HBW 12 - Foreword on Fossil Birds by Kevin J. Caley

View foreword in PDF format: 


  Today, there are estimated to be some 9800 species of birds on Planet Earth. This is only a window on the evolution of the class Aves. With those known in the fossil record, it can be increased to about 12,000 species, though about a third of these fossil species come from unknown lineages. Pierce Brodkorb estimated in 1959 that about 1,634,000 species of bird had existed over the 150 million years of avian evolution. We therefore know of well under 1%. However, our understanding of bird evolution is improving, particularly with the discovery of such fossil-rich sites as the Cretaceous deposits of northern China and the early Tertiary deposits of the Messel and Quercy, in Europe, and the Green River Formation of Wyoming, North America, while fossil collecting in other areas, such as Australia, has added to our knowledge of the diversity of the birds that once existed. What follows is a very brief summary of avian diversity, starting with the basal Urvögel, moving through the early Pygostylia and other non-carinate species, before ending with a broad survey of the modern birds— the Neornithes—which include all the bird species alive today. Before beginning, special mention should perhaps be made of certain fossils that are regarded as contentious. Of particular significance is one such fossil, Protoavis texensis, as this species was hailed by its discoverer to be the primordial bird (Chatterjee 1994, 1995). Originating in the Triassic, it would have knocked the most famous of all basal birds, Archaeopteryx, off its pedestal as the oldest bird, predating it by over 100 million years. However, severe doubt has been cast as to its avian status, with indications that it is actually an early coelurosaur—just as significant palaeontologically, but not as a bird (Witmer 2002). Archaeopteryx is still holding onto its title of ‘the oldest fossil bird’—for now.

 Avian Origins

  In 1861, a chicken-sized fossil from Bavaria turned the scientific world on its head. Named Archaeopteryx lithographica, this animal had the feathers and wings of a bird, but the body, tail and jaws of a reptile—even its wings had hands and claws, and so it was hailed as ‘proof’ that birds evolved from reptiles. It was not until much later that the true significance of Archaeopteryx was revealed, when comparisons were made with those of the coelurosaurian dinosaur Compsognathus. Indeed, so similar are these fossils that without its feather impressions, Archaeopteryx could have been regarded as just another compsognathid dinosaur, if a rather odd one. The discovery of Archaeopteryx spurred a debate over the origins of the class Aves that lasted until the late 1990s, when several small dinosaurs, from the primitive Sinosauropteryx to the dromaeosaurid Microraptor, were found in sites across northern China. Each dinosaur was covered by integuments that were obviously feathers (Chen et al. 1998, Xu, Zhou & Wang 2000, Xu, Zhou, Wang, Kuang et al. 2003). Thus, possession of feathers is not indicative of being a bird, but is a primitive (‘plesiomorphic’) state within the coelurosaurian theropods; even primitive relatives of Tyrannosaurus had protofeathers (Xu et al. 2004). Rather, it is the possession of flight, contour and down feathers, and the combination of a variety of other skeletal modifications that separate birds from their closest relatives within the Dinosauria. Archaeopteryx and its kin still act as a morphological bridge, bearing a mosaic of primitive reptilian features such as teeth and the long bony tail without a pygostyle alongside more modern bird-like features, including asymmetrical flight feathers on the wings. The more we find out about dinosaurs such as Microraptor and its relatives, the more we realise that the distinction between birds and dinosaurs is an arbitrary line based on no single, unique feature. We know that the pygostyle is not a uniquely avian feature either and is found in a number of oviraptorosaurs as well (e.g. Barsbold et al. 2000). We know that the wishbone, or ‘furcula’, is found in many theropods, including Tyrannosaurus, one of the most advanced members of that group. There is also evidence that pneumaticisation of the bone—the evolution of hollow bones supported by internal struts—is a theropod characteristic, not a purely avian one. However, no non-avian theropod found so far has all these traits in one skeleton, while the fusion of the trunk vertebrae and the supporting struts of the ribs are uniquely avian. The closest relatives of the birds, the Dromaeosauridae, possess a suite of other features that can shed light on how birds themselves came into being. If we take a closer look at the arboreal Microraptor gui, for example, we find that not only does this tiny dinosaur have flight feathers, but it has them on all four legs. These feathers are asymmetrical in form, like those of modern birds. Yet, this is a bona fide dinosaur, not a bird. Microraptor has given credence to the ‘tree-down’ hypothesis of bird origins, which had floundered through the lack of examples of arboreal dinosaurs. Until Microraptor was discovered, the alternative theory, the ‘ground-up’ concept, was gaining favour, despite there being so many problems as to how this could be the accurate scenario, given that all modern species of flying and gliding organisms are plesiomorphically arboreal. It has since been suggested that the initiation to flying came through these tiny animals jumping about in the branches of the trees of their forest home, with the hindwings forming feathered ‘edges’, which aided lift. The four-winged approach to gliding in this dromaeosaurid came as a complete surprise and led to the re-examination of Archaeopteryx and later forms (Christiansen & Bonde 2004, Zhang & Zhou 2004), which revealed that their leg feathers still bore some similarities with those of M. gui rather than with the more typical contour feathers that we see today on the legs of birds. Although there are still several opponents to the ‘birds-as-dinosaurs’ theory, general opinion is that the evidence for birds being descendents of theropod dinosaurs is overwhelming. Despite all of the recent discoveries in China, the Archaeopterygidae are still regarded as the oldest known group of birds, although they are not regarded as directly ancestral. There are still a large suite of reptilian features in the skeleton, including a reptilian physiology, a ‘reversed’ pelvis more similar to their Saurischian ancestors, an unfused rib cage, and a long, bony tail. Also unlike modern birds, the most notable feature is the lack of a horny beak. The recent description of a tenth specimen has revealed more about Archaeopteryx and may throw our interpretation of the evolution of birds through a new twist: this example reveals features not shown in the others, clarifying the position of the four toes in relation to each other and, more amazingly, the presence of a ‘switch-claw’, otherwise found only in the dromaeosaurs and the fossil bird Rahonavis. Cladistic study also indicates that this discussion on fossil birds should include the dromaeosaurs and the related troodontids as both would be classed as members of the Aves if this is defined as the clade including the ancestor of Archaeopteryx and the birds (Mayr, Pohl & Peters 2005, Mayr, Pohl, Hartman & Peters 2007), rather than lying outside the group phylogenetically (as per Sereno 1999), although there is still debate as to whether this is an accurate interpretation (Corfe & Butler 2006, Mayr & Peters 2006). Archaeopteryx itself had poorly-developed flight muscles and could not fly as well as modern birds. Its long tail shifted its centre of balance behind the wings, again unlike modern species. Its claws tell us two different tales. Those on the wings were adapted primarily for climbing, while those on the hindlimbs suggest a ground-based animal. Current consensus is that Archaeopteryx was primarily terrestrial, but may also have been a facultative percher, if it perched at all (Mayr et al. 2005). Archaeopteryx inhabited a coastal scrubland of low-growing conifers, cycads and similar species, intermingled with more tree-like species such as firs and ginkgos. This vegetation was concentrated around shallow coastal lagoons that abounded in Europe in the Late Jurassic. It may be that, at the coastal edge, as around Solnhofen, the habitat was more open than that known to exist further ‘inland’ during the period. Archaeopteryx may have utilised its limited flight capabilities in the open spaces, adopting a ground-hugging flight pattern (O’Farrell et al. 2002). Archaeopteryx was also unable to perform a sudden halt, instead needing a short runway to come to a stop. All this suggests that Archaeopteryx flew between areas of vegetation, landed on the ground, then clambered into the bushes to reach food, or to escape predators. For more than 100 years, Archaeopteryx was the only basal bird known to science. Then, in 1995, news came out of Liaoning Province, China, of Early Cretaceous birds that bore many similarities to Archaeopteryx, but had just as many differences. These were named Confuciusornis sanctus. About the size of the Eichstätt specimen of Archaeopteryx, these new birds were toothless, with evidence of jaws covered by a horny beak, while they possessed a short tail. Feather impressions were preserved around many of these fossils, revealing two types of tail feather: many had short tail feathers, but some had long strap-like tail streamers. These may represent immatures and adults or, the more popular interpretation, they represent females and males, respectively: C. sanctus may therefore be the oldest species of bird known to show sexual dimorphism. Like Archaeopteryx, Confuciusornis had forelimb claws, but in this species they were relatively more robust, although its climbing abilities have been questioned (Chiappe et al. 1999, Peters & Ji 1999). There are further modifications to the hand, in that digits I and III are strengthened, while digit II is more slender, probably being used in life to facilitate the bird’s gliding ability, suggesting that the species was better adapted to flight than Archaeopteryx, although neither had the capabilities of modern birds. Hundreds of specimens of Confuciusornis have now been found at the sites. It is known that the area of Liaoning was an extremely turbulent place in terms of volcanic activity during the Early Cretaceous, and it may be that a massmortality event has been preserved in the fossil record. Alternatively, this collection of fossils may represent a colony, indicating that Confuciusornis sanctus was, in fact, gregarious. A second, smaller, species, C. dui, has also been described, and is the earliest fossil bird recovered with its keratinous bill intact. Two others were also described within the genus, but reanalysis of the family has concluded that these were misinterpretations of fossils of C. sanctus (Chiappe et al. 1999). Confuciusornithidae does contain one other species that has stood up to analysis. In fact, it has been classified as separate generically: this is Changchengornis hengdaoziensis, from the Chinese Late Cretaceous (Ji et al. 1999). This family was recovered from deposits within what is known as the Yixian formation (Chiappe et al. 1999, Zhou et al. 2003). These deposits, together with those of the Jiufotang deposits, constitute the ‘Jehol Biota’ and cover much of eastern Asia. The palaeoecological picture of the region is one of a diverse biota, with a large variety of vertebrates and early forms of flowering plant. These are preserved in ways that suggest that they lived in a predominantly marshland or wet-forest environment and that, in some areas, for instance around Liaoning, catastrophic mass-mortality events occurred that allowed the preservation of whole communities. These Cretaceous deposits provide evidence for the co-existence of different higher groups of birds, exhibiting a mosaic of avian features that constitute different degrees of evolution. Contemporaneous with Confuciusornis, for instance, were: the basal enantiornithine Protopteryx; its more advanced relatives, such as Jibeinia; basal carinate (modern-type) species, such as Liaoningornis; and the more advanced waderlike Hongshanornis. Meanwhile, the more derived Changchengornis of the Late Cretaceous lived alongside primitive basal birds like Jeholornis. Perhaps special mention should be made of two of the basal birds, namely Jeholornis prima, and another, rather enigmatic, basal ‘bird’ from Madagascar, Rahonavis ostromi. While the exact position of Jeholornis within the class Aves has been questioned, the identity of Rahonavis even as a bird has come under scrutiny. Today, there are two camps, one which regards it as a bird (Chiappe 2002, Chiappe & Dyke 2006, Forster et al. 1998, Zhou & Zhang 2002b) and one which feels that it is better placed immediately outside this grouping (Holtz 2000, 2001), possibly within Dromaeosauridae itself (Clark et al. 2002, Makovicky et al. 2005). Some have suggested that dromaeosaurs may be more closely related to Confuciusornis and all other ‘Pygostylia’ than they are to Archaeopteryx, putting them within the radiation of what we call ‘birds’ (Mayr et al. 2005). Both birds had wing-claws and long reptilian tails, but there the similarity ends. The Chinese Jeholornis had a primitive lower torso, but possessed an advanced scapula similar to modern birds, suggesting a better capability of flight than Archaeopteryx. It was probably an arboreal species. The stomach contents are well preserved in this fossil, and indicate that Jeholornis fed on seeds (Zhou & Zhang 2002b). The remains of Rahonavis include a remarkable sickle-like claw on the second toe of the foot (Forster et al. 1998). At the time of discovery, this claw added a major feature to the debate on bird origins, for the only other animals that had been found with such a claw were the dromaeosaurs. It has now been described for the most recently discovered specimen of Archaeopteryx, providing more evidence for a relationship between the two groups (Mayr et al. 2005). The only other elements of Rahonavis to have survived are the basal vertebrae of a long tail and the upper arm bones of the right wing. Therefore any reconstruction of this species is purely supposition, although it might not be too far-fetched to suspect that the species was predatory. 

The ‘Opposite Birds’

 In 1981, Walker described a new major radiation of Cretaceous birds, which he called the Enantiornithes. He also nicknamed these the ‘opposite birds’, as one insignificantlooking, but immensely important, skeletal characteristic separates the enantiornithines from all other birds: the nature of the fusion of the tarsometatarsus. In modern birds, the fusion occurs towards the foot (distal) end, rather than the body end, i.e. it shows a ‘disto-proximal’ fusion. In the ‘opposite birds’, the fusion occurs at the body end, i.e. the opposite way round. Like the Urvögel and confuciusornithids, the Enantiornithes possessed wing-claws and a reptile-like growth-rate and physiology, both of which caused earlier researchers to group these three lineages into the Sauriurae. However, this interpretation was in error: the Enantiornithes are, in reality, the sister-group to a lineage containing all modern birds as well as such famous toothed species as Ichthyornis and Hesperornis (Chiappe & Dyke 2002). The phylogenetically important Enantiornithes are also important ecologically, for they were the dominant avian group in the Cretaceous. In their heyday towards the end of that period, the Enantiornithes occupied as diverse a set of niches as modern birds, while they were also distributed across every continent. The most primitive member of the group, the starling-sized Protopteryx fengningensis, was discovered in the same Early Cretaceous formations as the confuciusornithids (Zhang & Zhou 2000). Other primitive enantiornithines are known from Spain, including the thrush-sized Iberomesornis romeralli (Sanz et al. 1988) and the goldfinch-sized Eoalulavis hoyasi, and Australia, in the form of Nanantius eos (Molnar 1986). The enantiornithines possessed a mosaic of primitive and modern characteristics, a common feature among basal birds. Even Protopteryx exhibited a mosaic of features, including wing claws, and a keel and procoracoid to which powerful flight muscles were attached, and asymmetrical flight feathers (Zhang & Zhou 2000). However, Protopteryx is marked out among the Enantiornithes by the possession of some very unusual, barbless, strap-like tail feathers, the nature of which has been found in only a few other species, most notably as the elongated central tail feathers in Confuciusornis, whilst something superficially similar is found in a few members of the modern Paradisaeidae. Like Confuciusornis, the tail ribbons of Protopteryx may have had the primary role as a counterbalance to the body, as both birds possess remarkably short tails otherwise. It is possible that they may also have been used for display. What Protopteryx does share with other early enantiornithines is its small size, few small teeth, fairly well-developed wingclaws and a perching habit. Later species were much larger: for example, Enantiornis leali from the Late Cretaceous of Argentina had a wingspan of about 1m (Walker 1981). They were also toothless and had reduced wing-claws—in Halimornis thompsoni, a seabird of the Cretaceous American Interior Seaway, these claws were so reduced as to be vestigial (Chiappe et al. 2002). A few enantiornithines have provided us with direct evidence of their diet. Eoalulavis, for instance, as well as being celebrated as the earliest bird with an alula, is also the earliest bird so far discovered to still contain the remains of its last meal in its stomach, in this case small, aquatic, crustaceans (Sanz et al. 1996). For the majority, however, we have to look at their skeletal morphology and compare it with modern examples to work out what they may have fed on and how they lived generally. For instance, the possession of small teeth in some of the more primitive species may signify an insectivorous diet, as may have been the case with, for example, Protopteryx. Other species, such as Sinornis santensis and Boluochia zhengi, both from Cretaceous China, had adopted a raptorial lifestyle, and may have occupied the niche held today by shrikes and small falcons (Hou 1996, Zhou 1995).  Other Toothed Birds  Another group of birds arose in the Cretaceous that differed from their contemporaries through the lack of growth rings in their bones: in other words, their physiology was not of the reptilian type characteristic of the Enantiornithes, confuciusornithids and Urvögel, but was more avian in form (e.g. Chinsamy et al. 1998). These primitive ‘Ornithurae’ belonged to a new radiation that included the ancestors of the Neornithes, or modern birds. They contained some of the most famous of bird fossils other than Archaeopteryx, namely Ichthyornis and Hesperornis. There are a plethora of forms in addition to these marine species, among which are included Apsaravis ukhaana, from semi-arid dune systems of Mongolia, Patagopteryx deferrariisi, a flightless bird from Argentina (Chiappe 1996, Chiappe & Dyke 2006), and Gargantuavis philoinos of Europe, comparable in size to an ostrich (Buffetaut & Le Loeuff 1998). These toothed birds died out by the end of the Cretaceous. The most recent of the group, listed purely by its museum number, was a Belgian specimen dating from 800,000 years before the Cretaceous–Tertiary (or ‘K–T’) boundary. It was similar in form to Ichthyornis, if somewhat larger, but it was too fragmentary for proper placement (Dyke et al. 2002). Ichthyornis is perhaps the archetypal toothed flying bird. In a monograph on fossil birds in 1880, Marsh listed seven species in the genus, but these were based on a set of rather arbitrary decisions, one of which involved size differences between the specimens. The first major revision of the genus was performed by Julia Clarke in 2004, when she reduced these seven American species to one, namely I. dispar. At the same time, she showed that some of the bones attributed to the genus came from birds more closely related to modern birds than they were to Ichthyornis. Indeed, one specimen may be a close relative of the Galliformes, and therefore within the neornithine radiation, the more remarkable given its Late Cretaceous age. The interpretations of the general morphology of Ichthyornis remain similar to those described by Marsh, although it is now recognised that the American species was quite variable in size. In summary, Ichthyornis was a small Cretaceous toothed seabird with tern-like proportions, including powerful, elongate wings, an elongate bill and small legs and feet. Whether it was a pelagic species of the Cretaceous North American seaways, or whether the bodies merely floated to these deep sea positions from more coastal origins, remains a matter of conjecture. Ichthyornis has been described from the Old World and as far south as Antarctica (see Feduccia 1996), but as yet these specimens have not been compared directly with the American skeletons, so no conclusions can be made as to their relationships or even specific identities. The other major grouping, the Hesperornithes, were also marine birds, like Ichthyornis, but there the similarity ends. These were foot-propelled divers, and those so far discovered were all flightless, with vestigial wings and the lack of a sternal keel. The hesperornithids and their kin ranged in size ‘from that of a large grebe to the largest of the living penguins’ (Feduccia 1996): the American Hesperornis regalis has been estimated at 1·5m in length, while H. rossicus, from Kazakhstan and Sweden, was even larger. These birds have been likened ecologically to divers, with a diver-like morphology, similar postures, poor mobility on land and a fish-based diet. However, they also differ in a number of ways. For instance, they had a toe rotation convergent on that of the grebes and, like them, probably had lobed feet, rather than webbing. Unlike either group, the hesperornithines were more marine, possibly nesting in colonies on islets and isolated coastlines. Their flightlessness also sets them apart from the divers and most grebes: their bones were more dense than these neornithines, making them better adapted to diving. It may have meant that the birds ‘sat’ in the water lower than most modern birds, with comparisons being drawn with the subsurface swimming technique of anhingas. The hesperornithines included several species from a number of families, the most familiar of which is the American H. regalis, but with more slender relatives such as the Baptornithidae and the primitive Enaliornis. Although they disappeared at the end of the Cretaceous, their place was rapidly taken by diving birds from the neornithine groups in the early Tertiary, if not at the end of the Cretaceous.  Neornithes: Modern Birds  Of the modern bird orders, the ratites and tinamous, also called ‘palaeognaths’, are regarded as the most basal divergence (Dyke & van Tuinen 2004). However, evidence suggests that their age as a group is comparable to that of many groups of so-called ‘neognathous’ birds, which makes unravelling their history complicated. The most ancient palaeognaths come from at least the early Paleogene of North America and Europe, in the form of the Lithornithidae (Houde 1988). The fossil evidence to date suggests that palaeognaths originated in the Northern Hemisphere and is thus in contrast to evidence from the modern distribution of the ratites and tinamous, and also contra molecular studies (e.g. Cooper et al. 2001). When the genomes of two species of moa are included, the origins of the ratites are pushed back into the Cretaceous. The fossil Lithornithidae are an enigmatic group whose relationship to the other palaeognaths is poorly known, although they may have been related to the ancestor of the modern ratites (Dyke & van Tuinen 2004). The Lithornithidae sensu Houde (1988) is divided into two grades (see also Davies 2002). One, the Lithornis-Pseudocrypturus grade, consists of Pseudocrypturus cercanaxius from the Middle Eocene of Wyoming and six species of Lithornis from the Eocene of Europe (exemplified by L. vulturinus) and the Paleocene and Eocene of North America (e.g. L. promiscuus). Specimens have also been found in rocks of putative Late Cretaceous age from New Jersey, USA (Parris & Hope 2002), pushing the origin of this family back into the Cretaceous. These birds were superficially tinamou-like, but they differed in many ways, having a moderately long, sensitive bill and relatively long legs with four toes, the hallux (or ‘hind toe’) being well developed and adapted for perching. They also possessed an unreduced keel, indicating that they were capable of sustained flight, unlike any modern palaeognath. The second grade consists only of Paracathartes howardae, a much larger bird, more heavily set, and evidently a poor flier, adapted for a more cursorial way of living, and possessing a reduced hallux. Recent work on the lithornithids has confirmed their status as palaeognaths through observations on their eggshells (Grellet- Tinner & Dyke 2005), whilst new discoveries indicate that lithornithids are closely related to tinamous, rather than their similarity being purely a consequence of similar lifestyle (Leonard et al. 2005). The Lithornithidae died out by the Middle Eocene, when the only palaeognath present was the European Palaeotis  weigelti (family Palaeotididae), a more advanced species related to extant forms such as the ostriches (Houde & Haubold 1987), if not to all modern ratites (Dyke & van Tuinen 2004). Like the modern species, Palaeotis  was flightless, had long legs and lacked a hallux, consistent with a cursorial way-of-life. It also possessed a narrow bill, similar to that of a cassowary. Palaeotis  was similar in size to a small modern crane (Olson 1985). Originally described as a bustard, Palaeotis  was later redescribed as an ostrich, when it raised the possibility that the ratites were, in fact, offshoots of some neognathine lineage, with their resultant primitive states being the result of extreme neotony (retention of juvenile features into adulthood). Even the palaeognathous palate has been suggested to be a result of neotony. However, the discovery of the lithornithids and a reanalysis through morphological and molecular studies have revealed no relationship with modern neognathous birds. Combined fossil and Molecular studies suggest that the oldest ‘modern’ group of ratites are the rheas of South America (Cooper et al. 2001), with the moas of New Zealand being the next offshoot. However, other studies differ: in those that incorporate molecular and fossil work, the old associations reappear, although in a different ordering, with the ostrich/rhea partnership being the most recent split, the kiwi/moa partnership the oldest, and the cassowary/emu partnership splitting off somewhere inbetween. In this scenario, Palaeotis  evolved from the ancestor of all modern ratites (as per Dyke & van Tuinen 2004), and is not directly related to the ostriches, while the lithornithids would be ancestral to all of the above, with the tinamous being the most ancient group. This last is at odds with the fossil evidence for tinamous alone, as the oldest species found are Miocene in age (Bertelli & Chiappe 2005). The one drawback with the second molecular study is that the associated details that would allow us to see if this was a better phylogeny have not been published as yet, while arguments as to why the study differs from that of Cooper et al. (2001) have not been put forward. These are only two examples of the arguments currently raging over the origin of ratites, but what these studies have done is instigate debates about the need to link known fossils into molecular calibrations. Obviously, they have also highlighted the problems of ratite biogeography. If the ratites are of northern origin, as the discovery of Palaeotis  suggests, why are all modern ratites found on the remnants of Gondwana, with no relatives in Europe today? Conversely, if the ratites are of southern origin, as the molecular information suggests and which seems to be the more plausible explanation, where are the Gondwanan fossils? There is still a huge gap in the fossil record with respect to early Tertiary ratites. At least one of the mysteries in ratite evolution has been resolved recently, namely the origins of the two-toed ostriches of the genus Struthio. For a long while, the only early Struthio fossils known came from the Miocene of Moldavia, in the form of S. orlovi (see Feduccia 1996), while there was an extensive record of ostriches from the Pliocene and Pleistocene of Europe and Asia, including the widespread Struthio asiaticus, the European S. brachydactylus (which may also have existed in the Late Miocene), and the giant Georgian S. dmanisensis from the Early Pleistocene. At the same time, only thickset, graviportal ostriches (originally thought to belong to the Aepyornithidae), such as Diamantornis, were found in African deposits. It was theorised that Struthio had evolved in the grasslands of Asia and Europe, and that it was only later that it spread into Africa, when the graviportal species disappeared. Then, in the 1990s, a Lower Miocene ostrich, S. coppensi, was described from Namibia (Mourer-Chauviré et al. 1996), while a biostratigraphic series from the early Neogene was described from the same area, showing changes in egg type from ‘aepyornithoid’ (i.e. Diamantornis) to ‘Struthioid’ (i.e. Struthio). Thus, one graviportal ostrich at least was the direct ancestor of a modern Struthio (Senut et al. 1998). These discoveries have turned the older theories on their head, for now it is possible to state that the gracile, two-toed ostriches did originate in Africa, migrating out into Eurasia in the Late Miocene to take advantage of the spreading grasslands that subsequently dominated through the Pliocene and Early Pleistocene. The early fossil record of the other living ratites is not as full. Only the rheas have a fairly substantial record, mainly of Pliocene forms such as Hinasuri nehuensis and Pleistocene representatives such as Heterorhea dabbenei (Tambussi 1995). Possibly related to these are the Opisthodactylidae from the Upper Paleocene to Miocene, members of which include Diogenornis fragilis from southern Brazil and Opisthodactylus patagonicus from Argentina, but the fossils of many of these are rather fragmentary. All rheas are South American, so it is reasonable to assume that they originated there, rather than evolving elsewhere then migrating into South America. Likewise, the cassowaries and emus originated from within the range they occupy now, in Australia. However, they have a very limited fossil record, with a pygmy cassowary, Casuarius lydekkeri, from the Pleistocene of New Guinea and a number of emus from Australia (Vickers-Rich et al. 1988). The emus are of particular interest, as there exist two species that share characters of both the forest (cassowary) form and the bush (emu) form, namely Emuarius gidju and E. guljaruba, from the Oligocene. These are probably close to the basal stem lineage before the cassowaries and emus split and adapted to different habitat types (Boles 1992, 2005a; see also Feduccia 1996). Three modern ratite families are now extinct, namely the Aepyornithidae and the two families within the order Dinornithiformes. The elephantbirds (Aepyornithidae) were formerly thought to share a close relationship with the ostriches (see Feduccia 1996), but molecular techniques have pointed at a more derived ancestry with the kiwis, cassowaries and emus, and that the graviportal ostriches of Africa were probably not of this family, the Aepyornithidae being endemic to Madagascar (see Davies 2002). The family consisted of seven species in two genera: three species in Mullerornis, which died out during the Pleistocene, and the four larger species in Aepyornis, ranging from the Pliocene to early Holocene. These were heavy-bodied birds, but only one reached elephantine proportions, namely A. maximus, the Vouronsatrana (Marden 1967), standing at 2·7–3 m and estimated at 418–457 kg (Amadon 1947), making it the heaviest bird known: the largest extant ratite by comparison, the Ostrich (Struthio camelus) weighs a mere 100–135 kg. Like all the larger ratites, the aepyornithids were herbivorous browsers (Wetmore 1967); the giant size of the largest may have allowed it to graze the higher branches of bushes and small trees, out of reach of its smaller relatives. The last of the aepyornithids, A. maximus itself, disappeared at a time coincident with human arrival on the island, eventually becoming extinct by the end of the 16th century. It seems that human overexploitation of the environment caused its downfall, rather than direct persecution (Burney 1993, Burney et al. 2003). The speciose moas of New Zealand also succumbed to human colonisation, but in their case it was a combination of Maori hunting pressure and loss of habitat that wiped out most of the species by the 14th century (Holdaway 1989, Huynen et al. 2003, Worthy & Holdaway 2002). One species, Megalapteryx didinus, may have survived just long enough to be seen by the first European settlers. However, doubt has been cast on the reliability of these sightings, while archaeological evidence associated with Holocene moa bones indicate that moas were wiped out long before the European colonisation (Anderson 1989). The plethora of moa bones and the presence of mummies has meant that the Dinornithiformes have been extensively studied and written about, both in terms of their evolutionary history and ecology (see Worthy & Holdaway 2002). Although the oldest known moa fossils date to only 2·5 million years ago (mya), at the end of the Late Pliocene (Worthy et al. 1991), their DNA indicates that these flightless giants diverged from the other ‘modern’ lineages more than 80 mya, in the Late Cretaceous (Cooper et al. 2001). These studies also suggest that ratites invaded New Zealand on two different occasions, rather than the moas and kiwis sharing a common ancestor. The 10–11 species of moas are divided into two families: Emeidae (synonym: Anomalopterygidae), consisting of the majority of species, generally robust in form; and Dinornithidae, containing only the two species of tall moas, exemplified by the North Island Dinornis novaezealandiae, (which includes D. giganteus). The two families split from each other about 18 mya, when much of New Zealand became submerged, while the major diversification of species occurred about 4–10 mya, when New Zealand was experiencing major geological upheavals in combination with climatic cooling (Baker et al. 2005). Both moa groups are characterised by a great variation in size within the species, which has been attributed to reversed sexual dimorphism, for example in Pachyornis, Emeus and Megalapteryx (Worthy 1987, 1988), culminating in a recent combination of morphological and molecular analyses which have shown that, in Dinornis at least, the females were as much as twice the size of the males (Bunce et al. 2003). This has led to immense confusion over the number of species in the past, with up to 38 being recognised, a problem confounded by the lack of understanding that moas had a slow growth rate when compared to other birds (Turvey et al. 2005). Despite their ancient demise, moa signs can still be seen in New Zealand today, from (probable) ancient trackways (Horn 1989) to the better substantiated effects of these herbivorous birds on the evolution and ecology of many of New Zealand’s native plants (see, e.g., Bond et al. 2004, Greenwood & Atkinson 1977). The picture emerging is one of differing habitat and feeding preferences, with some species, such as Dinornis robustus and Pachyornis mappini, being widespread in a variety of habitats, the former primarily feeding on coarse twigs and fibrous shoots, the latter on tough leaves, while others, such as Anomalopteryx didiformis (another specialist of tough leaves) and D. novaezealandiae, were species dominating in the more heavily forested regions. Some, for instance Megalapteryx didinus, foraged along upland forest edges as well as in high-altitude grasslands (Worthy 2002). Finally, Euryapteryx and Emeus contained shrubland species that fed primarily on soft leaves and berries. Whilst the evolution of the ratites is tied up with the land, that of the next few groups is linked with water, in particular the sea. The ‘seabird’ morphology has occurred several times in the fossil record. We have already seen this in the toothed birds, where there are long-distance fliers, such as the enantiorn Halimornis, and diver (= gaviiform) types, such as Hesperornis, Baptornis and relatives. Even the gull morph has occurred several times, first in such toothed birds as Ichthyornis and later with the true gulls and within the Pelecaniformes. In addition, several seabird species, ancient and modern, have evolved heightened swimming skills at the cost of flight, itself an energy-costly form of locomotion. The most famous of this last group are probably the penguins. Today there are about 20 living species, but over 50 additional species are found in the fossil record, with many coming from South America and Antarctica, the majority originating from the early Tertiary. The oldest-named species, Waimanu manneringi, comes from the late Early Paleocene of New Zealand (Slack et al. 2006). At an estimated height of 80 cm, this species was about the size of a Yellow-eyed Penguin (Megadyptes antipodes). However, unlike that species, but like other ancient penguins, Waimanu possessed an elongate skull and bill. Other ancient penguins are renowned for their large size. The Late Paleocene Crossvalia unienwillia, for instance, was a penguin 1·4 m tall from Seymour Island (Tambussi et al. 2005), a place that has revealed 28% of the known fossil penguin species. The largest species, however, must be Anthropornis nordenskjoeldi, from the Late Eocene of Seymour Island, and Pachydyptes ponderosus a Late Eocene New Zealand penguin. George Gaylord Simpson (1946) estimated that these birds had a maximum standing height of 5′4′′ (1·6 m), although he admitted that there was a large degree of error involved in the sorts of estimates undertaken. These early penguins were similar in general appearance to modern species, and some species were already highly colonial, like their modern counterparts. Penguin relationships to other birds are, however, still ambiguous, with suggested connections with the tubenoses (Procellariiformes), through similarities of fledgling nasal morphology combined with other osteological features. More recently, G. Mayr (2005d) has suggested a possible close relationship between the penguins and an extinct group of cormorant-gannet relatives, the Plotopteridae, an amazing northern Pacific family of large to gigantic seabirds that, although cormorant- like in many respects, bore flipper-like wings similar to those of the penguins and which have been considered ecological equivalents since their discovery. The first species to be described was the Early Miocene Plotopterus joaquinensis of California, the coracoid of which Hildegarde Howard (1969) recognised as coming from an as yet undescribed form that ‘flew’ through the water. However, she could not state emphatically that the birds were flightless, although she did infer it. She estimated the size of the bird, likening it to Brandt’s Cormorant (Phalacrocorax penicillatus). Several more fossils have subsequently been described from sites in Japan (Olson & Hasegawa 1979), including the largest species, the Late Oligocene Copepteryx titan, estimated at 1·8 m in total length. However, the earliest species, Phocavis maritima, was found by Goedert (1988) in Late Eocene deposits in Oregon. The demise of the group in the Early Miocene has been attributed to the rise of the sea mammals, the proliferation of which may also have caused the extinction of the giant penguins in the Southern Hemisphere (Simpson 1971). However, rising sea temperatures may also have contributed to the extinction of these giant diving birds (Goedert 1988). The divers have also been suggested as the northern sister-group to the penguins through similarities in the bill morphology of some fossil penguins. However, like other groups, there are no intermediate forms that would suggest such a relationship. A substantial number of Gaviiformes have been recorded from the Miocene and Pliocene of Europe and North America. Many of these are currently placed in the genus Gavia, but Mayr (2004e) states that some of these birds are ‘clearly stem group representatives of the Gaviidae’. The marine Colymboides are the earliest definitive divers, with a range extending from the Eocene to the Early Miocene. They include the teal-sized C. minutus from the Late Oligocene/Early Miocene of France. Although these birds were recognisably divers, their leg morphology was less adapted to footpropelled swimming than modern species. Indeed, R.W. Storer, who studied C. minutus in 1956, pointed out that the differences observed between Colymboides and Gavia have a lot to do with the smaller size of Colymboides and its lesser degree of specialisation. Even so, a species found in Germany, C. metzleri, has revealed that even this primitive marine genus specialised on a diet of fish. Two putative divers, Neogaeornis and Polarornis, are of particular interest as they come from Upper Cretaceous deposits in Chile and Antarctica respectively (Chatterjee 2002, Olson 1992a); they may even be the same species (Mayr 2004e). The former consists of a single tarsometatarsus, while the remains of Polarornis are more extensive. However, even here, the reconstruction of much of the fossil has been questioned, whilst five of the six characters used to assign it to within the Neornithes are also found in more primitive Mesozoic birds. The relationship of Polarornis and Neogaeornis with the Neornithes, let alone the Gaviidae, is therefore still debatable and awaits further analysis, particularly given their physical location well outside the distribution of all other known divers. The next group in traditional classifications, the boobies, cormorants and darters, have a strong fossil record, with a total of about 80 species, examples coming from all continents except Antarctica. Of the three groups today, only the darters are primarily freshwater, and while the cormorants have freshwater examples, the boobies are exclusively marine. Although the similarity in habitat type is true for the fossil record for two of these families, the earliest species of sulid, Masillastega rectirostris, from the Middle Eocene of Messel, Germany, was at least partially freshwater. If this is indeed a sulid, it would be the only member of the family known from freshwater deposits  (Mayr 2002d). Known from a single skull, Masillastega possessed a longer bill in relation to other features than its modern counterparts; this is probably due to the specialization of modern species to the plunge-diving technique so characteristic of the Sulidae. The next oldest is Sula ronzoni, from the Early Oligocene of France, some 15 million years later, but this is a marine species, more typical of the family. The identity of other species in the family is not so controversial, yet they still reveal interesting facts. One is that the north Pacific had its own group of gannets (as opposed to boobies), although that lineage ended with the Californian Morus reyanus in the Late Pleistocene (see Olson 1985). The majority of fossil sulids come from the USA and Europe, although a few, for instance Ramphastosula ramirezi, have been discovered elsewhere (in this case, Peru). The discovery of cormorant fossils has also shown a concentration of specimens, and therefore species, in Europe and North America, although species have also been found in Africa, South-east Asia and Australia. The Phalacrocoracidae are remarkable in their overall similarity to each other, and such authorities as Feduccia have noted the lack of morphological diversity within the family, with the only bizarre species being the modern Galapagos Flightless Cormorant (Phalacrocorax harrisi)! The fossil darters are more interesting. The earliest purported species is Protoplotus beauforti, from the Eocene of Sumatra: this was a small bird, but already with the slender darter-like bill (Lambrecht, cited in Martin & Mengel 1975). However, it differed from modern anhingas in its limb proportions and small size, which contribute to its position outside of what we would term an ‘anhinga’ (Vickers-Rich, cited in Olson 1985) and has led to it being assigned its own family, Protoplotidae (van Tets et al. 1989). The next oldest, ‘true’ darters are Anhinga subvolans and Meganhinga chilensis, from the Early Miocene of Florida and Chile respectively. A. subvolans was originally described as a cormorant (Brodkorb 1956), but the condition of the humerus was noted as being similar to that of the darters, a group that tend to soar more than the cormorants (Becker 1986a). M. chilensis, on the other hand, was a flightless, giant species, like many of the Miocene anhingas from the Americas, which have also been placed in genera other than Anhinga by Noriega and his colleagues in papers from 1992 to 2004 (e.g. Noriega 1992, 2001, Noriega & Alvarenga 2002, Noriega & Piña 2004, Rinderknecht & Noriega 2002). The largest darter found to date is Macranhinga paranensis, another Upper Miocene species, this time from Argentina. Weighing an estimated 5·4 kg, it was almost four times the size of the modern American Anhinga (A. anhinga). Intriguingly, M. paranensis is thought to have hunted using a pursuit-diving technique like that of cormorants, rather than the more typical stalking style of other darters, while it also converged on the cormorants in terms of its flapping flight (Noriega 2001). It should be pointed out that whilst the anhingas listed here are all South American, fossil species are also known from North America, Africa and Australia.  There are a number of other small families within the Pelecaniformes sensu lato that deserve some mention. More recent cladistic analyses indicate that their placement within the order is purely through overall similarity of form, rather than any true phylogenetic relationship. This is complicated by the lack of early fossils, so that we are still in the dark as to what the ‘proto-sulid’ or ‘proto-pelican’ looked like. What we do find, though, are some tantalizing windows into an alternate world, where freshwater gannets lived, and where brackish-water, gull-like frigatebirds foraged on lakes in North America. One such frigatebird was the North American Limnofregata azygosternon, which fed on the fish of deep lakes in what is now Wyoming (Olson 1977). It is proposed that, like gulls today, an assemblage of different species of Limnofregata existed, taking advantage of the huge numbers of fish in this giant lake system (Olson & Matsuoka 2005). To date, we only know of two species in this genus, the afore-mentioned, short-billed L. azygosternon and the larger, longer-beaked, L. hasegawai, both of which are Lower Eocene in age. Their wings were not as long as modern species and, although these birds possessed reduced feet, they were not as small as in modern frigatebirds. Also unlike the marine frigatebirds, the toes were long and well-developed, indicating that Limnofregata was able to paddle, probably scavenging fish from the lake surface as many gulls do today. Several factors have been put forward for the extinction of these inland frigatebirds. One is the cooling of the climate that occurred in the early Middle Eocene, another, the infilling and desiccation of the Green River lakes, while the arrival of potential competitors such as the gulls may have been the final element that pushed frigatebirds into the pelagic niche that they occupy now. As old as the frigatebirds are the tropicbirds, constituting the families Prophaethontidae and Phaethontidae. For a long while, the only fossil species known was the Lower Eocene Prophaethon shrubsolei, first identified in the late 19th century from southern England, and reappraised by Harrison & Walker in 1976, when it was identified as a new family. Very recently, a new, Paleocene, species has been discovered, therefore older than Prophaethon, but equating to the age of other tentative Prophaethontid fossils found in North America. This new species, Lithoptila abdounensis, extends the fossil family biogeographically, as it was found in phosphatic deposits of the Ulen Abdou basin in Morocco; it is also the earliest African neornithine bird fossil discovered to date (Bourdon et al. 2005). Other tropicbirds, of the Phaethontidae, are putatively identified from Miocene deposits on both sides of the Atlantic. Although tropicbirds have a similar overall appearance to gannets, this may be purely a result of the plunge-diving technique that these birds specialise in, for the studies of Bourdon and her colleagues point to a relationship between the phaethontid lineages and the Procellariiformes, not the Pelecaniformes—something that has been suggested by a number of other studies. If the recent molecular analyses are correct, then even the pelicans do not belong with this grouping, being the most derived family within a lineage that includes the Shoebill and Hamerkop, usually considered Ciconiiformes (see van Tuinen et al. 2001). While there is some evidence for this in the osteology of modern species, the fossil representatives of these other families are too similar to modern forms to shed light on the relationships between the three groups. The fossil record of the pelicans themselves reveals a number of small species from the Tertiary, and several species from the Miocene and Pliocene: the Pelecanidae were widespread early in their fossil history.   Often portrayed as pelican-like, but in reality the most unusual group associated with the Pelecaniformes, is an extinct lineage known as the ‘pseudodontorns’, a name given to them because of the bony, tooth-like projections along the cutting edges of their immense mandibles (Feduccia 1996). These were giants among the seabirds, yet they were incredibly light, with elongate bones and wings specialised for gliding. For instance, Osteodontornis orri, from Late Oligocene and Miocene water off California, had a wingspan of 5·5–6 m (Olson 1985), whilst the smallest species were about the same size as modern albatrosses. This may not be a coincidence, for Olson suggests that, like extant albatrosses, pseudodontorns were probably pelagic species. According to Zusi & Warheit (1992), these gigantic gliding birds fed by taking prey on or from just below the surface whilst they were in flight, using a downward nod of the head, similar to the surface-plucking behaviour of frigatebirds, or sitting on the surface, using a lunging behaviour that can be seen in modern pelicans: the tooth-pegs on the bill may have enabled the birds to hold on to soft-bodied prey such as squid. The pseudodontorns first appear in the fossil record with Pseudodontornis tenuirostris from Late Paleocene deposits in the British Isles and Odontopteryx species from the Late Paleocene/Early Eocene of Morocco. The first Pacific records come from Middle Eocene deposits in Washington State (Bourdon 2005, Warheit 2002). These birds spread throughout the Atlantic and Pacific oceans very early in their history, and were found as far south as Antarctica. Their reign lasted 57–59 million years, with the last representatives of the group, members of the genus Pelagornis, coming from the Pliocene. Like the Pelecaniformes in general, there is a lot of controversy over the phylogeny of the pseudodontorns. They have been assigned to a separate order, the Odontopterygiformes, with an association with the Procellariiformes and Pelecaniformes, or as a separate suborder (Odontopterygia) within the Pelecaniformes itself. Within the latter, most recent authors list the species under one family, the Pelagornithidae, but with the proviso that the exact relationship to other Pelecaniformes is uncertain. Given that the Pelecaniformes are now generally regarded as being polyphyletic, it is perhaps no surprise to find other proposals have been put forward. The most recent cladistic analysis, for example, suggests that the pseudodontorns (as the order Odontopterygiformes) are the sister-taxon to the Anseriformes (Bourdon 2005). However, there are some problems with this study, as other anseriform relatives such as the gastornithids were not included in the analysis, which might have clarified the degree of this relationship.  The earliest fossils of the Procellariiformes come from the Cretaceous and Paleocene, but most are too fragmentary for proper identification, and thus are deemed unassignable (Dyke & van Tuinen, 2005). An isolated humerus from New Jersey Greensand deposits (Lower Paleocene) was listed as Tyttostonyx glauconiticus, and is regarded as incertae sedis by Kenneth Warheit (2002), although assigned to the family Tyttostonychidae and tentatively placed in the Procellariiformes by Olson & Parris (1987). The earliest definable procellariiform species in Europe are members of the Diomedioididae, reported from the Lower Oligocene of Germany, France & Belgium, as well as from Iran (Mayr et al. 2002, Peters & Hamedani 2000). However, it is not known to which of the extant groups they were most closely related. Beautifully preserved specimens of Diomedioides brodkorbi (Germany, Early Oligocene) show a peculiar foot morphology most similar to the modern storm-petrel, Nesofregata. Although twice the size of the latter, these birds may have shared similar feeding habits, flying into the wind and trailing their feet in the water as they searched for surface prey. All other tubenose fossils belong to extant families, some of which fill anomalies in their distribution. The Diomedeidae, for example, are absent today from the north Atlantic. However, at least five species existed there in the Pliocene, including Phoebastria/Diomedea anglica, also found in the Pacific, and P. rexsularum, endemic to the north Atlantic (Olson & Rasmussen 2001). A species known as P. howardae was also described, but this has since been synonymised with the Short-tailed Albatross (Phoebastria/Diomedea albatrus), and accounts for one of three modern species whose ranges extended into the north Atlantic. Bermuda records of this species are particularly interesting as they show a healthy colony existed until the Middle Pleistocene, when a momentary 20 m rise in sea level caused its extinction (Olson & Hearty 2003), thus removing albatrosses from the north Atlantic avifauna. The Procellariidae are the most speciose fossil-wise of the other tubenoses. The coast of Miocene California has been particularly fruitful in searching for procellariiform fossils, and several species of the modern genus Puffinus have been uncovered, while the genus Fulmarus was also more speciose during this period than it is now. Like fossil members of the Hydrobatidae and Pelecanoididae, these birds are generally indistinguishable from modern genera even if, at the species level, they can be separated. The long-legged waders are usually the next group in traditional classifications. However, special mention should be made of one group that is usually placed alongside the divers and penguins at the beginning of traditional classifications, namely the grebes, primarily because recent molecular and morphological studies suggest that they are closest to the flamingos and relatives (Mayr 2004f, Manegold 2006, van Tuinen et al. 2001), a group traditionally associated loosely with the Ciconiiformes. The grebes have a poor fossil record that sheds little light on any relationship. The earliest described species is Miobaptus walteri, from the Early Miocene of Czechoslovakia (Švec 1984), but this is already a typical grebe, as are the several species of Podiceps and Aechmophorus from the North American Pliocene. If correct, and given the fact that even flamingos swim on occasion, and that their closest fossil relatives, the Palaelodidae, were better adapted to swimming, it is suggested that the ancestor of the Phoenicopteriformes + Podicipediformes was aquatic, rather than a wader. However, the earliest members of the flamingo group, Juncitarsus gracillimus, from the Eocene of North America and the contemporaneous J. merkeli of Europe, were small, colonial, wader-like birds inhabiting saline environments (Feduccia 1996). Mayr suggests that further study may reveal the juncitarsids to be the sister-group to the flamingos + grebes, rather than being ancestral, if they are closely related to these two groups at all. The next oldest in terms of fossils are the Phoenicopteridae, with representatives in the Early Oligocene of France. The most surprising discovery, however, is the existence of several species in Australia, as well as a species of Phoeniconaias (P. siamensis) from the Miocene of Thailand. An Australian species which also belongs in this genus, P. gracilis, lived in the Lake Eyre region in the Early Pleistocene, and is the youngest flamingo known from that continent. Another Australian flamingo, the Late Oligocene/Early Miocene Phoenicopterus novaehollandiae, was a member of the most widespread genus today. Phoenicopterus, is the most speciose genus in the family generally, including several fossil representatives in the New World. Flamingos can be recognised by their extraordinary long legs and neck combined with a deeply decurved bill specially adapted to filter-feeding. In contrast, the closely-related Palaelodidae possessed a short, straight bill. Like the Phoenicopteridae, they were probably planktonivores, but had a more primitive morphology. In some areas, the palaelodids were extremely common and, like the Phoenicopteridae, were highly colonial. For instance, at least 477 individuals of Palaelodus ambiguus have been recovered from the Miocene deposits of St Gerand-le-Puy, France, along with large numbers of bones of three other species (see Feduccia 1996). Also like the Phoenicopteridae, the Palaelodidae have been found in Australia (P. kadimakari), in Europe and in North America, where the second genus in the family, Megapalaelodus, also existed.  The other long-legged waders are collected in the order Ciconiiformes. The most distinctive are probably the herons, the earliest fossils of which are already characteristically ardeid; these are Proardea amissa from the Upper Oligocene of Europe (Mayr 2005a), and Calcardea junei from Eocene deposits in North America. Two other families, described as ‘aberrant African forms’ by Feduccia (1996), the Hamerkop (Scopidae) and the Shoebill (Balaenicipitidae), may actually be close relatives to the pelicans (Mayr 2003c, van Tuinen et al. 2001), rather than having a close relationship with other Ciconiiformes (contra Feduccia 1977). The fossil record of these two families lends nothing to the debate: that of Balaenicipitidae is too fragmentary, although one species, Paludavis richae, does extend the family’s range into the Oriental region (Harrison & Walker 1982); that of Scopidae consists of one Early Pliocene species from South Africa, Scopus xenopus, ‘slightly larger than the living species’ (Feduccia 1996). Within traditional classifications, it is the Ciconiidae that are regarded as the closest relatives of the Shoebill, while the Hamerkop is of unknown relationships. The storks themselves are also suggested to be derived relatives of the cathartid vultures. The earliest species of stork known to date, Palaeoephippiorhynchus dietrichi, comes from the Late Eocene to Late Oligocene deposits of the Fayum series of Egypt (Boles 2005b, Feduccia 1996, Mayr 2005a); the exact age of this species appears to be contentious. Likewise, the Middle Eocene Chinese ‘stork’, Eociconia sangequanensis (Hou 1989), needs confirmation before it can be claimed as the earliest ciconiid (Unwin 1993), while a number of other potential ‘storks’ are too fragmentary to be identified properly. Of the later species, Leptoptilus falconeri is of particular interest, as the distribution of this giant marabou stretched across north Africa, Eurasia and into India. At 2 m in height and weighing an estimated 20 kg, this was one of the largest of the storks. Being predominantly terrestrial, L. falconeri probably had slightly reduced forelimbs, leading to a marginal reduction in flight ability (Louchart, Vignaud et al. 2005). Another giant stork, originally described as a Mycteria, is Ciconia maltha, a North American species from the Upper Pliocene and Pleistocene that possessed a large size range, from smaller than the wood-stork to larger than the Jabiru (Miller 1932). Fossil ibises are equally spectacular, with those of particular interest originating from the extremes of the fossil and subfossil record. The earliest identifiable ibis, the German Rhynchaeites messelensis, is also the smallest member of the family. This species is famous for its short metatarsus, which it shares with another, as yet, undescribed species from France (Mayr 2005a). At the other end of the timeline are a number of insular flightless species, with Xenicibis xympithecus from Jamaica (Olson & Steadman 1977) and two species of Apteribis (A. glenos and A. brevis) from the Hawaiian Islands. These ibises had short stout legs and, when their bones were first discovered, their identity as members of the Threskiornithidae were obscured; indeed, Xenicibis was originally described as a relative of the sloths and anteaters, rather than a bird! When the bones of the Hawaiian species were first examined, they were said to resemble most closely kiwis: it is thought that these birds searched the leaf litter for invertebrates much in the way that kiwis do today.  Another waterbird group, Anseriformes, is one of the best represented orders in the fossil record, with nearly 200 described fossil species of ducks, swans and geese. It is also one of the most widespread orders, with examples from all over the globe and from as far back as the Cretaceous. Early species were comparatively small, ducklike animals, while later species often became gigantic and flightless, particularly island forms. The earliest representatives are clustered within the Presbyornithidae, and bear many superficial similarities with the whistling ducks. The skull of Presbyornis bears a close resemblance to Stictonetta, a basal species within the Anatidae (Olson & Feduccia 1980). Like many early groups, their relationships have come under severe scrutiny, with affiliations suggested with the the flamingos, and the shorebirds and even such enigmatic groups as the Graculavidae, which led Feduccia to conclude that Early Paleocene birds all stem from a possible shorebird mosaic group with morphological similarities to a number of modern orders, be they ducks (as in the case of Presbyornis), ibises (for example the Messel ‘rail-ibis’, Rhynchaeites), or even flamingos (Juncitarsus). However, such a proposal has been rejected by many in the palaeontological and molecular communities (e.g. Dyke & van Tuinen 2004, Paton et al. 2002, van Tuinen et al. 2003), although Feduccia has since countered these arguments (Feduccia 2003a, 2003b). Even so, recent studies have moved the Presbyornithidae into the Anatoidea, as a group basal to the Anatidae (Ericson 1997), which must cast further doubt on this ‘transitional shorebird’ hypothesis. Presbyornithids were a widespread group, with examples from sites across the Northern Hemisphere. The most abundant fossils are of Presbyornis pervetus, a colonial, sexually dimorphic species from Eocene deposits of Utah and Wyoming (Ericson 2000). Another species, P. isoni, was a giant relative. Originally described from a single humerus from the Upper Paleocene of Maryland, it has now been found at other sites in North America and possibly also Britain (Benson 1999, Olson 1994). Teviornis gobiensis, another giant presbyornithid, but from Upper Cretaceous deposits in Mongolia, is also worth mentioning as it pushes the group—and, indeed, the Anatoidea generally—back before the infamous K–T boundary, thereby presenting important evidence that the Neornithes do have origins in the Mesozoic, rather than appearing after the K–T cataclysm (Kurochkin et al. 2002). Other ducks have been described from this period, most notably Vegavis iaai from the Antarctic Peninsula. However, these Cretaceous fossils are all very fragmentary, and can provide only ‘tantalising indications’, rather than irrefutable evidence (Clarke et al. 2005, Dyke & van Tuinen 2004, Noriega & Tambussi 1995).  The other notable proliferation within the fossil Anseriformes is of flightless forms. Many are island species, such as the giant swan Cygnus falconeri, uncovered from the European Pleistocene (Northcote 1982), but coastal examples also exist, for example the Californian ‘diving goose’ Chendytes lawi from the Late Pleistocene (Howard 1947, Miller 1925). New Zealand had a number of recently extinct anatids, many of which, again, were flightless. This avifauna included two Cnemiornis—giant, 1-m tall, goose-like birds—of the North (Cnemiornis gracilis) and South Islands (C. calcitrans), possibly related to shelducks, but with their modern ecological equivalent being the Cape Barren Goose (Cereopsis novaehollandiae) of southern Australia. However, it is the islands of Hawaii that have revealed the most bizarre forms. Here, primitive dabbling ducks (Anatinae) radiated through the early islands, evolving into large flightless, terrestrial ‘geese’ (Sorenson et al. 1999). Called ‘Moa Nalo’ in the native Hawaiian, these birds foraged in the forests, feeding on a variety of herbage. Four species have been described, belonging to three genera, including Thambetochen and Ptaiochen, with T. chauliodous, the most widespread species, from Molokai, Oahu and Maui, a related species, T. xanion found only on Oahu, and P. pau from Maui (Olson & James 1982). One species, Chelychelynechen quassus, had a bill particularly modified for such browsing, to the extent that it resembled more the jaws of a turtle than a duck’s bill, hence the popular name for the species, ‘Turtle-billed Moa-nalo’ (Olson & Wetmore 1976). Like 90% of the birds that were found on the Hawaiian islands, their demise occurred with colonisation by the Polynesian people. The Moa Nalo, like many island species worldwide, such as the Dodo (Raphus cucullatus), had no experience of such a successful terrestrial predator as Homo sapiens, nor of the other threats that came with him, so they had no way to cope with the onslaught; as a result, many species were wiped off the face of the earth within comparatively few years. In 1881 a very different type of bird was described by Lemoine from Paleocene/ Eocene deposits in France. Although missing much of the skeleton, he concluded that this was a huge animal, taller than a man. This he named Gastornis parisiensis. However, his reconstruction included a substantial amount of material that was not even avian (Martin 1992). For a long period, these birds were considered relatives of the terrorbirds—giant predatory ‘seriemas’ in the order Gruiformes from the Americas. It was not until a century after Lemoine’s description that his species and its relatives, known collectively as the Gastornithidae, were placed close to the Anseriformes (Andors 1991) and were identified as sharing a close relationship with the American Diatryma (Olson 1985)—so close, in fact, that the latter is now included in Gastornis (Buffetaut 1997). Gastornis (‘Diatryma’) gigantea, the most famous of the group, was a giant flightless bird, at least 2 m tall, found in North America during the Early Eocene, 58–51 mya, while its relatives lived in Europe throughout the Late Paleocene to Middle Eocene, 62–43 mya, at least. Fragmentary fossils indicate that gastornithids may have survived until about 40 mya. Examples were also present in Eocene Asia, in the form of Zhonguanus xichuanensis (Hou 1980), but this bird was a more primitive form to that of its larger cousins (Andors 1992). As much controversy surrounds how the birds lived as with their relationships. They were originally portrayed as terrifying giant predators that ran down and fed on the various mammals that were around during the early Tertiary. In 1991, Andors cast doubt on this long-held portrayal, pointing at the absence of a hook on the huge beak, the short toes and a physical appearance that made the animal far too heavy and large to be able to run down anything (Andors 1991, 1992)! Instead, he presented the bird as a particularly large browsing animal, feeding on bushes and shrubs in the lowlands and floodplains, comparing it in part with modern herbivorous species such as New Zealand’s surviving Takahe (Porphyrio mantelli), which also has a large bill. Even so, counter-arguments still exist suggesting that this was a giant predator that, although slow compared to modern predators, was still able to tackle those animals not in the prime of life through age, sickness or injury, whilst also scavenging on their remains (Witmer & Rose 1991). The same arguments also surround the mihirungs or ‘thunderbirds’ (Dromornithidae), another group of giant flightless birds rather ratite-like in form, and also now thought to be early offshoots of the Anseriformes (Murray & Megirian 1998), although their exact relationships with other basal groups, namely the Magpie-goose (Anseranas) and the screamers (Anhimidae), are still under question. The mihirungs occurred only in Australia, occupying a variety of niches (Murray & Vickers-Rich 2004). To date, eight species have been described, the youngest being the Pleistocene Genyornis newtoni, the oldest being the Late Oligocene/Early Miocene Barawertornis tedfordi. Trace and fragmentary fossils indicate that the dromornithids may have been around at least since the Eocene. The mihirungs included amongst their number one of the largest birds that ever lived, Dromornis stirtoni, standing at 3 m in height. Two species within this family are particularly interesting: one, Genyornis newtoni, may have come into contact with modern humans, with its extinction being connected with the spread of the aboriginal peoples through Australia. The other is the 15 million year old Bullockornis planei, because a lot of work has gone into examining whether this giant- billed animal fed on meat or plant material. Indeed, the latter species has been labelled in the popular press as the ‘Demon Duck of Doom’ (Pain 2000, Wroe 1999)! However, Murray and Vickers-Rich have maintained that because the mihirungs are basal anseriforms, they must have been predominantly herbivores. They argued that the mihirungs grew huge in response to the need to travel large distances between food patches. Physical evidence for their herbivory exists in the possession of large gizzard stone sets, together with, at best, a small terminal hook, both features that indicate a herbivorous, rather than carnivorous, diet (Murray & Megirian 1998). So, instead of meat, they may have fed on twigs, fibrous vegetation and hard-shelled seeds. The most unusual feature of the family are their toe claws, which have been drastically modified, becoming far more nail-like than any other bird species. These are not the raptorial claws of predators. The modification is so extreme that these ‘nails’ have been likened to the hooves found on cattle, with the mihirungs often being described as the only hooved birds on the planet (Vickers-Rich 1980).  There is general agreement among palaeontologists and molecular biologists that the Galliformes are the closest living relatives of the wildfowl and their kin. Like that group, the Galliformes have an extensive fossil record. The earliest examples were members of the now extinct Gallinuloididae, a group of basal ‘landfowl’ found across North America and Europe. Of these birds, the most intensively studied is Gallinuloides wyomingensis, even though it is known from only two specimens, both from the Lower Eocene deposits of the Green River Formation of Wyoming (Mayr & Weidig 2004). This was a small bird, about the size of a quail, but with relatively long legs and which probably had the appearance of a small cracid in life, as comparisons have been made in the past with that family (e.g. Olson 1985), even to the point of Gallinuloides being included in the family (e.g. Tordoff & Macdonald 1957). However, these characteristics are currently regarded as primitive traits in the Galliformes, rather than showing a true relationship between the two groups (Dyke 2003, Mayr 2005a, Mayr & Weidig 2004). A related species, Paraortygoides messelensis, has revealed two features about the gallinuloidids (Mayr 2006a): first, they are the sister-group to all other galliform birds, living and extinct; and, second, a large crop, characteristic of all other Galliformes, is absent in gallinuloidids. This latter feature—or, rather, its absence—suggests that the gallinuloidids fed only on soft plant material, rather than seeds and other coarse matter. Another basal galliform group, the Quercymegapodiidae, were also misinterpreted at one time as belonging to an extant group, in this case the primarily Australasian Megapodiidae. However, like the Gallinuloididae, the traits suggesting this relationship are now interpreted as plesiomorphic, and that there is no basis for the suggested relationship, other than that they are both galliform. Currently, members of the Quercymegapodiidae are known from Europe, from where the type genus Quercymegapodius was described, as well as from deposits in the Miocene of Brazil. A number of present-day groups within the Galliformes are also of interest, including some members of the more primitive groups, Cracidae and Megapodiidae. One of these is the famed Du (Sylviornis neocaledoniae), a flightless giant ‘megapode’ from New Caledonia; although usually included in Megapodiidae, it is regarded as being so aberrant as sometimes to be placed in its own family, Sylviornithidae. Reconstructions of the species show a bird almost three times the size of the modern Malleefowl (Leipoa ocellata) with a comb wattle on its head, the latter feature gleaned from the oral tradition of the islanders (see Feduccia 1996). This species was exterminated in the early Holocene by the first human colonists, like so many other island species in the Pacific and elsewhere. It was not the only giant megapode, as these existed elsewhere in Australasia at about the same time, with the flightless Megavitiornis altirostris from Fiji rivalling Sylviornis in size (Worthy 2000), and Leipoa gallinacea from Australia itself. While the story of the megapodes is one of expansion and diversification on many different islands and within Australia, that of the cracids of the Americas is one of a northerly origination, followed by a retraction, perhaps due to competition with other Galliformes, into the Neotropics. Procrax brevipes, for instance, was a bird from the Lower Oligocene of South Dakota (Tordoff & Macdonald 1957). The genus Ortalis was present in the Lower Miocene of Nebraska, through O. tantala, a small guan half the size of the modern species O. vetula; and the Middle Miocene of South Dakota, through O. pollicaris. Of the other Galliformes, it is perhaps surprising to discover that there was an abundance of turkey species in the past, including some tiny species. The miniature turkey Rhegminornis calobates, for instance, was so small that when it was discovered in Florida it was described as a jacana (Wetmore 1943a), and it was not until 1974 that its true identity as a Lower Miocene turkey was announced by Olson & Farrand. Other species within the group included members of Proagriocharis and Parapavo, as well as additional members of the two modern genera, together raising the family diversity to about 11 species. The Phasianidae and Tetraonidae also have many fossils, but these consist mainly of range extensions for genera still present today: for example, members of the genus Gallus have been discovered as far west as the Caucasus, in Georgia, while the genus Pavo had a repre- sentative, P. bravardi, in the European avifauna during the Upper Pliocene (Boev 2002, Mourer-Chauviré 1989a).  Raptorial birds have evolved several times throughout the fossil record, both within the toothed birds (e.g. Boluochia of the Enantiornithes) and within modern birds. The most obvious of these latter groups are the so-called ‘raptors’, currently within the order Falconiformes. However, this grouping is artificial, forcing together two groups of birds of very different origins. That of the Accipitridae and their relatives is still ambiguous. In the case of the Cathartidae, both molecular and morphological work point to an ancestry shared with the storks (Ciconiidae), rather than falconiform raptors. The oldest fossil in the group, that of the small Diatropornis ellioti, comes not from the Americas, but from the Quercy Fissure deposits of France, and is Late Eocene to Early Oligocene in age (Cracraft & Vickers-Rich 1972, Mourer-Chauviré 2002). Two other species come from early in the history of the Cathartidae: one, Phasmagyps patritus, originates from the Early Oligocene of Colorado, and therefore comes from within the range of modern species. The other, Oligocathartes olsoni, is from the Lower Oligocene of England. However, Mayr (2005a) regards the latter as too fragmentary to be identifiable and perhaps this species should be removed from the Cathartidae. Although the origins of the family may not have been in the New World, or, at least, their origins may have been shared with Europe (as seems to be the case with many groups), the majority of species are from the Americas. Hadrogyps aigialeus represents a small stocky ‘condor’ from the Middle Miocene of California and was probably a coastal species (Emslie 1988), while the next ‘large condor’ is the Late Miocene/Early Pliocene Perugyps diazi, from the Pisco formation of Peru. Together, this suggests that condors evolved in North America and spread to South America by the Late Miocene (Stucchi & Emslie 2005). Cathartid vultures were at their most diverse in the Pleistocene, when several species of condor existed in North America. The formerly widespread Gymnogyps californianus, which we call the ‘California Condor’, had a range that stretched throughout the USA in the Late Pleistocene, but its range contracted with the progressive extinction of the North American megafauna (Steadman & Miller 1987). Another species, G. kofordi, was described from the Early Pleistocene of Florida, while a larger form of the modern species, G. californianus amplus, was exumed from the La Brea tar pits, California. In addition, a closelyrelated condor, G. varonai, was found in Late Pleistocene deposits of Cuba. Alongside the Cathartidae in the Americas was another vulture-like group, the teratorns (Teratornithidae). Three of the four species were North American, the exception being Argentavis magnificens, whose fossil was discovered in Late Miocene deposits of Argentina, and which must have been a huge bird, with a wingspan of 6–8 m. Indeed, A. magnificens is considered to be the largest flying bird ever to have lived. Not only was this the largest of the teratorns, but it was also the oldest (Campbell & Tonni 1983), although it was not the first species in the family to be described. That honour goes to Teratornis merriami, which was also the most abundant of the group. Teratornis was a species whose temporal range extended from the Pliocene until the Late Pleistocene. It is probably typical of the group in many ways. Merriam’s Teratorn, as it has been named, was also a large bird, but only half the size of Argentavis, having a wingspan of a mere 3–4 m; if one compares this with the Andean Condor (Vultur gryphus), which can attain a span of about 3·2 m, one can begin to see the true size of these birds. Traditionally, the teratorns have been portrayed as giant versions of the Cathartidae, even sharing their scavenging habits, a perception reinforced by the shape of the skull and by discovery of over a hundred individals of T. merriami in the asphalt deposits of La Brea, a site that has produced hundreds of fossils of raptors and vultures, all thought to be attracted to the animals that became trapped there. However, instead of being scavengers, teratorns are now considered to have been active hunters, with T. merriami being a fish-eater and facultative scavenger (Hertel 1995). In the air, these birds may have soared like condors, but on the ground they were apparently more agile: in the smaller teratorns such as T. merriami, this agility may have helped them in their attempts to take off. How the giant A. magnificens took off is a matter for conjecture, but the age of the deposits in which it was found and the geographical position of the site indicate that its flight was enabled by the constant strong winds that crossed the South American plains in the Late Miocene, unhindered by the Andes, which were, at the time, undergoing upheaval. Another deduction from the age of this specimen is that teratorns probably evolved in South America, later spreading north into North America, before finally dying out at the end of the Pleistocene.  Just as cathartids have been found in the Paleogene deposits of Europe, so have true vultures been discovered in North America. The most diverse of these New World ‘Old World’ groups were the neophrons (Neophrontops spp.). Fossils of Neophrontops americanus, the first species to be described, have been found in Middle and Late Pleistocene deposits of western North America. It is one of the most recent species in the group: the oldest is probably N. vetustus, from the Middle Miocene of Nebraska (Wetmore 1943b), which is also the smallest. N. americanus is fairly typical of the American Neophrons and suggests that they were most similar to, and probably ecological equivalents of, the Egyptian Vulture (Neophron percnopterus), with N. americanus being very similar in size to the modern Neophron. Feduccia goes further than this, stating that the Egyptian Vulture’s lack of a fossil record and the similarity between the two genera ‘strongly suggests that the genus’ [= Neophron] ‘is derived from an invasion of the Old World by New World Neophrontops stock, possibly as late as Pleistocene time’ (Feduccia 1974). While Neophron may be of American origin, that of the other Old World vultures can equally be American or Old World: the oldest fossils for the group include Palaeohierax gervaisii from the Oligocene/Early Miocene of Europe and Palaeoborus rosatus, from the Early Miocene of South Dakota. Apart from Neophron and its relatives, vultures are among the largest of the living raptors. However, there are notable examples in the fossil record of eagle-like forms of gigantic proportions, truly terrors of the air! Among these, and probably the most famous, is Haast’s Eagle (Hieraaetus moorei) from the South Island, New Zealand (see Worthy & Holdaway 2002, for an extensive description). Until recently, the species was listed as Harpagornis moorei, but recent molecular research (Bunce et al. 2005) has shown that the closest relatives among those eagles studied were the Booted (Hieraaetus pennatus) and Little Eagles (H. morphnoides), the latter an Australopapuan species with a wingspan of only about 1 m and a weight of 0·5–1·3 kg. Hieraaetus moorei, on the other hand, had a wingspan of 2–3 m, with females weighing 10–15 kg—an order of magnitude larger than extant Hieraaetus, and 30–40% larger than the Harpy Eagle (Harpia harpyja), the largest extant eagle. The main reason why this bird became so large becomes apparent when examining the ecology of prehuman New Zealand, for this was a land without any land predators and with only a few bird predators, included among diurnal species being an endemic species of harrier (Circus eylesi), also now extinct, a hawk and a falcon (Holdaway et al. 2001). Additionally, without mammalian predators, the herbivorous birds of New Zealand grew to massive proportions, most notable being the moas. Thus, the eagle’s huge size was a response to a combination of competition with the resident raptors, a lack of competition with mammalian predators and the size of prey potentially available. Marks on the bones of moa up to 200 kg suggest that the eagle ambushed these large birds in forested and scrub areas, striking and gripping the moa’s pelvic area, then killing with a single strike of the powerful foot to the neck or head. Giant eagle-like raptors have been found on islands elsewhere, for example Stephanoaetus mahery of Madagascar, which, like H. moorei, was present until fairly recent times, and the appropriately named Titanohierax borrasi of Pleistocene Cuba. In each case, it is the lack of large mammalian predators and the availability of large prey that induces the evolution of gigantism in the respective avian top predators. The fossil record reveals several other hawks and eagles, from island forms to species in genera that today are regarded as tropical or southern endemics but which were found much further north in the past, for instance, in the Americas, Spizaetus and Buteogallus. This is also the case with the secretarybirds (Sagittariidae), in which two of the three fossil species known, Pelargopappus schlosseri from the Middle to Late Oligocene and P. magnus from the Early Miocene, were found not in Africa, as one would have expected, given the distribution of the extant species, but in France, indicating the modern species to be relictual (see Mourer-Chauviré 2003 for the first African birds). Compared to Sagittarius, the foot bones of Pelargopappus were more accipitrid-like in form. Mayr (2005a) has suggested that this is because the Sagittariidae are derived from the same ancestral group as the Accipitridae, and that the accipitrid characteristics observed are due to the more primitive nature of this genus, although Mourer-Chauviré & Cheneval, 1983 had previously considered the genus to be too advanced to be the ancestor of recent Sagittariidae. The morphology of the Sagittariidae, with their long legs and eagle-like bodies and head, is fairly bizarre, even when accounting for their rather specialised hunting strategy, but it is not unique: similar morphology was exhibited by an American accipitrid, Apatosagittarius terrenus, the fossils of which were uncovered from the Miocene of Nebraska (Feduccia & Voorhies 1989), suggesting that this species had adopted similar hunting methods.  Unlike the accipitrids and sagittariids, the origins and diversity of the Falconidae appear to be American, with a caracara-like falcon, Badiostes patagonicus, known from the lower Miocene of Patagonia, while several species belonging to the modern Polyborinae and Falconinae have also been described from North America. The earliest European record is of a species of Falco from the Late Miocene. A close relative of the Falconidae, Pediohierax ramenta, is known from the Middle Miocene of Nebraska. The falcon-sized Horusornis vianeyliaudae, described from the Upper Eocene of the Quercy Fissure Formations of France (Mourer-Chauviré 1991) brings up an interest- ing conundrum. Ecologically, it occupied the same niche as Polyboroides of Africa, Geranospiza of the Neotropics and Pengana from Tertiary deposits in Australia in that it could flex its leg backwards, thereby enabling it to ‘grope around in tree cavities or hollows for nestlings or small mammals’ (Feduccia 1996). From the evolutionary standpoint, its relationship to the other falconiform groups is ambiguous, as Horusornis shares features of its wing bones with the Accipitridae, and derived characters of its foot bones with the Accipitridae and Falconidae. These features alone would suggest that Horusornis is derived from the same ancestor as the Accipitridae and Falconidae. However, the distal end of its tibiotarsus bears similarities to that of owls, which clouds the issue. Such ‘mosaic species’ are, however, not that unusual. The unique Hoatzin is one such species and consequently has had a chequered history with respect to studies on its phylogenetic relationships. It has been described as a primitive galliform, a member of the gruiform radiation, placed in its own order and has even been connected with the cuckoos. Recently, Hughes (2000) has suggested that the turacos (Musophagidae) should be placed alongside the Hoatzin in the order Opisthocomiformes: they are the only extant families in which the young of some species have a well-developed wing-claw. Hughes also remarked that the Lower Eocene Foro panarium, a ground-based bird found in Wyoming, seems to confirm this placement, as this species ‘has a skull and mandible most like the Hoatzin, but shows some similarities to turacos in postcranial skeletal elements’. However, she suspected Foro to be a separate lineage that may stem from the same ancestor as the cuckoo and hoatzin + turaco lineages, which would agree with Olson’s placement of the bird in its own family, Foratidae (Olson 1992b). True hoatzins are South American and are found much later in the fossil record, with the Miocene Hoazinoides magdalenae not only laying claim to being the first hoatzin fossil to be described (Miller 1953), but also representing the most westerly occurrence in the family. Unfortunately, only part of the skull was found and, although very hoatzin-like in form, little else can be told of the bird, other than that it was less specialised than the modern species.  The Gruiformes are another group in which the relationships are not at all clear. Unlike the Opisthocomiformes, their fossil record is extensive, with several families present only as fossils. The Gruiformes have their fair share of giants, with the peculiar adzebills of New Zealand, and the aptly-named terrorbirds of the Americas. Possibly unsurprisingly, there are also a large number of fossil and subfossil rails, many of which disappeared at the hands of humankind. Among the earliest of the Gruiformes are the Messel-rails (Messelornithidae), which Mayr (2004g, 2005a) regards as related to the finfoots (Heliornithidae) and rails (Rallidae). Earlier authors (e.g. Hesse 1988, Livezey 1998) included them alongside the Sunbittern (Eurypyga helias) of the Neotropics. The family includes four species, the earliest of which, Messelornis russelli, occurred in the Upper Paleocene of France (Mourier-Chauviré 1995a). The family also occurred in North America, in the form of M. nearctica (Hesse 1992). The most famous member of the group, described in a monograph by A. Hesse in 1988, is M. cristata. This bird is known from a large number of fossils, originating from various Eocene/Oligocene sites in France and Germany, including Messel itself. Some of the articulated fossils of this species include a fleshy or horny crest preserved as an impression. None of the other species have shown the possession of a crest. The adzebills (Aptornithidae) of New Zealand are another enigmatic group, but of much more recent occurrence. In fact, these birds existed until as recently as 1000 years ago; like so many Polynesian species, they succumbed upon the arrival of humans and their animal associates (Gill & Martinson 1991). These flightless giants have been portrayed as bizarre takahe-like “rails” with large, downcurved bills and stout feet that enabled them to dig out petrel burrows, turn over litter, tear open rotting logs etc. in order to feed on large invertebrates, lizards and even tuatara and petrels (Holdaway 1989), or feed on berries, dig up tubers (Gill & Martinson 1991), or tear off pieces of tussock-grass (Feduccia 1996). In summary, they possessed a variable diet that changed between inland, montane and coastal sites. Two species are recognised, one for the North Island (Aptornis otidiformis), occurring in vegetation mosaic areas of open scrub, forest and grassland (Worthy 1999, Worthy & Swabey 2002) and a larger species (A. (o.) defossor) from more open, scrubland, sites on the South Island (Holdaway 1989, Worthy 1998, Worthy & Holdaway 1995). The latter species stood at 80 cm in height, roughly the same size as a male Great Bustard (Otis tarda) and that of Pachyornis, the second largest of the moas. It is the bill that has attracted the most interest, and has given the two species their common name, due to the very superficial similarity to the carpenter’s tool of the same name. The most recent study (Houde et al. 1997) indicates that the adzebills may be basal to the rail lineage, although other studies suggest relationships with the Kagu of New Caledonia. Aptornithidae is one family whose position is still unclear: one study has even come up with the suggestion that they may not be gruiform at all, but, instead, may be distant relatives of the Anseriformes and Galliformes (Weber & Hesse 1995).  The Cariamoidea would also appear to be an aberrant offshoot within the Gruiformes. However, this is deceiving, for, as well as being as old as the Messelornithidae, the fossil record shows a group that was more speciose than the two species we see now. The group consists of four families: Cariamidae (the only extant family); Idiornithidae; Bathornithidae; and Phorusrhacidae. Cariamidae has one valid species in the fossil record: Chunga incerta, from the Pliocene of Argentina. An Oligocene species has also been listed, but its relationship with the Cariamoidea has been called into question (Mourer-Chauviré 1981, Olson 1985). The Idiornithidae, predominantly from the Middle Eocene and Early Oligocene of Europe and Asia, and the Bathornithidae, from the Late Eocene and Early Oligocene of North America, are two associated groups that have been treated as subfamilies of the Cariamidae (as per Mourer-Chauviré 1981). The Bathornithidae have a chequered history, with the original specimens being allotted to the Cathartidae, Rallidae and Burhinidae, only being combined and elevated to family status in 1933 (by the original author, A. Wetmore). It was not until 1968 that the family finally found its proper placement alongside the Seriemas (see Olson 1985 for a more complete summary). The Bathornithidae includes two genera: Bathornis, which includes B. grallator from the Late Eocene of Wyoming, a bird originally described as a ‘terrestrial vulture’ by Wetmore in 1944; and Paracrax, including P. gigantea from the Late Oligocene of South Dakota. Within the related Idiornithidae, the genus Elaphrocnemus is of particular interest, as the wing elements were ‘almost perfectly intermediate between the seriemas and hoatzins’ (Olson 1985), leading to his conclusion that the Opisthocomidae and the Cariamoidea were related, a conclusion also reached by Mourer-Chauviré (1981). The Cariamoidea evolved on open grasslands as predatory birds, running down smaller prey from reptiles and other birds to mammals of various types. Some species may also have scavenged. The elegant, long-legged seriemas are one end of the extreme, while the phorusrhacid terrorbirds are the other, some reaching gigantic proportions, their most notable feature being a large head and high-ridged bill. Alvarenga & Höfling re-examined the classification of the terrorbirds in 2003 and concluded that the Phorusrhacidae occurred purely in the Americas. Most of the 17 species so far identified were South American, the majority inhabiting the pampas and dry grasslands of Argentina and Brazil; the exception is Titanis. Their origins are suggested to be in the Late Cretaceous or Early Paleocene, when South America was still an island continent. However, the earliest known species is the turkey-sized Paleopsilopterus itaboraiensis from the Middle Paleocene of the Argentinian Pampas, which suggests that other species are yet to be discovered. The family lasted until the Late Pliocene, when they had reached as far north as Florida and Texas, through the 2-m tall Titanis walleri (MacFadden et al. 2007). In the past, the family was also thought to have examples in Europe. However, these two fossils, Ameghinornis and Aenigmavis, have been removed from the Phorusrhacidae. The former was redescribed as a probable idiornithid and the latter as ‘Aves: incertae sedis’, with the suggestion that Aenigmavis was an arboreal, rather than a purely terrestrial, species (Mourer-Chauviré 1981). This observation in itself should have made people suspicious, given the terrestrial nature of the rest of the family. Gerald Mayr (2005b) has since reidentified both these specimens as belonging to Strigogyps, a member of the owl radiation, and therefore not even a cariamoid, although this has been countered by Peters (2007). Problems with identification aside, the phorusrhacids have proven to be an interesting family of predatory birds. As with others in the group, these ground-birds were generally built for speed, with members of the subfamily Phorusrhacinae being the top predators of the grassland and sparse scrubland areas they occupied. Those of the Brontornithinae were at least partial scavengers. Studies examining the locomotory abilities of these animals have estimated that the giants of the family, the Phorusrhacinae, could attain speeds of up to 50 km/hour (31 mph), while smaller species, such as the Argentinian Mesembriornis milneedwardsi from the Upper Pliocene, may have been even faster, with an ability to run at up to 97 km/hour (60 mph), although probably only for short bursts (Blanco & Jones 2005): these speeds may seem incredible, but today’s fastest cursorial bird, the ostrich, can attain similar speeds (see Folch 1992). One particular surprise has come with the discovery of the North American giant, Titanis walleri, in that it shows major modifications to the wing. Instead of the reduced wing one would expect in a flightless bird, its forelimbs were surprisingly well developed and equipped with manipulative thumb-claws, providing the adult bird with a two-clawed ‘hand’, possibly utilised in manipulating prey. Whether other members of the family possessed this adaptation or not is still unknown (Chandler 1994). Thus, T. walleri is the only species of bird known to date in which the adult possessed such a two-clawed ‘hand’, even though a few other, modern groups, namely the Hoatzin (Opisthocomus hoazin) and certain turacos, have wing-claws when young. A further three groups of ‘runners’ were related to the cranes and rails (Gruides). These were families whose members occupied the grasslands of the Northern Hemisphere from the Eocene to the Early Pliocene: the Geranoididae, a North American radiation in the Eocene and a family that we know little about; the Eogruidae, their contemporaneous Palearctic equivalents, of which one species at least, the Late Oligocene Sonogrus gregalis from Outer Mongolia, showed sexual dimorphism in size; and the Ergilornithidae, highly specialised forms from the Oligocene to Pliocene of the Central Asia steppes. These crane-sized birds were highly cursorial, so much so that they were probably flightless or nearly so (Olson 1985). More amazingly, ergilornithids show an evolutionary trend in the reduction of toes, to the extent that species such as the Late Miocene/Early Pliocene Amphipelargus maraghanus had just two toes on each foot, a feature shared only with the modern ostrich (Harrison 1981). This remarkable example of co-evolution is an adaptation for running at speed on open terrain, and such a reduction in toes is also shown by plains mammals, most notably the horses (Equidae). In life, these birds must have been amazing to watch, while the habitat that they occupied would have been all the more extraordinary, given that they shared their steppe home with an ostrich.  The origins of the Gruides as a whole are in the Eocene, if not the Paleocene. The cranes are represented in the fossil record by at least 20 species. However, many of these have been described only from fragmentary tibiotarsi, which makes their reconstruction difficult, to say the least. Indeed, the fossil record of the Gruidae is so poor for the early part of their history that doubt has been cast even over some of those bones that are accepted currently as gruid (Mayr 2005a). The oldest recognised cranes include Geranopsis hastingsiae from the Upper Eocene of southern England. Even this may turn out to belong to a different group: resemblances have been shown between the coracoid of this species and those of Anserpica kiliani, a possible magpiegoose from the Upper Oligocene of France. Modern cranes belong to two subfamilies, the Balearicinae and the Gruinae, and it is thought that these split at a very early stage in the evolution of the family. Today, the Balearicinae are restricted to two African species of Crowned Crane. However, the subfamily was at one time more widespread and more diverse. Balearica exigua, smaller than the modern species, was present in North America during the Miocene, alongside the member of another genus, Probalearica crataegensis. A member of this second genus, Probalearica mongolica, occurred during the Upper Miocene in Europe and the Pliocene of Asia. Probalearica, or at least P. crataegensis, was described by Olson (1985) as being similar to Balearica and he suggested that it may be part of that genus. Of these ‘crowned’ cranes, B. exigua is perhaps the best preserved, with several almost complete fossils uncovered from the ancient damp grassland beds of Nebraska, showing that the habitat for this species was similar to that of the African species (Feduccia & Voorhies 1992). Although the habitat may have been similar, no inferences can be made with regards to the plumage of these ancient balearicine cranes. Whether they bore the tufted crests of today’s species or not is unknown; it is more likely that this is a feature of the modern Balearica species through the isolation of their ancestor in the Afrotropics. Like these balearicines, the fossil Gruinae show many skeletal similarities with their extant relatives, although the Upper Miocene Camusia quintanai from the Mediterranean island of Menorca shares a number of plesiomorphic traits with the Balearicines, despite being placed in the Gruinae (Seguí 2002). Endemic island species have been found elsewhere, with the Pleistocene species Grus melitensis from the islands of Sicily and Malta and the Caribbean Grus cubensis, another Pleistocene species, but one distinguished by its flightlessness. The origins of the Gruidae remained a mystery until 2005, when Gerald Mayr published a paper describing what he called a ‘crane precursor’, Parvigrus pohli (Parvigruidae), from the Lower Oligocene of Lubéron in southern France, a time when southern Europe was a place of tropical forests (Mayr 2004c). To date, this is ‘the most substantial Paleogene fossil record of the Grues’ (i.e. limpkins and cranes), ‘and [is] among its oldest representatives’. Unlike modern cranes or limpkins, this chicken-sized fossil had a short beak and rail-like limb proportions. Parvigrus possesses many features shared with either the cranes or the limpkin and was what can best be described as a mosaic of the two groups. The paper also hints at the evolution of the cranes and limpkins, suggesting that the former evolved into the giant long-legged form that we know today as an adaptation to life on the grasslands and damp marshlands that appeared and spread towards the end of the Oligocene and throughout the Miocene (Blondel & Mourer-Chauviré 1998), while the limpkins evolved in response to a specialised niche, as an adaptation to eating snails in marshland conditions. Within the Gruidae, Parvigrus shares most features with the Balearicinae, adding further credence to the perception that this latter subfamily is the more primitive of the two living groups. The last two groups to discuss are the Otidides and Ralloidea, and these are vastly different in the number of fossil and subfossil representatives they possess. The few representatives of the bustards are Palearctic in distribution. The greatest number is listed for the Pliocene, for example Otis paratetrax from Moldavia and the tentative bustard Gryzaja odessana, from the Ukraine, while at least one species is listed for the Miocene of Europe. The Ralloidea may incorporate the Messelornithidae (already described), as well as the Heliornithidae (finfoots) and the Rallidae (rails) (Houde et al. 1997, Livezey 1998, Mayr 2005a). Heliornithidae is known only from single fossils of members of extant genera from the Miocene of the USA and the Middle Miocene of Chad. The latter is a species closely related to the Asiatic Masked Sungrebe (Heliopais personata), indicating a more widespread distribution for that genus in the past (Louchart, Mourer-Chauviré et al. 2005). Rallidae has a more substantial fossil record, although many species are known only from a few bones, so little can be diagnosed as to their living form. The earliest species are known from the Eocene, and include members of the genus Quercyrallus from France (Cracraft 1973) and a variety of American species, such as Eocrex primus, a relative of the gallinules from Colorado, and Palaeorallus troxelli from Wyoming (Wetmore 1931). More substantial records are known for later, European, genera, with, for example, Rupelrallus and Belgirallus from the Lower Oligocene (see Mayr 2005a, 2006b). The major groups such as the coots, gallinules and marsh-rails had evolved by the start of the Pliocene (Feduccia 1968, Wetmore 1957). A substantial number of these rails were flightless or nearly flightless, especially subfossil species from island habitats. Like so many island species, these Holocene rails fell to the advance of humans and their exotic familiars. Most notable among these were the giant coots of New Zealand and the Chatham Islands (Fulica prisca and F. chathamensis, respectively), and the Takahes, including the subfossil North Island form, Porphyrio m. mantelli, which Trewick lists as a species separate from the surviving South Island form hochstetteri (Trewick 1996, 1997). Europe also possessed at least one flightless island endemic whose extinction may be related to the colonisation of its island by early humans during the Pleistocene or early Holocene, namely the Ibizan Rallus eivissensis, closely related to R. aquaticus (McMinn et al. 2005). Other flightless rails have been described from the Americas with members of the genus Rallus in Bermuda (Olson & Wingate 2000, 2001) and members of the genus Nesotrochis from the Caribbean (see Livezey 1998). Before we leave the gruiforms and their relatives, we should perhaps mention the enigmatic buttonquails (Turnicidae), with their mosaic of characteristics. There is not much to tell fossil-wise, as no turnicid fossils have been found apart from Neogene examples of modern species. In 2000, Mayr described a stem group, the Turnipacidae, from Lower Oligocene deposits at Céreste, France. Two species were listed, Turnipax dissipata and (‘tentatively’) Cerestenia pulchrapenna. Evidence indicates that these two families may be an early offshoot of the Charadriiformes, rather than derived from Gruiformes or other ancestors. One theory on the evolution of the neornithines pivots around the existence of a group of birds with a mosaic of characteristics—the so-called ‘transitional shorebirds’ (Feduccia 1996)—and their subsequent diversification at the start of the Tertiary (Feduccia 2003a). This has drawn major disagreement from many phylogeneticists, particularly because of its reliance on ‘a limited fossil record of neornithine birds from the Cretaceous and Paleocene, classified to within just a handful of the modern orders’ (Dyke & van Tuinen 2004). Recent work combining the fossil and molecular records suggests that this may be a real phenomenon rather than an artefact of the fossil record (Ericsson et al. 2006). One such group that Feduccia included in the transitional shorebirds is the form-family Graculavidae, a globally widespread assemblage that show certain similarities to the modern Burhinidae. However, many examples are fragmentary, while the family has not yet been examined cladistically, which would reveal the relationships within the group and would provide evidence as to whether the Graculavidae should be included in the Charadriiformes in the first place. Whilst other charadriiform fossils have been described for the Paleogene, Mayr (2005a) states that, in Europe at least, modern-type Charadriiformes are not known before the Oligocene. When they do appear on the scene, the majority of these waders closely resemble modern forms, to the point that several belong to modern genera. For the more unusual diversifications within the order, one has only to examine the Alcidae, a group of pelagic species adapted to underwater pursuit diving, often at the expense of their flying capabilities. The auks, possibly because of their more robust skeletons, have a much stronger fossil record than most other charadriiform families, and have definitive examples as far back as the Middle Miocene. Even at this time alcids were present in both the Pacific and Atlantic oceans. Earlier fossils attributed to the group, namely Hydrotherikornis and Petralca, are considered by some authors to be of doubtful affinities, with Hydrotherikornis closer to the Procellariiformes and Petralca being too poorly described to justify inclusion in the Alcidae (see Warheit 2002). Another fossil with alcid affinities, as yet undescribed, has been found in the Lower Eocene deposits of England and, if this does indeed turn out to be an alcid, would be the earliest member of that group so far found. The major radiation of the auks occurred in the Late Miocene, with the appearance of several genera, modern and fossil. Most of these were Pacific species, including several from the upwellings off the coast of California. Like other seabird groups, flightlessness has occurred in more than one lineage within the family. Many are familiar with the Great Auks (Pinguinus spp.) of the north Atlantic, in particular the last species of the group, which was forced into extinction within the last 200 years (see Fuller 2002). However, there is a much older group, from the Late Miocene to Late Pliocene of the Pacific, so distinct that they form a separate subfamily, Mancallinae; indeed, Miller (1946) considered them a separate family, Mancallidae. These auks included Mancalla californiensis, a species from California about the size of the Great Auk that showed a more advanced state of evolution towards the penguin-like anatomy of underwater ‘fliers’ than was evident in the Atlantic Pinguinus. Warheit lists three genera in the mancalline radiation: Mancalla, with five named species, ranging from the Late Miocene to the Late Pliocene; Praemancalla, a less advanced group consisting of two, possibly three, species; and the primitive Alcodes, a poorly known mancalline progressing towards flightlessness that was found in the coastal region of California during the Middle to Late Miocene. It is not known why the mancallines became extinct, although a number of possibilities have been suggested, one being extinction through unsuccessful competition for breeding space and food with the Otariidae, the eared-seals, which were becoming abundant during this period (Konyukhov 2002). Sandgrouse have also been connected with the Charadriiformes, although their fossils do not give us any clues as to whether they are more closely related to the shorebirds or to the pigeons. Indeed, few fossil species have been recorded. The earliest described comes from Upper Oligocene deposits in France. These belong to two genera, Archaeoganga consisting of three species, and the more modern Leptoganga, with the species L. sepultus, from the Upper Oligocene/Early Miocene (see Mourer- Chauviré 1993). Nor can we look to the Columbidae to find these connections, as their early fossil record is also scant. The earliest definitive member of the Columbidae is Gerandia calcaria, a small dove from the Miocene of France. Another dove, Microena goodwini, was described from the Lower Eocene of the London Clay deposits (Harrison & Walker 1977), but this specimen consists solely of a tibiotarsus, which makes its identification difficult (Dyke & van Tuinen 2004). Indeed, Mayr regards it as a member of the swift/nightjar radiation, rather than a dove (Mayr 2005a). The later record of the Columbidae is more substantial, with a number of subfossil species recovered from Holocene archaeological sites in the Pacific. The giant Nutunaornis gigoura, from the island of Viti Levu (Fiji), characterizes the extinct pigeons of the Pacific in that it was the largest of its group; similar trends can be seen with the extinct Caloenas canacorum and Gallicolumba longitarsus from New Caledonia and Ducula davidi from Uvea. Nutunaornis is known from a number of bones, including leg and wing bones and part of the beak, from which it can be estimated that the bird was 30% larger than its nearest relatives, members of the genus Goura: this species approached the Mauritius Dodo (Raphus cucullatus) in size (Worthy 2001).  The cuckoos are another enigmatic group. They have been associated most often with the turacos in modern classifications, and with the occasional inclusion of the Hoatzin, although this is not widely accepted (see Sibley & Ahlquist 1990). The most recent (morphological) analysis of the cuckoos, Hoatzin and turacos, however, placed the last two together with the cuckoos on a separate lineage, but related through a distant ancestor, Foro, which may have been predominantly terrestrial in form (Hughes 2000). The fossil record of the Cuculidae itself is too scanty to support or refute this. Although the oldest confirmed species are two small North American species, Eocuculus cherpinae, an arboreal cuckoo from the Late Eocene of Colorado (Chandler 1999), and Neococcyx maccorquodalei, from the Early Oligocene of Saskatchewan, there are other, Old World, contenders for the title, including Dynamopterus from the Eocene/Oligocene of France (Mayr 2006c). However, because this genus is represented by two humeri, one at least (that of D. boulei) probably being a member of the raptor genus Aquilavis (Olson 1985), more skeletal elements are required before their identity as cuckoos can be corroborated (Mayr 2005a). Another relative of the cuckoos, Procuculus minutus, has been identified from Early Eocene deposits from the south of England, but its placement within the order is also only putative. As we see with so many groups, the difficulties of identification, either through interpretation of the fossils, the need to examine fossils using modern techniques, or even just the fragmentary nature of the fossils themselves, all lead to a confusion of phylogenetic identity. Such is the example of the Parvicuculidae, initially placed alongside the cuckoos, then moved to the catch-all zygodactyl family Primobucconidae in the 1980s, before the realisation that, in fact, they may be relatives of the caprimulgine radiation. What is probably more interesting—a better word may be ‘odd’—is that the majority of known cuckoo fossils are American, when the group is thought to have originated in the Old World. Indeed, according to Hughes (2000), the most primitive extant species come from South-east Asia in the form of the ground-cuckoos of the genus Carpococcyx. This new classification is different from that of the Peters Check-list used as the starting point for HBW, and the DNA–DNA hybridization classification, but the former makes assumptions as to the evolution of the zygodactyl foot and the occurrence of terrestriality in the family, while the latter had major omissions in its pair-wise comparisons: the studies of Hughes, admittedly, lack molecular work, so may also have flaws in the decisions made, even if it does make a convincing case. If we do assume that Hughes is correct, it is probably to the Old World tropics, with a focus on the east, that one would have to look in order to find early fossils relating to the group. Of the more modern, post-Oligocene, fossils, it is perhaps worth mentioning two in particular. One is the ground-cuckoo Cursoricoccyx geraldinae from the Lower Miocene of Colorado (Martin & Mengel 1984), which is the most northerly species of the Neomorphine cuckoos found to date, although this hints more at the climate in North America at the time rather than being a truly remarkable record. The second, Geococcyx (californianus) concklingi, is of a Pleistocene relative of the Greater Roadrunner (Geococcyx californianus), first found in the Donna County caverns of New Mexico by Hildegarde Howard in 1931. This was a large species, which, if the measurements she provides are typical, was more than 10% larger than the modern form of G. californianus. However, its wing bones, or, rather, the dimensions of the ulna, are 24% larger, the proportions to the rest of the fossil elements inferring that this bird was perhaps less adapted to cursoriality than is the modern form, with which it was contemporaneous. The turacos are younger than the cuckoos and are known from as far back as the Lower Oligocene, with an unnamed species listed for Bavaria (Ballmann 1970). Oligocene fragments from Egypt have also been assigned to the family, but Olson (1985) regards this evidence as unreliable. Further species are known from the Lower Miocene of Kenya (Musophaga africanus) and contemporaneously from France (M. meini). Thus, despite the fossil record for the Musophagidae being so meagre, we find that they had an ancient range that included Europe as well as Africa. This may be a surprise to many, until one realises that the early Paleogene was a warmer period than now, with extensive tropical forests across Europe, and elsewhere. Promusophaga from the Late Eocene of the USA and London Clay was also considered to be a turaco, but is in fact an example of Lithornis vulturinus, and is therefore a member of the palaeognathous family Lithornithidae (Leonard et al. 2005, Olson 1985). The parrots have been associated with both the Columbiformes and Cuculiformes or with a number of other zygodactylous species, in particular the Coliiformes, but their actual origins remain unclear (Mayr 2002b). Until recently, very few fossils were available, except for modern parrots from archaeological sites, which did shed some light on the diversity of Pacific parrots prior to human colonisation: for instance, the lories Vini sinotoi and V. vidivici from the Marquesas Islands (Steadman & Zarriello 1987), and some modern genera and species from Miocene and Pliocene sites in the Americas and Australia (see Mayr & Göhlich 2004 for a brief review). However, in recent years, parrot-like fossils have been discovered from the Eocene of Europe and North America. Some of these are reanalyses of known, but enigmatic, species, such as ‘Primobuccoolsoni, formerly placed in the Piciformes alongside the Galbuli (Feduccia & Martin 1976, Houde & Olson 1988): Mayr points out that this species may be better listed as ‘Pulchrapollia olsoni, owing to its possible conspecificity with Pulchrapollia gracilis (Mayr 2002b). Some of the new fossils may belong to the Psittacidae, e.g. Xenopsitta fejfari from the Czech Rebulic (Mlíkovský 1998). Others represent Eocene species from more primitive families. Pseudasturidae, for instance, is a group with representatives in Europe and the USA (Mayr 2002b; see also Dyke & Cooper 2000), for example Pseudasturides (formerly Pseudastur) macrocephalus, a Middle Eocene ‘parrot’ from Germany and the type species for the family (Mayr 2002b, 2004a), while others, such as Bavaripsitta ballmanni from the Middle Miocene of southern Germany (Mayr & Göhlich 2004), are as yet of unknown affinities. The oldest representatives of the Psittaciformes so far known thus occur in Europe and North America, rather than the Southern Hemisphere as one may have expected given the distribution of the majority of extant species. The most remarkable feature of the complete specimens of pseudasturids (‘Primobucco olsoni, Serudaptus pohli and Pseudasturides macrocephalus) is the lack of a parrot-like beak. Instead, they possess something more general in form, which explains the previous confusions. Another family, the Quercypsittidae, are known from the Upper Eocene of France (Mourer- Chauviré 1992), but the beak of this family is not known. Given that they are regarded as another family of primitive parrot-relatives, it is not illogical to suppose that these birds also possessed a generalist unhooked beak, rather than one resembling that of modern-day species. Other species included in the Psittaciformes at one time or another have now been assigned to other groups. Included among these is Palaeopsittacus from the London Clay (Harrison 1982): currently regarded as incertae sedis, this species shows similarities with the frogmouths (Mayr 2003b). One problematic find is that of a jaw fragment from the Cretaceous of Montana (described by Stidham in 1998). However, its identity as a bird, let alone as a parrot, has been contested by Dyke & Mayr (1998). Owls were also thought to have fossil representatives in the Cretaceous. However, these have since been shown to belong to a group of theropod dinosaurs! Currently, the honour of ‘oldest owl’ belongs to the Paleocene Ogygoptynx wetmorei, described from a fossil tarsometatarsus found in SW Colorado (Vickers-Rich & Bohaska 1976) and Berruornis, a large owl originally described from the Upper Paleocene of France. Both belong to fossil families, the Ogygoptyngidae being monotypic, while Berruornis joined several other owls in the Sophiornithidae, a European family that existed from the Upper Paleocene to the Oligocene, and included the Eocene/Oligocene Sophiornis quercynus. The fossil record of owls is fairly extensive compared to other land groups, with at least six families present, consisting of about 90 species. Several species come from the Paleogene of Europe and North America, including members of the family Protostrigidae, whose type genus, Protostrix, was synonymised with Minerva (Mourer- Chauviré 1983). The protostrigids are thought to be the sister-group to the tytonid and strigid lineages (Mourer-Chauviré 1987). Minerva spp. were large owls, about the size of Bubo, for which members of the genus were originally mistaken. The relationship among the different owl groups, particularly between these ancient groups and the modern families, still needs to be understood properly. What can be said is that owls were far more diverse in the early Paleogene than they are now. Recently, the supposed Eocene raptor Messelastur from Europe has also been connected with the owls, being similar to Tynskya eoceana, an ‘owl’ from the Green River Formation of Wyoming: together, they constitute a sister-group to the Strigiformes, rather than being ancestral to them, and, controversially, unite the clade (Messelasturidae + Strigiformes) as the sister-group to the raptors (Falconidae + Accipitridae) (Mayr 2005h). Of the modern families, Tytonidae is the older, and was more diverse during the Paleogene, with two subfamilies in existence prior to the modern subfamilies. The Necrobyinae consisted of Nocturnavis incerta, a species from the Eocene and Oligocene deposits of the Quercy Fissures in France, and Necrobyas, also from Quercy (Mourer- Chauviré 1987; see also Bruce 1999). The tarsometatarsus of these owls is similar to that of the modern barn-owls, but is shorter and stockier. Several other ‘modern’ owls have been described from Paleogene Europe, mostly from Quercy, including Selenornis henrici, member of a second tytonid subfamily, Selenornithinae. Modern-type tytonids are not known until much later, with the majority of species coming from the Pleistocene and Holocene. Sexual dimorphism existed in at least some of these fossil Tyto, for instance the Pleistocene T. neddi from the Caribbean island of Barbuda (Steadman & Hilgartner 1999), and T. mourerchauvireae, a large owl from the Middle Pleistocene of Sicily (Pavia 2004). Several of these extinct tytonids, including the Pleistocene T. balearica of the western Mediterranean, were giants in the family, while some were merely large versions of the typical barn-owl type. Studies on sites where groups of tytonid species existed suggest that competition between the different species caused them to evolve into different size-class niches, a phenomenon that also involved the related strigid owls. For instance, during the Pleistocene, Cuba possessed three Tyto owls, one being T. alba, one substantially larger (T. noeli) and the last, T. riveroi, being a true giant, the size of an eagle-owl. Also present in this ‘series’ was a truly huge, probably flightless, strigid owl, Ornimegalonyx oteroi (Arredondo 1976, 1982). Gigantism therefore is not restricted to the Tytonidae, and nor are they all fossils, for they include such species as the Great Grey Owl, Strix nebulosa. Most famous of these giants is probably the La Brea Owl, originally described as a form of the Great Horned Owl (Bubo virginianus) but later recognised as being a species in its own right, not related to Bubo, but to the Barred and Spotted Owls, with which it was compared (Howard 1933) and thus given the name of Strix brea: to date, it does not seem to have been compared with the even larger S. nebulosa (Feduccia 1996). There is one special group of ‘normal-size’ owls that were specialist hunters of birds. These were members of the genus Gallistrix, which inhabited the islands of Hawaii until fairly recently. They possessed short, rounded wings and long legs similar to those of an Accipiter, occupying the niche these hawks would have occupied had they been on the islands. Each of the major islands had its own species of Gallistrix, with G. auceps on Kauai, G. erdmani on Maui, G. geleches on Molokai and G. orion on Oahu. As with so much of the avifauna of the Hawaiian Islands, human colonisation was their downfall, despite their abundance in the fossil deposits. Today, only two owl species exist on the islands, the Hawaiian endemic subspecies of the Short-eared Owl (Asio flammeus), thought to be a recent colonist (see, e.g., Burney et al. 2001 and Olson & James 1982), and the introduced Common Barn-owl (Tyto alba). The superficially owl-like Caprimulgiformes are also an ancient lineage, with the oldest current examples originating from the Eocene. The Paleogene saw the existence of species belonging to modern families as well as others that have no modern equivalents. Included among these are the enigmatic Fluvioviridavidae, consisting of two species from the Eocene of North America and Europe (Mayr 2005i, Mayr & Daniels 2001), although these have still not been placed definitively in a known order. An extinct group of definite nightjar-relatives, the Archaeotrogonidae, were found in European deposits. Assigned to the trogons from the initial description of Archaeotrogon in 1892, similarities with the Caprimulgiformes were noted in 1980, when Cécile Mourer-Chauviré described the family as incertae sedis. It was not until 1995 that she moved the family into the Caprimulgiformes. Archaeotrogonidae consists of five small species. Four belong to the genus Archaeotrogon of the Quercy fissures, with A. venustus having the greatest temporal range, from the Late Eocene to the Late Oligocene, the other three being restricted to the Oligocene (Mourer-Chauviré 1995b). The fifth species is Hassiavis laticauda, a putative Archaeotrogon from the Middle Eocene of Messel, and, at about 13 cm in length, the smallest as well as the oldest species in the group. This last species already showed adaptations to catching insects on the wing. In particular, it had a short, wide beak, most similar to that of the owlet-nightjars, although this may be the expression of plesiomorphic characteristics, rather than any sign of direct relationships (Mayr 2004c). Additionally, Hassiavis is preserved with its feathers, and those of the tail are distinctly barred, as in the majority of modern Caprimulgiformes. Paleogene Europe hosted species of at least two other extant families, both now restricted to the tropics. Three species of potoo (Nyctibiidae) have been recorded from Europe, with Euronyctibius kurochkini, identified from an incomplete humerus from the Upper Eocene/Oligocene deposits of France (Mourer-Chauviré 1989b), and members of the genus Paraprefica from the Middle Eocene of Germany (Mayr 1999b, 2005m). The latter have a mosaic of characters, with the derived skull and tarsometatarsus of modern species, but otherwise showing a more primitive structure, and thus indicating their status as stem-lineage representatives. The only other nyctibiid fossils are of a modern species, within the current, South American range of the family. A similar tale can be told for the frogmouths (Podargidae), with two age groups of fossils, one from Quaternary deposits of Australia, the other from the early Paleogene of Europe. The European frogmouths consist of two species, with an additional species regarded as ‘putative’. The first of these three to be discovered was originally considered to be a parrot. However, Palaeopsittacus georgei, as it is known, may be a stem-lineage representative of the Podargidae and is very similar to the French Quercypodargus olsoni, which is a primitive frogmouth. Paleogene fossils pertaining to the oilbirds and owlet-nightjars have also been putatively identified from isolated bones from Europe. However, further specimens are required to confirm their presence. The only confirmed oilbird fossils (Prefica nivea) come from the Lower Eocene of the Green River Formation of Wyoming. The earliest confirmed representative of the Aegothelidae is a small, partially associated specimen from north-west Queensland. However, it is during the Quaternary that we find some surprises pertaining to this family, including a species from the Pleistocene of New Zealand, Aegotheles novaezeelandiae, a more terrestrial species than those that remain today, although it is similar to, but longer legged than, the New Caledonian Owlet-nightjar (A. savesi), an endangered, species also known from a few Pleistocene fossils. As with many of New Zealand’s birds, A. novaezeelandiae became extinct with the advent of Polynesian colonisation, although its actual demise seems to have been due to the introduction of the Pacific rat (Rattus exulans), which initiated its decline 1000 years prior to widespread Polynesian settlement (Holdaway et al. 2002). The most widespread and speciose of the families today is the Caprimulgidae. However, this family has a very poor fossil record, and little can be said as to the evolution or even prehistorical extinctions within the group. The origins of the Caprimulgiformes have recently been discovered to be closely tied to the evolution of another aerial group, the Apodiformes. The major twist in the story came when Mayr (2002a) discovered that the owlet-nightjars may be more closely related to the Apodiformes than they are to the other caprimulgiform families. The discovery began with the analysis of an Eocene family of basal swifts, Aegialornithidae, which bore striking resemblances to recent Caprimulgidae and Aegothelidae, as well as resembling another contemporaneous ‘caprimulgiform’ family, Archaeotrogonidae. The close relationship of the whole clade, Cypselomorphae, has been further bolstered by the discovery of the aerial-hawking Protocypselomorphus manfredkelleri from the Messel; it appears to sit phylogenetically outside this clade as its sister-taxon (Mayr 2005j). The phylogenetic position of Archaeotrogonidae, however, remains uncertain. Both the Aegialornithidae and Archaeotrogonidae were amongst the most abundant small birds in the Upper Eocene of Europe. Temporally, they existed from the Middle Eocene to the Lower Oligocene, with fossils found in Germany and France (Mayr 2003a, Mourer-Chauviré 1980, Peters 1998); indeed, their temporal range may even have extended as far back as the Early Eocene, as recent finds in North America suggest (Olson 1999). The early swifts, such as Aegialornis gallicus of the Quercy fissures, were similar to modern tree-swifts in general form (Mayr 2003a). At least two other groups of swifts overlap with these birds. One, Eocypselus vincenti, is a tiny species known from an incomplete specimen from the Lower Eocene London Clay. It is one of the most primitive swifts, and is placed in its own family, the Eocypselidae, although it has also been placed alongside the Hemiprocnidae (Mourer-Chauviré 1988): like the aegialornithids, this latter resemblance may be more due to superficial, plesiomorphic similarities than to any true relationship. The Apodidae have records in the Early Eocene of Denmark and the Middle Eocene of Germany, in the form of Scaniocypselus wardi and S. szarskii respectively. Like modern species, these birds (or, S. szarskii, at least) had long wings and a short tail that was hardly forked. Given their structure, they probably hunted much in the way of modern species. However, the most remarkable discovery has to be that hummingbirds and at least some of their ancestors existed in Europe in the Paleogene, if not elsewhere in the Old World. Primitive hummingbirds were first described in the late 1980s from a tiny bird discovered in Lower Oligocene deposits in the Caucasus. This was named Jungornis tesselatus, and allotted a new family, Jungornithidae (Karhu 1988); another specimen of Jungornis was recently discovered dating to the Late Eocene of France, extending both the temporal and geographical ranges of the genus (Mourer-Chauviré & Sigé 2006). Since then, other species have been described, for instance Argornis caucasicus, from the Upper Eocene, a less specialised bird than Jungornis, and Parargornis messelensis from Germany. These were initially placed in the Jungornithidae, but doubt has been cast over this: they may represent separate lineages on the evolutionary line towards modern-type hummingbirds. Other than Parargornis, these stem-hummingbirds are known only from wing bones of single specimens. The Parargornis fossil is particularly striking, as it is a fully feathered specimen (Mayr 2003), and thus can be used to gain a lot more information about these tiny ‘micro-swifts’. Parargornis had a swiftlike bill, but a long tail like that seen in some modern hummingbirds. However, its wings were broad and rounded, unlike any other apodiform known. At a length of 13 cm (with the long tail feathers), this bird was tiny, like others in the group. What is apparent from the wing and tail is that the bird was not adapted to gliding, although it may have captured insects on the wing. The suggestion is that Parargornis fed in dense vegetation and may have adopted a flycatcher-like hunting strategy, with the tail enabling great manouvrability. None of these, though, are true hummingbirds. The modern, trochilid, nectar-feeding hummingbirds probably diversified in the Neotropics in the Late Miocene (as per Bleiweiss 1998a, 1998b), but their origins are thought to extend back to the Paleocene (Bleiweiss 1998c). However, in 2004 a totally unexpected find was described from the Early Oligocene of Germany (Mayr 2004d), when a ‘basal hummingbird’ was described and given the rather appropriate name of Eurotrochilus inexpectatus. Other than a few bones from the late Quaternary, this is the only other fossil that can be attributed to the family. About the size of a Rufous-breasted Hermit (Glaucis hirsutus), Eurotrochilus resembled modern hummingbirds in several ways, not only in size, but in bill type and wing form, together inferring nectarivory over other forms of foraging, although it must be stressed that there is no direct evidence to ‘confirm’ this observation (Mayr 2007). What this fossil does do is provide an Old World origin for the Trochilidae, contrary to what their current distribution would suggest, and it brings into focus the possibility that some Old World flowers of ancient lineage may have evolutionary traits related to a coevolution with hummingbirds, for example the ericaceous Agapetes from the Himalaya (see Mayr 2004d). Evidence from Europe and North America suggests that during the early Tertiary the passerines were largely absent. Consequently, the niches that they would subsequently take over were open to occupancy by other, non-passerine, groups for at least 30–40 million years before their arrival and domination. One group, that including Parargornis, had moved into a niche now occupied by flycatchers and some warblers: Parargornis may even have foliage-gleaned, much as the prediction for the original proto-hummingbird has been described, hovering below and around the leaves, picking off the insects it found. A myriad of zygodactyl forms and other so-called ‘nearpasserines’ occupied other ‘passerine’ niches. This diversity of small birds can be exemplified best through exploration of the mousebirds (Coliiformes), which today contains a single family (Coliidae) of only six species, all confined to the continent of Africa. However, the fossil record lists 15 species, in two main families, from the Eocene of Europe and North America. The second family is the Sandcoleidae, originally described from the Paleocene and Lower Eocene of North America (Houde & Olson 1992), when the several species, including the Lower Eocene Sandcoleus copiosus, were given a family, and order, of their own. One of their number, Eobucco, was a ubiquitous genus in North America, with E. brodkorbi from the Middle Eocene (Feduccia & Martin 1976) and at least two other species from Wyoming (Houde & Olson 1988). These birds bore many mousebird-like features, from the long, tapered tail without strengthened shafts to a large pygostyle (a unique feature of the Coliiformes). Like Sandcoleus, Eobucco had a slightly curved beak, larger than that of modern colies, giving both genera a superficial resemblance to some cuckoos. Sandcoleids, like the Coliidae (see de Juana 2001), were notable for their foot structure, as it is thought that they could rotate their toes either into a pamprodactyl position (like many swifts, probably as an aid to clinging onto vertical surfaces) or have their toes in a zygodactyl, ‘yoke-toed’, position. In general, the sandcoleids were generalists feeding on seeds, fruits and similar items. Of particular interest is Chascacocolius, described as a sandcoleid, but with some osteological differences to the type genus, and possessing two known species, one, C. oscitans, from the Lower Eocene of Wyoming (Houde & Olson 1992), the other, C. cacicirostris, a larger bird recently described from the Middle Eocene of Messel, Germany (Mayr 2005f). Only the latter is represented by a complete, articulated, skeleton: it reveals that at least C. cacicirostris possessed a pointed, conical bill most similar in form to modern icterids such as the caciques (Cacicus spp.). It may be that the species fed in the same way, plunging its bill into fruit and then opening it slightly (a method known as ‘gaping’), thereby breaking it open. Several other coly-like birds are known from Eocene and Oligocene Europe. Selmes absurdipes has been described from a number of specimens from Middle Eocene deposits of Messel, including two articulated skeletons as well as numerous isolated bones (Mayr & Mourer-Chauviré 2004a). This species has a generalised beak, as in the Sandcoleidae, and a more elongate foot than modern colies, although it is thought to possess the same grasping ability. Originally, it was described as a sandcoleid (as per Peters 1999), but Mayr and Mourer-Chauviré have since indicated that this is a primitive mousebird closely related to the Coliidae. Oligocolius brevitarsus, is also closely related to modern mousebirds. This species, from the Lower Oligocene of Germany (Mayr 2000b) has more strongly developed wings than modern mousebirds and a shorter tarsomet:atarsus: together, these suggest a bird adapted for a more arboreal lifestyle, with adaptations towards more sustained flight than modern species. Other coliids recorded include Masillacolius brevidactylus, a pamprodactyl species from the Messel, two species in the genus Primocolius (P. minor and the larger P. sigei) from the Quercy deposits of France, while the only fossil representatives of modern genera are found in Africa, coming from the Early Pliocene (Vickers-Rich & Haarhoff 1985). The trogons are another group where there are several examples in the fossil record, but their origins are not clarified by their early Tertiary representatives, even though these are close to the origination time of the group. Nor do they clarify their relationships with the Coraciiformes (within which they are sometimes placed), nor with the Coliiformes nor the Steatornithidae, with both of which they have recently been linked. Also, contrary to what one would expect given the modern distribution of the family, the oldest fossils known are European. The oldest articulated specimen was from the Middle Eocene of Messel, Germany (Mayr 2005e), and was tentatively placed in Primotrogon. This species, P. pumilio, resembled modern trogons in its overall morphology and in possessing a heterodactyl foot, in which the second toe is permanently turned backwards: this unique foot structure is found only in trogons. This species is very similar to Primotrogon wintersteini, from the Middle Oligocene of France. However, the two species differ in size, with P. pumilio being about 10 cm long, and therefore the smallest member of the Trogoniformes, and P. wintersteini being roughly 120% larger. They are separated not only by a considerable period of time (some 16 million years), but also by habitat, the former occurring in tropical forest, the latter living in arid woodland or scrub: further studies may reveal that they belong to separate genera. The skulls of these early trogons differ significantly from their modern counterparts: in particular, that of P. wintersteini is narrower, with a narrower beak and smaller eye-sockets, suggesting that the bird was less adapted to foraging on flying insects than modern species (Mayr 1999). P. wintersteini is generally similar to the Asian trogons, which are also less reliant on insects, and which take a substantial amount of fruit in their diet, unlike the more specialised African species (Collar 2001). A much larger species, Septentrogon madseni, has been found recently in the deposits of the Fur Formation of Denmark, dated as latest Paleocene to earliest Eocene (Kristoffersen 2002), but this bird consists only of part of the skull, enough of which exists to ascertain that the species was also different from modern forms. There are a number of later examples of trogons, but these are very similar to modern species. Trogons do not occur in the Americas until the Pleistocene, which is in agreement with the suggestion that the Neotropical species are the most derived of the group. The Coraciiformes sensu lato are regarded by many as a polyphyletic group. For instance, HBW divides the Coraciiformes into the Trogoniformes and Coraciiformes, while Sibley, Ahlquist and Monroe went further, splitting from the latter the Bucerotiformes, Upupiformes and Alcediniformes (Sibley & Ahquist 1990, Sibley & Monroe 1990). Mayr has suggested that the Cuckoo-roller (Leptosomidae) should also be removed from the Coraciiformes (Mayr 2002c). It is argued that the roller-like appearance of the Leptosomidae is more to do with the expression of a ‘generalist, primitive- percher’ morphology than being an indication of true close relationships within the order. The current distribution of the Leptosomidae, the island of Madagascar, is relictual, as its ancient distribution stretched at least as far north as Europe. Including these fossils, Leptosomidae consists of two genera, the extant Malagasy Leptosomus, and the European Plesiocathartes, the first species of which, P. europaeus, was originally described in 1908 as a cathartid vulture from an incomplete tarsometatarsus from Middle Eocene to Oligocene deposits of France (Mayr 2002c). It is only recently that other specimens of the genus have been discovered, P. kelleri from the Middle Eocene of Messel and P. gaillardi from the Early Miocene of Spain, plus isolated bones from elsewhere in Europe. The most basal group within the Coraciiformes, however, is the Primobucconidae. This is an enigmatic family whose scientific history is fairly complex. Its establishment in the mid-1970s saw the inclusion of many small arboreal species from the Eocene of North America, most of which (with the exclusion of Primobucco mcgrewi) have since been removed to families as diverse as Sandcoleidae and Pseudasturidae. The primobucconids were considered by the original authors to be related to the puffbirds (Feduccia & Martin 1976), but they were moved to the Coraciiformes in 1988, being placed close to the ground-rollers (Houde & Olson 1988). Currently, studies indicate that the primobucconids were a primitive family of tiny rollers (Mayr & Mourer-Chauviré 2004b). The most recent discoveries, two specimens of the Middle Eocene species P. frugilegus from Germany, show that at least this species was a seed-eater, all the more remarkable considering that modern rollers are all carnivorous. However, there are indications that the primitive trait within Mayr’s Coraciiformes and ‘Alcediniformes’ is of a generalist feeder, rather than a carnivore: primitive members of the kingfisher radiation, the motmots, incorporate vegetable material (in this case fruit) in their diet, while the other basal roller family, Eocoraciidae (the Messel Eocoracias brachyptera), was a seed-eater. One more family of rollers existed, again in Europe: these were the Geranopteridae, whose members extend from the Late Eocene to the Early Miocene. Like the Eocoraciidae, these birds bore a mosaic of roller and ground-roller characteristics, although, in this case, there are more similarities to the latter (Mayr & Mourer-Chauviré 2000). The fossil record of the alcedinidine and meropine groups in comparison is poor. The earliest fossil that can be attributed to them is Quasisyndactylus longibrachis, a tiny Middle Eocene tody-like bird from the Messel (Mayr 2004c). This fossil bears a mosaic of characters, some of which are shared with the kingfishers and their kin, others shared with the related Coracii. Its tody-like bill, however, is probably due to convergence with the todies rather than to any direct relationship, as a similar bill type has evolved independently among the motmots. Fossils of ‘true’ todies (Todidae) are present in the early Tertiary, but not from the current range of the family. Rather, they come from: North America, in the form of Palaeotodus emryi, from the Lower Oligocene of Wyoming; and from France, with P. escampsiensis from the Upper Eocene, and P. itardiensis from the Oligocene, the identity of this latter group requiring further study. These birds were larger than modern species, with P. emryi having proportionately longer wings, suggesting a greater power of flight than in Todus (Feduccia 1996). The Lower Oligocene Protornis glarniensis from Switzerland was another member of what is today regarded as a purely Neotropical group. Formerly considered a motmot, this species is now interpreted as belonging to a more primitive lineage, related to both the todies and the motmots (Cracraft 1980, Mayr 2005a). True motmots are known only from the Americas, with a Miocene record of a humerus that is indistinguishable from three of the six currently recognised genera, while fossils of modern species have been discovered from prehistoric sites in the Neotropics (Becker 1986b). Of the other groups, the true kingfishers (Alcedinidae) possess a number of ancient examples. For example, the oldest halcyonine, which might even be a member of the modern genus Todiramphus, was discovered recently in the Miocene Riversleigh deposits of north-west Queensland (Boles 1997). The final coraciiform families to consider are the hornbills and hoopoes. The fossil record of the former consists of two ground-hornbill species, Bucorvus brailloni, from Middle Miocene deposits in the Atlas Mountains of Morocco (Brunet 1971, Olson 1985), and the European Euroceros bulgaricus from the Late Miocene (Boev & Kovachev 2007), both of which appear to be primitive relatives to the modern species. Other fossils have been attributed to the hornbills, including Geiseloceros, formerly regarded as the earliest member of the group. However, it had a very short wing, unlike all modern species, and is thought now to be a member of the Idiornithidae (Mayr 2002e). One coraciiform was originally considered a member of the hornbills, namely Cryptornis antiquus, from the Upper Eocene of France, but it is regarded by Mayr and Mourer-Chauviré (2000) as a possible member of the roller-family Geranopteridae, and is possibly conspecific with Geranopterus alatus, also from Eocene Europe. However, Cryptornis is too poorly preserved to make a proper analysis and its correct placement awaits a better preserved specimen. This leaves us with little to go on as to the appearance of the early hornbills. However, the same cannot be said for the hoopoes, where there is at least one new family (Messelirrisoridae) from the Middle Eocene of Europe, the latter being regarded as the earliest known representatives of the hoopoe/wood-hoopoe lineage. Messelirrisorids were tiny hoopoe-like birds with curved bills, and may have had a similar diet to the modern forms (Mayr 2000c). The family includes three species, Messelirrisor parvus, M. halcyrostris, and the largest species, M. grandis. The Messelirrisoridae occupied a niche not utilised by either hoopoe families today, being adapted for foraging in branches of trees, unlike upupid hoopoes, but were not adapted to tree-climbing, therefore unlike the woodhoopoes. They were an abundant group during the Eocene, with several specimens found in the lacustrine deposits of the Messel, plus isolated bones from sites in Britain and France. However, there is a huge gap between these early fossils and the only upupid fossil species known (the flightless Pleistocene/Holocene Upupa antaios of St Helena: Olson 1975), while Miocene species have been described for the Phoeniculidae, including a tiny specimen described from the Lower Miocene of Bavaria, which may be another messelirrisorid. The earliest members of the Piciformes sensu stricto also appear to be European, with the earliest record being a fragmentary tarsometatarsus from the Lower Oligocene of Belgium, the next record being of a slender tarsometatarsus from a tiny moderntype piciform from the Upper Oligocene of Germany that was not as well adapted for climbing as modern species (Mayr 2001b). Unfortunately, it is too fragmentary to be given a name, or even allocation to family within the order. The Miopiconidae, tiny birds so far only found in Miocene deposits of Morocco, are the nearest relatives to this group. The oldest articulated fossil of a barbet is that of the recently described Rupelramphastoides knopfi, a tiny bird from the Lower Oligocene of Germany whose overall resemblance to toucans is likely to be another example of convergence on a similar niche (Mayr 2005g). Despite its appearance, this species is regarded as a primitive omnivore or generalized insectivore within the Piciformes (Mayr 2006d). Another genus, Capitonides, is a member of the Capitonidae, overall being similar to the modern genus Trachyphonus. However, unlike the latter, which are exclusively African, Capitonides europaeus and C. protractus both come from the Miocene of Europe. Woodpeckers are comparatively late in the fossil record, being known primarily from Pliocene and Pleistocene deposits. For instance, Pliopicus brodkorbi was a small, slender, Melanerpine woodpecker from the Lower Pliocene of Kansas (Feduccia & Wilson 1967), contemporaneous with another, larger, species, Palaeonerpes shorti from Nebraska. The latter was a species similar to Melanerpines in form, but may have been part of an earlier lineage of woodpeckers, now extinct (Cracraft & Morony 1969). Pleistocene species include among their number such forms as Dendrocopos submajor, the ancestral species to the modern D. major. There are a number of other small arboreal fossils whose relationships were unclear until recently. Most of these were discovered in Europe or North America in the past decade. The majority of this motly collection of birds originate from the Paleocene to Eocene epochs, although one family, the Zygodactylidae, occurred in the Lower Miocene. Some families, for example Primoscenidae, were among the most abundant of small birds during the early Tertiary. Of all these odd little birds, perhaps special mention should be made of Gracilitarsus mirabilis (Gracilitarsidae). This species possessed a body plan very different from any living bird (Mayr 2001a). Although it had a long metatarsus like all these little birds, its toes were unusually short. Its claws, however, were rather deep in cross-section, indicating a possible adaptation for hanging onto vertical surfaces. Gracilitarsus also possessed a ‘swallow-like wing length’, suggesting rapid flight capabilities, although no interpretation can be made of the full wing length since the remiges are poorly preserved; the tail feathers are lacking. G. mirabilis possessed a bill very similar in shape to that of Anthreptes sunbirds, yet it may also have been rhynchokinetic—that is, possessing a flexible tip that can be flexed upwards by muscles present in the bill, as seen today in snipe, pigeons and humming- birds; generally, this type of bill suggests a nectarivorous bird. As its discovery predates Eurotrochilus, this bird holds the prize for being the first described probable nectarivore from the early Tertiary. Like Eurotrochilus, its phylogenetic relationships show a connection with birds present only in South America. Now Gracilitarsus has been shown to be a primitive piciform-relative (Mayr 2005k). The Zygodactylidae and Primoscenidae are more enigmatic, and while Mayr states that they are ‘unquestionably closely related’ the most recent cladogram presented (Mayr 2004b) shows two possible alternatives, one in which the two families are split, with the Primoscenidae being basal to the Piciformes + Galbuliformes, the other in which the two cluster together and are positioned as the closest relatives of the Passeriformes. Further specimens are required to resolve this fossil quandary. The fossil record of the Passeriformes is particularly sparse when compared to the other major orders. In recent years, a number of passerine specimens have been identified, although, to date, it is only the more recent forms that have been complete enough and distinctive enough to be assignable to identifiable taxa. One major problem with any attempt to classify passerines on the basis solely of skeletal material is that the skeleton itself shows a remarkable uniformity within the order. It is only with very recent techniques that microstructural differences have proven to show up features diagnostic for even the major groupings. It has aided the partial identification of some of the older fossils. Having said that, several passerines are worthy of mention. Although many can not be given names, some of these relate to the age and southern derivation of the group, while others show what can happen when passerines become isolated on islands. One of the major controversies when dealing with the Passeriformes has been the timing of their appearance and spread. Due to the modern-day distribution of the suboscines, traditionally regarded as the oldest passerine suborder, it has been surmised that the Passeriformes have a southern origin, and that it is only with the advent of the oscines that the Passeriformes spread to the Northern Hemisphere. It was therefore timely that, in the mid-1990s, Australian passerine fossils of Early Eocene age were discovered from the Murgon range in south-east Queensland, thereby predating the oldest (Miocene) European fossils known at that time. However, like many passerine fossils, these were rather fragmentary in nature, and their identification to a recognisable taxon within the Passeriformes has proven impossible (Boles 1995). Recent studies have shown that the Acanthisittidae are the only extant members of a previously unrecognised group of ancient passerines (Acanthisittia) which are the sister- group to all other living passerines (Eupasseres) (Ericson, Christidis et al. 2002, Ericson, Irestedt & Johansson 2003). In 2004 and 2006, Mayr and Manegold reported on German and French passerines from the Early Oligocene, 10 million years before the time when the Passeriformes were thought to have entered Europe. These birds, originally described as suboscines, held further surprises, in particular that they fell outside the Eupasseres and could belong to a more primitive group. There were yet more surprises, as these primitive forms showed a finch-like bill for feeding on seeds or fruit, not insects, thus upsetting yet another hypothesis about passeriform evolution, which relied on an insectivorous ancestral form. Other European fossils have since been investigated more thoroughly and some, too, fall outside the living passerine groups, showing a wide radiation of so-called primitive passerines in Europe in the early Tertiary (Manegold et al. 2004, Mayr & Manegold 2006a). Some specimens belonging to this ancient European radiation have recently been given the name ‘Weislochia weissi’ (Mayr & Manegold 2006b). The origins of the Passeriformes are still thought to be on the southern continent of Gondwana, as per the original observations. Many ‘later’ species of passerines are attributable to modern families, for instance the Miocene Menura tyawanoides of Riversleigh, Australia, and even among these we find geographical surprises, such as records of drongo in Pleistocene Europe. However, it is the island species that show the most interesting traits among the Passeriformes. For instance, only three species of passerine are truly flightless. One, the Stephens Wren (Traversia lyalli), still existed during European colonisation of New Zealand. The second is another New Zealand Wren, popularly-called the Longbilled Wren (Dendroscansor decurvirostris), a species whose overall appearance might have been similar to a rather small, fat-bodied treecreeper. This species probably became extinct during the Maori colonisation, although the species is only known from three sites, all on the South Island (Millener & Worthy 1991, Worthy 1998). Many of the New Zealand wrens show a trend towards flightlessness, so to find two truly flightless species among their number is probably not such a surprise, unlike the third species. This was an emberizid, described as the ‘Long-legged Bunting’ (Emberiza alcoveri) and found on the island of Tenerife in the Canaries. It occupied the Laurel forests of the island at least until the early Holocene. Larger than any living Emberiza, E. alcoveri was 39% heavier than one of its closest relatives, Cabanis’s Bunting  (Emberiza cabanisi) of Africa, and possessed longer legs and shorter wings. Like that species, it was a seed-eater, although in this particular case it was capable of feeding on the harder seeds characteristic of the island. E. alcoveri probably nested on the ground, in common with many island species where ground predators do not exist. It was this trait that led to the final demise of the species, when its island home was invaded by the first human colonists, who not only destroyed its fragile habitat, but also brought with them such exotic predators as cats and rats (Rando et al. 1999). Among the other passerines, it is perhaps the Hawaiian honeycreepers that attract the most attention, firstly because of their extremes in bill shape, but also because their historical and prehistorical interactions with humankind make sober reading. Today, there are 23 species on the islands, but this is only what remains since the huge destructive forces of Polynesian and, later, European colonisation wiped out many of their relatives: one estimation suggests that as much as 64% of the native avifauna has disappeared from the Hawaiian islands to date. Some amazing adaptations existed among these ‘pre-human’ birds, and none more so than among the Drepanididae. For instance, the speciose genus Hemignathus has at least two additional species in the fossil record, one of which, H. vorpalis, was a giant (James & Olson 2003). H. vorpalis possessed an incredible, elongated upper mandible that the bird probably used to probe into deep crevices and sift through leaf litter to extract its insect prey. This bird became extinct on the island of Hawaii about 3000 years ago. A further 14 species of fossil and subfossil honeycreeper have so far been described, bringing the world’s total up to 48; species are still being described (e.g. James 2004, James & Olson 2005). Several other species became extinct during European colonisation (see Fuller 2002), but whether this was due to European pressure or just the tail end of the original colonisation remains to be discovered. In reality, it was probably a mixture of the two, with the European influences putting the last nail in the coffin for many species, but causing the extinction of others that were still common. It is only recently that it has been realised how destructive the initial human colonisation of the Hawaiian islands was, and how diverse the original ‘pristine’ Hawaiian fauna had been. 


Although this overview has tried to be comprehensive in terms of the diversity of birds it has covered, the reader must be made aware that to do the subject justice would require far more space than can be given in HBW. Indeed, a whole book could be devoted to the subject (and many have been!). However, two features stand out above all the others. One is that of interpreting biogeographical patterns. Without the knowledge from fossils, what would our interpretation of such families as the swifts and nightjars been like? For instance, would we ever have discovered from other sources that hummingbirds used to be present in Europe? Probably not, although the discovery of nectarivorous birds in the European early Tertiary does clear up the little mystery of bird-adapted nectar plants, even if their presence was only investigated after the nectarivores were discovered. Also interesting in terms of biogeography is the closeness of the avifaunal assemblages of Europe and North America, which has provided the possibility that land bridges existed at various times between these two regions. The second feature that must stand out, other than the huge variety of bird forms that have existed, is the difficulty in interpreting those forms both phylogenetically and, even harder, in terms of habit and ecology. Although these latter can be tackled with educated guesses, we will never know for certain in the majority of cases what these birds ate, how they behaved, or even what they looked like in glorious technicolor. New fossils are being discovered every day, or so it seems. While I have attempted to keep up with the most important of these, I know that I have not covered in any detail the relationships of the basal groups. As with any study, this article is already out of date before it has even gone into print, so please treat this as it should be—a celebration of the diversity that are the birds! Acknowledgements  I would like to thank Jane Cliff, Bob Deavin and Phil Harris for reading through the draft manuscript. I would also like to thank Gerald Mayr and C. David Polly for the help that they gave with respect to particular questions I encumbered them with, as well as Barbara Sinclair for translating some of the German citations, and Chen Zehang for her advice on the correct citation of the Chinese papers. I thank Frank Steinheimer for his thorough examination of the references, and the staff at Lynx, who have made writing this article a really pleasurable experience! 

Text and illustrations by Dr Kevin J. Caley 


Articulated fossil: a complete fossil in which all the bones lie in the correct position,as they did when the animal died.

Biogeography: the study of distributions and ranges in organisms.

Carinate: of those birds whose primitive state involves the possession of a carina, orkeel, on the sternal plate.

Cladistics: a philosophy of classification that arranges organisms only by their order of branching in an evolutionary tree, and not by purely morphological similarity. Thus, with cladistics, birds and theropods are placed together, whereas more traditional classifications would separate them.

Cursorial: of terrestrial animals adapted for speed rather than strength. Derived characters: those which have evolved beyond the original form. For instance, the bill of a flamingo is highly derived from the standard ‘generalist’ bill morphology.

Fossil: the remains of a creature in which the bone and tissue have been replaced through a process of mineralisation, becoming rock-like in form. Trace fossils exist when tracks or faeces undergo a similar process. Particular environmental conditions must be met for fossilisation to occur.

Furcula: the ‘wishbone’, formed by the fusion of the two clavicles, or collar bones. Found in theropods, including birds.

Genome: the hereditary information encoded within an organism’s DNA—its genetic make-up.

Gondwana: an ancient southern landmass encompassing what is now South America, India, Australia, New Zealand, Africa, Madagascar and Antarctica. Other continents were fused in a northern landmass, Laurasia.

Graviportal: megafaunal animals with massive bones and bodies, with columnar bones to carry weight; in consequence, their movements are restricted. Built for strength rather than speed in the first instance.

Humerus: upper arm/wing bone.

Integuments: hair, feathers, nails, skin, etc.—the natural outer covering of an organism.

Jehol Biota: consisting of the Yixian and Jiufotang Deposits, Early Cretaceous deposits from Liaoning, in the north of China. Famous for the discovery of Confuciusornis.

Based on two different radiations of pterosaur, one group akin to that seen in the Late Jurassic of Europe, one to that seen in the Early Cretaceous of Brazil.

Keel/Carina: extension of the sternal plate on which the muscles of the wing are attached. In flightless species, this is often much reduced or even vestigial.

K–T boundary: Cretaceous–Tertiary boundary (from the German).

Megafauna: term used to encompass the giants of the fauna, e.g. elephant, rhino,ostrich.

Mosaic species: species in which characters of a variety of different groups are exhibited. Often, these mosaic species are basal examples before such groups split to become independent lineages.

Mummy: preserved specimens with flesh and integuments intact in such a way as that the organic material has not been replaced by rock minerals.

Neoaves: a taxonomic unit for all neognathous birds and including their ancestors, but excluding the Galloanserae.

Neognath: (‘new jaw’) any bird from the modern (neornithine) radiation that possesses the neognath palate. Among the neornithines, this excludes the tinamous, lithorns and ratites.

Neotony: retaining juvenile features into the adult phase of a life cycle, for instance the chick-like appearance of the adult ostrich.

Osteology: bone form; literally the study of bones.

Palaeognath: (‘old jaw’) a lineage of birds including the tinamous, lithorns and ratites.

Palaeognaths and neognaths are distinguished by the anatomy of the palate.

Pamprodactyl: having all toes facing forwards, like the feet of a swift.

Plesiomorphic: primitive traits, found in the basal group and signifying an evolutionary lineage if found in more advanced species.

Polyphyletic: a taxon is said to be polyphyletic if the groups it contains possess different origins and come from different lineages.

Pre-human birds: birds that existed on islands prior to human colonisation. Most commonly used with reference to Pacific islands.

Procoracoid: the procoracoid bones lie posterior to the clavicles. They give support and stability to the pectoral girdle.

Proto-feathers: filamentous integumentary structures thought to be the precursors to true feathers. In reconstructions, often portrayed as hair-like or loosely structured.

Pygostylia: After Chiappe (1996), this includes all birds from Confuciusornis to Neornithes. The major characteristic of the group is the fusion of some or all of the tail bones into a solid structure, or pygostyle; this structure is not restricted to birds.

Relictual distribution: the distribution observed today, in which only part of an original range for the taxon in question is still occupied.

Relictual species: a species that is the last of its group, effectively a sort of living fossil.

Saurischian: ‘lizard-hipped’, including the sauropods (e.g. Diplodocus) and theropods. Birds originate from this group, rather than the other major dinosaur lineage, the ornithischian, or ‘bird-hipped’, group. The latter terms are essentially misnomers.

Subfossil: buried bones and tissue in which the mineralisation process has begun but is not complete.

Switch-claw: the enlarged claw of the inner toe of the Dromaeosauria and troodontids, usually held in a flexed position rather than flat on the ground like the other toes.

Tarsometatarsus: compound bone between the tibiotarsus and phalanges of a bird, formed by the fusion of the distal tarsal bones and the metatarsals.

Theropods: mainly bipedal carnivorous dinosaurs that include the ancestor of the birds (Aves). This grouping includes the Coelurosauria, among other groups. Characterised by thin-walled, hollow bones, hands and feet with three main fingers and toes.

Tibiotarsus: the large bone between the top leg bone (the femur) and the foot bones (tarsometatarsus) in a bird. Formed through the fusion of the long bone with the proximal bones of the ankle.

Urvögel: the earliest birds known, usually used in reference to Archaeopteryx and its closest relatives, if they are regarded as separate species.

Vegetation mosaic: a mixed zone in which patches of different habitat are intermingled with each other.

Zygodactyl: ‘yoke-toed’, in which two toes point forwards and two backwards (phalanges 1 and 4), as in a woodpecker.



Alvarenga, H.M.F. & Höfling, E. (2003). Systematic revision of the Phorusrhacidae (Aves: Ralliformes).

Papéis Avulsos Zool. 43(4): 55–91.

Amadon, D. (1947). An estimated weight of the largest known bird. Condor 49: 159–164.

Anderson, A. (1989). On evidence for the survival of moa in European Fiordland. New Zealand J. Ecol.

12(Suppl.): 39–44.

Andors, V.A. (1991). Paleobiology and relationships of the Giant Groundbird Diatryma (Aves:

Gastornithiformes). Pp. 563–571 in: Bell, B.D., Cossee, R.O., Flux, J.E.C., Heather, B.D., Hitchmough,

R.A., Robertson, C.J.R. & Williams, M.J. eds. (1991). Acta XX Congressus Internationalis Ornithologici

Christchurch, New Zealand 2–9 December 1990. Vol. 1. New Zealand Ornithological Congress Trust

Board, Wellington, New Zealand.

Andors, V.A. (1992). Reappraisal of the Eocene Groundbird Diatryma (Aves: Anserimorphae). Pp. 109–

126 in: Campbell (1992).

Arredondo, O. (1976). The great predatory birds of the Pleistocene of Cuba. Smithsonian Contrib.

Paleobiol. 27: 169–187.

Arredondo, O. (1982). Los Strigiformes fósiles del pleistoceno cubano. Bol. Soc. Venez. Cienc. Nat. 37:

33–55. In Spanish.

Baker, A.J., Huynen, L.J., Haddrath, O., Millar, C.D. & Lambert, D.M. (2005). Reconstructing the

tempo and mode of evolution in an extinct clade of birds with ancient DNA: the giant moas of New

Zealand. Proc. Natl. Acad. Sci. USA 102(23): 8257–8262.

Ballmann, P. (1970). Ein neuer Vertreter der Musophagidae (Aves) aus dem Chattium [=Chattian] von

Gaimersheim bei Ingolstadt (Bayern). Mitt. Bayer. Staatssamml. Paläont. Hist. Geol. 10: 271–276. In


Barsbold, R., Osmólska, H., Watabe, M., Currie, P.J. & Tsgotbaatar, K. (2000). Nowy owiraptorozaur

(Dinosauria, Theropoda) z Mongolii: pierwszy dinozaur z pygostylem. [New oviraptorosaur (Dinosauria,

Theropoda) from Mongolia: the first dinosaur with a pygostyle]. Acta Palaeontol. Polonica 45(2): 97–

106. In Polish.

Becker, J.J. (1986a). Reidentification of “Phalacrocorax” subvolans Brodkorb as the earliest record of

Anhingidae. Auk 103(4): 804–808.

Becker, J.J. (1986b). A fossil motmot (Aves: Momotidae) from the Late Miocene of Florida. Condor 88:


Benson, R.D. (1999). Presbyornis isoni and other Paleocene birds from North Dakota. Smithsonian Contrib.

Paleobiol. 89: 253–259.

Bertelli, S. & Chiappe, L.M. (2005). Earliest tinamous (Aves: Palaeognathae) from the Miocene of

Argentina and their phylogenetic position. Nat. Hist. Mus. Los Angeles County (Contrib. Sci.) 502: 1–20.

Blanco, R.E. & Jones, W.W. (2005). Terror birds on the run: a mechanical model to estimate its maximum

running speed. Proc. Royal Soc. London (Ser. B Biol. Sci.) 272: 1769–1773.

Bleiweiss, R. (1998a). Tempo and mode of hummingbird evolution. Biol. J. Linn. Soc. 65: 63–76.

Bleiweiss, R. (1998b). Origin of hummingbird faunas. Biol. J. Linn. Soc. 65: 77–97.

Bleiweiss, R. (1998c). Fossil gap analysis supports Early Tertiary origin of trophically diverse avian

orders. Geology 26: 323–326.

Blondel, J. & Mourer-Chauviré, C. (1998). Evolution and history of the Western Palearctic avifauna.

Tree 13(12): 488–492.

Boev, Z. (2002). Fossil record and disappearance of peafowl (Pavo Linnaeus) from the Balkan Peninsula

and Europe (Aves: Phasianidae). Hist. Nat. Bulgarica 14: 109–115.

Boev, Z. & Kovachev, D. (2007). Euroceros bulgaricus gen. nov., sp. nov. from Hadzhidimovo (SW

Bulgaria) (Late Miocene) – the first European record of hornbills (Aves: Coraciiformes). Géobios

(Villeurbanne) 40: 39–49.

Boles, W.E. (1992). Revision of Dromaius gidju Patterson and Rich 1987 from Riversleigh, northwestern

Queensland, Australia, with a reassessment of its generic position. Pp. 195–208 in: Campbell (1992).

Boles, W.E. (1995). The world’s oldest songbird. Nature (London) 374: 21–22.

Boles, W.E. (1997). A kingfisher (Halcyonidae) from the Miocene of Riversleigh, northwestern Queensland,

with comments on the evolution of kingfishers in Australo-Papua. Mem. Queensland Mus. 41:


Boles, W.E. (2005a). A new emu (Dromaiinae) from the Late Oligocene Etadunna formation. Emu 101(4):


Boles, W.E. (2005b). A review of the Australian fossil storks of the genus Ciconia (Aves: Ciconiidae),

with the description of a new species. Records Austr. Mus. 57: 165–178.

Bond, W.J., Lee, W.G. & Craine, J.M. (2004). Plant structural defences against browsing birds: a legacy

of New Zealand’s extinct moas. Oikos 104: 500–508.

Bourdon, E. (2005). Osteological evidence for sister group relationship between pseudo-toothed birds

(Aves: Odontopterygiformes) and waterfowls (Anseriformes). Naturwissenschaften 92: 586–591.

Bourdon, E., Bouya, B. & Iarochene, M. (2005). Earliest African Neornithine bird: a new species of

Prophaethontidae (Aves) from the Paleocene of Morocco. J. Vert. Paleo. 25(1): 157–170.

Brodkorb, P. (1956). Two new birds from the Miocene of Florida. Condor 58: 367–370.

Brodkorb, P. (1959). How many species of birds have existed? Bull. Florida State Mus. Biol. Sci. 5(3):


Bruce, M.D. (1999). Family Tytonidae (barn-owls). Pp. 34–75 in: del Hoyo, J., Elliott, A. & Sargatal, J.

eds. (1999). Handbook of the Birds of the World. Vol. 5. Barn-owls to hummingbirds. Lynx Edicions,


Brunet, J. (1971). Oiseaux miocènes de Beni Mellal (Maroc): un complément à leur étude. Notes Mém.

Serv. Géol. Maroc 31: 109–111. In French.

Buffetaut, E. (1997). New remains of the giant bird Gastornis from the Upper Paleocene of the eastern

Paris Basin and the relationships between Gastornis and Diatryma. Neues Jahrb. Geol. Paläontol.

(Monatshefte) 1997: 179–190.

Buffetaut, E. & Le Loeuff, J. (1998). A new giant ground bird from the Upper Cretaceous of southern

France. J. Geol. Soc. London 155(Special Suppl.): 1–4.

Bunce, M., Szulkin, M., Lerner, H.R.L., Barnes, I., Shapiro, B., Cooper, A. & Holdaway, R.N. (2005).

Ancient DNA provides new insights into the evolutionary history of New Zealand’s extinct Giant Eagle.

PLOS Biol. 3(1, e9): 044–046.

Bunce, M., Worthy, T.H., Ford, T., Hoppitt, W., Willerslev, E., Drummond, A. & Cooper, A. (2003).

Extreme reversed sexual size dimorphism in the extinct New Zealand moa Dinornis. Nature (London)

425: 172–175.

Burney, D. (1993). Late Holocene environmental changes in arid southwestern Madagascar. Quaternary

Res. (San Diego) 40: 98–106.

Burney, D.A., James, H.F., Burney, L.P., Olson, S.L., Kikuchi, W., Wagner, W.L., Burney, M.,

McCloskey, D., Kikuchi, D., Grady, F.V., Gage, R. & Nishek, R. (2001). Fossil evidence for a diverse

biota from Kaua’i and its transformation since human arrival. Ecol. Monogr. 71(4): 615–641.

Burney, D.A., Robinson, G.S. & Burney, L.P. (2003). Sporomiella and the Late Holocene extinctions.

Proc. Natl. Acad. Sci. USA 100(19): 10800–10805.

Campbell, K.E. ed. (1992). Papers in Avian Paleontology Honoring Pierce Brodkorb – Proceedings of

the Second International Symposium of the Society of Avian Paleontology and Evolution. Natural History

Museum of Los Angeles County (Contributions in Science) 36. Natural History Museum of Los Angeles

County, Los Angeles. xxxvii, 491 pp.

Campbell, K.E. & Tonni, E.P. (1983). Size and locomotion in teratorns (Aves: Teratornithidae). Auk

100: 390–403.

Chandler, R.M. (1994). The wing of Titanis walleri (Aves: Phorusrhacidae) from the Late Blancan of

Florida. Bull. Florida State Mus. (Biol. Sci.) 36(6): 175–180.

Chandler, R.M. (1999). Fossil birds of Florissant, Colorado: with a description of a new genus and species

of cuckoo. Pp. 49–53 in: Santucci, V.L. & McClelland, L. eds. (1999). National Park Service Paleontological

Research. Vol. 4. US National Park Service, Geological Resources Division, Lakewood, Colorado.

Chatterjee, S. (1994). Protoavis from the Triassic of Texas: the oldest bird. J. Orn. 135: 330.

Chatterjee, S. (1995). The Triassic bird Protoavis. Archaeopteryx 13: 15–31.

Chatterjee, S. (2002). The morphology and systematics of Polarornis, a Cretaceous loon (Aves: Gaviidae)

from Antarctica. Pp. 125–155 in: Zhou Zhonghe & Zhang Fucheng (2002a).

Chen Peiji, Dong Zhiming & Zhen Shuonan (1998). An exceptionally well-preserved theropod dinosaur

from the Yixian Formation of China. Nature (London) 391: 147–152.

Chiappe, L.M. (1996). Late Cretaceous birds of southern South America: anatomy and systematics of

Enantiornithes and Patagopteryx deferrariisi. Pp. 203–244 in: Arratia, G. ed. (1996). Contributions of

Southern South America to Vertebrate Paleontology. Münchner Geowissenschaftliche Abhandlungen Reihe

A 30. Friedrich Pfeil, Munich, Germany.

Chiappe, L.M. (2002). Basal bird phylogeny: problems and solutions. Pp 448–472 in: Chiappe & Witmer


Chiappe, L.M. & Dyke, G.J. (2002). The Mesozoic radiation of birds. Ann. Rev. Ecol. Syst. 33: 91–124.

Chiappe, L.M. & Dyke, G.J. (2006). The early evolutionary history of birds. J. Paleont. Soc. Korea

22(1): 133–151.

Chiappe, L.M. & Witmer, L.M. eds. (2002). Mesozoic Birds: Above the Heads of Dinosaurs. University

of California Press, Berkeley, California.

Chiappe, L.M., Ji Shuan, Ji Qiang & Norell, M.A. (1999). Anatomy and systematics of the

Confuciusornithidae (Theropoda: Aves) from the Late Mesozoic of northeastern China. Bull. Amer. Mus.

Nat. Hist. 242: 1–89.

Chiappe, L.M., Lamb, J.P. & Ericson, P.G.P. (2002). New enantiornithine from the marine Upper Cretaceous

of Alabama. J. Vert. Paleo. 22(1): 170–174.

Chinsamy, A., Martin, L.D. & Dodson, P. (1998). Bone microstructure of the diving Hesperornis and

the volant Ichthyornis from the Niobrara Chalk of western Kansas. Cretaceous Research 19: 225–235.

Christiansen, P. & Bonde, N. (2004). Body plumage in Archaeopteryx: a review, and new evidence from

the Berlin specimen. Compt. Rend. Palevol 3: 99–118.

Clark, J.M., Norell, M.A. & Makovicky, P.J. (2002). Cladistic approaches to bird origins. Pp. 31–64 in:

Chiappe & Witmer (2002).

Clarke, J.A. (2004). Morphology, phylogenetic taxonomy, and systematics of Ichthyornis and Apatornis

(Avialae: Ornithurae). Bull. Amer. Mus. Nat. Hist. 286: 1–179.

Clarke, J.A., Tambussi, C.P., Noriega, J.I., Erickson, G.M. & Ketcham, R.A. (2005). Definitive fossil

evidence for the extant avian radiation in the Cretaceous. Nature (London) 433: 305–308.

Collar, N.J. (2001). Family Trogonidae (trogons). Pp. 80–129 in: del Hoyo, J., Elliott, A. & Sargatal,

J. eds. (2001). Handbook of the Birds of the World. Vol. 6. Mousebirds to hornbills. Lynx Edicions,


Cooper, A., Lalueza-Fox, C., Anderson, S., Rambaut, A., Austin, J. & Ward, R. (2001). Complete

mitochondrial genome sequences of two extinct moas clarify ratite evolution. Nature (London) 409:


Corfe, I.J. & Butler, R.J. (2006). Comment on “a well preserved Archaeopteryx specimen with theropod

features”. Science (Washington, D.C.) 313: 1238.

Cracraft, J. (1973). Systematics and evolution of the Gruiformes (class Aves): 3. Phylogeny of the suborder

Grues. Bull. Amer. Mus. Nat. Hist. 151(1): 1–128.

Cracraft, J. (1980). Phylogenetic theory and methodology in avian paleontology: a critical appraisal. Pp.

9–16 in: Campbell, K.E. ed. (1980). Papers in Avian Paleontology Honoring Hildegarde Howard. Natural

History Museum of Los Angeles County (Contributions in Science) 330. Natural History Museum of

Los Angeles County, Los Angeles. xxxviii, 260 pp.

Cracraft, J. (2001). Avian evolution, Gondwana biogeography and the Cretaceous-Tertiary mass extinction

event. Proc. Royal Soc. London (Ser. B Biol. Sci.) 268: 459–469.

Cracraft, J. & Morony, J.J. (1969). A new Pliocene woodpecker, with comments on the fossil Picidae.

Amer. Mus. Novit. 2400: 1–8.

Cracraft, J. & Vickers-Rich, P. (1972). The systematics and evolution of the Cathartidae in the Old

World Tertiary. Condor 74: 272–283.

Davies, S.J.J.F. (2002). Ratites and Tinamous: Tinamidae, Rheidae, Dromaiidae, Casuariidae, Apterygidae,

Struthionidae. Bird families of the World. Oxford University Press, Oxford & New York.

Dyke, G.J. (2003). The phylogenetic position of Gallinuloides Eastman (Aves: Galliformes) from the

Tertiary of North America. Zootaxa 199: 1–10.

Dyke, G.J. & Cooper, J.H. (2000). A new psittaciform bird from the London Clay (Lower Eocene) of

England. Palaeontology 43: 271–285.

Dyke, G.J. & Mayr, G. (1998). Did parrots exist in the Cretaceous? Nature (London) 399: 317–318.

Dyke, G.J. & van Tuinen, M. (2004). The evolutionary radiation of modern birds (Neornithes): reconciling

molecules, morphology and the fossil record. Zool. J. Linn. Soc. 141(2): 153–177.

Dyke, G.J., Dortangs, R.W., Jagt, J.W., Mulder, E.W.A., Schulp, A.S. & Chiappe, L.M. (2002). Europe’s

last Mesozoic bird. Naturwissenschaften 89: 408–411.

Elzanowski, A. (2001). A new genus and species for the largest specimen of Archaeopteryx. Acta

Palaeontol. Polonica 46(4): 519–532.

Elzanowski, A. (2002). Archaeopterygidae (Upper Jurassic of Germany). Pp. 129–159 in: Chiappe &

Witmer (2002).

Emslie, S.D. (1988). An early condor-like vulture from North America. Auk 105: 529–535.

Ericson, P.G.P. (1997). Systematic relationships of the Palaeogene family Presbyornithidae (Aves:

Anseriformes). Zoo. J. Linn. Soc. 121: 429–483.

Ericson, P.G.P. (2000). Systematic revision, skeletal anatomy and paleoecology of the New World Early

Tertiary Presbyornithidae (Aves: Anseriformes). Paleobios (Berkeley) 20(2): 1–23.

Ericson, P.G.P., Anderson, C.L., Britton, T., Elzanowski, A., Johansson, U.S., Kallersjo, M., Ohlson,

J.I., Parsons, T.J., Zuccon, D. & Mayr, G. (2006). Diversification of Neoaves: integration of molecular

sequence data and fossils. Biol. Letters 2(4): 543–547

Ericson, P.G.P., Christidis, L., Cooper, A., Irestedt, M., Jackson, J., Johansson, U.S. & Norman,

J.A. (2002). A Gondwana origin of passerine birds supported by DNA sequences of the endemic New

Zealand wrens. Proc. Royal Soc. London (Ser. B Biol. Sci.) 269: 235–241.

Ericson, P.G.P., Irestedt, M. & Johansson, U.S. (2003). Evolution, biogeography, and patterns of diversification

in passerine birds. J. Avian Biol. 34: 3–15.

Feduccia, J.A. (1968). The Pliocene rails of North America. Auk 85(3): 441–453.

Feduccia, J.A. (1974). Another Old World vulture from the New World. Wilson Bull. 86(3): 251–255.

Feduccia, J.A. (1977). The Whalebill is a stork. Nature (London) 266: 719–720.

Feduccia, J.A. (1996). The Origin and Evolution of Birds. Yale University Press, New Haven & London.

Feduccia, J.A. (2003a). ‘Big bang’ for Tertiary birds? Tree 18(4):172–176.

Feduccia, J.A. (2003b). Response to Dyke, and van Tuinen et al.: ‘Big bang’ for Tertiary birds? Tree

18(9): 443–444.

Feduccia, J.A. & Martin, L.D. (1976). The Eocene zygodactyl birds of North America (Aves: Piciformes).

Smithsonian Contrib. Paleobiol. 27: 101–110.

Feduccia, J.A. & Voorhies, M.R. (1989). Miocene hawk converges on Secretarybird. Ibis 131: 349–354.

Feduccia, J.A. & Voorhies, M.R. (1992). Crowned cranes (Gruidae: Balearica) in the Miocene of Nebraska.

Pp. 239–248 in: Campbell (1992).

Feduccia, J.A. & Wilson, R.L. (1967). Avian fossils from the Lower Pliocene of Kansas. Occas. Pap.

Mus. Zool. Univ. Mich. 655: 1–6.

Folch, A. (1992). Family Struthionidae (ostrich). Pp. 76–82 in: del Hoyo, J., Elliott, A. & Sargatal, J. eds.

(1992). Handbook of the Birds of the World. Vol. 1. Ostrich to ducks. Lynx Edicions, Barcelona.

Forster, C.A., Sampson, S.D., Chiappe, L.M. & Krause, D.W. (1998). The theropod ancestry of birds:

new evidence from the Late Cretaceous of Madagascar. Science (Washington, D.C.) 279: 1915–1919.

Fuller, E. (2002). Extinct birds. Pp. 11–68 in: del Hoyo, J., Elliott, A. & Sargatal, J. eds. (2002). Handbook

of the Birds of the World. Vol. 7. Jacamars to woodpeckers. Lynx Edicions, Barcelona.

Gill, B. & Martinson, P. (1991). New Zealand’s Extinct Birds. Random Century New Zealand, Auckland,

New Zealand.

Goedert, J.L. (1988). A new Late Eocene species of Plopteridae (Aves: Pelecaniformes) from northwestern

Oregon. Proc. Calif. Acad. Sci. 45(6): 97–102.

Greenwood, R.M. & Atkinson, I.A.E. (1977). Evolution of divaricating plants in New Zealand in relation

to moa browsing. Proc. New Zealand Ecol. Soc. 24: 21–33.

Grellet-Tinner, G. & Dyke, G.J. (2005). The eggshell of the Eocene bird Lithornis. Acta Palaeontol.

Polonica 50(4): 831–835.

Harrison, C.J.O. (1981). A re-assignment of Amphipelagus majori from Ciconiidae (Ciconiiformes) to

Ergilornithidae (Gruiformes). Tertiary Res. 3(3): 111–112.

Harrison, C.J.O. (1982). The earliest parrot: a new species from the British Eocene. Ibis 124: 23–210.

Harrison, C.J.O. & Walker, C.A. (1976). A reappraisal of Prophaethon shrubsolei Andrews (Aves).

Bull. Brit. Mus. (Nat. Hist.) Geol. 27: 1–30.

Harrison, C.J.O. & Walker, C.A. (1977). Birds of the British Lower Eocene. Tertiary Res. Spec. Pap. 3: 1–52.

Harrison, C.J.O. & Walker, C.A. (1982). Fossil birds from the Upper Miocene of northern Pakistan.

Tertiary Res. 4: 53–69.

Hertel, F. (1995). Ecomorphological indicators of feeding behaviour in recent and fossil raptors. Auk

112: 890–903.

Hesse, A. (1988). The † Messelornithidae – a new family of Gruiformes (Aves: Gruiformes: Rhynocheti)

from the Tertiary of Europe and North America. J. Orn. 129(1): 83–95. In German with English summary.

Hesse, A. (1992). A new species of Messelornis (Aves: Gruiformes: Messelornithidae) from the Middle

Eocene Green River Formation. Pp. 171–178 in: Campbell (1992).

Holdaway, R.N. (1989). New Zealand’s pre-human avifauna and its vulnerability. New Zealand J. Ecol.

12(Suppl.): 11–25.

Holdaway, R.N., Jones, M.D. & Athfield, N.R.B. (2002). Late Holocene extinction of the New Zealand

Owlet-nightjar Aegotheles novaezealandiae. J. Roy. Soc. New Zealand 32(4): 653–667.

Holdaway, R.N., Worthy, T.H. & Tennyson, A.J.D. (2001). A working list of breeding bird species of

the New Zealand region at first human contact. New Zealand J. Zool. 28(2): 119–187.

Holtz, T.R. (2000). A new phylogeny of the carnivorous dinosaurs. Gaia (Lisboa) 15 (1998): 5–61.

Holtz, T.R. (2001). Arctometatarsalia revisited: the problem of homoplasy in reconstructing theropod

phylogeny. Pp. 99–122 in: Gauthier, J.A. & Gall, L.F. eds. (2001). New Perspectives on the Origin and

Early Evolution of Birds. Proceedings of the International Symposium in Honor of John H. Ostrom.

Peabody Museum of Natural History, New Haven, Connecticut.

Horn, P.L. (1989). Moa tracks: an unrecognised legacy from an extinct bird? New Zealand J. Ecol.

12(Suppl.): 45–50.

Hou Lianhai (1980). A new form of the Gastornithidae from the Lower Eocene of the Xichuan, Honan.

Vertebrata Palasiatica 18: 111–115. In Chinese with English summary.

Hou Lianhai (1989). A Middle Eocene bird from Sangequan, Xinjiang. Vertebrata Palasiatica 27: 65–

70. In Chinese with English summary.

Hou Lianhai (1996). [Mesozoic Birds of China]. Phoenix Valley Provincial Aviary of Taiwan, Taipei. In

Chinese. [Translation byWill Downs (2001), Bilby Research Center, Northern Arizona University. URL:

http://ravenel.si.edu/paleo/paleoglot/files/Hou_00.pdf (download 24 June 2007)].

Houde, P.W. (1988). Paleognathous Birds from the Tertiary of the Northern Hemisphere. Publications of

the Nuttall Ornithological Club 22. Nuttall Ornithological Club, Cambridge, Massachusetts. 148 pp.

Houde, P.W. & Haubold, H. (1987). Palaeotis weigelti restudied: a small Middle Eocene ostrich (Aves:

Struthioniformes). Palaeovertebrata 17: 27–42.

Houde, P.W. & Olson, S.L. (1988). Small arboreal nonpasserine birds from the Early Tertiary of western

North America. Pp. 2030–2036 in: Ouellet, H. ed. (1988). Acta XIX Congressus Internationalis

Ornithologici. Vol. 2. University of Ottawa Press, Ottawa.

Houde, P.W. & Olson, S.L. (1992). A radiation of coly-like birds from the Early Eocene of North America

(Aves: Sandcoleiformes new order). Pp. 137–160 in: Campbell (1992).

Houde, P.W., Cooper, A., Leslie, E., Strand, A.E. & Montaño, G.A. (1997). Phylogeny and evolution

of 12 S rDNA in Gruiformes (Aves). Pp. 121–158 in: Mindell, D.P. ed. (1997). Avian Molecular Evolution

and Systematics. Academic Press, San Diego & London.

Howard, H. (1931). A new species of road-runner from Quaternary cave deposits in New Mexico. Condor

33: 206–209.

Howard, H. (1933). A new species of owl from the Pleistocene of Rancho La Brea, California. Condor

35: 66–69.

Howard, H. (1947). Wing elements assigned to Chendytes. Condor 49: 76–77.

Howard, H. (1969). A new avian fossil from Kern County, California. Condor 71: 68–69.

Hughes, J.M. (2000). Monophyly and phylogeny of cuckoos (Aves, Cuculidae) inferred from osteological

characters. Zoo. J. Linn. Soc. 130: 263–307.

Huynen, L., Millar, C.D., Scofield, R.P. & Lambert, D.M. (2003). Nuclear DNA sequences detect

species limits in ancient moa. Nature (London) 425: 175–178.

James, H.F. (2004). The osteology and phylogeny of the Hawaiian finch radiation (Fringillidae:

Drepanidini), including extinct taxa. Zoo. J. Linn. Soc. 141: 207–255.

James, H.F. & Olson, S.L. (2003). A giant new species of Nukupuu (Fringillidae: Drepanidini:

Hemignathus) from the island of Hawaii. Auk 120(4): 970–981.

James, H.F. & Olson, S.L. (2005). The diversity and biogeography of koa-finches (Drepanidini:

Rhodacanthis), with descriptions of two new species. Zoo. J. Linn. Soc. 144: 527–541.

Ji Qiang, Chiappe, L.M. & Ji Shuan (1999). A new Late Mesozoic confuciusornithid bird from China.

J. Vert. Paleo. 19(1): 1–7.

de Juana, E. (2001). Family Coliidae (mousebirds). Pp. 60–77 in: del Hoyo, J., Elliott, A. & Sargatal,

J. eds. (2001). Handbook of the Birds of the World. Vol. 6. Mousebirds to hornbills. Lynx Edicions,


Karhu, A. (1988). Novoye semeystvo strizheobraznykh iz paleogena Yevropy. [A new family of apodiform

birds of Paleogene Europe]. Paleontologicheskii Zhurnal 3: 78–88. In Russian.

Konyukhov, N.B. (2002). [Possible ways of spreading and evolution of alcids]. Izv. Ross. Akad. Nauk

Ser. Biol. 5: 552–560. In Russian with English translation in: Biol. Bull. (Moscow) 29(5): 447–454.

Kristoffersen, A.V. (2002). An early Paleogene trogon (Aves: Trogoniformes) from the Fur Formation,

Denmark. J. Vert. Paleo. 22: 661–666.

Kurochkin, E.N., Dyke, G.J. & Karhu, A.A. (2002). A new presbyornithid bird (Aves, Anseriformes)

from the Late Cretaceous of southern Mongolia. Amer. Mus. Novit. 3386: 1–11.

Lemoine, V. (1881). Sur le Gastornis edwardsii et le Remiornis heberti de l’Éocène inférieur des environs

de Reims. Compt. Rend. Séances Acad. Sci. Paris 93: 1157–1159. In French.

Leonard, L., Dyke, G.J. & van Tuinen, M. (2005). A new specimen of the fossil palaeognath Lithornis

from the Lower Eocene of Denmark. Amer. Mus. Novit. 3491: 1–11.

Livezey, B.C. (1998). A phylogenetic analysis of the Gruiformes (Aves) based on morphological characters,

with an emphasis on the rails (Rallidae). Phil. Trans. Roy. Soc. London (Ser. B) 353: 2077–2151.

Louchart, A., Mourer-Chauviré, C., Vignaud, P., MacKaye, H.T. & Brunet, M. (2005). A finfoot

from the Late Miocene of Toros Menalla (Chad, Africa): paleobiogeographical and palaeoecological implications.

Palaeogeogr. Palaeoclimatol. Palaeoecol. 222: 1–9.

Louchart, A., Vignaud, P., Likius, A., Brunet, M. & White, T.D. (2005). A large extinct marabou stork

in African Pliocene hominid sites, and a review of the fossil species of Leptoptilos. Acta Palaeontol.

Polonica 50(3): 549–563.

MacFadden, B., Labbs-Hochstein, J., Hulbert, R.C. & Baskin, J.A. (2007). Revised age of the late

Neogene terrorbird (Titanis) in North America during the Great American Interchange. Geology 35(2):


Makovicky, P.J., Apesteguia, S. & Agnolin, F.L. (2005). The earliest dromaeosaurid theropod from

South America. Nature (London) 437: 1007–1011.

Manegold, A. (2006). Two additional synapomorphies of grebes Podicipedidae and flamingos Phoenicopteridae.

Acta Orn. (Warszawa) 41(1): 79–82.

Manegold, A., Mayr, G. & Mourer-Chauviré, C. (2004). Miocene songbirds and the composition of the

European passeriform avifauna. Auk 121(4): 1155–1160.

Marden, L. (1967). Madagascar: island at the end of the earth. Natl. Geogr. 132(4): 443–487.

Marsh, O.C. (1880). Odontornithes: a Monograph on the Extinct Toothed Birds of North America.

United States Geological Exploration of the 40th Parallel. Vol. 7. U.S. Government Printing Office,

Washington, D.C.

Martin, L.D. (1992). The status of the Late Paleocene birds Gastornis and Remiornis. Pp. 97–108 in:

Campbell (1992).

Martin, L.D. & Mengel, R.M. (1975). A new species of anhinga (Anhingidae) from the Upper Pliocene

of Nebraska. Auk 92: 137–140.

Martin, L.D. & Mengel, R.M. (1984). A new cuckoo and a chachalaca from the Early Miocene of

Colorado. Pp. 171–177 in: Mengel, R.M. ed. (1984). Papers in Vertebrate Paleontology Honoring Robert

Warren Wilson. Carnegie Museum of Natural History Special Publication 9. Carnegie Museum of Natural

History, Pittsburgh, Pennsylvania. v, 186 pp.

Mayr, G. (1999a). A new trogon from the Middle Oligocene of Céreste, France. Auk 116(2): 427–434.

Mayr, G. (1999b). Caprimulgiform birds from the Middle Eocene of Messel (Hessen, Germany). J. Vert.

Paleo. 19: 521–532.

Mayr, G. (2000a). Charadriiform birds from the Early Oligocene of Céreste (France) and the Middle

Eocene of Messel (Hessen, Germany). Géobios (Villeurbanne) 33(5): 625–636.

Mayr, G. (2000b). A new mousebird (Coliiformes: Coliidae) from the Oligocene of Germany. J. Orn.

141: 85–92.

Mayr, G. (2000c). Tiny hoopoe-like birds from the Middle Eocene of Messel (Germany). Auk 117:


Mayr, G. (2001a). A new specimen of the tiny Middle Eocene bird Gracilitarsus mirabilis (new family:

Gracilitarsidae). Condor 103: 78–84.

Mayr, G. (2001b). The earliest fossil record of a modern-type piciform bird from the Late Oligocene of

Germany. J. Orn. 142: 2–6.

Mayr, G. (2002a). Osteological evidence for paraphyly of the avian order Caprimulgiformes (nightjars

and allies). J. Orn. 143: 82–97.

Mayr, G. (2002b). On the osteology and phylogenetic affinities of the Pseudasturidae – Lower Eocene

stem-group representatives of parrots (Aves, Psittaciformes). Zoo. J. Linn. Soc. 136: 715–729.

Mayr, G. (2002c). A new species of Plesiocathartes (Aves: ?Leptosomidae) from the Middle Eocene of

Messel, Germany. Paleobios (Berkeley) 22(1): 10–21.

Mayr, G. (2002d). A skull of a new pelecaniform bird from the Middle Eocene of Messel, Germany. Acta

Palaeontol. Polonica 47(3): 507–512.

Mayr, G. (2002e). Avian remains from the Middle Eocene of the Geiseltal (Sachsen-Anhalt, Germany).

Pp. 77–96 in: Zhou Zhonghe & Zhang Fucheng (2002a).

Mayr, G. (2003a). A new Eocene swift-like bird with a peculiar feathering. Ibis 145: 382–391.

Mayr, G. (2003b). A postcranial skeleton of Palaeopsittacus Harrison, 1982 (Aves incertae sedis) from

the Middle Eocene of Messel (Germany). Oryctos 4: 75–82.

Mayr, G. (2003c). The phylogenetic affinities of the Shoebill (Balaeniceps rex). J. Orn. 144: 157–175.

Mayr, G. (2004a). Pseudasturides n. gen., a replacement name for the stem group parrot Pseudastur

Mayr 1998 (preoccupied by Pseudastur Blyth 1850). Senckenbergiana Lethaea 83: 2.

Mayr, G. (2004b). The phylogenetic relationships of the Early Tertiary Primoscenidae and Sylphornithidae

and the sister taxon of crown group piciform birds. J. Orn. 145: 188–198.

Mayr, G. (2004c). New specimens of Hassiavis laticauda (Aves: Cypselomorphae) and Quasisyndactylus

longibrachis (Aves: Alcediniformes) from the Middle Eocene of Messel, Germany. Courier Forschungsinst.

Senckenberg 252: 23–38.

Mayr, G. (2004d). Old World fossil record of modern-type hummingbirds. Science (Washington, D.C.)

304: 861–864.

Mayr, G. (2004e). A partial skeleton of a new fossil loon (Aves, Gaviiformes) from the Early Oligocene

of Germany with preserved stomach content. J. Orn. 145: 281–286.

Mayr, G. (2004f). Morphological evidence for sister group relationship between flamingos (Aves:

Phoenicopteridae) and grebes (Podicipedidae). Zoo. J. Linn. Soc. 140: 157–169.

Mayr, G. (2004g). Phylogenetic relationships of the Early Tertiary Messel rails (Aves, Messelornithidae).

Senckenbergiana Lethaea 84(1/2): 317–322.

Mayr, G. (2005a). The Paleogene fossil record of birds in Europe. Biol. Rev. Cambridge Philos. Soc.

80(4): 515–542.

Mayr, G. (2005b). “Old World phorusrhacids” (Aves, Phorusrhacidae): a new look at Strigogyps

(“Aenigmavis”) sapea (Peters 1987). Paleobios (Berkeley) 25(1): 11–16.

Mayr, G. (2005c). A chicken-sized crane precursor from the Early Oligocene of France.

Naturwissenschaften 92: 389–393.

Mayr, G. (2005d). Tertiary plotopterids (Aves, Plotopteridae) and a novel hypothesis on the phylogenetic

relationships of penguins (Spheniscidae). J. Zool. Syst. Evol. Res. 43(1): 61–71.

Mayr, G. (2005e). New trogons from the Early Tertiary of Germany. Ibis 147: 512–518.

Mayr, G. (2005f). A new Eocene Chascacocolius-like mousebird (Aves: Coliiformes) with a remarkable

gaping adaptation. Org. Div. Evol. 5(3): 167–171.

Mayr, G. (2005g). A tiny barbet-like bird from the Lower Oligocene of Germany: the smallest species

and earliest substantial fossil record of the Pici (Woodpeckers and allies). Auk 122(4): 1055–1063.

Mayr, G. (2005h). The postcranial osteology and phylogenetic position of the Middle Eocene Messelastur

gratulator Peters, 1994 – a morphological link between owls (Strigiformes) and falconiform birds? J.

Vert. Paleo. 25(3): 635–645.

Mayr, G. (2005i). A Fluvioviridavis-like bird from the Middle Eocene of Messel, Germany. Can. J. Earth

Sci. 42(11): 2021–2037.

Mayr, G. (2005j). A new cypselomorph bird from the Middle Eocene of Germany and the early diversification

of avian aerial insectivores. Condor 107(2): 342–352.

Mayr, G. (2005k). Phylogenetic affinities and composition of the Early Eocene Gracilitarsidae (Aves,

?Piciformes). Neues Jahrb. Geol. Paläontol. (Monatshefte) 1: 1–16.

Mayr, G. (2005m). The Palaeogene Old World potoo Paraprefica Mayr, 1999 (Aves, Nyctibiidae): its

osteology and affinities to the New World Preficinae Olson, 1987. J. Syst. Palaeo. 3(4): 359–370.

Mayr, G. (2006a). New specimens of the Early Eocene stem galliform Paraortygoides (Gallinuloididae),

with comments on the evolution of a crop in the stem lineage of Galliformes. J. Orn. 147(1): 31–37.

Mayr, G. (2006b). A rail (Aves, Rallidae) from the Early Oligocene of Germany. Ardea 94(1): 23–31.

Mayr, G. (2006c). A specimen of Eocuculus Chandler, 1999 (Aves, ?Cuculidae) from the Early Oligocene

of France. Géobios (Villeurbanne) 39(6): 865–872.

Mayr, G. (2006d). First fossil skull of a Palaeogene representative of the Pici (woodpeckers and allies)

and its evolutionary implications. Ibis 148(4): 824–827.

Mayr, G. (2007). New specimens of the Early Oligocene Old World hummingbird Eurotrochilus

inexpectatus. J. Orn. 148(1): 105–111.

Mayr, G. & Daniels, M. (2001). A new short-legged landbird from the early Eocene of Wyoming and

contemporaneous European sites. Acta Paleontologica Polonica 46: 393–402.

Mayr, G. & Göhlich, U.B. (2004). A new parrot from the Miocene of Germany, with comments on the

variation of hyptarsus morphology in some Psittaciformes. Belg. J. Zool. 134(1): 47–54.

Mayr, G. & Manegold, A. (2004). The oldest European fossil songbird from the Early Oligocene of

Germany. Naturwissenschaften 91: 173–177.

Mayr, G. & Manegold, A. (2006a). A small suboscine-like passeriform bird from the Early Oligocene of

France. Condor 108(3): 717–720.

Mayr, G. & Manegold, A. (2006b). New specimens of the earliest European passeriform bird. Acta

Palaeontol. Polonica 51(2): 315–323.

Mayr, G. & Mourer-Chauviré, C. (2000). Rollers (Aves: Coraciiformes s.s.) from the Middle Eocene of

Messel (Germany) and the Upper Eocene of the Quercy (France). J. Vert. Paleo. 20(3): 533–546.

Mayr, G. & Mourer-Chauviré, C. (2004a). Unusual tarsometatarsus of a mousebird from the Pliocene

of France and the relationships of Selmes Peters, 1999. J. Vert. Paleo. 24(2): 366–372.

Mayr, G. & Mourer-Chauviré, C. (2004b). Osteology and systematic position of the Eocene

Primobucconidae (Aves, Coraciiformes sensu stricto), with first records from Europe. J. Syst. Palaeo.

2(1): 1–12.

Mayr, G. & Peters, D.S. (2006). Response to comment on “a well-preserved Archaeopteryx specimen

with theropod features”. Science (Washington, D.C.) 313: 1238c.

Mayr, G. & Weidig, I. (2004). The Early Eocene bird Gallinuloides wyomingensis – a stem group representative

of Galliformes. Acta Palaeontol. Polonica 49(2): 211–217.

Mayr, G., Peters, D.S. & Rietschel, S. (2002). Petrel-like birds with a peculiar foot morphology from the

Oligocene of Germany and Belgium (Aves: Procellariiformes). J. Vert. Paleo. 22(3): 667–676.

Mayr, G., Pohl, B. & Peters, D.S. (2005). A well-preserved Archaeopteryx specimen with theropod

features. Science (Washington, D.C.) 310: 1483–1486.

Mayr, G., Pohl, B., Hartman, S. & Peters, D.S. (2007). The tenth skeletal specimen of Archaeopteryx.

Zool. J. Linn. Soc. 149(1): 97–116.

McMinn, M., Palmer, M. & Alcover, J.A. (2005). A new species of rail (Aves: Rallidae) from the Upper

Pleistocene and Holocene of Eivissa (Pityusic Islands, western Mediterranean). Ibis 147: 706–716.

Millener, P.R. & Worthy, T.H. (1991). Contributions to New Zealand’s Late Quaternary avifauna. II:

Dendroscansor decurvirostris, a new genus and species of wren (Aves: Acanthisittidae). J. Roy. Soc. New

Zealand 21: 179–200.

Miller, A.H. (1953). A fossil Hoatzin from the Miocene of Colombia. Auk 70: 484–489.

Miller, L. (1925). Chendytes, a diving goose from the California Pleistocene. Condor 27: 145–147.

Miller, L. (1932). The Pleistocene storks of California. Condor 34: 212–126.

Miller, L. (1946). The Lucas Auk appears again. Condor 48: 32–36.

Mlíkovský, J. (1998). A new parrot (Aves: Psittacidae) from the Early Miocene of the Czech Republic.

Acta Soc. Zool. Bohemicae 62: 335–341.

Molnar, R.E. (1986). An enantiornithine bird from the Lower Cretaceous of Queensland, Australia. Nature

(London) 322: 736–738.

Mourer-Chauviré, C. (1980). The Archaeotrogonidae from the Eocene and Oligocene deposits of

“Phosphorites du Quercy”, France. Pp. 17–31 in: Campbell, K.E. ed. (1980). Papers in Avian Paleontology

Honoring Hildegarde Howard. Natural History Museum of Los Angeles County (Contributions in Science)

330. Natural History Museum of Los Angeles County, Los Angeles. xxxviii, 260 pp.

Mourer-Chauviré, C. (1981). Première indication de la présence de Phorusracidés [sic], famille d’oiseaux

géants d’Amérique du Sud, dans le Tertiaire européen: Ameghinornis nov. gen. (Aves, Ralliformes) des

Phosphorites du Quercy, France. Géobios (Villeurbanne) 14: 637–647. In French.

Mourer-Chauviré, C. (1983). Minerva antiqua (Aves, Strigiformes), an owl mistaken for an edentate

mammal. Amer. Mus. Novit. 2773: 1–11.

Mourer-Chauviré, C. (1987). Les Strigiformes (Aves) des Phosphorites du Quercy (France): systématique,

biostratigraphie et paléobiogéographie. Docum. Lab. Géol. Lyon 99: 89–135. In French.

Mourer-Chauviré, C. (1988). Les Aegialornithidae (Aves: Apodiformes) des Phosphorite du Quercy.

Comparaison avec la forme de Messel. Courier Forchungsinst. Senckenberg 107: 369–381. In French

with English and German summaries.

Mourer-Chauviré, C. (1989a). A peafowl from the Pliocene of Perpignan, France. Palaeontology 32:


Mourer-Chauviré, C. (1989b). Les Caprimulgiformes et les Coraciiformes de L’Éocène et de l’Oligocène

des Phosphorites du Quercy et description de deux genres nouveaux de Podargidae et Nyctibiidae. Pp.

2047–2055 in: Ouellet, H. ed. (1988). Acta XIX Congressus Internationalis Ornithologici. Vol. 2. University

of Ottawa Press, Ottawa. In French with English summary.

Mourer-Chauviré, C. (1991). Les Horusornithidae nov. fam., Acciptiriformes (Aves) à articulation

intertarsienne hyperflexible de l’Éocène du Quercy. Géobios (Mém. Spéc.) 13: 183–192. In French.

Mourer-Chauviré, C. (1992). Une nouvelle famille de Perroquets (Aves, Psittaciformes) dans l’Éocène

supérieur des phosphorites du Quercy, France. Géobios (Mém. Spéc.) 14: 169–177. In French.

Mourer-Chauviré, C. (1993). Les gangas (Aves, Columbiformes, Pteroclidae) du Paléogène et du Miocène

inférieur de France. Palaeovertebrata 22(2/3):73–98. In French with English summary.

Mourer-Chauviré, C. (1995a). The Messelornithidae (Aves: Gruiformes) from the Paleogene of France.

Courier Forschungsinst. Senckenberg 181: 95–105.

Mourer-Chauviré, C. (1995b). Les Garouillas et les sites contemporains (Oligocène, MP 25) des

Phosphorites du Quercy (Lot, Tarn-et-Garonne, France) et leurs faunes de Vertébrés. 3. Oiseaux.

Palaeontographica (Abt. A) 236(1/6): 33–38. In French with English and German summaries.

Mourer-Chauviré, C. (2002). Revision of the Cathartidae (Aves, Ciconiiformes) from the Middle Eocene

to the Upper Oligocene Phosphorites du Quercy, France. Pp. 97–111 in: Zhou Zhonghe & Zhang Fucheng


Mourer-Chauviré, C. (2003). Birds (Aves) from the Middle Miocene of Arrisdrift (Namibia). Preliminary

study with description of two new genera: Amanuensis (Accipitriformes, Sagittariidae) and Namibiavis

(Gruiformes, Idiornithidae). Geol. Survey Namibia Mem. 19: 103–113. In French and English.

Mourer-Chauviré, C. & Cheneval, J. (1983). Les Sagittariidae fossiles (Aves, Accipitriformes) de

l’Oligocène des Phosphorites du Quercy et du Miocène inférieur du Saint-Gérand-le-Puy. Géobios

(Villeurbanne) 16: 443–459. In French.

Mourer-Chauviré, C. & Poplin, F. (1985). Le mystère des tumulus de Nouvelle-Calédonie. La Recherche

(Paris) 16: 1094. In French.

Mourer-Chauviré, C. & Sigé, B. (2006). Une nouvelle espèce de Jungornis (Aves, Apodiformes) et de

nouvelles formes de Coraciiformes s.s. dans l’Éocène supérieur du Quercy. Strata (Toulouse) Ser. 1, no.

13: 151–159. In French.

Mourer-Chauviré, C., Senut, B., Pickford, M., Mein, P. & Dauphin, Y. (1996). Ostrich legs, eggs and

phylogeny. South Afr. J. Sci. 92: 492–495.

Murray, P.F. & Megirian, D. (1998). The skull of dromornithid birds: anatomical evidence for their

relationship to Anseriformes (Dromornithidae, Anseriformes). Records South Austr. Mus. 31: 51–97.

Murray, P.F. & Vickers-Rich, P. (2004). Magnificent Mihirungs: the Colossal Flightless Birds of the

Australian Dreamtime. Indiana University Press, Bloomington, Indiana.

Noriega, J.I. (1992). Un nuevo género de Anhingidae (Aves: Pelecaniformes) de la Formación Ituzaingó

(Mioceno superior) de Argentina. Notas Museo Plata (Palaeontol.) 109: 217–223.

Noriega, J.I. (2001). Body mass estimation and locomotion of the Miocene pelecaniform bird Macranhinga.

Acta Palaeontol. Polonica 46(2): 247–260.

Noriega, J.I. & Alvarenga, H.M.F. (2002). Phylogeny of the Tertiary giant anhingas (Pelecaniformes:

Anhingidae) from South America. Pp. 41–49 in: Zhou Zhonghe & Zhang Fucheng (2002a).

Noriega, J.I. & Piña, C.I. (2004). Nuevo material de Macranhinga paranensis Noriega, 1992 (Aves:

Pelecaniformes: Anhingidae) del Mioceno Superior de la Formación Ituzaingó, provincia de Entre Ríos,

Argentina. Ameghiniana 41(1): 115–118. In Spanish.

Noriega, J.I. & Tambussi, C. (1995). A Late Cretaceous Presbyornithidae (Aves: Anseriformes) from

Vega Island, Antarctic Peninsula: paleobiogeographic implications. Ameghiniana 32: 57–61.

Northcote, E.M. (1982). Size, form and habit of the extinct Maltese swan Cygnus falconeri. Ibis 124:


O’Farrell, B., Davenport, J. & Kelly, T. (2002). Was Archaeopteryx a wing-in-ground effect flier? Ibis

144: 686–688.

Olson, S.L. (1975). Paleornithology of St. Helena Island, south Atlantic Ocean. Smithsonian Contrib.

Paleobiol. 23: 1–49.

Olson, S.L. (1977). A Lower Eocene frigatebird from the Green River Formation of Wyoming

(Pelecaniformes: Fregatidae). Smithsonian Contrib. Paleobiol. 35: 1–33.

Olson, S.L. (1985). The fossil record of birds. Pp. 79–238 in: Farner, D.S., King, J.R. & Parkes, K.C. eds.

(1985). Avian Biology. Vol. 8. Academic Press, New York & London.

Olson, S.L. (1992a). Neogaeornis wetzeli Lambrecht, a Cretaceous loon from Chile (Aves: Gaviidae). J.

Vert. Paleo. 12: 122–124.

Olson, S.L. (1992b). A new family of primitive landbirds from the Lower Eocene Green River Formation

of Wyoming. Pp. 127–136 in: Campbell (1992).

Olson, S.L. (1994). A giant Presbyornis (Aves: Anseriformes) and other birds from the Paleocene Aquia

Formation of Maryland. Proc. Biol. Soc. Washington 107: 429–435.

Olson, S.L. (1999). Early Eocene birds from eastern North America: a faunule from the Nanjemoy Formation

of Virginia. Virginia Div. Mineral Resource Publ. 152: 123–132.

Olson, S.L. & Farrand, J. (1974). Rhegminornis restudied: a tiny Miocene turkey. Wilson Bull. 86(2):


Olson, S.L. & Feduccia, A. (1980). Presbyornis and the origin of the Anseriformes (Aves:

Charadriimorphae). Smithsonian Contrib. Zool. 323: 1–24.

Olson, S.L. & Hasegawa, Y. (1979). Fossil counterparts of giant penguins from the north Pacific. Science

(Washington, D.C.) 206: 688–689.

Olson, S.L. & Hearty, P.J. (2003). Probable extirpation of a breeding colony of Short-tailed Albatross

(Phoebastria albatrus) on Bermuda by Pleistocene sea-level rise. Proc. Natl. Acad. Sci. USA 100(22):


Olson, S.L. & James, H.F. (1982). Fossil birds from the Hawaiian Islands: evidence for wholesale extinction

by man before Western contact. Science (Washington, D.C.) 21: 633–635.

Olson, S.L. & James, H.F. (1991). Descriptions of thirty-two new species of birds from the Hawaiian

Islands. Part 1. Non-Passeriformes. Ornithological Monographs 45. American Ornithologists’ Union,

Washington, D.C. 88 pp.

Olson, S.L. & Matsuoka, H. (2005). New specimens of the Early Eocene frigatebird Limnofregata

(Pelecaniformes: Fregatidae), with the description of a new species. Zootaxa 1046: 1–15.

Olson, S.L. & Parris, D.C. (1987). The Cretaceous birds of New Jersey. Smithsonian Contr. Paleobiol.

63: 1–22.

Olson, S.L. & Rasmussen, P.C. (2001). Miocene & Pliocene birds from the Lee Creek Mine, North

Carolina. Smithsonian Contr. Paleobiol. 90: 233–265.

Olson, S.L. & Steadman, D.W. (1977). A new genus of flightless ibis (Threskiornithidae) and other

fossil birds from cave deposits in Jamaica. Proc. Biol. Soc. Washington 90(2): 23–27.

Olson, S.L. & Wetmore, A. (1976). Preliminary diagnoses of extraordinary new genera of birds and

Pleistocene deposits in the Hawaiian Islands. Proc. Biol. Soc. Washington 89: 247–258.

Olson, S.L. & Wingate, D.B. (2000). Two new species of flightless rails (Aves: Rallidae) from the Middle

Pleistocene ‘crane fauna’ of Bermuda. Proc. Biol. Soc. Washington 113: 356–368.

Olson, S.L. & Wingate, D.B. (2001). A new species of large flightless rail of the Rallus longirostris /

elegans complex (Aves: Rallidae) from the Late Pleistocene of Bermuda. Proc. Biol. Soc. Washington

114: 509–516.

Pain, S. (2000). The demon duck of doom. New Scientist 166: 36–39.

Parris, D.C. & Hope, S. (2002). New interpretations of birds from the Navesink and Hornerstown Formations,

New Jersey, USA (Aves: Neornithes). Pp. 113–124 in: Zhou Zhonghe & Zhang Fucheng (2002a).

Paton, T., Haddrath, O. & Baker, A.J. (2002). Complete mitochondrial DNA genome sequences show

that modern birds are not descended from transitional shorebirds. Proc. Royal Soc. London (Ser. B Biol.

Sci.) 269: 839–846.

Pavia, M. (2004). A new large barn owl (Aves, Strigiformes, Tytonidae) from the Middle Pleistocene of

Sicily, Italy, and its taphonomic significance. Géobios (Villeurbanne) 37: 631–641.

Peters, D.S. (1998). Erstnachweis eines Seglers aus dem Geiseltal (Aves: Apodiformes). Senckenbergiana

Lethaea 78: 211–212. In German with English summary.

Peters, D.S. (1999). Selmes absurdipes, new genus, new species, a sandcoleiform bird from the Oil Shale

of Messel (Germany, Middle Eocene). Pp. 217–222 in: Olson, S.L. ed. (1999). Avian Paleontology at the

Close of the 20th Century – Proceedings of the 4th International Meeting of the Society of Avian

Paleontology and Evolution, Washington DC, 4–7 June 1996. Smithsonian Contributions to Paleobiology

89. Smithsonian Institution, Washington, D.C.

Peters, D.S. (2007). The fossil family Ameghinornithidae (Mourer-Chauviré 1981): a short synopsis. J.

Orn. 148(1): 25–28.

Peters, D.S. & Hamedani, A. (2000). Frigidafons babaheydariensis n. sp., ein Sturmvogel aus dem

Oligozän des Irans (Aves: Procellariidae). Senckenbergiana Lethaea 80: 29–37. In German with English


Peters, D.S. & Ji Qiang (1999). Mußte Confuciusornis klettern? J. Orn. 140(1): 41–50. In German with

English summary.

Poplin, F. (1980). Sylviornis neocaledoniae n.g., n.sp. (Aves), ratite éteint de la Nouvelle-Calédonie.

Compt. Rend. Séances Acad. Sci. Paris Ser. D, no. 290: 691–694. In French.

Poplin, F. & Mourer-Chauviré, C. (1985). Sylviornis neocaledoniae (Aves, Galliformes, Megapodiidae),

oiseau géant éteint de l’Ile des Pins (Nouvelle-Calédonie). Géobios (Villeurbanne) 18: 73–97.

Rando, J.C., López, M. & Seguí, B. (1999). A new species of extinct flightless passerine (Emberizidae:

Emberiza) from the Canary Islands. Condor 101(1): 1–13.

Rinderknecht, A. & Noriega, J.I. (2002). Un nuevo género de Anhingidae (Aves: Pelecaniformes) de la

Formación San José (Plioceno-Pleistoceno) del Uruguay. Ameghiniana 39:183–191.

Sanz, J.L., Bonaparte, J.F. & Lacasa, A. (1988). Unusual Early Cretaceous birds from Spain. Nature

(London) 331: 433–435.

Sanz, J.L., Chiappe, L.M., Perez-Moreno, B.P., Buscalioni, A.D., Moratalla, J.J., Ortega, F. & Poyato-

Ariza, F.J. (1996). A new Lower Cretaceous bird from Spain: implications for the evolution of flight.

Nature (London) 382: 442–445.

Seguí, B. (2002). A new genus of crane (Aves: Gruiformes) from the Late Tertiary of the Balearic Islands,

western Mediterranean. Ibis 144(3): 411–422.

Senut, B., Dauphin, Y. & Pickford, M. (1998). Nouveaux restes aviens du Néogène de la Sperrgebiet

(Namibie): complément à la biostratigraphie avienne des éolianites du désert de Namib. Compt. Rend.

Séances Acad. Sci. Paris 327: 639–644. In French.

Sereno, P. (1999). The evolution of dinosaurs. Science (Washington, D.C.) 284: 2137–2147.

Sibley, C.G. & Ahlquist, J.E. (1990). Phylogeny and Classification of Birds: a Study in Molecular Evolution.

Yale University Press, New Haven & London.

Sibley, C.G. & Monroe, B.L. (1990). Distribution and Taxonomy of Birds of the World. Yale University

Press, New Haven & London.

Simpson, G.G. (1946). Fossil penguins. Publications of the Scarritt Expeditions 33. Bull. Amer. Mus. Nat.

Hist. 87(1): 1–100.

Simpson, G.G. (1971). A review of the pre-Pliocene penguins of New Zealand. Bull. Amer. Mus. Nat.

Hist. 144(5): 319–378.

Slack, K.E., Jones, C.M., Ando, T., Harrison, G.L., Fordyce, R.E., Arnason, U. & Penny, D. (2006).

Early penguin fossils, plus mitochondrial genomes, calibrate avian evolution. Mol. Biol. Evol. 23(6):


Sorenson, M.D., Cooper, A., Paxinos, E.E., Quinn, T.W., James, H.F., Olson, S.L. & Fleischer, R.C.

(1999). Relationships of the extinct moa-nalos, flightless Hawaiian waterfowl, based on ancient DNA.

Proc. Royal Soc. London (Ser. B Biol. Sci.) 266: 2187–2193.

Steadman, D.W. & Hilgartner, W.B. (1999). A new species of extinct barn owl (Aves: Tyto) from Barbuda,

Lesser Antilles. Pp. 75–82 in: Olson, S.L. ed. (1999). Avian Paleontology at the close of the 20th Century

– Proceedings of the 4th International Meeting of the Society of Avian Paleontology and Evolution, Washington

DC, June 4–June 7, 1996. Smithsonian Contributions to Paleobiology 89. Smithsonian Institution

Press, Washington, D.C. viii, 344 pp.

Steadman, D.W. & Miller, N.G. (1987). California Condor associated with spruce-jack pine woodland in

the Late Pleistocene of New York. Quaternary Res. (San Diego) 28: 415–426.

Steadman, D.W. & Zarriello, M.C. (1987). Two species of parrots (Aves: Psittacidae) from archaeological

sites in the Marquesas Islands. Proc. Biol. Soc. Washington 100: 518–528.

Stidham, T.A. (1998). A lower jaw from a Cretaceous parrot. Nature (London) 396: 29–30.

Storer, R.W. (1956). The fossil loon, Colymboides minutus. Condor 58: 413–426.

Stucchi, M. & Emslie, S.D. (2005). A new condor (Ciconiiformes, Vulturidae) from the Late Miocene/

Early Pliocene Pisco Formation, Peru. Condor 107: 107–113.

Švec, P. (1984). Further finds of grebe Miobaptus walteri in the Miocene of Bohemia. Casopis Mineral.

Geol. 29: 167–170.

Tambussi, C. (1995). Fossil Rheiformes of Argentina. Courier Forschungsinst. Senckenberg 181: 121–129.

Tambussi, C.P., Reguero, M.A., Marenssi, S.A. & Santillana, S.N. (2005). Crossvallia unienwillia, a

new Spheniscidae (Sphenisciformes, Aves) from the Late Paleocene of Antarctica. Géobios (Villeurbanne)

38: 667–675.

van Tets, G.F., Vickers-Rich, P. & Marino-Hadiwardojo, H.R. (1989). A reappraisal of Protoplotus

beauforti from the Early Tertiary of Sumatra and the basis of a new pelecaniform family. Publ. Geol. Res.

Develop. Centre Ministry Mines Energy Indonesia (Paleo. Ser.) 5: 57–75.

Tordoff, H.B. & Macdonald, J.R. (1957). A new bird (family Cracidae) from the Early Oligocene of

South Dakota. Auk 74: 174–184.

Trewick, S.A. (1996). Morphology and evolution of two Takahe: flightless rails of New Zealand. J.

Zool., London 238: 221–237.

Trewick, S.A. (1997). Flightlessness and phylogeny amongst endemic rails (Aves: Rallidae) of the New

Zealand region. Phil. Trans. Roy. Soc. London (Ser. B) 352: 429–446.

van Tuinen, M., Butvill, D.B., Kirsch, J.A.W. & Hedges, S.B. (2001). Convergence and divergence in

the evolution of aquatic birds. Proc. Royal Soc. London (Ser. B Biol. Sci.) 268: 1345–1350.

van Tuinen, M., Paton, T., Haddrath, O. & Baker, A. (2003). ‘Big bang’ for Tertiary birds? A reply.

Tree 18(9): 442–443.

Turvey, S.T., Green, O.R. & Holdaway, R.N. (2005). Cortical growth marks reveal extended juvenile

development in New Zealand moa. Nature (London) 435: 940–943.

Unwin, D.M. (1993). Aves. Pp. 717–737 in: Benton, M.J. ed. (1993). The Fossil Record 2. Chapman &

Hall, London.

Vickers-Rich, P. (1980). The Australian Dromornithidae: a group of extinct large ratites. Nat. Hist. Mus.

Los Angeles County (Contrib. Sci.) 330: 93–103.

Vickers-Rich, P. & Bohaska, D.J. (1976). The world’s oldest owl: a new strigiform from the Paleocene

of southwestern Colorado. Smithsonian Contrib. Paleobiol. 27: 87–93.

Vickers-Rich, P. & Haarhoff, P.J. (1985). Early Pliocene Coliidae (Aves, Coliiformes) from

Langebaanweg, South Africa. Ostrich 56(1/3): 20–41.

Vickers-Rich, P., Plane, M. & Schroeder, N. (1988). A pygmy cassowary (Casuarius lydekkeri) from

Late Pleistocene bog deposits at Pureni, Papua New Guinea. J. Austr. Geol. Geophys. 10: 377–389.

Walker, C.A. (1981). A new subclass of birds from the Cretaceous of South America. Nature (London)

292: 51–53.

Warheit, K.I. (2002). The seabird fossil record and the role of paleontology in understanding seabird

community structure. Pp. 17–55 in: Schreiber, E.A. & Burger, J. eds. (2002). Biology of Marine Birds.

CRC Press, Boca Raton, Florida.

Weber, E. & Hesse, A. (1995). The systematic position of Aptornis, a flightless bird from New Zealand.

Courier Forschungsinst. Senckenberg 181: 293–301.

Wetmore, A. (1931). Two primitive rails from the Eocene of Colorado and Wyoming. Condor 33: 107–109.

Wetmore, A. (1933). A second specimen of the fossil bird Bathornis veredus. Auk 50: 213–214.

Wetmore, A. (1943a). Fossil birds from the Tertiary deposits of Florida. Proc. New England Zool. Club

22: 59–68.

Wetmore, A. (1943b). Two more fossil hawks from the Miocene of Nebraska. Condor 45: 229–231.

Wetmore, A. (1944). A new terrestrial vulture from the Upper Eocene deposits of Wyoming. Ann. Carnegie

Mus. 30: 58–69.

Wetmore, A. (1957). A fossil rail from the Pliocene of Arizona. Condor 59: 267–268.

Wetmore, A. (1967). Re-creating Madagascar’s giant extinct bird. Natl. Geogr. 132(4): 488–493.

Witmer, L.A. (2002). The debate on avian ancestry: phylogeny, function and fossils. Pp. 3–30 in: Chiappe

& Witmer (2002).

Witmer, L.M. & Rose, K.D. (1991). Biomechanics of the jaw apparatus of the gigantic Eocene bird

Diatryma: implications for diet and mode of life. Paleobiology 17: 95–120.

Worthy, T.H. (1987). Sexual dimorphism and temporal variation in the North Island moa species

Euryapteryx curtus (Owen) and Pachyornis mappini Archy. Rec. Natl. Mus. New Zealand 3: 59–70.

Worthy, T.H. (1988). A re-examination of the moa genus Megalapteryx. Notornis 35: 99–108.

Worthy, T.H. (1998). The Quaternary fossil avifauna of Southland, South Island, New Zealand. J. Roy.

Soc. New Zealand 28(4): 537–589.

Worthy, T.H. (1999). Changes induced in the New Zealand avifauna by climate during the last glacial /

interglacial period. Pp. 111–123 in: Olson, S.L. ed. (1999). Avian Paleontology at the close of the 20th

Century – Proceedings of the 4th International Meeting of the Society of Avian Paleontology and Evolution,

Washington DC, June 4–June 7, 1996. Smithsonian Contributions to Paleobiology 89. Smithsonian

Institution, Washington, D.C. viii, 344 pp.

Worthy, T.H. (2000). The fossil megapodes (Aves: Megapodiidae) of Fiji with descriptions of a new

genus and two new species. J. Roy. Soc. New Zealand 30(4): 337–364.

Worthy, T.H. (2001). A giant flightless pigeon gen. et sp. nov. and a new species of Ducula (Aves:

Columbidae), from Quaternary deposits in Fiji. J. Roy. Soc. New Zealand 31(4): 763–794.

Worthy, T.H. (2002). The youngest giant: discovery and significance of the remains of a Giant Moa

(Dinornis giganteus) near Turangi, in central North Island, New Zealand. J. Roy. Soc. New Zealand 32(1):


Worthy, T.H. & Holdaway, R.N. (1995). Quaternary fossil faunas from caves on Mt Cookson, north

Canterbury, South Island, New Zealand. J. Roy. Soc. New Zealand 25(3): 333–370.

Worthy, T.H. & Holdaway, R.N. (2002). The Lost World of the Moa: Prehistoric Life of New Zealand.

Canterbury University Press, Christchurch, New Zealand.

Worthy, T.H. & Swabey, S.E.J. (2002). Avifaunal changes revealed in Quaternary deposits near Waitomo

Caves, North Island, New Zealand. J. Roy. Soc. New Zealand 32(2): 293–325.

Worthy, T.H., Edwards, A.R. & Millener, P.R. (1991). The fossil record of moas (Aves: Dinornithiformes)

older than the Otira (last) Glaciation. J. Roy. Soc. New Zealand 21: 101–118.

Wroe, S. (1999). The bird from hell? Nature Australia 26(7): 56–63.

Xu Xing, Norell, M.A., Kuang Xuewen, Wang Xiaolin, Qi Zhao & Jia Chengkai (2004). Basal

tyrannosauroids from China and evidence for protofeathers in tyrannosauroids. Nature (London) 431:


Xu Xing, Zhou Zhonghe & Wang Xiaolin (2000). The smallest known non-avian theropod dinosaur.

Nature (London) 408: 405–408.

Xu Xing, Zhou Zhonghe, Wang Xiaolin, Kuang Xuewen, Zhang Fucheng & Du Xiangke (2003).

Four-winged dinosaurs from China. Nature (London) 421: 335–340.

Zhang Fucheng & Zhou Zhonghe (2000). A primitive enantiornithine bird and the origin of feathers.

Science (Washington, D.C.) 290: 1955–1959.

Zhang Fucheng & Zhou Zhonghe (2004). Leg feathers in an Early Cretaceous bird. Nature (London)

431: 925.

Zhang Fucheng, Zhou Zhonghe, Xu Xing & Wang Xiaolin (2002). A juvenile coelurosaurian theropod

from China indicates arboreal habits. Naturwissenschaften 89: 394–398.

Zhou Zhonghe (1995). Discovery of a new enantiornithine bird from the Early Cretaceous of Liaoning,

China. Vertebrata Palasiatica 33: 99–113.

Zhou Zhonghe & Zhang Fucheng eds. (2002a). Proceedings of the 5th Symposium of the Society of

Avian Paleontology and Evolution, Beijing, 1–4 June 2000. Science Press, Beijing.

Zhou Zhonghe & Zhang Fucheng (2002b). A long-tailed, seed-eating bird from the Early Cretaceous of

China. Nature (London) 418: 405–409.

Zhou Zhonghe, Barrett, P.M. & Hilton, J. (2003). An exceptionally preserved Lower Cretaceous ecosystem.

Nature (London) 421: 807–814.

Zusi, R.L. & Warheit, K.I. (1992). On the evolution of intraramal mandibular joints in pseudodontorns

(Aves: Odontopterygia). Pp. 351–360 in: Campbell (1992