The distinctive features of birds, from beaks to feathers, provide a stark separation between avians and other animal groups. But how did the features of the bird skull evolve? In a paper in Nature, Field et al.1 present a computerized reconstruction of the skull of a pivotal early bird that brings avian evolution into sharper focus.
In the late 1800s, the palaeontologist Othniel C. Marsh and his field crews made many of the first reported discoveries of ancient dinosaurs and mammals from western North America, amassing a fossilized ‘bestiary’ that dwarfed what was then known from Europe2. Marsh’s treasures were constantly in the headlines, perhaps never more so than when he published3 his monograph Odontornithes in 1880, which reported several previously undescribed fossil birds of the mid-Cretaceous period (around 80 million to 87 million years ago) from the shores of Kansas and nearby states. Familiar yet strange in many ways, these creatures were so archaic that they retained teeth and substantial bony tails, thus providing clues to the reptilian origin of birds. When Charles Darwin received a copy of the monograph from Marsh, the letter that he wrote back to Marsh said: “Your work on these old birds and on the many fossil animals of N. America has afforded the best support to the theory of evolution, which has appeared within the last 20 years” (see go.nature.com/2hhjxrd).
The specimens Marsh presented in Odontornithes were predominantly from two contrasting bird genera: Hesperornis, which was flightless and essentially wingless, standing 1.3–1.8 metres tall and comparable to today’s loons, and a tern-like bird called Ichthyornis, which had an average wingspan of about 60 centimetres3. However, neither was closely related to living loons or terns. Both birds had many sharp, curved teeth, which were absent only from the front part of the upper jaw, and their beaks were covered by a horny sheath. Unfortunately, the excavated bones, being small, fragile and of an elaborate architecture, were badly crushed, and proved challenging to prepare. The restoration, mounting and illustration of the specimens were, shall we say, somewhat overenthusiastic. The specimens could be convincingly described only after the mounts had been disassembled and prepared afresh more than a century later1,4.
Fast forward to the twenty-first century, and in the past 20 years some of the most sensational dinosaur discoveries have been the seemingly endless reports of ‘feathered’ dinosaurs and newly identified early birds, mainly from Cretaceous deposits in China. These specimens are closer to Archaeopteryx (the earliest known bird, from the Late Jurassic of Germany about 145 million years ago) than to Hesperornis and Ichthyornis5. These discoveries have shown that the evolution of feathers, from hair-like down to flight feathers, broadly paralleled the sequence of development of the features of a single feather in living birds6. Such insights suggest a plausible sequence for the evolution of wings and flight in birds, whereby newly hatched ancient dinosaurs flapped their incipient wings as a way of boosting their ability to scale steep inclines when evading predators7.
But many questions persist about the anatomical changes in early bird evolution, and this is where the work of Field and colleagues comes in. Present-day birds have skulls that are different in many ways from those of all other animals, including the dinosaurs from which they evolved. Bird snouts are lightweight, usually narrow and sometimes quite long. Indeed, the bones of the bird snout are relatively light and fragile compared with those of other animals, and these structures are covered by a strong beak made of the protein keratin, which enables birds to access various foods, such as seeds or carcasses. Inside the beak is a complex of bones that corresponds to the human palate; but unlike ours, the bird bones have mobile connections to each other and to the surrounding skull and jaw bones. This system of mobility is an elaboration of the basic dinosaurian one, and is key to accommodating the diverse feeding habits of birds.
Moreover, ‘bird brain’ is not the insult you might think. Bird brains are larger relative to their body size than is the case for reptiles, and the relative size of bird brains is comparable to that of placental mammals. As birds evolved from their dinosaur ancestors, the bones that protect the brain enlarged to keep pace with the changes in brain size. The bones of the skull roof and cheek region are also comparatively larger than the equivalent structures in their dinosaur ancestors, whereas the adductor muscles of the bird jaw are reduced. But in what order did these features evolve, and how did they shape avian evolution?
Ichthyornis is closely related to living birds, but retains many features of the earliest birds. No Ichthyornis skull material had been uncovered since Marsh’s discoveries in the 1870s. But Field et al. describe four new three-dimensionally preserved specimens with skull remains, and they image them in 3D, along with some overlooked skull bones from Marsh’s original specimens. The authors used a standard technique called high-resolution computed tomography, in which a reconstruction of each bone is compiled by taking extremely thin cross-sectional images all the way through the bone, like slicing a salami sausage and reassembling it. This enables the internal anatomy and outer shape to be visualized. The images of all the separate bones are assembled, and a computer program enables the bone images to be manipulated, allowing analysis of how the bones might have moved.
The resulting skull images (Fig. 1) show that the beak of Ichthyornis has some features that place it between the earliest birds and living birds: the beak was small, had not yet evolved a bony shelf structure in the palate and was limited to the tip of the jaw. However, the probable mobility of the Ichthyornis skull seems to be more like that of living birds. The brain would have been much like those of today’s birds, but the cheek region, bounded by bones of the skull roof and the side of the skull, has characteristics that are closer to those of dinosaurs, such as the retention of a large bony chamber for the adductor muscles that close the jaw. Therefore, several key features of the brain and palate evolved before the jaw muscles became reduced and the familiar features of the beak of living birds evolved.
This study raises many questions that remain to be answered. For example, were there functional changes that went along with reducing the jaw muscles from the ancestral dinosaurian condition? Did this change reflect a change in diet? And what ecological habits are correlated with the loss of teeth from the front part of the upper jaw and the evolution of the horny beak that covers it? Hesperornis was probably a diver that hunted fishes and invertebrates in the water column, whereas Ichthyornis seems to have been more a surface skimmer or perhaps a shallow plunger like a tern or gull2–4. How did these different predatory approaches favour the same pattern of tooth reduction, which also happened independently in other early bird groups? How did the mobility of the bones of the palate against adjacent skull bones in Ichthyornis compare with the ranges of motion in the palates of dinosaurs and living birds, and what might these evolutionary changes suggest about the diet and mode of feeding of Ichthyornis?
Whatever the answers to these questions turn out to be, Field and colleagues’ beautifully rendered 3D scans and reconstructions of this iconic fossil avian, along with their comparisons of these structures with those of earlier and later birds, provide an important resource to aid our understanding of early bird evolution.
Nature 557, 36-37 (2018)