Department of Vertebrate Paleontology, Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, CA 90007, USA. Email: [email protected]
Resúmen. – Los ancestros más cercanos de las aves. – El origen de las aves, el clado originado a partir delancestro común de Archaeopteryx del Jurásico tardío y las aves vivientes, ha estado inmerso dentro de ungran debate científico durante toda la historia de la biología evolutiva. Si bien muchas hipótesis diferentessobre el origen de las aves han sido propuestas en los últimos dos siglos, hoy en día existe un enorme con-senso en favor de la idea de que las aves evolucionaron a partir de dinosaurios terópodos clasificados den-tro de los Maniraptora. El sustento osteológico de esta hipótesis es enorme. Los esqueletos de dinosauriosmaniraptores como los dromaeosáuridos, troodóntidos y oviraptóridos, comparten muchas similitudescon aquellos de las aves. Además, una serie de espectaculares descubrimientos realizados durante la últimadécada ha brindado diversas líneas de evidencia que complementan el ya inmenso cúmulo de característi-cas osteológicas que sustenta la hipótesis Maniraptora. Esta reciente evidencia deriva fundamentalmentedel estudio de la morfología de los huevos y de la anatomía tegumentaria, pero también incluye inferenciasde comportamiento basadas en un pequeño, pero extraordinario, número de fósiles. Todos estos descubri-mientos han documentado la presencia de atributos tales como plumas, empollamiento, ovodeposiciónsecuencial (autocrónica), y otras características avianas en dinosaurios maniraptores basales. La evidenciadisponible sugiere fuertemente que las aves deben ser clasificadas dentro de los terópodos y que muchosatributos previamente considerados como únicos de las aves (desde el comportamiento de empollamientoa la capacidad de volar) evolucionaron por primera vez en dinosaurios maniraptores. Si bien los críticos dela hipótesis Maniraptora han resaltado problemas temporales y ontogenéticos, dichas objeciones son clara-mente irrelevantes. Los dos argumentos más frecuentemente utilizados, la llamada “paradoja temporal” yla homología de los dedos de la mano aviana, se encuentran embebidos en inconsistencias lógicas. Quizásmás importante es el hecho de que los críticos de la hipótesis Maniraptora han sido incapaces de formularuna hipótesis filogenética alternativa que pueda explicar la enorme similitud entre terópodos no-avianos yaves, dentro del marco de la parsimonia cladista.
Abstract. – The origin of birds, the clade originating from the common ancestor of the Late JurassicArchaeopteryx and extant birds, has been at the center of a heated debate throughout the history of evolu-tionary biology. Although many disparate hypotheses of bird origins have been proposed in the last twocenturies, an overwhelming consensus exists in support of the idea that birds evolved from maniraptorantheropod dinosaurs. Osteological support for this hypothesis is plentiful. The skeletons of such manirapto-ran dinosaurs as dromaeosaurids, troodontids, and oviraptorids, share a great deal of similarity with thoseof birds. In addition, a series of spectacular discoveries in the last decade has provided new lines of evi-dence that supplement the already overwhelming osteological data. This recent evidence is derived prima-rily from the study of egg morphology and integumentary anatomy but also includes behavioral inferencesbased on a handful of rare fossils. These discoveries have documented the presence of feathers, broodingbehavior, autochronous ovideposition, and other avian attributes among basal maniraptoran dinosaurs.The available evidence strongly supports the classification of birds within theropods and indicates thatmany avian attributes previously thought to be unique to birds (from brooding behavior to flight) firstevolved among maniraptoran dinosaurs. Although dissenters of the Maniraptoran hypothesis of bird ori-
101
CHIAPPE
gins have countered by highlighting temporal and developmental limitations, these criticisms are clearlyspurious. The most frequently voiced arguments, the so called “temporal paradox” and the homology ofthe digits of the avian hand, are tainted by logical inconsistencies. Perhaps the most important is the factthat these dissenters have been unable to produce alternative phylogenetic hypotheses that could explain,within the methodological framework of cladistic parsimony, the vast amount of similarity between non-avian theropods and birds. Accepted 11 December 2003.
Birds diversified more than 150 million yearsago. Their oldest known records are still fromthe Late Jurassic of southern Germany, whereArchaeopteryx was first discovered in the mid-19th century. Identifying the closest relativesto the group’s ancestor (the most recentcommon ancestor of Archaeopteryx andmodern birds) has been a matter of scientificdebate and scrutiny throughout the historyof evolutionary biology. As early as the18th century, birds were placed immediatelyahead of flying fishes in the ‘chain of being’postulated by the naturalists of thattime. With the advent of evolutionarythinking, especially after Darwin’s theoryof evolution by natural selection, moreexplicit hypotheses of relationships wereformulated. Indeed, in post-Darwiniantimes, birds were considered to be mostclosely related to a variety of extinctand extant lineages of reptiles, includingturtles, lizards, crocodylomorphs (moderncrocodiles and its Triassic relatives), adiversity of basal archosaurs and archosauro-morphs (e.g., the Triassic Euparkeria,Longisquama, and Megalancosaurus), pterosaurs(pterodactyls and their kin) as well as thero-pod and ornithischian dinosaurs (Fig. 1).Today, however, although most of thesehypothetical relationships have been aban-doned, the theropod hypothesis is receivingnearly universal acceptance. Hypothesesidentifying crocodylomorphs, basal archo-saurs, or basal archosauromorphs as the
closest relatives of birds have occasionallyresurfaced in the recent literature butthese have been used more as defaulthypotheses than as real alternatives tothe theropod origin of birds. Indeed, aclose examination of these “alternative”hypotheses reveals a lack of empirical supportbecause character evidence in supportof these hypotheses has also been discoveredamong theropod dinosaurs. Furthermore,these “alternative” hypotheses have continuedto be framed outside modern systematicmethods (i.e., cladistics), and thus alsolack the rigor of current phylogenetichypotheses.
Today, the debate on bird ancestry hasbeen resolved. The uncertainties that led tothis long controversy, both empirical andmethodological, have been clarified. The clos-est relatives of birds can be found amongtheropod dinosaurs, the carnivorous preda-tors that ruled the Mesozoic ecosystems. Thehistory of this fascinating scientific debate hasbeen summarized in a number of recentreviews, among them those of Witmer (1991,2002), Padian & Chiappe (1998), Chiappe(2001), and Prum (2002). A discussion inSpanish can also be found in Chiappe & Var-gas (2003). In this paper, addressed to theornithological community, I hope to conveythe message that the scientific hypothesis oftheir diverse disciplines, from ecology tobehavior to systematics, will greatly benefitfrom incorporating the notion that modernbirds are highly specialized, short-tailed, andflighted theropod dinosaurs.
102
CLOSEST RELATIVES OF BIRDS
HISTORY OF THE THEROPODHYPOTHESIS OF BIRD ORIGINS
The first suggestion that birds could havebeen related to dinosaurs came soon after the
publication of Darwin’s “Origin of Species”.Similarities in the structure of the tarsus ledGegenbaur (1864) to place the small, LateJurassic theropod Compsognathus in an interme-diate position between birds and other rep-
FIG. 1. Cladogram illustrating the diversity of hypotheses of the origin of birds (within Archosauromor-pha). BA, hypotheses that relate birds to some basal archosaurs or basal archosauromorphs; CRO,hypotheses in support of a crocodylomorph origin of birds; OR, a hypothesis in favor of a common ori-gin between birds and ornithischian dinosaurs; PT, a hypothesis supporting the ancestry of birds frompterosaurs (flying reptiles); TH, hypotheses in favor of the origin of birds from theropod dinosaurs; thishypothesis is the one endorsed in the present study. Modified from Chiappe & Vargas (2003).
103
CHIAPPE
tiles. At about the same time, Cope (1867)compared the tarsus of the Jurassic theropodMegalosaurus to that of an ostrich and, on thebasis of this and on similarities in the elonga-tion of the neck vertebrae and the lightness ofthe skull, he argued for a close relationship
between theropods and birds. Despite theseinitial considerations, it was Huxley (1868,1869) who championed the 19th century dis-cussions of the origin of birds from theropoddinosaurs. In 1869, Huxley argued that “if thewhole hind quarters, from the ilium to the
FIG. 2. Some osteological synapomorphies in support of the maniraptoran origin of birds. Modified fromChiappe & Vargas (2003).
104
CLOSEST RELATIVES OF BIRDS
toes, of a half-hatched chicken could be sud-denly enlarged, ossified, and fossilized as theyare, they would furnish us with the last step ofthe transition between Birds and Reptiles; forthere would be nothing in their characters toprevent us referring them to the Dinosauria.”(Huxley 1869). During the second half of the19th century, the theropod hypothesis of birdorigins was one of a pool of other hypotheses.Just as today, detractors argued that the simi-larities between birds and theropod dinosaurscould well be explained by convergence [seeSeeley’s discussion in Huxley (1869)]. In theearly 20th century, with the discovery of moregeneralized, Triassic archosaurs, in particularthe South African Euparkeria (Broom 1913),the theropod hypothesis lost ground. Thero-pods were deemed as too specialized to be theancestors of birds (Heillman 1926). Such anotion dominated the field for several decades(see Romer 1966) until the early 1970s, whenrenewed interest on the origin of birds tookplace (see Witmer 1991). Among this newwave of interest was Ostrom’s work onArchaeopteryx (Ostrom 1973, 1976), who revi-talized Huxley’s hypothesis of the theropodorigin of birds. During the thirty years thathas passed since Ostrom’s initial work on birdorigins, a large quantity of fossil documenta-tion supporting the dinosaurian ancestry ofthe group has been accumulated. Today,despite disagreement regarding the specifictheropod clade phylogenetically closest tobirds (e.g., dromaeosaurids, troodontids, ovi-raptorids), an overwhelming consensus existsin support to the notion that birds evolvedfrom maniraptoran theropods (Chiappe &Dyke 2002).
EVIDENCE FOR THE MANIRAPT-ORAN ORIGIN OF BIRDS
Several lines of evidence converge in supportof the hypothesis that the closest relatives ofbirds are to be found among maniraptoran
theropod dinosaurs (Chiappe 2001). Themost visible evidence of this hypothesis isbased on comparisons of the osteology,behavior, oology, and integument of birdswith that of a variety of nonavian theropods.These lines of evidence are summarizedbelow.
Osteology. A multitude of derived osteologicalcharacters are shared by all, or some, nona-vian maniraptoran theropods and birds (Fig.2). Comparisons between these taxa aregreatly assisted by the many newly discoveredbasal birds (Padian & Chiappe 1998, Chiappe& Dyke 2002), which possess a skeletal mor-phology only slightly modified from theancestral maniraptoran condition. Some ofthese derived characters are the presence ofrostral, dorsal, and caudal tympanic recesses(air spaces connected to the ear region), ven-tral processes on cervicothoracic vertebrae,ossified ventral segments of thoracic ribs,forelimbs that are more than half the lengthof hindlimbs, a semilunate carpal bone allow-ing swivel-like movements of the wrist, clavi-cles fused into a wishbone (probably asynapomorphy of a more inclusive theropodgroup), a pubic peduncle of the ilium longerthan the ischiadic peduncle (these pedunclesform the front and rear borders of the hip-socket), a vertically to caudoventrally orientedpubis ending in a boot-like expansion thatprojects only caudally, an ischium two-thirdsor less the length of the pubis, a femur with afeeble fourth trochanter (the attachment ofthe caudofemoralis longus muscle), and manyother characters distributed over the entireskeleton (Novas & Puerta 1997, Holtz 1998,Sereno 1999, Clark et al. 2002). Birds alsoshare a number of derived characters withmore inclusive theropod clades such as theCoelurosauria, Tetanurae, and Neotheropoda,and evolutionary trends towards the modernavian condition can be seen when these areexamined across cladograms of theropods
105
CHIAPPE
(e.g., forelimb elongation, pubic rotation,braincase amplification, stiffening of the tail).Indeed, many osteological features previ-ously thought to be exclusively avian, such asa furcula, laterally facing glenoids, large bonysterna, uncinate processes on ribs, have nowbeen discovered among nonavian manirap-torans (Padian & Chiappe 1998).
Behavior. Evidence of the behavior of extinctorganisms is rarely preserved in the fossilrecord. A handful of extraordinary discover-ies, however, have shed light on the nestingconduct of certain nonavian maniraptorans.Several skeletons of Late Cretaceous ovirap-torids from the Gobi Desert, belonging toboth Oviraptor (Osborn 1924) and Citipati
FIG. 3. A partial skeleton of the oviraptorid Citipati from the Late Cretaceous of the Gobi Desert brood-ing a clutch of its own eggs (A) and an interpretation of the posture of this animal when in life (B). Modi-fied from Clark et al. (1999).
106
CLOSEST RELATIVES OF BIRDS
(Clark et al. 2001), have been discovered ontop of their clutches of eggs (Fig. 3). Thespecimens of Citipati show that the animalsadopted a posture similar to those of brood-ing birds, with their legs tucked inside an openspace at the center of the clutch and theirforelimbs surrounding the periphery of theclutch (Clark et al. 1999). An oviraptoridembryo inside an egg of comparable mor-
phology to those in these clutches stronglysupports the idea that these specimens werebrooding their own nest (Clark et al. 1999).These discoveries have forced the reinterpre-tation of inferences made decades ago on thebasis of the holotype of Oviraptor philoceratops(Osborn 1924) which, because it had alsobeen found on top of a clutch of eggs, hadbeen interpreted (and consequently named) as
FIG. 4. Some oological synapomorphies in support of the maniraptoran origin of birds. Note the pres-ence of two or more layers in the eggshell of oviraptorids, troodontids, and birds, and the asymmetricshape of the egg in the last two lineages. Modified from Chiappe & Vargas (2003).
107
CHIAPPE
an egg-predator (Norell et al. 1995). A similardiscovery of a Late Cretaceous troodontidskeleton from Montana in an identical brood-ing position suggests that, regardless of itsspecific function (e.g., protection, incubation),the typical avian nesting behavior (i.e., adultssitting on top of their nests) was widespreadamong nonavian maniraptorans.
Oology. The general morphology and micro-structure of calcified eggs is specific to certaingroups of extant and extinct reptiles(Mikhailov 1997, Grellet-Tinner 2000). Untilrecently, the precise characteristics of the egg-shell microstructure of nonavian theropodsremained elusive due to the absence of diag-nostic embryonic material. The discovery ofthe Gobi oviraptorid embryo provided thefirst definitive evidence of a nonavian thero-pod egg (Clark et al. 1999). Since then, othernonavian maniraptorans embryos have beenfound. These include other species of ovirap-torids (Weishampel et al. 2000), therizinosau-rids (Manning et al. 2000), and troodontids(Varricchio et al. 2002). Comparative studiesbetween the eggshell microstructure of theseeggs and those of extant birds have revealedfeatures exclusively common to them (Fig. 4)(Mikhailov 1992, Varricchio et al. 1997, Zelen-itzky et al. 2002, Grellet-Tinner & Chiappe2004). One of these features involves thepresence of more than one distinct micro-structural layer, most commonly distin-guished by the differential disposition of thecalcitic crystals (Grellet-Tinner & Chiappe2004). The dinosaurian eggshell is character-ized by the presence of shell units, of whichthe inner portion is formed by a crystallinestructure that radiates from a core, oftentermed the organic core. In nonavian thero-pods, these units also possess an externalzone with a more spongeus microstructuralappearance, although in thin sections thislayer exhibits a more laminar appearance (thislayer is often referred as the squamatic zone).
In birds, this external zone may grade into, orbe completely separated from, a third, outer-most zone and, in paleognaths, even a fourthzone can be recognized (Grellet-Tinner2000). Even though up to now only two lay-ers have been found in the eggshell of non-avian theropods, no other group of nonavianreptiles possesses a similar zonation. Addi-tional character states shared by the eggs ofnonavian maniraptorans and those of birdsinclude a reduction in the porosity of theshell, a relative increase in the volume of theegg (with respect to the size of the adult), andthe presence of a longer axis (eggs that areelongated) (Zelenitzky et al. 2002). Anotheroological feature easily recognizable is thepresence of asymmetrical eggs, those in whichone pole is narrower than the other. Whileturtles, crocodiles, and nearly all non-aviandinosaurs are characterized for having sym-metrical eggs, the eggs of birds and those oftroodontid maniraptoran theropods have onepole that is narrower than the other. Furthersimilarities between nonavian maniraptoransand birds involve the mode of ovideposition.Unlike in the clutches of other dinosaurswhere eggs have no spatial arrangement, intheropod clutches, the eggs are clearlyarranged in pairs. Such a pairing is suggestiveof autochronous ovideposition, a mode ofdeposition in which eggs are not laid en massebut sequentially, at discrete time intervals. Thepairing of eggs found in fossil egg-clutchesattributed to theropod dinosaurs thus sug-gests that it could have taken several days fora theropod female to lay its egg-clutch(Varricchio et al. 1997, Grellet-Tinner & Chi-appe 2004), a condition shared with birds.
Integument. Feathers have always been thequintessential bird feature. In recent times,however, feathers have been found in a vari-ety of maniraptoran theropods, whose skele-tons are found in a series of Early Cretaceouslacustrine deposits in the northeastern corner
108
CLOSEST RELATIVES OF BIRDS
of China (Fig. 5). Carbonized remains offeathers are now known for the therizinosau-rid Beipiaosaurus (Xu et al. 1999), the oviraptor-
osaur Caudipteryx (Ji et al. 1998, Zhou & Wang2000), the dromaeosaurids Sinornithosaurus(Xu et al. 1999) and Microraptor (Xu et al. 2000
FIG. 5. The integument of nonavian maniraptorans (A, B) and a cladogram (C) illustrating the phyloge-netic relationships of known feathered nonavian theropods. A represents the oviraptorosaur Caudipteryx.The inset highlights the vaned feathers attached to the distal end of the forelimb. Note also the tuft offeathers attached to the tail of this dinosaur. B represents filamentous feathers surrounding the tail of thedromaeosaurid Sinornithosaurus. Modified from Chiappe & Dyke (2002).
109
CHIAPPE
2003), and the long-armed Protarchaeopteryx (Jiet al. 1998), which phylogenetic placementamong maniraptorans is less well-known.Integumentary structures interpreted as feath-ers have also been found in more primitivetheropods such as the basal coelurosaurianSinosauropteryx (Chen et al. 1998, Currie &Chen 2001), also from the same Early Creta-ceous rocks of China. While the latter exhibitsfeathers that are filament-like, with a minimaldegree of branching, Caudipteryx, Protarchaeop-teryx, Sinosauropteryx, and Microraptor displaypennaceous feathers with distinct shafts andvanes. Down-like feathers also cover portionsof the skeletons of all these taxa. A fan-shaped cluster of pennaceous feathers isattached to the distal part of the tail of Caudip-teryx (Fig. 5). Frond-like tails similar to thosein Archaeopteryx are present in the dromaeo-saurids Sinosauropteryx and Microraptor,although these pennaceous feathers are morerestricted to the distal half of the tail. Longpennaceous feathers are also attached to thetip of the forelimbs of Caudipteryx (Ji et al.1998) (Fig. 5) while in the tiny Microraptor gui,they form a wing of essentially modern design(Xu et al. 2003). There is also the remarkablepresence in the latter taxon of pennaceousfeathers attached to the distal half of thehindlimb. Such an attribute has been used toargue that Microraptor gui was able to glideusing these hindlimb feathers as an additionalairfoil (Xu et al. 2003). Despite functionalconsiderations that make this idea untenable(Chiappe & Vargas 2003), there is little doubtthat with such a small wing loading, Microrap-tor gui was able to fly [see Padian & Chiappe(1998) for a recent discussion on the origin offlight]. The presence of feathers in so manycoelurosaurian taxa suggests that these integ-umentary appendages evolved in the com-mon ancestor of the group if not earlier.Given the evidence at hand, the presence ofsimple, filament-like feathers is considered asa synapomorphy of Coelurosauria while the
presence of more derived, pennaceous feath-ers is interpreted as synapomorphic ofManiraptora. These discoveries not only doc-ument the presence of feathers outside birdsbut also, suggest that some nonavian thero-pod dinosaurs (e.g., Microraptor gui) may havebeen able to fly.
CRITICISMS OF THE MANIRAPT-ORAN HYPOTHESIS OF BIRDORIGINS
The evidence summarized above is so com-pelling that the idea that birds are the descen-dants of a maniraptoran ancestor has beenaccepted by a great number of evolutionarybiologists. Nonetheless, the maniraptoranhypothesis of bird origins is not exempt ofcritics, even though it is fair to say that theserepresent only a tiny fraction of specialists.Concerns have been expressed primarilyhighlighting apparent inconsistencies withinthe known fossil record and with the inferredhomology of certain structures. These appar-ent inconsistencies are briefly discussedbelow.
INCONSISTENCIES WITHIN THEKNOWN FOSSIL RECORD
This criticism highlights the chronologicalgap between the oldest known bird, the LateJurassic Archaeopteryx, and the Cretaceousnonavian maniraptorans that are typicallyused in discussions of bird origins (e.g.,Deinonychus, Velociraptor, Oviraptor). This argu-ment has become known as the “temporalparadox” since it highlights the inconsistencyof arguing that birds evolved from creaturesthat lived several million years after their ownorigin (Feduccia 1996, 1999). However, exam-ination of the theoretical basis and supportingevidence of the “temporal paradox” indicatesthat this argument stems from philosophicalmisconceptions, disregards critical fossil evi-
110
CLOSEST RELATIVES OF BIRDS
dence, and it constitutes an artifact caused bynot considering all alternative hypotheses ofbird origins at the same time (Brochu &Norell 2000).
In the first place, the “temporal paradox”stems from a philosophical misconceptionbecause none of these Cretaceous dinosaurs isregarded as the direct ancestor of birds(Padian Chiappe 1998, Witmer 2002). Inmodern times, the hypothesis of a manirapto-ran ancestry of birds has been framed as a cla-distic hypothesis that postulates the existenceof a most recent common ancestor of theseCretaceous dinosaurs and Archaeopteryx thatobviously existed before the divergence of theoldest of these taxa, Archaeopteryx (e.g., Gau-thier 1986, Forster et al. 1998, Holtz 1998, Se-reno 1999, Norell et al. 2001, Clark et al. 2002).Thus, in contrast to what has been claimed bythe proponents of the “temporal paradox”,the maniraptoran hypothesis does set the ori-gin of birds in pre-Jurassic times.
In addition, the absence of the pre-Creta-ceous maniraptorans the “temporal paradox”seems to highlight has long been proved to bemistaken. Late Jurassic maniraptorans havebeen known for decades, even if from frag-mentary remains (Padian & Jensen 1980), anda lower jaw of a maniraptoran therizinosaurid,Eshanosaurus deguchiianus, has been discoveredin the Early Jurassic of China (Xu et al. 2001).This and the fact that the stratigraphic rangesof theropod groups containing the cladeManiraptora have been recently extendedback by many millions of years [e.g., basal tet-anurans are now known from the Late Trias-sic; Arcucci & Coria (2003)], suggest that thedivergence of maniraptorans occurred muchearlier than in the Late Jurassic. The fact thatmaniraptoran fossils are exceedingly rare inpre-Cretaceous times may be related to a clearbias against small-sized dinosaurs of Jurassicage (most maniraptorans are of relativelysmall size) and the much smaller volume ofJurassic outcrops than those of Cretaceous
age (Clark et al. 2002).Finally, the “temporal paradox” appears to
exist only if one considers the temporal gapbetween the 100-million-year old Deinonychus,to take an example of a well-known dromaeo-saurid, and the 150-million-year old Archaeop-teryx. Yet, as shown by the statistical test ofBrochu & Norell (2000), when other recordsof well-known maniraptorans are included(e.g., the 125-million-year old dromaeosauridsSinornithosaurus and Microraptor) and whenhypotheses of bird origins are comparedagainst each other, placing birds withingroups indicated by hypotheses (e.g., crocody-lomorphs or more basally within archosaurs;see Fig. 1) other than the theropod hypothesismay increase the temporal disparity by asmuch as 15%. Thus, when the “temporal par-adox” is considered in the context of currentalternative hypotheses of bird origins (Fig. 1),the maniraptoran hypothesis is temporally themost consistent (Brochu & Norell 2000).
INCONSISTENCIES WITH THEINFERRED HOMOLOGY OF SOMESTRUCTURES
Embryology of the avian hand. Opponents to themaniraptoran origin of birds (e.g., Feduccia1999, 2003) have highlighted the conflictbetween the correspondence of the threewing digits of birds with respect to the ances-tral pentadactyl hand of tetrapods, as indi-cated by embryological studies, and thatinferred from the palaeontological evidence.Embryological investigations of extant birdshave identified five precartilaginous conden-sations of which only those in positions II,III, and IV develop into the three osseousdigits of the adult hand (Feduccia & Nowicki2002, Larsson & Wagner 2002). This observa-tion has often been extrapolated to include allbirds, even the Late Jurassic Archaeopteryx, andthe three fingers of the avian hand have beenidentified as homologous to digits II–IV of
111
CHIAPPE
the ancestral pentadactyl hand (Burke &Feduccia 1997, Feduccia 1999, Feduccia &Nowicki 2002). In contrast, inferences basedon the transformation of the hand asobserved from fossils representing differentstages of dinosaur evolution have identifiedthe homology of the three fingers of Archaeop-teryx as those corresponding to digits I, II,and III of the ancestral pentadactyl hand.This paleontological evidence shows a trendof reduction of the outermost two digits (VIand V) from the most basal theropods, where
these digits are abbreviated but still present,to tetanuran theropods bearing a tridactylhand (Padian & Chiappe 1998). The threedigits of the latter theropods have the samephalangeal formula as digits I, II, and III ofthe primitive five-fingered theropods, thusindicating that the three fingers of tetanurans(a group that also includes Maniraptora) cor-respond to digits I–III of the ancestral penta-dactyl hand. The remarkable similarity inmorphology, proportions, and phalangeal for-mula of the manual digits of certain nonavian
FIG. 6. Chondrification and ossification patterns of the right hand of the scincid lizards Hemiergis perioniand H. quadrilineata – an example of a homeotic frame shift in the development of the hand of a tetrapod.H. perioni includes morphs with three and four manual digits (A, B). In these morphs, digital condensa-tions II and III develop into the two anteriormost digits of the adult hand, which have three and four pha-langes, respectively. Adults of H. quadrilineata have only two manual digits, the anteriormost of them withthree phalanges and the other one with four phalanges (C). While in H. perioni the two anteriormost man-ual digits, those with three and four phalanges, ossify from condensations II and III, in H. quadrilineata,these digits ossify from condensations III and IV. The morphological similarity between the mature digitsof this species and digits II and III of other Hemiergis species is such that the positional identity of the twodigits of H. quadrilineata can only be verified through ontogenetic studies. Modified from Shapiro (2002).
112
CLOSEST RELATIVES OF BIRDS
maniraptorans (e.g., Velociraptor, Deinonychus)to those of Archaeopteryx has extended thisconclusion to this and to other basal birds.
Two different issues are involved in thiscontroversy. On the one hand is the questionof whether there is empirical basis for extrap-olating the ontogenetic development of mod-ern birds to Archaeopteryx. On the other handis whether the maniraptoran ancestry of birdscan be sustained even if nonavian theropodsdeveloped their manual digits through a devel-opmental pathway different than that of mod-ern avians. The extrapolation of theembryogenesis of the hand of extant birds toarchaic avian lineages including Archaeopteryxappears unwarranted given that the hand ofmodern birds is highly transformed and thatembryological evidence is unavailable foreither Archaeopteryx or any other basal avianlineage. Indeed, the fact that the hand ofArchaeopteryx is remarkably similar to that ofnonavian maniraptorans such as dromaeosau-rids (Ostrom 1976) suggests that if any devel-opmental trajectory is to be extrapolated tothis Late Jurassic bird, it should be the oneinferred for dromaeosaurid theropods. Thesecond issue, namely whether embryogeneticdifferences should take precedence over theenormous volume of evidence supporting thephylogenetic relationship between birds andcertain lineages of nonavian maniraptorans, isalso problematic. Certainly, such an approachwould be in direct conflict with the parsimonymethods endorsed by modern systematictechniques. If the maniraptoran origin ofbirds is to be rejected because the digits of thehand of living birds have ontogenetic trajecto-ries different from those inferred for extinctmaniraptorans, the unquestionable similari-ties in the osteology, plumage, oology, andbehavior of all these organisms would have tobe explained in the context of evolutionaryconvergence. Nonetheless, the apparentincongruence between the manual osteogene-sis of modern birds and that of their nonavian
theropod relatives can be explained withoutresorting to a different phylogenetic hypothe-sis. Wagner & Gauthier (1999) have arguedthat homeotic frame shifts could have led to adevelopmental pattern in which digits thatpreviously ossified from condensations I–IIIbecame ossified from condensations II–IV.Homeotic frame shifts are relatively commonamong other vertebrate lineages. An illustra-tive example involves the development of thehand of the two-toed earless skink (Hemiergisquadrilineata), an Australian scincid lizard (Sha-piro 2002) (Fig. 6). Many studies have docu-mented the fact that ontogenetic trajectoriesdo evolve and that these transformationscould occur without affecting either the mor-phology or function of the developing struc-ture (Wagner & Misof 1993, Mabee 2000, Hall2003). These structures are homologous, evenif their development pathways are different.
Lung structure and ventilation. Interpretations ofsoft structures supposed to indicate visceralcompartmentalization in the early Cretaceousbasal coelurosaurian theropod Sinosauropteryxprima (Chen et al. 1998) played a paramountrole in Ruben et al.’s (1997) claim that nona-vian theropods had a crocodile-like, hepatic-piston mechanism for lung ventilation. Rubenet al. (1997) questioned the close relationshipbetween birds and nonavian theropods on thebasis of this interpretation, because accordingto these authors, the transition from the croc-odile-like pulmonary system to the flow-through lung system of birds would haverequired the evolution of a diaphragmatic her-nia in the alleged partition that would havecompromised the efficiency of the pulmonarysystem of the transitional forms.
As in the case of other critics to the thero-pod hypothesis of bird origins, this argumentis based on problematic interpretations.Detailed studies of the skeleton of the speci-men of Sinosauropteryx prima used by Ruben etal. (1997) have demonstrated that the struc-
113
CHIAPPE
ture interpreted as a septum separating theabdominal cavity from the thoracic cavity is apreservational artifact (Currie & Chen 2001).Furthermore, the osteological correlates ofthe avian flow-through lung are well-knownamong nonavian theropods, in particular thevertebral pneumatization for the entrance ofpulmonary air sacs. The presence of intracos-tal articulations delimiting costal and ventralribs and the relatively large ossified sternalplates of nonavian maniraptorans also suggestthat the coordinated costal and sternal move-ments that ventilate the lungs of extant birdsmay have already been present in these dino-saurs (Clark et al. 1999). There is undeniableevidence for the presence of skeletal struc-tures correlated to the avian system of lungstructure and ventilation among nonaviantheropods.
CONCLUSION
Recent fossil discoveries have drasticallychange the volume of available evidence fordeciphering the historical relationships ofbirds. This new evidence has shown thatmany of the features previously considered tobe avian trademarks first evolved withintheropod dinosaurs. Criticisms leveled by theopponents of the theropod ancestry of birdsare empirically and methodologically mislead-ing; no alternative phylogenetic hypothesishas been framed within rigorous cladisticmethods. The notion that the closest relativesof birds must be found among maniraptorantheropods is today indisputable.
REFERENCES
Arcucci, A. B. and R. A. Coria. 2003. A new Trias-sic carnivorous dinosaur from Argentina.Ameghiniana 40: 217–228.
Brochu, C., & M. A. Norell. 2000. Temporal con-gruence and the origin of birds. J. Vertebr. Pale-ontol. 20: 197–200.
Broom, R. 1913. On the South-African pseudosu-chian Euparkeria and allied genera. Proc. Zool.Soc. London 1913: 619–633.
Burke, A. C., & A. Feduccia. 1997. Developmentalpatterns and the identification of homologiesin the avian hand. Science 278: 666–668.
Chen, P. J., Z. M. Dong, & S. N. Zhen. 1998. Anexceptionally well-preserved theropod dinosaurfrom the Yixian Formation of China. Nature391: 147–152.
Chiappe, L. M. 2001. The rise of birds. Pp. 102–106 in Briggs, D. E. C., & P. R. Crowther (eds.).Paleobiology II. Blackwell Science, Oxford,UK.
Chiappe, L. M., & L. M.Witmer. 2002. MesozoicBirds: Above the Heads of Dinosaurs. Univ. ofCalifornia Press, Berkeley, California.
Chiappe, L. M., & G. Dyke. 2002. The Mesozoicradiation of birds. Annu. Rev. Ecol. Syst. 33:91–124.
Chiappe, L. M., & A. Vargas. 2003. Emplumandodinosaurios: La transición evolutiva de terópo-dos a aves. Hornero 18: 1–11.
Clark, J. M., M. A. Norell, & L. M. Chiappe. 1999.An oviraptorid skeleton from the Late Creta-ceous of Ukhaa Tolgod, Mongolia, preservedin an avian-like brooding position over an ovi-raptorid nest. Am. Mus. Novit. 3265: 1–36.
Clark, J. M., M. A. Norell, & R. Barsbold. 2001.Two new oviraptorids (Theropoda: Ovirapto-rosauria), Upper Cretaceous Djadokhta forma-tion, Ukhaa Tolgod, Mongolia. J. Vertebr. Pale-ontol. 21: 209–213.
Clark, J. M., M. A. Norell, & P. J Makovicky. 2002.Cladistic approaches to bird origins. Pp. 31–61in Chiappe, L. M., & L. M. Witmer (eds.).Mesozoic birds: Above the heads of dinosaurs.Univ. of California Press, Berkeley, California.
Cope, E. D. 1867. An account of the extinct rep-tiles which approached the birds. Proc. Acad.Nat. Sci. Phila. 1867: 234-235.
Currie, P. J., & P. J. Chen. 2001. Anatomy of Sino-sauropteryx prima from Liaoning, northeasternChina. Can. J. Earth Sci. 38: 1705–1727.
Feduccia, A. 1996. The origin and evolution ofbirds. Yale Univ. Press, New Haven, Connecti-cut.
Feduccia, A. 1999. The origin and evolution ofbirds, 2nd ed. Yale Univ. Press, New Haven,
114
CLOSEST RELATIVES OF BIRDS
Connecticut.Fediccia, A. 2003. Bird origins: problem solved but
the debate continues…Trend Ecol. Evol. 18:9–10.
Feduccia, A., & J. Nowicki. 2002. The hand ofbirds revealed by early ostrich embryos. Natur-wissenshaften 98: 391–393.
Forster, C. A., S. D. Sampson, L. M. Chiappe, & D.W. Krause. 1998. The theropodan ancestry ofbirds: New evidence from the Late Cretaceousof Madagascar. Science 279: 1915–1919.
Galis, F., M. Kundrát, & B. Sinervo. 2003. An oldcontroversy solved: bird embryos have five fin-gers. Trends Ecol. Evol. 18: 7–9.
Gauthier, J. 1986. Saurischian monophyly and theorigin of birds. Mem. Calif. Acad. Sci. 8: 1–55.
Grellet-Tinner, G. 2000. Phylogenetic interpreta-tion of eggs and eggshells of Paleognathae. Pp.61–73 in Bravo, A. M, & T. Reyes (eds.). Firstsymposium on dinosaur eggs and babies. Isonai Conca Dellá, Catalonia, Spain.
Grellet-Tinner, G., & L. M. Chiappe. 2004. Dino-saur eggs and nesting: Implications for under-standing the origin of birds. Pp. 185–214 inCurrie, P. J., E. B. Koppelhus, & M. S. Shugar(eds.). Feathered dragons: studies on the transi-tion from dinosaurs to birds. Indiana Univ.Press, Bloomington, Indiana.
Hall, B. K. 2003. Descent with modification: theunity underlying homology and homoplasy asseen through an analysis of development andevolution. Biol. Rev. 78: 409–433.
Heilmann, G. 1926. Origin of birds. Witherby, Lon-don, UK.
Holtz, T. R., Jr. 1998. A new phylogeny of the car-nivorous dinosaurs. Gaia 15: 5–61.
Huxley, T. H. 1868. On the animals which are mostnearly intermediate between birds and reptiles.Ann. Mag. Natur. Hist. 4: 66–75.
Huxley, T. H. 1869. Further evidence of the affinitybetween the dinosaurian reptiles and birds.Quart. J.Geol. Soc. Lond. 26: 12–31.
Ji, Q., P. Currie, M. A. Norell, & S. A. Ji. 1998. Twofeathered dinosaurs from northeastern China.Nature 393: 753–761.
Larsson, H. C., & G. P. Wagner. 2002. Pentadactylground state of the avian wing. J. Exp. Zool.294: 146–151.
Mabee, P. M. 2000. The usefulness of ontogeny in
interpreting morphological characters. Pp. 84–114 in Wiens, J. J. (ed.). Phylogenetic analysis ofmorphological data. Smithsonian InstitutionPress, Washington. D.C..
Manning, T. W., K. A. Joysey, & A. R. I. Cruicks-hank. 2000. In ovo tooth replacement in a ther-izinosaurid dinosaur. Pp. 129–134 in Bravo, A.M. & T. Reyes (eds.). First symposium on dino-saur eggs and babies. Isona i Conca Dellá, Cata-lonia, Spain.
Mikhailov, K. E. 1992. The microstructure of avianand dinosaurian eggshell: phylogenetic implica-tions. Pp. 361–373 in Campbell, K. E. (ed.).Papers in avian paleontology, honoring PierceBrodkorb. Natural History Museum of LosAngeles County, Sci. Ser. No. 36, Los Angeles,California.
Mikhailov, K. E. 1997. Fossil and recent eggshell inamniotic vertebrates: fine structure, compara-tive morphology and classification. Spe. Pap.Palaeontol. 56: 1–80.
Norell, M. A., J. M. Clark, L. M. Chiappe, & D.Dashzeveg. 1995. A nesting dinosaur. Nature378: 774–776.
Norell, M. A., J. M. Clark, & P. J. Makovicky. 2001.Phylogenetic relationships among coelurosau-rian theropods. Pp. 49–68 in Gauthier, J. A., &L. F. Gall (eds.). New perspectives on the originand early evolution of birds: Proceedings of theinternational symposium in honor of John H.Ostrom. Peabody Museum of Natural History,Yale Univ., New Haven, Connecticut.
Novas, F. E., & P. Puerta. 1997. New evidence con-cerning avian origins from the Late Cretaceousof Patagonia. Nature 387: 390–392.
Osborn, H. F. 1924. Three new Theropoda, Proto-ceratops zone, central Mongolia. Am. Mus.Novit. 144: 1–12.
Ostrom, J. H. 1973. The ancestry of birds. Nature242: 136.
Ostrom, J. H. 1976. Archaeopteryx and the origin ofbirds. Biol. J. Linn. Soc. 8: 91–182.
Padian, K., & L. M. Chiappe. 1998. The origin andearly evolution of birds. Biol. Rev. 73: 1–42.
Jensen, J. A., & K. Padian. 1980. Small pterosaursand dinosaurs from the Uncompahgre fauna(Brushy Basin member, Morrison formation:?Tithonian), Late Jurassic, Western Colorado. J.Paleontol. 63: 364–373.
115
CHIAPPE
Romer, A. S. 1966. Vertebrate paleontology. 3rd. ed.Univ. of Chicago Press, Chicago, Illinois.
Ruben, J. A., T. D. Jones, N. R. Geist, & W. J. Hille-nius. 1997. Lung structure and ventilation intheropod dinosaurs and early birds. Science278: 1267–1270
Sereno, P. C. 1999. The evolution of dinosaurs. Sci-ence 284: 2137–2147.
Shapiro, M. D. 2002. Developmental morphologyof limb reduction in Hemiergis (Squamata: Scin-cidae): Chondrogenesis, osteogenesis, and het-erochrony. J. Morphol. 254: 211–231.
Varricchio, D. J., F. Jackson, J. Borkowski, & J. R.Horner. 1997. Nest and egg clutches of thedinosaur Troodon formosus and the evolution ofavian reproductive traits. Nature 385: 247–250.
Varicchio, D. J., J. R. Horner, & F. D. Jackson. 2002.Embryos and eggs for the Cretaceous thero-pod Troodon formosus. J. Vertebr. Paleontol. 22:564–576.
Wagner, G. P., & J. A. Gauthier. 1999. 1, 2, 3 = 2, 3,4: A solution to the problem of homology ofthe digits of the avian hand. Proc. Natl. Acad.Sci. USA 96: 5111–5116.
Wagner, G. P., & Y. Misof. 1993. How can a charac-ter be developmentally constrained despitevariation in developmental pathways? J. Evol.Biol. 6: 449–455.
Weishampel, D. B., D. E. Fastovsky, M. Watabe, R.Barsbold, & K. Tsogtbaatar. 2000. New embry-
onic and hatchling dinosaur remains from theLate Cretaceous of Mongolia. J. Vertebr. Pale-ontol. 20: 78A.
Witmer, L. M. 1991. Perspectives on avian origins.Pp. 427–466 in Schultze, H.-P., & L. Trueb(eds.). Origins of the higher groups of tetra-pods: controversy and consensus. Cornell Univ.Press, Ithaca, New York.
Xu, X., X. Wang, & X. Wu. 1999. A dromaeosaurdinosaur with filamentous integument from theYixian Formation of China. Nature 401: 262–266.
Xu, X., Z. Zhou, & X. Wang. 2000. The smallestknown non-avian theropod dinosaur. Nature408: 705–708.
Xu, X, Z. Zhou, X. Wang, X. Kuang, F. Zhang, &X. Du. 2003. Four-winged dinosaurs fromChina. Nature 421: 335–340.
Xu, X, Z. Zhao, & J. M. Clark. 2001. A new theriz-inosaur from the Lower Jurassic lower Lufengformation of Yunnan, China. J. Vertebr. Pale-ontol. 21: 477–483.
Zelenitsky, D. K., S. P. Modesto, & P. J. Currie.2002. Bird-like characteristics of troodontidtheropod eggshell. Cretaceous Res. 23: 297–305.
Zhou, Z., & X. Wang. 2000. A new species of Cau-dipteryx from the Yixian formation of Liaoning,northeast China. Vertebr. Palasiat. 38: 111–127.
