An Archaeopteryx-like theropod dinosaur newly

BCAS Vol.25 No.4 2011 Archaeopteryx: Dinosaur or Bird? By SONG Jianlan (Staff Reporter) An Archaeopteryx-like theropod dinosaur newly found from western Liaoning Province in northeastern China would make an unusual, if not unwelcome, gift for the 150 th birthday of Archaeopteryx, the oldest bird as long-believed by paleontologists: Named as Xiaotingia zhengiis, the new species carries some critical traits suggesting that Archaeopteryx might have actually been a dinosaur. Naturally this breaking news stirred intense controversies. Was The Oldest Bird a bird? If not, what makes a bird? With these questions in mind, the author joined an exploration in search of the real first bird along with the paleontologists at the Institute of Vertebrate Paleontology and Paleoanthropology (IVPP) under CAS. Archaeopteryx not a bird This Archaeopteryx-like skeleton, as reported by dinosaur expert Prof. XU Xing at IVPP in the July 28 issue of Nature, represents an evolutionary path to a more dinosaurian end among the close relatives of early avians, and unfortunately Archaeopteryx was revealed to be following this backwards path. It seems that The First Bird got lost on its way to the perch of a bird and turned to another direction. Therefore Archaeopteryx should be given a taxonomic status as a dinosaur as seen by XU and his collaborators, rather than a bird as having been believed by paleontologists for 150 years. Photo: By courtesy of Dr. XU at IVPP The fossil of Xiaotingia zhengiis. The newly found dinosaurian species from northeastern China poses a challenge to Archaeopteryx s taxonomic position as The Oldest Bird. 266 Bulletin of the Chinese Academy of Sciences

Vol.25 No.4 2011 Science Highlights This surprising conclusion, if proven true, will also shake many well-established beliefs concerning the origin of birds and issues related to their early evolution, like the origin of flight and feathers. It is also expected to refuel the long-lasting debate between evolutionists and creationists, given this benchmark specimen s special role as a transitional fossil in confirming Charles Darwin s legendary theory articulated in his On the Origin of Species by Means of Natural Selection. Xu s team named this new species, from the strata of about 160 million years old in West Liaoning as Xiaotingia zhengiis, following the name of Mr. ZHENG Xiaoting, founder and head of the Tianyu Museum of Natural History in Shandong Province of China, in recognition of his great efforts to protect and preserve important fossilized vertebrate animals. About 800 grams in weight, the Xiaotingia had a confusing composition of characteristics very close to those of primitive birds, including its sharp, conical teeth and long and robust arms, which formerly thought to be diagnostic of Avialae. However it also had specialized feet with extremely long second-toes identical to those of all Deinonychosauria, a clade of theropod dinosaurs. Like several previously identified species from the same area by IVPP experts, it had four wings, with long flight feathers attached to the hind limbs. Based on their accumulated findings of small theropods and early birds from China, XU s team closely analyzed the morphology of Archaeopteryx, and made anew a comprehensive systematic analysis between birdlike dinosaurs and early birds, combining the new clues from the Xiaotingia specimen. They discovered that once the new species was included into the analysis, the resulting phylogenetic tree would be revised and led to some surprising conclusions, though they also admitted in the paper that their hypothesis was only weakly supported by the available data. The study revealed an interesting phenomenon, in that the overall morphology of Archaeopteryx, the Xiaotingia zhengiis and Anchiornis Huxleyi were more similar to that of Deinonychosauria, rather than other primitive birds. Among many others, the features commonly shared by them were shallow snouts and expanded regions behind their eye sockets. The quantitative phylogenetic analysis based on the traits of these species suggested that both Xiaotingia zhengiis and Archaeopteryx fall into a primitive clade of Deinonychosauria, a dinosaurian taxon. The most exciting result from our research is that both Xiaotingia zhengiis and Archaeopteryx were primitive deinonychosaurs rather than birds. In other words, Archaeopteryx was a kind of Velociraptor ancestor rather than a bird ancestor, XU emphasized in a news release of IVPP. Deinonychosaurs were a branch of dinosaurs closely related to birds. This family thrived across a vast range of lands, spreading from Late Jurassic Asia to Late Cretaceous South America and Africa. Famous representatives from this family include Velociraptor appearing in the film Jurassic Park, and the four-winged Microraptor from northeastern China. Now Archaeopteryx might actually have been a cousin of Velociraptor and Microraptor. The discovery of Xiaotingia zhengiis led to a revised understanding of phylogenetic relationships between Archaeopteryx and its close relatives. Shown is a simplified chart based on the latest analysis as published in Nature. By courtesy of Dr. XU Bulletin of the Chinese Academy of Sciences 267

BCAS Vol.25 No.4 2011 The comprehensive morphological analysis of these taxa also revealed an interesting episode in their early evolution: a family of carnivorous theropods became vegetarian, and this feeding habit was inherited by their descendants, birds. As a surprising contrast, Archaeopteryx represented a reverse evolutionary trend from vegetarian to carnivorous embodied by Deinonychosauria, which went extinct by the end of the Cretaceous epoch. The game changers Before the discovery of Xiaotingia zhengiis, disputes did exist over Archaeopteryx s ability to fly, behavior or even some of its traits, but its taxonomic position had never been challenged. What has overthrown it off the throne of The First Bird this time? Previously bone histology studies revealed that the Archaeopteryx might have grown in a way quite different from modern birds, but such results did not shake its position as a bird, introduced XU: The exact evolutionary position of Archaeopteryx shall be diagnosed through systematic analysis; therefore we adopted the cladistic method and made an accurate phylogenetic analysis. Actually XU and his IVPP colleagues have found so many missing-links between birds and dinosaurs that now the ancestors of the two families look very similar. How do they tell a bird from a dragon in the first instance? We arrange the data, or measurements of the traits of a species into a matrix. After a series of calculations we get the results showing the relationships of these birdlike dinosaurs and birds and hence know how to classify it, explained XU in response to my curiosity. In other words, nowadays scientists are not judging the identity of a species by simply comparing one or two single traits, like the ability to glide or the existence of feathers. XU and his IVPP colleagues identification and analyses of many feathered dinosaurs and early birds from China since the 1990s contributed a lot to this breath-taking finding: These previous studies have rendered them more sufficient and better woven evidence to explore broader possibility. The main players here are an assemblage of birdlike dinosaurs oviraptorosaurs and deinonychosaurs (including troodontids and dromaeosaurids) and dinosaurlike birds that belong to the avialans, commented Dr. Lawrence Witmer, Professor of Anatomy and Chang Ying- Chien Professor of Paleontology from Ohio University of the USA in his commentary published along with XU s paper in Nature, admitting that the Xiaotingia zhengiis s great phylogenetic value cannot be denied in understanding the elusive relationships between the basal clades of dinosaurs and early birds. Archaeopteryx a bird again? Naturally the subversive systematic position of Picture by XING Lida and LIU Yi; by courtesy of Dr. XU A reconstructed image of the Xiaotingia zhengiis. 268 Bulletin of the Chinese Academy of Sciences

Vol.25 No.4 2011 Science Highlights Archaeopteryx triggered a fierce contention, to a great extent due to the complex social and scientific meanings hinged on Archaeopteryx. Firstly unearthed from Bavaria, Germany, Archaeopteryx was identified in 1861, just two years after the publication of Charles Darwin s famous book, On the Origin of Species by Means of Natural Selection, and soon gained its great influence as an important piece of evidence for Darwin s evolution theory. Meanwhile as the earliest bird known to human beings, it had since been put at the heart of studies on birds origin, attracting continual interest from paleontologists all over the world. It has since been a symbol of evolution, a textbook example of a transitional fossil and, above all, the oldest and most primitive bird, remarked Dr. Witmer. This finding, he predicted, was likely to be met with considerable controversy, if not outright horror, given its historical and sociological significance, as well as its crucial position as a guide for how birds began their journey of evolution. Scholars would need to reevaluate the results from research on related issues, including the origin of birds and flight, he emphasized. In the eye of some scholars, it is still too early to deprive Archaeopteryx of the crown of The First Bird. They argue that with some critical issues on early evolution of dinosaurs and birds unsolved, other possibility exists. For example, Archaeopteryx might be able to fly, but its Deinonychosauria descendants lost this ability and degenerated to a more dinosaurian end of evolution. In that case, Archaeopteryx should be considered as a bird, so should all Paraves, which includes both its flying and flightless descendants. Adding more drama to the debate, on October 26, Dr. Michael Lee at the South Australian Museum reported in the Royal Society journal Biology Letters a different result from a phylogenetic analysis. Archaeopteryx was a bird again argued Dr. Lee. Lee gave greater weights to some special traits of the Archaeopteryx, presuming such traits might have evolved more slowly than others and hence the same amount of change in them might represent longer time of evolution. His research was evaluated as more sophisticated by the Guardian from UK. This quick U-turn might be a great relief for supporters of Archaeopteryx as the first bird and meanwhile a bitter frustration for those who have just embraced the new taxonomic position of Archaeopteryx, or the revised starting point for birds origin. XU, however, seemed to have taken this very well, with great calm as a scientist. It is not surprising at all, he responded to this news: Different methodologies often lead to different conclusions. Australian scholars came to the conclusion based on possibility-based methods, such as Bayesian and maximum likelihood methods, yet what we adopted in the analysis was the principle of parsimony. Meanwhile he emphasized: What matters here is whether or not the methodology fits to the research question. Bayesian and maximum likelihood methods, which are widely applied to phylogenetic analyses based on molecular biological data, are believed to be unsuitable for morphology data by many scholars. In the field of paleontology, phylogenetic analyses are generally based on morphologic data rather than molecular biological data, so is our work on Archaeopteryx. The principle of parsimony, instead, is the mainstream method applied to phylogenetic analyses based on morphologic data. In this sense, our result is more reliable. According to XU, some competing research groups applied Bayesian and maximum likelihood methods to analyze the morphologic data of Archaeopteryx, but their work, which came to conclusions different from XU s, failed to pass the reviewing procedure of Nature. Actually XU, one of the reviewers of Lee s manuscript contributed to the Biology Letters, recommended the Royal Society journal publish the work. Strictly speaking their method is problematic, and it recovers a very unusual phylogeny, but seeing the question through a different lens is beneficial to the exploration, he explained: With more diverse opinions joining in the debate, we would be able to approach this open question more wisely. From another perspective Prof. ZHOU Zhonghe, prestigious paleornithologist, CAS Member, Foreign Associate of the National Academy of Sciences of USA and head of IVPP, gave a reason for the rise of the controversy. Part of the reasons we often have a lot debates on the phylogenetic relationships of extinct animals is: the phylogenetic analysis we currently resort to does not provide a final solution to our questions. It is probably the best method we can have for now, yet it is obviously not as strict as some methods in physics or mathematics. Although Lee et al. (2011) claimed to have used a more sophisticated statistical method, there is no way to determine whether his result is more reliable. Bulletin of the Chinese Academy of Sciences 269

BCAS Vol.25 No.4 2011 Reconstruction by ZHAO Chuang and XING Lida; Picture by courtesy of Dr. Xu An artist impression of the Anchiornis Huxleyi, a member of the assemblage of bird-like dinosaurs that knocked the Archaeopteryx off the perch of The Oldest Bird. Personally I am more inclined to believe Archaeopteryx remains the oldest bird as concluded by Lee et al., ZHOU commented: Yet, I have to admit that I am not as sure today as I used to be. Lee et al. (2011) s result is also strange to me because Epidexipteryx and Epidendrosaurus have been referred to as birds more advanced than Archaeopteryx, which I do not think likely. With professional caution he concluded: I believe it is premature to accept either hypothesis. A lot more work needs to be done in order to have a phylogenetic tree to be generally accepted by the community. What makes a bird? To determine whether or not Archaeopteryx is a bird, one has to know what differentiates a bird from other clades. What makes a bird anyway? Unfortunately it is difficult to give a definition for Aves, according to Dr. Jingmai O Connor from the Dinosaur Institute, Natural History Museum of LA County, USA, currently a postdoctoral researcher at Prof. ZHOU Zhonghe s group at IVPP. For this reason we cannot say now whether or not Archaeopteryx was a bird, she remarked. The new study by Lee et al. (2011) does not end the debate just as the Xu et al. (2011) paper did not mean definitively that Archaeopteryx was no longer a bird in the first place. She advanced that without further closer investigation it would be hard to judge the systematic position of Archaeopteryx: Both phylogenetic positions are possible and new fossil discoveries may further change our perspective. These two papers have also brought to light a weakness of studies that are bird specific and rely on Archaeopteryx as the basal-most taxon, a hypothesis that we now know may not be true, she continued, and further pointed out that with the phylogenetic position of Archaeopteryx unresolved, further research on Mesozoic birds would be hindered. I generally agree with Jingmai, remarked Prof. ZHOU. He emphasized that both XU and Lee s work are interesting and welcome addition to the current knowledge on bird origin and early evolution, but it is hard to determine which provides a better solution to the question of whether or not Archaeopteryx is the oldest bird. Nevertheless, he pointed out: The difficulty is exciting and reflects a progress of our discoveries and understanding, as new fossil evidence has clearly narrowed the gap between 270 Bulletin of the Chinese Academy of Sciences

Vol.25 No.4 2011 Science Highlights the earliest birds and their closest ancestors although we probably will never know the answer for sure. In fact, Xiaotingia is not the first monster awakened by IVPP scientists to obscure the line between a bird and a dinosaur; the definition for a bird became vague long before its discovery. As early as in the 1990s, IVPP scholars discoveries of a series of feathery dinosaurs already broke the myth that feathers were exclusively possessed by birds. More recently, XU et al. identified and named Anchiornis Huxleyi in 2008, noting its great similarities to Archaeopteryx. The newly found Xiaotingia specimen came from the same area, and provides more crucial clues shedding light on the elusive relations between Archaeopteryx and its close relatives from the Jurassic strata of China. If Archaeopteryx is not a bird, Xiaotingia is no more than the last straw to have overwhelmed it. If we had clear defining characters, we could say whether or not Archaeopteryx had them and the debate would be over. Fifty years ago, birds were defined as feathered reptiles with a furcula, retroverted pubis and a reversed hallux (DeBeer, 1954). However, we now know many dinosaurs had feathers, a furcula, and a retroverted pubis, and some basal birds did not have fully a reversed first digit. Currently, there are no autapomorphies (unique characters) of Aves but a diagnosis could be made by the unique combination of features, such as a dorsal process on the ischium, elongate forelimb, feathers, proportions of the metacarpals, etc., suggested Dr. O Connor. Concerning what makes a bird, XU gave an answer somewhat out of my expectation. To some extent the definition of a bird could be arbitrary. We can easily include Archaeopteryx in the family of birds if we like, but what really matters here is not whether it is a bird, but the fact that it should be attributed to a different position on the evolution tree. Such a change removed it from the branch evolving towards modern birds, and placed it on the one of Deinonychosauria. This has implications to our future research on many issues, including the origin of flight. In search of The First Bird Now Archaeopteryx has to celebrate its 150 th birthday with its own identity unknown. Paradoxically, at this point it might be even more interesting to ask where the research on the origin and early evolution of birds will go, with greater uncertainty if without Archaeopteryx at the bottom of the lineage of birds. Now it seems that scientists need to trek a rockier way to find the authentic first bird. If we do need a starting point from which to investigate the evolution of birds, where should we draw this line? If Archaeopteryx is not the beginning, where is it? Scholars at IVPP have provided some clues for the search of the real first bird, no matter how vague they are. As indicated by their long-time studies, the most promising candidates might be found in the taxa such as Epidexipteryx, Jeholornis and Sapeornis, which were identified as the most basal fossil birds early in the past decade. If Archaeopteryx is not a bird, then the oldest and most basal fossil birds would be those from the Early Cretaceous Chinese Jehol Group, proposed Dr. O Connor: If Archaeopteryx is not a bird, it may still be closely related, belonging to the avian sister-group. Also she suggested: The large number of different birds known from the Jehol Fauna strongly suggests that birds originated earlier and had already diversified by the Early Cretaceous. Now the first bird is set free in the evolutionary forests, with enigmatic feathers and colors. Maybe just as what the scientists have suggested, what is most important is not to capture it, but to understand the bigger picture of the mysterious dinosaurian-avian transition. References de Beer, G.R., 1954. Archaeopteryx lithographica: a study based upon the British Museum specimen. Trustees of the British Museum, London. Lee, M. S. Y. & Worthy, T. H., 2011. Likelihood reinstates Archaeopteryx as a primitive bird, Biol. Lett., published online before print Oct. 26, 2011, doi:10.1098/rsbl.2011.0884 Witmer, L. M., 2011. An icon knocked from its perch, Nature, 475, 458 459. Xu, X., et al., 2011. An Archaeopteryx-like theropod from China and the origin of Avialae, Nature, 475, 465 470. Bulletin of the Chinese Academy of Sciences 271