Royal British Colombia Museum Victoria Canada 2University College Dublin Dublin Ireland

In our world today there is no living animal that looks like a bird but is actually something else. If it has feathered wings, a wishbone, and walks on its toes, then it is definitely a bird. In the Mesozoic world life was much more diverse. Among the giant dinosaurs, there were dozens of small, feathered animals that had wishbones and walked on their toes. Some of these were so small that the feathers on their limbs could lift them into the air. Somewhere among those ancient feathered animals, there was just one that was the earliest ancestor of modern birds. The rest, and all of their varied descendants, were "also rans" that disappeared in the events that carried away the giant dinosaurs and many other lineages at the end of the Cretaceous Period. Equally mysterious, and even more poorly known, are the birds, or groups of birds, that survived that great extinction to become the ancestors of today's 10,000 or so species.

The precise nature of the earliest birds has been one of the most contentious topics in biology for almost 150 years. At first the discovery of Archaeopteryx seemed to endorse Charles Darwin's newly published theory (Huxley, 1868). A reptile-like animal with feathers looked like the perfect link between two major lineages. However, concerns about apparent differences between birds and dinosaurs arose in the early 20th century (Heilman, 1926) and generated a heated debate that lasted until the beginning of the 21st century.

The issue was not with Archaeopteryx as the earliest known bird but the suite of differences that distinguish birds from dinosaurs. Until recently, birds seemed unlikely as relatives of the iconic giant dinosaurs of the Mesozoic; indeed as discussed in later chapters the nature of their clavicles and furculae remained contentious until the end of the 20th century (Chure & Madsen, 1996; Makovicky & Currie, 1998). In addition, Richard Owen, the noted Victorian authority on all things anatomical, declared that the two groups could not be related. Dinosaurs had simplified their hands by losing the two outer digits (IV and V), whereas birds lost the innermost and the outermost digits (I and V) (see Makovicky & Zanno, Chapter 1, this volume).

We can blame shortcomings in the fossil record for the duration of this controversy. Dinosaurs were thought to be giants because the bones of giants are more likely to be fossilized than the skeletons of small animals. Fragile structures like bird skeletons are easily destroyed and scattered long before they can be preserved. Fortunately the fossil record has been greatly enhanced and reinterpreted since 1980. John Ostrom (1976) led the way by describing the specialized bird-like wrist in the theropod Deinonychus and within a few years, thousands of new specimens were discovered in unexplored areas of China and other remote parts of the world (see O'Connor et al., Chapter 3, this volume). This new material included fossils of dozens of small, feathered animals in both avian and nonavian lineages.

During that same period at the end of the 20th century, biomolecular techniques were brought to bear on the genetic history of living birds. Analyses of nuclear and mitochondrial DNA allowed the construction of family trees that established some new connections between groups and demolished long-held beliefs in other relationships. At the time pioneering biomolecular analyses revealed that the ostriches and their relatives (Paleognathae) represent a very ancient lineage branch from the living birds (Sibley & Ahlquist, 1990; see Figure 0.1). Similarly the highly aquatic ducks were shown to be close relatives of the entirely terrestrial chickens and both belong to the Galloanserae, a group that split from the base of the main lineage shortly after the divergence of the paleognaths (Sibley & Ahlquist, 1990; Hackett et al., 2008).

Recently, genomic analyses have revealed the genetic history of the chicken, providing insight into the ancient history of birds and hinting at events in deep time when the avian lineage diverged from that of mammals (International Chicken Genome Sequencing Consortium, 2004). To no one's surprise, chickens completely lack genes for making tooth enamel giving scientific credibility to the adage, "scarce as hens' teeth." Recently, genetic studies of Hox gene activity in the embryo have contributed to discussions of the vexed question of the bird's lost digits (Wagner, 2005; Vargas et al., 2008;Young et al., 2009).

In spite of amazing advances, biomolecular studies still have one great shortcoming. They cannot be applied directly to the interpretation of organisms known only as fossils. However, they

Figure 0.1 Cartoon to illustrate the basics of our current consensus regarding the pattern of the evolution of birds, relative to the Cretaceous-Paleogene (K-Pg) boundary. [This figure appears in color as Plate 0.1]

might lead us to the identity of extant lineages whose antecedents crossed the Cretaceous-Paleo-gene (K-Pg) boundary and gave rise to modern birds (Figure 0.1). The fossil record of terrestrial animals in the Paleocene is particularly weak and biomolecular insights may be our best chance of understanding post-Mesozoic survival.

In spite of the new approaches to cladistic analyses of morphological characters and a vast array of newly discovered fossils, Archaeopteryx remains secure in its position as the earliest bird. It is no longer the oldest known feathered animal. (Xu & Zhang, 2005) and Anchiornis from China (Hu et al., 2009) have been placed in the Mid-Jurassic, some 10 million years (myr) earlier than Archaeopteryx (see Makovicky and Zanno, Chapter 1, this volume). Both fossils reveal feathered animals as small as Archaeopteryx but their feathers are symmetrical and not aerodynamically shaped. Such feathers cannot generate the thrust needed for forward flight and the aerial performance of these animals may have been limited to parachuting to the ground. Perhaps the plumage in these animals was more important as insulation.

These early feathered animals suggest that control of the third dimension was a theme with a long history in dinosaur evolution. The parasaggital stance of early forms raised them above their squamosal contemporaries, greatly increasing their field of view and allowing effective attacks from above. Pedopenna, Anchiornis, and Archaeopteryx represent a later development of small, feathered animals that may have kept their plumage clean by living in the forest canopy, such as it was in the Jurassic. The feathers offered a low overall density that reduced the risks of living far above the ground and allowed these animals to parachute, glide, or even fly down onto unsuspecting prey.

Archaeopteryx preserves the earliest evidence for feathered wings with asymmetric feathers capable of generating some useful thrust in flapping flight, but anatomically it is more similar to smaller-bodied dinosaurs, such as Velociiaptor, Deinonychus or Troodon, than to any modern bird. It had none of the specialized skeletal struc tures found in later birds that anchor large flight muscles and its aerial capabilities may have been extremely limited.

The birds that followed Archaeopteryx, in the Early Cretaceous, were also very dinosaur-like (Figure 0.1). Taxa such as Jeholornis and Raho-navis retained independent fingers in their wings, long bony tails, and, in some cases, sharp little teeth. Surprisingly, other fossils of Early Cretaceous birds, such as those of Gansus and Apsaravis already show the fused hand bones (carpometacarpus) and footbones (tibiotarsus) that are typical of modern birds. Gansus also had a significant keel on its breastbone, which suggests that it was also capable of sustained flapping flight. The vertebrae at the end of its vertebral column had fused to become a short pygostyle similar the one that carries the tail fan in modern birds. The pygostyle and sophisticated wrist joints imply that some Cretaceous birds were able to use their tail fans and wing tips for aerial maneuvers and control.

The skeleton of Gansus differed from those of its contemporaries in another important way that has been passed down to its descendants. The tips of its pubic bones met but were not fused. By the Mid-Cretaceous (100 million years ago (Ma)), the pubic bones of ornithurine birds, such as Ichthyornis, had lost even that level of connection. It and all more recent birds have no bony structures that cross the lower abdomen. The eggs and the gut are held in place solely by flexible connective tissue and there are no skeletal constraints on egg size. The absence of skeletal constraints on egg-size allows the production of large eggs and ultimately the altricial young, intensive nestling care, and elaborate nest structures that characterize the crown clades of living birds (Dyke & Kaiser, 2010).

Although primitive versions of modern birds appear in the Early Cretaceous, 120 Ma (Figure 0.1), their fossils are rather rare and we know little about their global distribution. The rich fossil beds of the Jehol Formation in China suggest that early representatives of modern birds successfully competed for resources with a variety of other feathered animals. However, the primitive pygostylian Confuciusornis is its most abundant avian fossil, even though it soon disappeared without known descendants. Somewhat later in the Cretaceous, birds must have competed for resources with a variety of small, feathered, arborial theropods, best represented by Microraptor gui. The exceptional diversity of the Jehol fauna may not have been a widespread phenomenon but there, and elsewhere, early ornithine birds competed for air space with other major lineages that achieved global distribution in the Cretaceous.

The membrane-winged pterosaurs were one of the most diverse groups of flying animals in the Cretaceous. They came in a wide variety of sizes and had already achieved global diversity long before the time of Archaeopteryx. Pterosaurs may have been able to exclude birds from some specific habitats but, in spite of their aerial habits, we have no evidence of interactions with early birds. They may have co-existed, just as bats and birds have little interaction today.

More importantly the Neornithes had to compete with a highly successful group of look-alikes called 'opposite birds' or Enantiornithes (Figure 0.1). Enantiornithines were not recognized until 1981, when Cyril Walker (1981) noticed subtle variations in the features in bird fossils from several different parts of the world. The Enan-tiornithines were capable of sophisticated flight based on a triosseal pulley system in the shouder, but the structure of their individual bones was somewhat different. Enantiornithines had a boss on the coracoid that articulated with a facet on the scapula. In ornithurine birds, a boss on the scapula articulates with a facet on the coracoid.

For many years the Mesozoic fossil record for ornithurine birds was so sparse that it was easy to believe that they did not begin to achieve their modern diversity until after the great extinction at the end of the Cretaceous. Not only did potential competitors among the small, feathered dinosaurs and the aerobatic pterosaurs die out 65 Ma but Enantiornithes disappeared as well. In many places fossils of enan-tiornithine birds are abundant enough to suggest that they prevented or delayed the diversification of ornithurine birds into some of the Cretaceous habitats. The enantiornithines appear to have been particularly abundant in terrestrial habitats, particularly in forests, where modern ornithurine birds are currently so successful. However, both fossils and tracks show that Mesozoic ornithurine were able to exploit aquatic and transitional shoreline habitats in which enantiornithine fossils were rare or absent. The very early ornithurine Gansus was fully aquatic (You et al., 2006), while abundant remains of the later graculavids present evidence for a wide variety of shorebird-like species in the Mid-Cretaceous.

It has proven exceptionally difficult to find fossils of Mesozoic birds that can be related to extant lineages. Not only are ornithurine fossils scarce but most consist solely of isolated elements or broken fragments. The handful of Meso-zoic species, such as Ichthyornis and Hesperornis, that are represented by nearly complete skeletons have become quite famous but appear to have no modern relatives. Unfortunately, the discovery of many well-preserved and nearly complete specimens of early birds in China has done little to clarify the situation. Most of them represent archaic, long-tailed birds with the rudimentary wings and unfused feet of other dinosaur lineages. A great many of them had teeth, a feature that clearly distinguishes them from any extant lineage. None of them have a uncontroversially "modern" anatomy and none can be linked to living forms.

For a while it seemed as though the Cretaceous avifauna would remain dominated almost entirely by creatures only remotely related to modern birds, but the story began to change early this century. First with the discovery of a 100 Ma wing in Mongolia's Gobi Desert (Kuroch-kin et al., 2002) and then with the re-description of a fossil from Cretaceous deposits in Antarctica (Clarke et al., 2005). Both these fossil birds have turned out to be the oldest known representatives of the living order Anseriformes (Figure 0.1). The detached wing was given the name Teviornis and is a member of a group of long-legged duck-like birds called presbyornithids. The more complete Antarctic fossil was named Vegavis. It is well enough preserved to be included in evolutionary studies that rely on anatomical characteristics taken from living birds. Both Teviornis and Vegavis appear to be early diverging members of the anseriform family tree.

Other named fossils from the Cretaceous that might just turn out to be representatives of living lineages - Palintropus, Lonchodytes, Tytthos-tonyx - remain contentious, either in the interpretation of their identifying characteristics or in the precise age of the matrix from which they were extracted. The determination of their exact relationship to modern forms must await further analysis or the discovery of new material. If these candidates are eventually demonstrated to be representatives of extent lineages, it will be as members of basal groups. None have characteristics attributable to any of the crown clades of birds.

The two major problems facing paleontologists over the next few years will be the same as those for the past century: "What did the ancestor of modern birds look like?" and "Where did the living groups birds come from?" However, the potential answers to those questions have already changed because our knowledge of Meso-zoic birds has expanded greatly. Even in the late 20th century we were looking for a more reptilian Ichthyornis to answer the first question and a less-toothy version to answer the second. Now the answer to the first question needs fossils from the Late Jurassic or the very early Cretaceous to tell us what the earliest modern birds looked like, an ancestor of Gansus that had not achieved sustained flight. Unfortunately, to find such a fossil one must be extremely lucky and appropriate Jurassic strata are exceedingly rare globally. It may be easier to answer the second question. All we need are examples of a few recognizably modern groups from some of the richest fossil beds in the world.

Was this article helpful?

0 0

Post a comment