快猫短视频

WHAT made the first birds fly? Were they cold-blooded tree-living reptiles that evolved feathers for gliding, or fast-running warm-blooded dinosaurs who evolved feathers for insulation? And, after the initial development of flight, when did the evolutionary 鈥渇ine-tuning鈥� of adaptations take place? 鈥淭his issue of bird flight has to be one of the most intriguing questions in vertebrate evolution,鈥� says John Ostrom, former professor of geology at Yale University, and godfather of the warm-blooded dinosaur theory of bird origins. 鈥淲hat is it,鈥� he asks, 鈥渢hat cultivates perfectly normal, ground-living creatures into flying up into the air?鈥�

Debate over the origin of flight which has been ruffling human feathers for decades, is rife again after a series of exciting fossil finds. Discoveries in Patagonia and Madagascar last year, and the 鈥渇eathered dinosaur鈥� from China, seem to strengthen the dinosaur link. But opponents of a dinosaur origin for birds remain sceptical. They maintain that feathers are made for flying, and can only have evolved on an animal that developed aerial locomotion in the trees-not in a ground-dwelling dinosaur. They suspect that the apparently feather-like remains will turn out to be something else. Meanwhile, discoveries in China and Spain are revolutionising our estimates of the diversity and sophistication of early Cretaceous birds.

Any theory of the evolution of birds and bird flight must meet two requirements. The evolutionary stages before flight must be functional-each adaptation along the way must have been viable and perhaps provided its owner with some extra advantage. And any scenario must match the known fossil record. Unfortunately, a patchy fossil record means that neither of these requirements can be entirely met on either side of the argument-leaving speculation to fill the gaps.

Wing and a prayer

Archaeopteryx, the 鈥淔irst Bird鈥�, nestles at the centre of the debate. All known specimens of this 150-million-year-old feathered fossil come from Solnhofen, Germany. Archaeopteryx is part bird, part reptile, combining bird-like feathered wings with a reptile-like toothed jaw, clawed fingers, and a long, bony tail. The creature may have been able to fly, although its small flight muscles raise serious doubts about its prowess. Present-day birds share eight characteristic flight adaptations (see 鈥淔lying Kit鈥�), of which Archaeopteryx has only two-feathers and a pair of collarbones modified to form a furcula or wishbone.

Most palaeontologists agree with Ostrom, that Archaeopteryx and all other birds are descended from dinosaurs-a view first suggested by Thomas Huxley in 1868. Archaeopteryx鈥檚 skeleton is remarkably like a small theropod dinosaur-a miniature Velociraptor or a dromeosaur. Theropods were bipedal, predatory creatures that included Tyrannosaurus amongst their ranks. In fact, it is so like one particular small theropod, Compsognathus, that-until its feathers were noticed-two of the seven specimens of Archaeopteryx were initially misidentified as members of this genus.

Jacques Gauthier of Yale University believes the dinosaur theory, based on his cladistic analysis of hundreds of anatomical traits of early birds, dinosaurs and reptiles. Cladistics is a technique that groups organisms on the basis of shared features to create a hierarchical classification that reflects evolutionary relationships. His 1986 study nests birds securely within the theropod dinosaurs. Gauthier says simply: 鈥淏irds are as much dinosaurs as humans are mammals.鈥�

The main opponent of the dinosaur theory championed by Ostrom and Gauthier is ornithologist Alan Feduccia of the University of North Carolina. 鈥淥rnithologists just see Archaeopteryx as a primitive bird, with no connection to dinosaurs or anything; these dinosaurologists see it as a little dinosaur,鈥� he says. 鈥淲ell, I鈥檝e studied bird skulls for 25 years and I don鈥檛 see any similarity whatsoever. I just don鈥檛 see it.鈥� How certain is he that birds are not descended from dinosaurs? 鈥淭he theropod origin of birds, in my opinion, will be the greatest embarrassment of palaeontology of the 20th century,鈥� he declares.

Larry Martin, an expert in the anatomy of archaic birds at the University of Kansas, also opposes the dinosaur theory. Martin initially accepted that features such as the bony structures of the wrist, hand, ankle and hindlimb proved a close affinity between birds and dinosaurs. But after re-examining these characters in the mid-1970s he argued that many palaeontologists are misled into finding similarities by their ignorance of avian anatomy. 鈥淭o tell you the truth, if I had to support the dinosaur origin of birds with those characters, I鈥檇 be embarrassed every time I had to get up and talk about it,鈥� he says.

In his new book, The Origin and Evolution of Birds, published in November 1996, Feduccia reiterates his arboreal theory. He argues that Archaeopteryx evolved from a tree-living non-dinosaur that initially leapt, then glided, from perch to perch. With Martin, he suggests this creature belonged to one of the poorly known ancient groups of reptiles such as the crocodylomorphs-鈥淚magine a small, bipedal crocodile up a tree . . .鈥� says Martin-or the pseudosuchians, a primitive group ancestral to dinosaurs and flying pterosaurs, which appeared about 225 million years ago.

But Feduccia and Martin have a problem. Neither their hypothetical ancestor nor transitional forms linking it to known fossil birds have yet been found. And, although they rightly argue that cladistic analyses are only as good as the data upon which they are based, no cladistic study has yet suggested a non-theropod bird ancestor.

This is how the debate stood until two fossil finds, reported last year, tipped the balance further in favour of the dinosaur-bird link. The first, by Fernando Novas of the Argentine Museum of Natural Sciences in Buenos Aires, is an unnamed 90-million-year-old dromeosaurid dinosaur from Patagonia. Although fragmentary and much more recent than the oldest birds, this theropod fossil has a shoulder joint which tucked its arms up against its body. The discovery shows that the bird-like folding mechanism existed in nonflying dinosaurs and could have evolved prior to flight.

The second is the fragmentary find of a 75-million-year-old archaic bird from Madagascar. This unnamed turkey-sized creature sported a wicked-looking, enlarged claw on each foot. Catherine Forster of the State University of New York at Stony Brook and her colleagues, who found the remains, say this bird simply retained the slashing Velociraptor-like claws of its dinosaur ancestors.

But the debate is not just an exercise in comparative anatomy. Protagonists also argue about whether or not dinosaurs were warm-blooded, or 鈥渆ndothermic鈥�, and whether the early birds started flying either by running and flapping, or instead by gliding from trees.

Ostrom argues that dinosaurs were far from the sluggish cold-blooded stereotypical reptile. He believes that bird ancestors were not only accomplished bipeds, but evolved feathers for insulation. These creatures chased insect prey, using their feathered arms to prolong leaps into the air-a lifestyle powered by a warm-blooded metabolism. This paved the way for the development of powered, flapping flight. A warm-blooded animal, whose metabolism can ensure a constant, long-lasting and abundant supply of energy is far more likely to specialise in energetically expensive forms of locomotion-like fast running or flying than an 鈥渆ctothermic鈥� or cold-blooded animal. Ecto-therms generally produce less energy and tire more rapidly, making energetic locomotion impractical.

Varied evidence supports the idea of endothermic dinosaurs. Footprints suggest that dinosaurs moved rapidly and that their legs, like those of mammals, were tucked underneath their bodies rather than sprawled out to the sides like those of reptiles. Also, dinosaurs were able to live in high-latitude habitats and coped with a wide range of temperatures unlike reptiles which favour warmer climates. Unlike other reptiles, they migrated over huge distances, grew rapidly and laid down bone with a complex, interwoven structure like that of birds or mammals.

But physiologist John Ruben of Oregon State University questions early needs to be warm-blooded to achieve powered flight. Ruben says that while energy production in warm-blooded animals works better under normal aerobic, oxygen-burning conditions, cold-blooded creatures fare better- producing maybe twice as much energy per kilogram of muscle-in short-lived bursts fuelled by anaerobic, oxygen-free, metabolism.

In 1991, he looked at whether a cold-blooded Archaeopteryx, with flight muscles forming only seven per cent of its weight, could fly. James Marden, a physiologist at the Pennsylvania State University, has shown that endothermic birds need flight muscles weighing at least 16 per cent of their body weight to take off and sustain flight. But Ruben calculated that an ectothermic Archaeopteryx, relying solely on bursts of anaerobic metabolism, could have flown. Although even very short flights would have required an hour or more of rest to replenish oxygen supplies, Ruben鈥檚 arguments add to his endorsement of Feduccia鈥檚 conviction that feathers evolved for flight, not insulation. In partisan style, Feduccia hailed Ruben as a 鈥渃omrade in the war against `hot-blooded dinos鈥�.

Feduccia argues that flight began 鈥渢ree-down鈥�, and that parachuting and gliding by a reptile with feathered forelimbs would have allowed the gradual development of the muscles and skeletal structure necessary for later powered, flapping flight. He says that feathers are well adapted to flight, and points out that in birds that have lost flight, the feathers have degenerated to more hair-like structures. Even if dinosaurs needed to keep warm, he suggests, they would have developed hair, not feathers. What has made the debate over feathers, endothermy and avian ancestry so long-lasting has been the existence of plausible, but inconclusive counter-arguments to all these claims.

But late last year, this debate received a stimulating shock. At the meeting of the Society for Vertebrate Palaeontology in New York, Chen Pei-Ji of the Nanjing Institute of Geology and Palaeontology and Philip Currie of the Royal Tyrrell Museum of Palaeontology in Alberta showed a photograph and a drawing of a recently discovered, and flightless, 鈥渇eathered dinosaur鈥�, sending their colleagues and the media into a flap (快猫短视频, 19 October 1996, p 7). The fossil, found in Liaoning province in northeastern China, is between 121 and 142 million years old, and may be only slightly younger than Archaeopteryx. Its metre-long skeleton is very similar to that of Compsognathus. But most amazing are the traces of something very like feathers, running along the backbone and down the sides of the body, arms and legs.

Feathered halo

Ji Qiang of the Chinese Geology Museum in Beijing named the fossil Sinosauropteryx prima but has yet to publish a full account. Currie, perhaps the only Westerner to have seen the original, initially suspected that the so-called feathers would turn out to be nothing more than dendrites, branching mineral deposits common on many fossils. But he was quickly convinced otherwise. 鈥淲ithin 10 seconds, that was out of my mind,鈥� he says, recalling his reaction on seeing the specimen. He says he saw a 鈥渟ort of halo鈥� around the skeleton, made by 鈥渟oft, pliable tissues鈥� that had become wet and 鈥渃lumped up鈥� in places.

Only detailed microscopic studies will resolve whether these tissues have the quill, barbs and tiny barbules typical of modern feathers. 鈥淵ou can argue whether they are feathers or feather-like scales for now, but these tissues are there for a purpose and it鈥檚 hard to imagine what they are for if not insulation,鈥� says Currie. 鈥淚 think the most parsimonious explanation is that they are feathers.鈥� To him, the evidence of a link between hot-blooded dinosaurs and birds is now 鈥渙verwhelming鈥�.

Feduccia and Martin, however, remain deeply dubious about the identification of the alleged feathers. 鈥淭here is a 99 per cent chance it鈥檚 incorrect,鈥� says Feduccia, unable to conceive how a tissue so well designed for flight could have evolved initially to serve another purpose. 鈥淓verything about them indicates an aerodynamic function,鈥� he says. 鈥淭hey鈥檙e lightweight, they鈥檙e excellent airfoils, they produce high lift at low speeds, and they have a Velcro-like quality that lets them be reassembled. Feathers have an almost magical construction which is all aerodynamic in function. It would be gross evolutionary overkill to produce feathers like this for insulating a hot-blooded dino.鈥� But this view will be seriously challenged if it turns out that the Chinese dinosaur, like Archaeopteryx, has completely modern feathers.

Archaeopteryx is the tantalising start-but only the start- of the story of how modern birds and bird flight developed. Fine-tuning of flight adaptations was in progress over the next 85 million years. According to some palaeontologistsArchaeopteryx could be the ancestor of all other birds, but others believe it is ancestral only to the extinct enantiornithines or 鈥渙pposite birds鈥�. These take their name from the sequence in which their foot bones fuse together during development-from the bones nearest the body to those farthest from it-which is opposite to the sequence found in modern birds. After Archaeopteryx, enantiornithines became the dominant, archaic land-based birds until they disappeared with the dinosaurs about 65 million years ago.

Living at the same time were the ornithurines, a second group of archaic birds, with a modern sequence of foot-bone fusion. Some were aquatic, foot-propelled divers with teeth, while others were flying land birds. Only a few ornithurines-a small group known as the transitional shorebirds because they have similar adaptations to flamingos, ducks and ibises-seem to have survived the mass extinctions 65 million years ago, eventually giving rise to modern birds (Neornithes). Like their descendants, all the transitional shore birds were toothless, with a more modern shape and style of fusion in the bones in their skull and pelvis.

However, recent fossil finds are filling in the gaps in the story of how and when the other adaptations were acquired in the 85 million years separating the appearance of Archaeopteryx and the Neornithes, about 50 million years ago.

Probably the oldest of these recent finds is Confuciusornis sanctus. It was found in 1994 at the same site as Sinosauropteryx and also dates to between 121 and 142 million years ago. Lianhai Hou and Zhonghe Zhou at the Institute of Vertebrate Palaeontology and Palaeoanthropology in Beijing and their colleagues first described Confuciusornis late in 1995. It is an opposite bird, with modern feathers arranged in a wing (快猫短视频, Science, 21 October1995, p 19). Its backward-pointing first toe, or hallux, and strongly curved claws indicate an early adaptation for perching on branches. Confuciusornis is toothless, and has the earliest horny beak of any bird. It also has the earliest pygostyle, a short bony remnant of a tail that supports the tail feathers.

Oriental mysteries

In November 1996, another fossil bird from the same site and age, Liaoningornis, was described in a paper in Science by Hou, Martin, Zhou and Feduccia. Liaoningornis is the oldest known ornithurine, and differs significantly from its enantiornithine contemporary, Confuciusornis. Both species have feathered wings and feet adapted for perching, but Liaoningornis has a primitive toothed jaw like a reptile, suggesting its diet was very different from that of the beaked Confuciusornis.

Liaoningornis鈥檚 other flying adaptations are more modern than those of Confuciusornis. Its sternum is well developed, with the oldest known keel for anchoring flight muscles. And this, together with a capacious and strengthened ribcage, suggests Liaoningornis might have benefited from the air sacs and improved respiratory system typical of modern birds. The combination of these characteristics in a bird from 121 to 142 million years ago strengthens arguments for an early link between endothermy and the energy-demanding flapping flight later perfected by modern birds.

Meanwhile, fossils from the Las Hoyas site in Cuenca province, Spain, described by Jos茅 Sanz of the Universidad Auton贸ma of Madrid and colleagues, sketch in details of further flight apparatus modifications. In 1992, they described Iberomesornis romerali, an enantiornithine from 120 to 130 years ago. This sparrow-sized creature had a more modern shoulder joint than Archaeopteryx, and a perching foot. But it still had teeth and a primitive pelvis in which the bones were not fused together, as well as a relatively long and primitive pygostyle of between 10 and 15 vertebrae.

In August 1996 Sanz and colleagues from Madrid, together with Luis Chiappe of the American Museum of Natural History, described the 115-million-year-old Eoalulavis hoyasi from the same site. About the size of a goldfinch, this fossil enantiornithine preserves in extraordinary detail the feathers on its wings and body and even the remains of its last crustacean meal in its belly. Its flight adaptations include an unusually narrow sternum with a faint keel, a robust furcula, and a still more modern shoulder than Iberomesornis. Yet its fingers are tipped with primitive claws of an unknown function.

But the most significant adaptation is its small feathered first digit, making this the oldest recorded alula or bastard wing. The alula plays an important aerodynamic role in modern birds, allowing the great manoeuvrability and slow flying speed they need for landing on branches. Until this discovery, alulas (which require extraordinary fossil preservation if they are to be detected) had only been found in Neornithes. Eoalulavis, only some 25 to 30 million years younger than Sinosauropteryx and Archaeopteryx, illustrates the surprisingly early diversity and sophistication of flight adaptations. It also indicates that the alula evolved more than once since the enantiornithines left no descendants.

So it was not lack of adaptations for flight that doomed the enantiornithines and the early ornithurines to extinction at the end of the Cretaceous. These archaic birds presumably died from the same causes as the dinosaurs- dwindling ecological niches brought on by climatic change, possibly accelerated by the impact of a meteor. The avian and mammalian survivors found themselves in possession of a world that was relatively empty and ripe for a great spate of evolutionary diversification-well known from the Tertiary fossil record.

Archaeopteryx played a vital part in proving to early doubters of evolution that reptiles had indeed taken to the air. But with the new rush of fossil finds, palaeontologists are confident we will not have to wait another 140 years to find out how they did it, and exactly where their ancestry lies.

* * *

Flying Kit

There are eight key adaptations for flight present in all modern flying birds:

• Feathers to form a wing and maintain a stable body temperature.
• A modified clavicle the furcula (or wishbone) to anchor the large flight muscles and stabilise the shoulder joint.
• A bony, keeled sternum (or breastbone), also for anchoring flight muscles.
• A hand reduced to three fused fingers: a small first finger forming a small alula (or bastard wing); a large second finger; and a small third finger.
• A pygostyle (or ploughshare bone) made of between four and seven fused tail vertebrae to support the tail feathers.
• A shoulder joint that enables the chest muscles to flip the wings into a near-vertical position above the back, to begin the downstroke.
• A folding forelimb to tuck the feathered forearm against the body for safety and insulation.
• A highly efficient respiratory system (to fuel the endothermic metabolism provides the energy for flight) in which the bronchi and air sacs create a continuous, unidirectional airflow through the lungs, enhancing oxygen and carbon dioxide exchange.