the SSFs have a mechanosensory tactile function, similar to bristles surrounding the mouth in some living birds [27]. This inference is sup- ported by the fact the SSFs are longest and most robust at the base of digit III. If this were the case, this would increase the utility of the elongated toe as a trophic specialization for probing insect burrows in trees or some similar feeding strategy. These inferences remain speculative in the absence of directly analogous species among the Neornithes and data from destructive histochemical sampling (to identify Herbst corpus- cles, nerve endings associated with tactile feathers in birds [27]). However, given that most feathers have some sensory function [28, 29] and that bristle-like feather structures in diverse anatomical locations have all been demonstrated to have such a mechanosensory function [30, 31], this hypothesis Figure 3. Cladogram Depicting the Hypothetical Position of seems to be a plausible explanation for the function of the long Elektorornis chenguangi gen. et sp. nov. HPG-15-2 The Nelson strict consensus tree is derived from 32 equal-length trees digit III of the new taxon. Tactile bristles on the feet in (consistency index, 0.445; retention index, 0.653). Elektorornis may have aided in prey detection, together with the elongated third digit producing a unique foraging structure. The unique pedal morphology of HPG-15-2 further eluci- HPG-15-1 (metatarsal III measures 7.7 mm inHPG-15-2 and is esti- dates the radiation of Cretaceous enantiornithines revealing mated at 8 mm in HPG-15-1), which was considered a hatchling [4]. some of the possibly unique morphospaces and feeding stra- Elektorornis chenguangi is diagnosed based on the presence tegies once utilized by Aves but since abandoned. of a remarkably elongated third digit. Similar digital proportions are not observed in any Mesozoic bird and, to our knowledge, STAR+METHODS are not present in any living bird (Table S2; Figures S3A– S3C). Some arboreal passerines have a similarly elongated Detailed methods are provided in the online version of this paper pedal digit III, but in our survey of taxa, we found none with and include the following: comparable proportions in all digits and relative to the length of the tarsometatarsus such that the third digit alone is strikingly d KEY RESOURCES TABLE elongated (Table S2). The proportionately long hallux, large d CONTACT FOR REAGENT AND RESOURCE SHARING recurved claws, and distal elongation of the pedal phalanges d METHOD DETAILS present in HPG-15-2 all suggest an arboreal ecology, which is B Specimen documentation conﬁrmed through a principle components analysis (PCA) B Phylogenetic analysis comparing neornithine pedal length proportions to that of B Principal Components Analysis Elektorornis (Figure 4G; Table S3)[22, 23]. These features are d QUANTIFICATION AND STATISTICAL ANALYSIS shared by nearly all enantiornithines, suggesting that this clade d DATA AND SOFTWARE AVAILABILITY

Current Biology 29, 2396–2401, July 22, 2019 2399 Figure 4. Comparative Drawings of Cretaceous Bird Pedal Proportions and PCA Analysis of the Pedal Proportions in Elektorornis and Select Neornithines (A) Elektorornis, (B) Eopengornis, (C) Rapaxavis, (D) Sulcavis, (E) Qiliania, (F) Gansus (all feet not to scale, but drawn so that the tarsometatarsus length is approximately equal), and (G) results of a principal components analysis of select ratios of pedal measurements (PC1 & PC2). Different locomotor substrates are indicated by different colors (red, ground; green, branches; dark blue, vertical rocks; light blue, wading; purple, vertical trunks); Elektorornis HPG-15-2 is indicated by the black circle. See also Tables S1, S2, and S3 and Figure S3.

SUPPLEMENTAL INFORMATION AUTHOR CONTRIBUTIONS

Supplemental Information can be found online at https://doi.org/10.1016/j. L.X., J.O., and R.M. designed the project; L.X., J.O., R.M., L.C., M.B., L.F., cub.2019.05.077. H.H., and J.F. performed the research; and L.X., J.O., L.C., R.M., and N.C. wrote the manuscript. ACKNOWLEDGMENTS DECLARATION OF INTERESTS We thank Chunyong Chou and Qiuxia Lin from China University of Geoscien- ces, Beijing, for measuring extant birds and R. Poulin for helpful discussions. The authors declare that they have no competing interests. This research was funded by the National Natural Science Foundation of China (nos. 41888101, 41790455, and 41772008); the National Geographic Society, Received: January 23, 2019 US (no. EC0768-15); the Fundamental Research Funds for the Central Univer- Revised: March 11, 2019 sities (no. 2652017215); and Natural Sciences and Engineering Research Accepted: May 31, 2019 Council of Canada (2015-00681). Published: July 11, 2019; corrected online: July 25, 2019

2400 Current Biology 29, 2396–2401, July 22, 2019 REFERENCES 20. O’Connor, J.K., Li, D.-Q., Lamanna, M., Wang, M., Harris, J.D., Atterholt, J.A., and You, H.-L. (2016). A new Early Cretaceous enantiornithine (Aves: 1. Ross, A., Mellish, C., York, P., and Crighton, B. (2010). Burmese amber. In Ornithothoraces) from northwestern China with elaborate tail ornamenta- Biodiversity of Fossils in Amber from the Major World Deposits, D. Penney, tion. J. Vertebr. Paleontol. 36, e1054035. ed. (Siri Scientific Press), pp. 208–235. 21. Chiappe, L.M., and Calvo, J.O. (1994). Neuquenornis volans, a new Late 2. Cruickshank, R.D., and Ko, K. (2003). Geology of an amber locality in the Cretaceous bird (Enantiornithes: Avisauridae) from Patagonia, Argentina. Hukawng Valley, northern Myanmar. J. Asian Earth Sci. 21, 441–455. J. Vertebr. Paleontol. 14, 230–246. 3. Xing, L.-D., Cockx, P., McKellar, R.C., and O’Connor, J. (2018). 22. Hopson, J.A. (2001). Ecomorphology of avian and nonavian theropod Ornamental feathers in Cretaceous Burmese amber: resolving the enigma phalangeal proportions: implications for the arboreal versus terrestrial of rachis-dominated feather structure. Journal of Palaeogeography 7, origin of bird flight. In New Perspectives on the Origin and Early 1–18. Evolution of Birds, J. Gauthier, and L.F. Gall, eds. (Peabody Museum of 4. Xing, L.-D., O’Connor, J.K., McKellar, R.C., Chiappe, L.M., Tseng, K.-W., Natural History), pp. 211–235. Li, G., and Bai, M. (2017). A mid-Cretaceous enantiornithine (Aves) hatch- 23. Hu, H., O’Connor, J.K., and Zhou, Z. (2015). A new species of pengornithi- ling preserved in Burmese amber with unusual plumage. Gondwana Res. dae (aves: enantiornithes) from the lower cretaceous of China suggests a 49, 264–277. specialized scansorial habitat previously unknown in early birds. PLoS 5. Xing, L.-D., Caldwell, M.W., Chen, R., Nydam, R.L., Palic, A., Simo˜ es, T.R., ONE 10, e0126791. McKellar, R.C., Lee, M.S.Y., Liu, Y., Shi, H.-L., et al. (2018). A mid- Cretaceous embryonic-to-neonate snake in amber from Myanmar. Sci. 24. O’Connor, J.K., Chiappe, L.M., and Bell, A. (2011). Pre-modern birds: avian Adv. 4, eaat5042. divergences in the Mesozoic. In Living Dinosaurs: the Evolutionary History of Birds, G.D. Dyke, and G. Kaiser, eds. (J. Wiley & Sons), pp. 39–114. 6. Xing, L., Stanley, E.L., Bai, M., and Blackburn, D.C. (2018). The earliest direct evidence of frogs in wet tropical forests from Cretaceous 25. Huang, W.-S. (2006). Ecological characteristics of the skink, Mabuya long- Burmese amber. Sci. Rep. 8, 8770. icaudata, on a tropical East Asian island. Copeia 2006, 293–300. 7. Daza, J.D., Stanley, E.L., Wagner, P., Bauer, A.M., and Grimaldi, D.A. 26. Stettenheim, P.R. (2000). The integumentary morphology of modern (2016). Mid-Cretaceous amber fossils illuminate the past diversity of trop- birds—an overview. Am. Zool. 40, 461–477. ical lizards. Sci. Adv. 2, e1501080. 27. Cunningham, S.J., Alley, M.R., and Castro, I. (2011). Facial bristle feather 8. Xing, L., McKellar, R.C., Wang, M., Bai, M., O’Connor, J.K., Benton, M.J., histology and morphology in New Zealand birds: implications for function. Zhang, J., Wang, Y., Tseng, K., Lockley, M.G., et al. (2016). Mummified J. Morphol. 272, 118–128. precocial bird wings in mid-Cretaceous Burmese amber. Nat. Commun. - 7, 12089. 28. Gewecke, M., and Woike, M. (1978). Breast feathers as an air current sense organ for the control of flight behaviour in a songbird (Carduelis 9. Xing, L.-D., O’Connor, J.K., McKellar, R.C., Chiappe, L.M., Bai, M., Tseng, spinus). Z. Tierpsychol. 47, 293–298. K.-W., Zhang, J., Yang, H.-D., Fang, J., and Li, G. (2018). A flattened enantiornithine in mid-Cretaceous Burmese amber: Morphology and preserva- 29. Brown, R.E., and Fedde, M.R. (1993). Airflow sensors in the avian wing. tion. Science Bulletin 63, 235–243. J. Exp. Biol. 179, 13–30. 10. Xing, L., McKellar, R.C., O’Connor, J.K., Bai, M., Tseng, K., and Chiappe, 30. Seneviratne, S.S., and Jones, I.L. (2010). Origin and maintenance of me- L.M. (2019). A fully feathered enantiornithine foot and wing fragment pre- chanosensory feather ornaments. Anim. Behav. 79, 637–644. served in mid-Cretaceous Burmese amber. Sci. Rep. 9, 927. 31. Kane, S.A., Van Beveren, D., and Dakin, R. (2018). Biomechanics of the 11. Shi, G., Grimaldi, D.A., Harlow, G.E., Wang, J., Wang, J., Yang, M., Lei, W., peafowl’s crest reveals frequencies tuned to social displays. PLoS ONE Li, Q., and Li, X. (2012). Age constraint on Burmese amber based on U-Pb 13, e0207247. dating of zircons. Cretac. Res. 37, 155–163. 32. Goloboff, P.A., Farris, J.S., and Nixon, K.C. (2008). TNT, a free program for 12. Hu, H., and O’Connor, J.K. (2017). First species of Enantiornithes from phylogenetic analysis. Cladistics 24, 774–786. Sihedang elucidates skeletal development in Early Cretaceous enantiornithines. J. Syst. Palaeontology 15, 909–926. 33. Baumel, J.J., and Witmer, L.M. (1993). Osteologia. In Handbook of Avian Anatomy: Nomina Anatomica Avium, Second Edition, Volume 23, J.J. 13. Chiappe, L.M., and Walker, C.A. (2002). Skeletal morphology and Baumel, A.S. King, J.E. Breazile, H.E. Evans, and J.C. Vanden Berge, systematics of the Cretaceous Euenantiornithes (Ornithothoraces: eds. (Nuttall Ornithological Club), pp. 45–132. Enantiornithes). In Mesozoic Birds: Above the Heads of Dinosaurs, L.M. Chiappe, and L.M. Witmer, eds. (University of California Press), pp. 240–267. 34. Lucas, S.G., and Stettenheim, P.R. (1972). Avian Anatomy: Integument 14. Chiappe, L.M. (1993). Enantiornithine (Aves) tarsometatarsi from the Part I, Volume 362 (U.S. Department of Agriculture). Cretaceous Lecho Formation of northwestern Argentina. Am. Mus. 35. Dove, C.J. (2000). A descriptive and phylogenetic analysis of plumula- Novit. 3083, 1–27. ceous feather characters in Charadriiformes. Ornithol. Monogr. 51, 2000. 15. Ji, S.-A., Atterholt, J.A., O’Connor, J.K., Lamanna, M., Harris, J., Li, D.-Q., 36. Xu, X. (2018). Mosaic Evolution in Birds: Brain vs. Feeding Apparatus You, H.-L., and Dodson, P. (2011). A new, three-dimensionally preserved (Science Bulletin). enantiornithian (Aves: Ornithothoraces) from Gansu Province, northwestern China. Zool. J. Linn. Soc. 162, 201–219. 37. Goloboff, P.A., Carpenter, J.M., Arias, J.S., and Esquivel, D.R.M. (2008). Weighting against homoplasy improves phylogenetic analysis of morpho- 16. Sanz, J.L., and Buscalioni, A.D. (1992). A new bird from the Early logical data sets. Cladistics 24, 1–16. Cretaceous of Las Hoyas, Spain, and the early radiation of birds. Palaeontology 35, 829–845. 38. Wang, M., O’Connor, J.K., Pan, Y., and Zhou, Z. (2017). A bizarre Early 17. Novas, F.E., Agnolıń, F.L., and Scanferla, C.A. (2010). New enantiornithine Cretaceous enantiornithine bird with unique crural feathers and an orni- bird (Aves, Ornithothoraces) from the Late Cretaceous of NW Argentina. thuromorph plough-shaped pygostyle. Nat. Commun. 8, 14141. C. R. Palevol 9, 499–503. 39. Zheng, X., O’Connor, J.K., Wang, X.,Wang,Y.,andZhou,Z.(2018). 18. Chiappe, L.M. (1992). Enantiornithine (Aves) tarsometatarsi and the avian af- Reinterpretation of a previously described Jehol bird clarifies early finities of theLate Cretaceous Avisauridae. J. Vertebr. Paleontol. 12, 344–350. trophic evolution in the Ornithuromorpha. Proc. Biol. Sci. 285, 20172494. 19. Atterholt, J., Hutchison, J.H., and O’Connor, J.K. (2018). The most complete enantiornithine from North America and a phylogenetic analysis of 40. del Hoyo, J., Elliot, A., and Sargatal, J. (2008). Handbook of the Birds of the the Avisauridae. PeerJ 6, e5910. World, Volumes 1–13 (Lynx Edicions).

Current Biology 29, 2396–2401, July 22, 2019 2401 STAR+METHODS

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER Software and Algorithms TNT [32] https://cladistics.org/tnt/ R 3.5.1 R Core Team (2013) https://cran.r-project.org/doc/FAQ/R-FAQ.html Amira 5.4 Visage Imaging, San Diego, USA https://amira.software.informer.com VG Studiomax 2.1 Volume Graphics, Heidelberg, Germany https://www.volumegraphics.com/en/products/ vgstudio-max.html

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources should be directed to and will be fulﬁlled by the Lead Contact, Jingmai O’Connor ([email protected]).

METHOD DETAILS

Specimen documentation HPG-15-2 was scanned with a MicroXCT 400 (Carl Zeiss X-ray Microscopy, Inc., Pleasanton, USA) at the Institute of Zoology, Chinese Academy of Sciences. The specimen was scanned with beam strength of 60 KV, and an absorption contrast and a spatial resolution of 32.0386 mm. Based on the obtained image stacks, structures of the specimen were reconstructed and isolated using Amira 5.4 (Visage Imaging, San Diego, USA). The subsequent volume rendering and animations were performed using VG Studiomax 2.1 (Volume Graphics, Heidelberg, Germany). Damage to the specimen was not permitted, which precluded subsampling for mela- nosome observations or detailed chemical mapping. Specimen observations and microphotography were conducted with a Leica SZ 12.5 stereomicroscope equipped with a Canon EOS Rebel T6i camera; an Olympus CH30 compound microscope equipped with a Sony NEX-5 camera; and a Visionary Digital imaging system, which consists of a computer driven stage with a Canon EOS 5D DSLR camera equipped with a Canon MP-E 65 mm lens and studio lighting. Submersion in glycerin was used to reduce refraction and optical distortions due to curvature in amber surfaces. The resulting images were combined with Helicon Focus software to increase depth of field at high magnifications. Lighting included UV fluorescence (395 nm) to map flow lines within the resin. Final figures were prepared with Photoshop CS6 (Adobe, San Jose, USA) and Illustrator CS6 (Adobe, San Jose, USA). Terminology follows that of Baumel and Witmer [33] for osteology, Lucas and Stettenheim [34] for integumentary structures, and Dove [35] for feather microstructure and pigmentation.

Phylogenetic analysis We explored the phylogenetic position of HPG-15-2 using the recently published Atterholt et al. [19]matrix(scoringsforHPG-15-2are as follows: 0?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????????? ?????0??11 ?01000???1 1001001000 020001[01]1?0 011?????0? 11). The final matrix consists of 42 Mesozoic birds using the Dromaeosauridae as the outgroup. These 43 operational taxonomic units were scored across 252 characters, 31 of which are ordered, and analyzed using TNT [32]. Early avian evolution is extremely homoplastic [24, 36], thus we explored the effects on the matrix of using implied weighting at various k values (k = 1–20) [37]. We found that resolution was improved by using implied weighting and that the results stabilized at k values higher than 12. In the presented analysis we used the recommended k value of 16 [37]. We conducted a heuristic search using tree-bisection reconnection (TBR), retaining the single shortest tree from every 1,000 followed by a second round of TBR. The first round produced four trees with a length of 20.96; a second round of TBR produced 32 trees of the same length. The results largely agree with recent analyses [19, 38, 39] and confirm our morphology-based hypothesis that HPG-15-2 is an enantiornithine. The equal length trees differ only in the relative placement of some derived enantiornithines, including Elektorornis.Inthe strict consensus tree (Figure 3) Elektorornis is resolved in a polytomy of derived enantiornithines consisting of Halimornis, Eoenantiornis, Elsornis, Enantiophoenix, and the Avisauridae (Consistency Index: 0.445; Retention Index: 0.653). These preliminary results suggest the Hukawng avifauna is more derived than the Jehol avifauna, consistent with its age.

Principal Components Analysis In order to determine the ecomorphospace of Elektorornis we performed a Principal Components Analysis (PCA) on the ratios of the pedal measurements of HPG-15-2 together with a sampling of extant neornithines in which the substrate interaction is known (Tables S2 and S3). Substrate locomotor bins were based on information from [40]. The data were analyzed using the princomp() e1 Current Biology 29, 2396–2401.e1–e2, July 22, 2019 function in R 3.5.1 and the following length ratios: the length of digit III relative to the length of the metatarsal III (as a proxy for tarso- metatarsal length); digit II length relative to digit III length; digit IV length relative to digit III length; and digit I length relative to digit II length. Birds were divided into the following categories of substrate interaction: ground (terrestrial); branches (arboreal); wading; and rocks (vertical) and trunks (vertical) for climbing species. Different substrate groups separated from each other along PC1 and PC2 axis (Figure 4G). HPG-15-2 falls within the overlapping morphospace occupied by terrestrial (ground) and arboreal (branches) species but clumps with arboreal taxa conﬁrming inferences based on morphology that strongly suggests Elektorornis, like other enantiornithines, was arboreal.

QUANTIFICATION AND STATISTICAL ANALYSIS

The PCA was run using the princomp() function in R 3.5.1.

DATA AND SOFTWARE AVAILABILITY

The holotype specimen of Elektorornis chenguangi has been deposited in the Hupoge Amber Museum (Tengchong City, China) under the collection number HPG-15-2.

Current Biology 29, 2396–2401.e1–e2, July 22, 2019 e2