THE past FEW DECADES have seen a revolution in bird systematics – the study of bird relationships. Molecular techniques assessing genetic differences between species have forced some radical changes in the way we think birds evolved – and hence the way we classify them. This process is ongoing, but sufficient consensus has emerged for several field guides and bird books to take the plunge and adopt the new order. Peter Ryan explains some of the major changes. iologists attempt to classify birds Evolutionary relationships can be and other organisms in a hier- inferred from the pattern of shared CLASSIFYING BIRDS Barchy that reflects their evolution- derived characters – features common to ary relationships. This is termed natural a group that are not shared with more Every species is placed in a hierarchical classification, and implies that species in distantly-related species. The trick is to structure that should reflect its evolutionary the same genus are more closely related identify which sets of characters con- history. For example, the Black-winged Stilt to each other than those in other gen- vey a meaningful evolutionary signal. is classified as follows: era, and that families in the same order Many birds share the same character- are more closely related to each other istics simply because they live in the Domain: Eukarya (as distinct from Prokarya, than to other families, etc. There are same environment or feed on the same the bacteria and archaebacteria) other ways to group birds – by colour, type of food. This is termed convergent Kingdom: Animalia (as distinct from Plantae, size or even alphabetically – but these evolution. Relationships are also con- Fungi and Protista) are artificial classifications which, albeit founded by reversals in character states, useful for some purposes, are imposed whereby some of a related group of Phylum: Chordata (as distinct from all the rather than inherent. The advantage of organisms have reverted to the ancestral invertebrate phyla) using a natural classification system is condition. Class: Aves (all birds, although strictly birds that it tells us about the fundamental Initial attempts to infer the relation- are a subset of dinosaurian reptiles) relationships among birds. The main ships among birds were based largely on disadvantage is that as our knowledge their shape and structure, but it soon Order (-iformes): Charadriiformes (waders, improves, so the classification is likely to became apparent that characters subject gulls, terns, auks and sandgrouse) change. Such changes have even taken to selection for immediate survival, such Family (-idae): Recurvirostridae (avocets and place at the highest levels of classifica- as bill, wing and leg shape, carry little stilts) tion. Whereas there used to be two king- deep evolutionary signal. Focus then doms of organisms, animals and plants, switched to more conserved characters, Genus: Himantopus (the stilts) we now recognise six kingdoms, divided such as bones, muscles and other inter- Species: himantopus (the Black-winged Stilt) into two domains. nal structures. These formed the basis Orders and families take their names for the classification of birds into orders from one of their typical genera, but and families, whereas external appear- are distinguished by specific suffixes: ance and structure were used to group relationships. Different genes evolve at -iformes for orders and -idae for families. similar species into genera. different rates, providing biologists with Other levels in the hierarchy may be Bird taxonomy was hotly debated a toolkit they can use to assess the rela- recognised, such as Superfamily (-oidea), during the late 19th and early 20th tionships among all levels of life. For the Subfamily (-inae) and Tribe (-ini, between centuries, but a fairly stable order was most part, the new genetic evidence has subfamily and genus). Sometimes the established by the 1950s, reflecting at supported the relationships among birds, level at which a group is pegged is best a consensus based on the available at least at the level of orders, but there debatable. For example, the avocets and evidence, or at worst a compromise have been some major changes in the stilts are often treated as a subfamily between opposing factions. This resulted relationships among and within orders. Recurvirostrinae of a more in clusive in the familiar sequence in which birds Charadriidae, the plovers and lapwings. were listed in bird books. Sibley’S ‘tAPESTRy’ The past few decades have seen a Charles Sibley devoted much of his revolution in taxonomy thanks to the career to using molecular markers to The unique structural adaptations for development of techniques that detect infer evolutionary relationships. During life as a flightless, wing-propelled diver and analyse differences among organ- the 1980s he used a relatively crude resulted in penguins being placed at the isms in the most fundamental building technique called DNA-DNA hybridisa- start of the sequence of modern birds, blocks of life – their genes. Because most tion to conduct pair-wise comparisons TONY CAMACHO but they don’t reflect the penguins’ genes mutate at random, they provide between the DNA of birds from virtually true relationships. power ful tools for inferring evolutionary all major groups. Together with Jon 66 BIRD SYSTEMATICS AFRICA – BIRDS & BIRDING AUGUST/SEPTEMBER 2005 BIRD SYSTEMATICS 67 Ahlquist, he amassed enough informa- Bird) and the South American tinamous, Until recently, it was assumed that the shorebirds and diurnal raptors are more tion to propose an entirely new view on are represented in Africa only by the break-up of Gondwanaland took place closely related than previously thought, avian relationships. ostriches. The striking similarity of the too long ago to have influenced the radia- leading to the lumping of several orders Sibley confirmed that the most basic large, flightless ratites in all the southern tion of the more modern neognaths, of birds. Indeed, Sibley and Ahlquist division among birds is between the continents suggests that they evolved despite intriguing similarities between advocate lumping them all into a single Palaeognathae (‘ancient jaws’) and on the ancient southern super-continent some widely scattered birds. For exam- enlarged order, Ciconiiformes. Neognathae (‘new jaws’), which have Gondwanaland, which broke up dur- ple, the flightless Kagu of New Caledonia Some orders proved not to be natural fundamentally different palate structures. ing the Cretaceous, about 120–85 million is surprisingly similar to the Sunbittern groups. Among the former Pelecani- The former group, comprising the ratites years ago, to form Africa, Australia, South from Central and South America. The formes, characterised by four webbed (ostriches, emus, cassowaries and kiwis, America, Antarctica, Madagascar and the problem was that fossils, mainly from toes, cormorants, darters, gannets and as well as the extinct moas and Elephant Indian subcontinent. the northern hemisphere, suggest that boobies are closely related, but pelicans, modern birds only appeared approxi- tropicbirds and frigatebirds are not. Struthioniformes (ostriches, rheas, emu, cassowaries mately 65 million years ago, imme- Similar errors were detected at other and kiwis) PALAEO- diately after the Cretaceous–Tertiary taxonomic levels. Families such as the gnatHAE Tinamiformes (tinamous) extinction which saw the demise of the cuckoos and barbets were split because PETER RYAN dinosaurs. However, differences in gene they are not natural groups, and new Craciformes (megapodes, curassows and guans) GALLO- sequences suggest that the neognath light was shed on the long-debated Rheas, such as this Greater Rhea, bear a striking Galliformes (guineafowls, quail, pheasants and francolins) ANSERAE orders diverged at least 80–90 million positions of several enigmatic taxa. For resemblance to ostriches, even though South America Anseriformes (screamers, ducks and geese) years ago, when the southern continents example, the Shoebill Balaeniceps rex is broke away from Africa some 110 million years ago. Turniciformes (buttonquail) were still close together. The appearance a derived pelican, and the New World of modern bird fossils in the northern vultures and condors are related to Piciformes (honeyguides, woodpeckers, barbets and toucans) hemisphere coincides with the collision storks, and are merely convergent with families: the Muscicapoidea (thrushes, structure. From an African perspective, Galbuliformes (jacamars and puffbirds) between the Indian plate and Asia some Old World vultures. fly catchers, starlings and allies), the several unexpected relationships have Buceratiformes (hornbills) 65 million years ago. Perhaps the most exciting result was Sylvioidea (tits, larks, warblers, babblers, emerged. The elminias, a group of small, Upupiformes (hoopoes, woodhoopoes and scimitarbills) Within the neognaths, there are two the development of a framework for the bulbuls, white-eyes and allies) and the fan-tailed ‘flycatchers’, are merely con- major groupings: a basal radiation largest avian order, the Passeriformes. Passeroidea (sunbirds, weavers, sparrows, vergent with the crested-flycatchers and Trogoniformes (trogons) of gamebirds and waterfowl, the Gallo- With more than 5 700 species, pas- pipits, canaries, buntings and allies). paradise-flycatchers of the Corvoidea, Coraciiformes (rollers, motmots, todies, kingfishers and anserae, with all remaining birds in the serines make up some 60 per cent of and fall