ArticlePDF Available

Abstract and Figures

The transformation from a long reptilian tail to a shortened tail ending in a pygostyle and accompanied by aerodynamic fanning rectrices is one of the most remarkable adaptations of early avian evolution. However, no fossils directly capture this transition, and information regarding the structural morphology and the early evolution of the pygostyle in Mesozoic birds and their integuments is relatively limited. Here we provide a review of the pygostyle morphology of Early Cretaceous birds with comparison to the structure in living birds. This study emphasizes the convergent evolution of distally co-ossified caudal vertebrae in non-avian maniraptorans and early birds. There further exist distinct differences in pygostyle morphology between Sapeornithiformes, Confuciusornithiformes, Enantiornithes, and Ornithuromorpha. The morphology of the pygostyle and rectrices in early ornithuromorphs appear similar to that of extant birds, whereas the pygostyle in more primitive birds does not appear morphologically capable of supporting the rectricial bulbs and musculature necessary to control an aerodynamic fan-shaped tail. The rectricial bulbs and rectricial fan appear to have coevolved with the plough-shaped pygostyle early in the evolution of the Ornithuromorpha. This study also shows that the confuciusornithiform pygostyle was more similar to that of enantiornithines than previously recognized, consistent with the presence of nearly identical ornamental tail feathers in both groups.
Content may be subject to copyright.
55卷 第3
古 脊 椎 动 物 学 报
Morphological coevolution of the pygostyle and tail
feathers in Early Cretaceous birds
WANG Wei1,2 Jingmai K. OCONNOR1
(1 Key Laboratory of Vertebrate Evolution and Human Origins of Chinese Academy of Sciences, Institute of Vertebrate
Paleontology and Paleoanthropology, Chinese Academy of Sciences Beijing 100044
(2 University of Chinese Academy of Sciences Beijing 100049)
Abstract The transformation from a long reptilian tail to a shortened tail ending in a pygostyle
and accompanied by aerodynamic fanning rectrices is one of the most remarkable adaptations
of early avian evolution. However, no fossils directly capture this transition, and information
regarding the structural morphology and the early evolution of the pygostyle in Mesozoic
birds and their integuments is relatively limited. Here we provide a review of the pygostyle
morphology of Early Cretaceous birds with comparison to the structure in living birds. This
study emphasizes the convergent evolution of distally co-ossied caudal vertebrae in non-avian
maniraptorans and early birds. There further exist distinct differences in pygostyle morphology
between Sapeornithiformes, Confuciusornithiformes, Enantiornithes, and Ornithuromorpha.
The morphology of the pygostyle and rectrices in early ornithuromorphs appear similar to that
of extant birds, whereas the pygostyle in more primitive birds does not appear morphologically
capable of supporting the rectricial bulbs and musculature necessary to control an aerodynamic
fan-shaped tail. The rectricial bulbs and rectricial fan appear to have coevolved with the plough-
shaped pygostyle early in the evolution of the Ornithuromorpha. This study also shows that the
confuciusornithiform pygostyle was more similar to that of enantiornithines than previously
recognized, consistent with the presence of nearly identical ornamental tail feathers in both
Key words Jehol Biota, Aves, rectricial bulb, rectrices
Citation Wang W and J K. O’Connor, 2017. Morphological coevolution of the pygostyle and tail
feathers in Early Cretaceous birds. Vertebrata PalAsiatica, 55(3):
1 Introduction
Avian origins have been studied for over a century but the hypothesis that “birds are
maniraptoran theropod dinosaurs” only became universally accepted in the past two decades
following the discovery of abundant feathered maniraptorans in northeastern China (both non-
avian and avian) (Xu et al., 2014). To be concise, we will hereinafter use “theropod” to refer to
Advanced online publication
non-avian theropod dinosaurs and “maniraptoran” for non-avian maniraptoran theropods. To
date, hundreds of specimens of derived maniraptorans and birds have been collected from the
Late Jurassic Tiaojishan Formation and the Early Cretaceous Jehol Group (Zhou, 2004, 2014;
Wang et al., 2015). The Jehol Group, consisting of the 125 Ma Yixian Formation and the 120
Ma Jiufotang Formation, has also produced thousands of early bird specimens representing
half the known Mesozoic taxonomic diversity. Together with the Protopteryx-horizon of the
Huajiying Formation, the fossils from the Jehol Group form the spectacularly diverse Jehol
Biota (Pan et al., 2013). Together, these faunas preserve a wealth of paleobiological evidence
(such as stomach contents, soft tissue structures, and behavior) that greatly elucidate the
dinosaur-bird transition and the early evolution of birds (Xu et al., 2014).
Despite this wealth of new data, there still exist significant gaps in our understanding
of the structural transformation of birds from the theropod condition. One such evolutionary
transformation is the reduction of the long boney tail (present in theropods and the basalmost
birds Archaeopteryx and Jeholornis) into an abbreviated tail ending in a pygostyle, a
compound element formed through fusion of the distalmost caudal vertebrae. All birds with
a pygostyle form a monophyletic clade, the Pygostylia (Chiappe, 2002), which excludes only
the long bony-tailed birds, Archaeopteryx and the Jeholornithiformes (Jeholornis and kin).
The earliest record of the pygostyle is in the 130.7 Ma Protopteryx-horizon of the Huajiying
Formation, the earliest stage of the Jehol Biota, with the rst appearance of three pygostylian
lineages, the Confuciusornithiformes, Enantiornithes, and Ornithuromorpha (Zhang and Zhou,
2000; Zhang et al., 2008a; Wang et al., 2015). Unfortunately, no intermediate morphotypes
are known. The only specimen that may potentially elucidate this important transition is the
holotype of Zhongornis haoae from the Jiufotang Formation (Gao et al., 2008). Although
initially described as having an intermediate tail morphology, the only known specimen of
this taxon is a juvenile and poor preservation obscures interpretations. Descriptions of the tail
vary from 14 to 20 vertebrae (Gao et al., 2008; O’Connor and Sullivan, 2014) and thus the
transmutation of the tail during early avian evolution remains poorly understood. Historically,
the pygostyle has been considered a uniquely avian feature (Haeckel, 1883). However, more
recently pygostyle-like structures have been discovered in three clades of maniraptorans:
Therizinosauroidea, Oviraptorosauria, and Scansoriopterygidae (Xu et al., 2003; Barsbold et
al., 2000a, b; Zhang et al, 2008b).
In Neornithes (the clade that includes all living birds) the tail complex consists of two
parts: the tail feathers and the uropygium, the latter consisting of the caudal vertebrae and
associated soft tissue (Fisher, 1959; Baumel, 1988; Baumel et al., 1990; Gatesy and Dial, 1993,
1996a, b; Balmford et al., 1993; Thomas, 1997). The aerodynamic tail feathers, called rectrices,
are stiff and asymmetrical and overlapped dorsally and ventrally by rows of short tail coverts
(called tectrices or deck feathers). There are usually 5–6 free caudal vertebrae with expanded
transverse processes followed by a small mediolaterally compressed pygostyle (Baumel and
Witmer, 1993). The blade-like dorsal portion, the pygostyle lamina (lamina pygostyli), is
Advanced online publication
inferred to form through the fusion of the neural spines and arches. The cone shaped pygostyle
base (basis pygostyli) is formed through the fusion of the vertebral bodies (centra) (Baumel et
al., 1990; Baumel and Witmer, 1993). The pygostyle base has ventral processes that are weakly
excavated for muscle attachment. Paired broadipose tissue structures called rectricial bulbs lie
on both sides of the pygostyle and predominantly attach to the pygostyle lamina; the calami of
the rectrices are imbedded in this tissue, except for the medial pair which attach directly onto
the dorsodistal end of the pygostyle (Baumel, 1988). Spiraling around the surface of each bulb
is a striated muscle, the rectricial bulb muscle (m. bulbi rectricium), primarily responsible for
controlling the spread of the tail feathers (Baumel, 1988; Baumel and Witmer, 1993; Gatesy
and Dial, 1996a, b). In some birds (e.g., in Piciformes and Coliiformes) the ventral processes
are laterally expanded forming the ventrally concave pygostyle disc (discus pygostyli) that
provides an expanded area for the attachment of the well-developed caudal depressor muscle (m.
depressor caudae) present in these species; the disc is inferred to be derived from rudimentary
transverse processes (Baumel, 1988; Baumel and Witmer, 1993).
Except for the ightless paleognathus birds (the “ratites”), all modern birds share nearly
the same tail complex formed by pygostyle, rectricial bulbs and rectrices. Although in many
taxa the rectrices are modified for display, most birds have an aerodynamic fan consisting
of 10–12 feathers (Thomas and Balmford, 1995). This sophisticated tail complex plays a
significant role in avian flight. The rectricial fan produces lift to supplement the wings, the
force of which can be adjusted by controlling the surface area of the rectricial fan by adjusting
the spread of the feathers (Thomas and Balmford, 1995; Thomas 1997). This supplemental
lift is particularly important during takeoff, landing (for decelerating and braking) and slow
ight (Pennycuick, 1968; Spedding et al., 1984; Tucker, 1992; Gatesy and Dial, 1993). Tail
fanning is also used when turning, acting as a paddle or rudder (Baumel et al., 1990; Thomas,
1993, 1997), and the presence of a tail helps to maintain stability and balance, and streamline
the body (Thomas, 1993). Because the tail complex functions as an integrated whole, the
uropygium and integument are morphologically correlated and one can be used to predict the
other (Felice and O’Connor, 2014). Furthermore, pygostyle morphology can be used to predict
ight or foraging style in both extinct and extant birds (Felice, 2014). These results support the
hypothesis that pygostyle and rectricial morphology co-evolve (Clark et al., 2006).
Despite the large number of specimens uncovered from the Jehol Biota over the past
several decades (Zhou, 2004, 2014; Xu et al., 2014), the early evolution of the avian tail
complex is poorly understood. Several short discussions have been published regarding
specic taxa: Jeholornis (O’Connor et al., 2012, 2013), Eopengornis (Wang X et al., 2014),
Shanweiniao (O’Connor et al., 2009), Yixianornis (Clarke et al., 2006) and Iteravis (Zhou et
al., 2014). However, the only research focused on the morphological changes that occur in the
derived maniraptoran tail during early avian evolution predates most important discoveries
(Gatesy and Dial, 1993, 1996a, b). The pygostyle and tail feathers, when preserved, are
typically only briefly described (Chiappe et al., 1999; Clarke et al., 2006; O’Connor et al.,
Advanced online publication
2009; Hu et al., 2015), and very little information is available regarding the soft tissue (rectricial
bulbs and muscle), although potential traces are reportedly preserved in the ornithuromorph
Iteravis (Zhou et al., 2014) and the enantiornithine Feitianius (O’Connor et al., 2015).
Extensive interspecific comparison is lacking and the relationship between skeletal features
and integument has only been explored in the pengornithid enantiornithines (Hu et al., 2015;
O’Connor et al., 2016). Here we describe in detail the morphological characteristics of the
pygostyle and tail feathers in Mesozoic birds and closely related theropods. We discuss
morphological differences between clades, potential function, and place this information in the
context of recent phylogenetic analyses in order to test hypotheses regarding the origin of the
extant avian tail fanning complex.
Institutional abbreviations BMNH, Beijing Natural History Museum, Beijing, China;
CAGS, Chinese Academy of Geological Sciences, Beijing, China; CNU, Capital Normal
University, Beijing, China; DHNM, Dalian Natural History Museum, Dalian, China; IVPP,
Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences,
Beijing, China; LPM, Liaoning Paleontology Museum, Shenyang, China; STM, Tianyu
Natural History Museum of Shandong, Pingyi, China.
2 Materials and methods
This study mainly utilized the collections of the IVPP, supplemented by data from the
collections of the STM. Only taxonomically identifiable specimens with pygostyle or tail
integument preserved were included in this study. Because the morphology of the caudal
skeleton is affected by post-hatching development (Heers et al., 2014; Zhou et al., 2013),
juvenile specimens (such as the holotype of Zhongornis haoae) were excluded from this study.
Anatomical nomenclature primarily follows the “Handbook of Avian Anatomy” using English
equivalents for the Latin terminology (Baumel and Witmer, 1993).
Preserved in nely laminated lacustrine sediments, almost all specimens from the Jehol
Biota are compressed, preserved nearly two-dimensionally in slabs. This preservation limits
the information available from individual elements to a single exposed surface (e.g., dorsal,
lateral, ventral or dorsolateral etc.) preventing the use of geometric morphometric analysis
methods (Zelditch et al., 2012). Thus, three simple but reliable measurements were chosen: the
total length of the pygostyle (TLP), the mid-length width of the pygostyle (MLW, for dorsally/
ventrally preserved specimens), and the mid-length height of the pygostyle (MLH, for laterally
preserved specimens). In order to reduce the effects of body size, we normalized the TLP using
the total length of the humerus (TLH) and the total length of femurs (TLF), two most reliable
and widely used proxies for body size and mass in most birds (Liu et al., 2012; Hone, 2012).
Most of the data (27 specimens) were collected directly from specimens measured using
stainless hardened digital calipers; measurements from four specimens were taken from the
literature (Sapeornis CAGS-03-07-08 (Yuan, 2005), Zhouornis (Zhang et al., 2013), Sulcavis
(O’Connor et al., 2013), Shanweiniao (O’Connor et al., 2009)).
Advanced online publication
3 Description and analysis
3.1 Description of pygostyles and tail feathers
Non-avian theropoda Non-maniraptoran theropods have typical reptilian tails usually
consisting of dozens of free caudal vertebrae, e.g., about 35 in Tyrannosaurus (Persons and
Currie, 2011a) and 42 in Carnotaurus (Persons and Currie, 2011b). The proximal vertebrae
have short centra with neural arches and transverse processes present. These processes are
completely reduced in the distal vertebrae, which have elongate centra with the length more
than four times the centra height/width in some taxa (e.g., Sinosauropteryx, Ornitholestes)
(Pittman et al., 2013). The fusion of distal caudal vertebrae is unknown in non-maniraptoran
theropods, and rare in non-avian maniraptoran theropods. The most primitive (Fig. 1) known
pygostyle-like structure is present in the basal therizinosaur Beipiaosaurus (IVPP V 11559) (Xu
et al., 2003). The complete tail of Beipiaosaurus (V 11559) consists of 30 vertebrae. Preserved
Fig. 1 Phylogeny of Maniraptora and
related Theropoda dinosaurs, modied
from Brusatte et al., 2014
in lateral view, the last five caudals are well co-ossified
into a rod-like element with a slightly convex dorsal
margin formed by the prezygopophyses. The distal sixth
and seventh caudal centra are completely fused to each
other but not to the other five fused vertebrae. The tail
plumage consists of lamentous (without a fused rachis)
“proto-feathers” with no proximodistal differentiation (Xu
et al., 2003). No composite laments are observed on the
tail of Beipiaosaurus, which falls outside Pennaraptora
(the derived clade of maniraptorans in which pennaceous
feathers are present) (Brusatte et al., 2014).
Some members of the Oviraptorosauria (basal most
clade in the Pennaraptora) also possess fused terminal
caudals. Some members of this bird-like clade have at one
time been considered secondarily ightless birds although
most similarities are derived cranial features and probably
related to the presence of herbivory (Dyke and Norell, 2005). In one specimen of the primitive
taxon Similicaudipteryx (IVPP V 12056) the distal most caudal vertebrae (at least two) are co-
ossied into a dagger-like mass (He et al., 2008). This morphology is also observed in derived
oviraptorosaurs, such as Nomingia, Citipati and Conchoraptor (Barsbold et al., 2000a, b;
Persons et al., 2013). In the very primitive Caudipteryx the distal ve caudal vertebrae articulate
tightly through the presence of well-developed prezygopophyses and the last three of these are
ankylosed but remain unfused (IVPP V 22606). These vertebrae lack neural spines and transverse
processes, which are present in the proximal caudal vertebrae. Tail feathers are preserved in at
least three primitive oviraptorosaurs: Protarchaeopteryx, Caudipteryx and Similicaudipteryx. The
holotype of Caudipteryx has an estimated eleven pairs of tail feathers attached to the caudalmost
Advanced online publication
ve or six vertebrae, which form the incipient pygostyle. Although the tail feathers in all these
three taxa are pennaceous (vaned), no specimen preserves evidence of hooklets on the barbules (Ji
et al., 1998). In Caudipteryx V 22606 two layers of tail feathers are evident: one shorter layer of
rachis-less body feathers and another longer layer of pennaceous rectrices (Fig. 2).
One member of the Scansoriopterygidae, Epidexipteryx, has a reduced tail consisting
of 16 vertebrae ending in a rod-like structure approximately formed by the distal ten caudal
vertebrae which are unfused. The distal ten vertebrae are reduced to simple centra without
processes. Two pairs of elongate feathers project from the pygostyle in Epidexipteryx;
described as “ribbon-like” (Zhang et al., 2008b), these tail feathers resemble the poorly
preserved “rachis-dominated” feathers present in some enantiornithines. Pennaceous tail
feathers are otherwise unknown in the Scansoriopterygidae.
Basalmost birds With the exceptions of Archaeopteryx and Jeholornis, all Mesozoic
birds have a pygostyle. Archaeopteryx is the basalmost bird with a tail composed of 21–22
articulated caudal vertebrae (Elzanowski, 2002; Foth et al., 2014) suggesting that the
Fig. 2 Photographs and camera lucida
drawings of distal caudal vertebrae of
Caudipteryx sp. (IVPP V 22606)
Abbreviations: ch. chevron人字骨;
cv. caudal vertebra尾椎; . laments
丝状羽毛; pi. pinnules羽枝;
poz. postzygopophysis后关节突;
prz. prezygopophysis前关节突
Scale bar=5mm
plesiomorphic avian tail was composed of no fewer than
20 caudal vertebrae. Pennaceous tail feathers, presumably
attached through ligamental connections, line the lateral
surfaces of the tail, overlapping to form a frond-like
morphology (Gatesy and Dial, 1996b; Mayr et al., 2007).
The 11th specimen of Archaeopteryx preserves a notch
in the caudal margin of the tail frond. The proximal
rectrices are asymmetric and shorter than the symmetrical
overlapping distal rectrices (Feduccia and Tordoff, 1979;
Foth et al., 2014). The slightly more derived Jeholornis
has 27 free caudal vertebrae (IVPP V 13350) forming
a tail longer than that of Archaeopteryx (Zhou and
Zhang, 2002a, 2003a). Jeholornis possesses a unique tail
plumage with two functionally discriminate rectricial
pterylae (O’Connor et al., 2012, 2013). One consists of a
short fan-shaped array of rectrices presumably embedded
in the soft tissue dorsal to the short proximal caudal
vertebrae, and the second is distally located, consisting
of laterally oriented cranially curved rectrices forming a
frond-like arrangement without an extensive aerodynamic
surface. Vane symmetry is equivocal due to preservation.
Although the lift generated by the tails in these two taxa
are insignicant when compared with that of the extant pigeon, the tail complex in Jeholornis is
relatively more efcient than that of Archaeopteryx, producing similar lift while at the same time
incurring much less estimated drag (O’Connor et al., 2013).
Advanced online publication
The Sapeornithiformes and the Confuciusornithiformes are the basalmost known avian
lineages with a pygostyle (Zhou and Zhang, 2002b; Zhang et al., 2008a). Despite published
diversity, the Sapeornithiformes is considered a monospecific clade with all taxa referable to
Sapeornis chaoyangensis (Pu et al., 2013; Gao et al., 2012). The pygostyle is best preserved in
two specimens, IVPP V 13275 and V 13276, both preserving this element in left dorsolateral
view. In the cranial three vertebrae of the pygostyle, neural spines and transverse processes can
be faintly distinguished; the subsequent vertebrae are more completely fused and individual
processes cannot be identified. The dorsal surface is interpreted as forming a weak pygostyle
lamina (Fig. 3). On each lateral surface of the lamina there is a laterally projecting ridge inferred
to form through fusion of the transverse processes. More than ten additional specimens of
sapeornithiforms preserving the pygostyle were examined at the STM. In roughly a third of these
specimens fusion of the pygostyle is incomplete, suggesting that the pygostyle fully co-ossies
late during ontogeny. Tail feathers are rarely preserved, present in only one published specimen
(STM 16-18); this specimen preserves multiple elongate rectrices preserved in lateral view,
interpreted as forming a long, strongly graded tail-fan (Zheng, et al., 2013; Wang X et al., 2014;
Xu et al., 2014). Two other specimens (STM 15-18,16-5) preserve this feature in dorsoventral
view revealing a tail consisting of 8-10 narrow, symmetrical, non-overlapping feathers.
Distinct from that of sapeornithiforms, the pygostyle in confuciusornithiforms is more
strongly co-ossified, proportionately longer and more robust (Martin et al., 1998; Chiappe
et al., 1999). This detailed study of the confuciusornithiform pygostyle reveals a more
Fig. 3 Photographs and camera lucida drawings of pygostyles of Sapeornithiformes
A. Sapeornis, IVPP V 13276, in dorsal view; B. Sapeornis, IVPP V 13275, in dorsal-lateral view
Abbreviations: dr. dorsal ridge背侧脊; llr. left lateral ridge左侧脊; ns. neural spine棘突;
rlr. right lateral ridge右侧脊; tp. transverse process横突. Scale bars=3 mm
Advanced online publication
complex morphology than previously recognized (Martin et al., 1998; Chiappe et al., 1999).
The pygostyle has a pair of dorsolateral ridges (Fig. 4A) and another pair of proximoventral
processes (Fig. 4B). The proximoventral processes project cranially beyond the proximal
articular surface (Fig. 4A, B) and are proximally restricted whereas the dorsolateral processes
extend the entire length dorsally dening a deep concavity. The ventral surface is keeled; the
height of the keel is obscured by compression but it appears that it forms a majority of the
pygostyle corpus (Fig. 4B). The caudal margin is rounded and slightly laterally expanded
(Fig. 4C). In lateral view, foramina are present in some specimens (e.g., IVPP V 16066, STM
13-58). The number and position of these foramina are variable, and we suggest these may
represent a subadult feature, subject to ontogenetic change and lost with maturity. Numerous
confuciusornithiform specimens preserve a single pair of rectrices that are elongate, rachis-
dominated and only distally pennaceous (racket-plumes). Other specimens preserve only
short rachis-less body feathers surrounding the pygostyle (Fig. 4A, C). These racket-plumes
are interpreted as sexually dimorphic ornaments present only in males, evolved under sexual
selection (Chiappe et al., 2008).
Enantiornithes Enantiornithes is the most diverse group of Cretaceous birds. Although
their pygostyle and tail feathers do not vary to the extent observed in extant birds, accordingly
they show a greater morphological diversity than other known Cretaceous clades. The caudal
vertebrae are well fused; sutures and individual processes cannot be recognized in adults. The
three-dimensionally preserved pygostyle of Halimornis (Chiappe et al., 2002) clearly reveals
that a typical enantiornithine pygostyle consists of paired dorsolateral ridges demarcating a
deeply incised dorsal surface (in the case of Halimornis, incised forming a deep V-shaped
excavation); ventrally, the body is very narrow, with a pair of ventrolateral processes. The
dorsolateral ridges project cranially beyond the proximal articular surface forming the dorsal
fork described by Chiappe et al. (2002). Together with the smaller ventrolateral ridges, the
pygostyle appears X-shaped in proximal view (Fig. 5). The distal end of the pygostyle is
tapered, formed by a constriction in the mediolateral width of the dorsal surface and the fact
the ventrolateral processes do not extend the entire length of the pygostyle.
Although most if not all known enantiornithine taxa share this unique general
morphology, appreciable variation within this template is observed. The basal Pengornithidae
(Zhou et al., 2008; Wang X et al., 2014; Hu et al., 2015) possess all the characteristic
enantiornithine features but their pygostyle is proportionately shorter and broader and the distal
end is untapered: the ventrolateral processes are proximally restricted and the dorsal surface
maintains an even width (Pengornis IVPP V 15336; Fig. 5A). The dorsal surface is broadly
concave, rather than deeply incised as in Halimornis. The caudal margin of the pygostyle has
a medial cleft, unique to this family of enantiornithines. Compared to the typical condition,
the bohaiornithid pygostyle (Fig. 5B, C) is more slender and the ventrolateral processes
taper gently rather than end abruptly as in most enantiornithines (Wang M et al., 2014).
The Longipterygidae, the family that includes Longipteryx, Rapaxavis, and Shanweiniao,
Advanced online publication
Fig. 4 Photographs and camera lucida drawings of pygostyles of Confuciusornithiformes
A. Confuciusornis, IVPP V 11640, in right doesolateral view; B. Jinzhouornis, IVPP V 12352, in ventral view;
C. Confuciusornis, IVPP V 13156, in dorsal view
Abbreviations: ldlr. left dorsolateral ridge左背侧脊; lpvp. left proximoventral process左腹侧前突;
rdlr. right dorsolateral ridge右背侧脊; rpvp. right proximoventral process右腹侧前突; vk. ventral keel腹侧脊
Scale bars=3 mm
is characterized by a pygostyle that is proportionately larger and more robust than other
enantiornithines (O’Connor et al., 2011). In Rapaxavis the distal end of the pygostyle caudal
to the ventrolateral processes forms a spade-like expansion that is distally tapered (Fig. 5D).
Other simpler morphologies have been reported (e.g., Vescornis) (Zhang et al., 2004) but in the
absence of additional evidence these are interpreted as due to poor or incomplete preservation.
In some enantiornithine specimens (e.g., Eoenantiornis IVPP V 11537, Longipteryx IVPP
V 12325) rectrices are distinctly absent, with only rachis-less body feathers present, similar
to some confuciusornithiform specimens. Many specimens preserve a single pair of elongate
rachis-dominated tail feathers (e.g., Protopteryx IVPP V 11665, Dapingfangornis LPM 00039,
Bohaiornis LPM B00167, Parapengornis IVPP V 18687) (Zhang and Zhou, 2000; Li et al.,
Advanced online publication
2006; Hu et al., 2011; Hu et al., 2015). Paraprotopteryx (STM V 001) preserves two pairs of
feathers (Zheng et al., 2007). These feathers vary with regards to the extent of the pennaceous
vane, being fully pennaceous in Parapengornis and Eopengornis and distally restricted in
all other specimens in which preservation is clear (Dapingfangornis, GSGM-07-CM-001,
Fig. 5 Photographs and camera lucida drawings of pygostyles of enantiornithines
A. Pengornis, IVPP V 15336, in left dorsolateral view; B. Parabohaiornis, IVPP V 18690, in left laterodorsal
view; C. Sulcavis, BMNH-Ph 00805, in left lateral view; D. Rapaxavis, DNHM D 2522, in ventral view
Abbreviations: ldlr. left dorsolateral ridge左背侧脊; lpdp. left proximodorsal process; lpvp. left proximoventral
process左腹侧前突; lvlr. left ventrolateral ridge左腹侧脊; rdlr. right dorsolateral ridge右背侧脊;
rpdp. right proximodorsal process; rvlr. right ventrolateral ridge右腹侧脊. Scale bars=3 mm
Advanced online publication
Paraprotopteryx STM V 001). The shape of the distally restricted pennaceous portion varies
between taxa being distinctly racket-shaped in Dapingfangornis and Paraprotopteryx whereas
the distal expansion is more gradual in other specimens (GSGM-07-CM-001) (O’Connor et al.,
Ornithuromorpha The Ornithuromorpha contains all living birds (Neornithes) and
extinct taxa more closely related to living birds than to the Enantiornithes (Chiappe, 2002;
O’Connor and Zhou, 2013; Jetz et al., 2012; Jarvis et al., 2014; Wang et al., 2015). Several
Early Cretaceous fossil ornithuromorphs preserve evidences of an advanced tail complex.
The pygostyle is most commonly preserved in lateral view (e.g., holotype of Yixianornis
grabaui IVPP V 12631 (Zhou and Zhang, 2001), holotype of Piscivoravis lii IVPP V 17078
(Zhou et al., 2013), holotype of Iteravis huchzermeyeri IVPP V 18958 (Zhou et al., 2014)
and Gansus yumenensis CAGS-IG-04-CM-002 (You et al., 2006)), revealing a ploughshare-
shape approximately like that observed in extant birds and proportionately smaller than in
other Early Cretaceous avian taxa (Fig. 6). The ratio of dorsoventral MLH compared to TLP
in Jehol ornithuromorphs is larger (varies from 0.37 in Yixianornis to 0.60 in Hongshanornis)
than in other Cretaceous clades (Confuciusornis 0.25, enantiornithine Longipteryx 0.17)
with the exception of Sapeornis (0.45). Because all available Early Cretaceous basal
ornithuromorph specimens preserve the pygostyle in lateral view, their width cannot be
successfully measured. We suggest the predominance of lateral preservation implies that
the pygostyle is mediolaterally thin, as in neornithines. Generally, the pygostyle in Early
Cretaceous ornithuromorphs consists of a pygostyle lamina and a narrow pygostyle base, with
the transverse processes forming uneven low ridges on the lateral surface (Fig. 6). In lateral
view the pygostyle is upturned (Fig. 6) so that it is angled relative to the distal free caudals;
this morphology is also observed in doves, falcons and many other living birds (Baumel et al.,
1988). Although somewhat similar to sapeornithiforms in overall shape, the ornithuromorph
pygostyle is mediolaterally thinner and proportionately shorter than in this taxon and
signicantly differs from the robust morphology in other Mesozoic avian lineages. Non-feather
dark traces preserved surrounding the pygostyle in Yixianornis V 12631 and Iteravis V 18958
have been interpreted as the carbonaceous remains of the rectricial bulbs (Clarke et al., 2006;
Zhou et al., 2014).
The basalmost ornithuromorph preserving tail feathers is the holotype of Schizooura lii
(IVPP V 16861), which preserves a forked tail, considered by aerodynamic models to be the
most efcient tail shape (Zhou et al., 2012, Thomas et al., 1993). All other ornithuromorph
specimens with caudal integument preserve the impression of an aerodynamic tail fan
formed by multiple (at least 6-10) rectrices. Fan-shaped arrays of rectrices are present in
Hongshanornithidae (Hongshanornis V 14533, DNHM D2945/6; Tianyuornis STM 7-53;
Archaeornithura STM 7-145), Songlingornithidae (Yixianornis: V 12631; Yanornis: STM 9-51)
and Piscivoravis (V 17078) (Chiappe et al., 2014; Zheng et al., 2014; Zhou et al., 2013; Wang
et al., 2015). The rachis and individual barbs are discernible in the rectrices of some exquisitely
Advanced online publication
Fig. 6 Photographs and camera lucida drawings of pygostyles of ornithuromorphs
A. Yixianornis, IVPP V 12631, in left lateral view; B. Piscivoravis, IVPP V 17078, in left lateral view;
C. Iteravis, IVPP V 18958, in left lateral view
Abbreviations: pb. pygostyle base尾综骨基部; pl. pygostyle lamina尾综骨板. Scale bars=3 mm
preserved specimens (e.g., Yixianornis: V 12631, Piscivoravis: V 17078). At least eight
rectrices are preserved in the holotype specimen of Yixianornis grabaui (Clarke et al., 2006);
a minimum of six feathers were reported in the holotype specimen of Piscivoravis lii (Zhou
et al., 2013); and more than ten are present in a referred specimen of Hongshanornis DNHM
D2945 (Chiappe et al., 2014). The caudal margins of these tail fans are weakly graduated.
Although the exact number of rectrices cannot be determined in any specimen, more than two
rectrices are present in all known basal ornithuromorphs with the possible exception of the
poorly preserved tail in the holotype of Iteravis huchzermeyeri (Zhou et al., 2014).
Advanced online publication
3.2 Data analysis
Pygostyle data was collected from 31 specimens belonging to 20 genera and representing
at least 8 families (Table 1). Almost all ornithuromorph pygostyles are preserved laterally and
do not provide data on the MLW. The width was measured from the holotype of Schizooura lii,
but lateral compression of the pygostyle in this specimen and specimens of Sapeornis (IVPP V
13275 and V 13276) prevents accurate measurements. Selected linear pygostyle measurements
were compared to determine if clades can be differentiated on the basis of pygostyle absolute
sizes and relative proportions.
The Mesozoic birds from Jehol can be divided into four groups: Sapeornithiformes,
Confuciusornithiformes, Enantiornithes and Ornithuromorpha, based on the phylogeny. We
choose TLH and TLF, which are commonly used, both as the appropriate proxies of avian
body size (Liu et al., 2012; Hone, 2012). The result (Table 1) indicates that TLP/TLH values
in sapeornithiforms (0.11–0.17) and ornithuromorphs (0.15–0.27) are distinctly lower than
those of confuciusornithiforms (0.45–0.54) and most of enantiornithines (0.34–0.66, exclude
those of Pengornithidae). The result can be visible in Fig. 7A: scatter points belonging
to sapeornithiforms and ornithuromorphs are more closed to TLH-axis, while those of
confuciusornithiforms and enantiornithines are near TLP-axis. In the enantiornithines, low
TLP/TLH values can also be found in the Pengornithidae (0.18–0.30). Consistent result also
can be conrmed when the proxy is changed into TLF (see TLP/TLF in Table. 1; Fig. 7B). The
two aspects of results both suggest proportionally shorter pygostyles relative to body size in
sapeornithiforms and ornithuromorphs, especially in the latter.
Clades cannot be easily distinguished by MLW/TLP values (Table 1; Fig. 7C). Because
the data of sapeornithiforms and ornithuromorphs are too questionable and limited (only one
datum of each) to be considered, and the ratio of MLW/TLP present in confuciusornithiforms
Fig. 7 Scatter diagram of metrical data (Table 1) of fossil pygostyles
A. TLP-TLH; B.TLP-TLF; C. TLP-MLW; D. TLP-MLH. In accord with the phylogenetic result, all the data
can also be separated into four parts by each parameter-couple in rectangular plane coordinate systems
Advanced online publication
Table 1 Measurements of pygostyles, humeri and femurs of Mesozoic avian fossils (mm )
Genera Specimen TLP MLW MLH TLH TLF TLP/
(the only family of
IVPP V 13275
21.3 7.5 122.9 74.1 0.17 0.29 0.35
IVPP V 13276
13.7 6.1 120.4 72.1 0.11 0.19 0.45
20.0 130.0 0.15
(the only family of
IVPP V 13175
30.4 7.6 67.2 55.3 0.45 0.55 0.25
IVPP V 16066
27.5 7.2 54.0 44.5 0.51 0.62 0.26
IVPP V 10932A
25.7 4.6 48.7 36.3 0.53 0.71 0.18
IVPP V 13172
28.9 3.6 53.1 45.8 0.54 0.63 0.12
IVPP V 11640
32.6 8.4 64.5 0.51 0.26
IVPP V 13156
30.5 4.1 60.4 47.1 0.50 0.65 0.13
IVPP V 13171
30.7 6.3 62.3 55.5 0.49 0.55 0.21
IVPP V 12352
14.8 5.9 53.3 30.5 0.47 0.49 0.24
Protopterygidae Protopteryx
IVPP V 11665
11.2 1.8 28.0 23.7 0.40 0.47 0.16
Pengornithidae Pengornis
IVPP V 15336
19.3 5.3 63.5 48.1 0.30 0.40 0.27
IVPP V 18632
11.7 3.2 45.8 34.6 0.26 0.34 0.27
IVPP V 18687
9.4 5.0 51.7 39.8 0.18 0.24 0.53
Bohaiornithidae Longusunguis
IVPP V 17964
22.7 3.5 41.8 35.8 0.54 0.63 0.15
IVPP V 18691
17.8 3.0 43.9 36.0 0.41 0.49 0.17
IVPP V 18690
21.9 2.8 47.3 37.5 0.46 0.58
IVPP V 17963
20.2 4.4 51.7 42.6 0.39 0.47 0.22
17.3 50.6 44.5 0.34 0.39
BMNH-Ph 00805
19.6 46.5 41.3 0.42 0.47
Longipterygidae Longipteryx
IVPP V 12325
22.5 4.7 44.6 28.1 0.50 0.80 0.21
IVPP V 12552
26.3 4.4 40.1 0.66 0.17
IVPP V 11309
13.4 2.6 2.2 23.1 19.5 0.58 0.69 0.19 0.16
DNHM 1878
12.4 22.4 17.6 0.55 0.70
Hongshanornithidae Hongshanornis
IVPP V 14533
5.2 3.1 25.6 24.9 0.20 0.21 0.60
Songlingornithidae Yixianornis
IVPP V 12631
11.1 4.1 49.4 41.0 0.22 0.27 0.37
IVPP V 13278
10.0 64.8 0.15
IVPP V 17078
14.8 6.0 80.6 56.0 0.18 0.26 0.41
(undetermined) Iteravis
IVPP V 18958
6.8 3.1 45.7 35.2 0.15 0.19 0.46
IVPP V 16861
15.1 5.0 56.8 43.6 0.27 0.35 0.33?
Citations: Sapeornis CAGS-03-07-08 (Yuan, 2005), Zhouornis (Zhang et al., 2013), Sulcavis (O’Connor
et al., 2013), Shanweiniao (O’Connor et al., 2009).
(0.12–0.24) and enantiornithines (0.15–0.53) are indistinguishable. The single datum point
representing the ornithuromorphs, taken from the holotype of Schizooura lii, is not reliable
due to deformation. However, we hypothesize that wide spread lateral preservation in this
clade precisely reflects their very low width values. The presence of robust ventrolateral
and/or dorsolateral processes may contribute to the width in confuciusornithiforms and
enantiornithines (Figs. 4, 5). Higher MLH/TLP values are interpreted from sapeornithiforms
(0.45) and ornithuromorphs (0.37–0.60), indicating proportionately taller and shorter
pygostyles compared to confuciusornithiforms (MLH/TLP value is 0.21–0.26) and
enantiornithines (MLH/TLP value is 0.16–0.22) (Table 1; Fig. 7D). This greater dorsoventral
depth is produced by the prominent neural spines in sapeornithiforms and the dominant
pygostyle lamina in ornithuromorphs, which both are dorsally projecting (Figs. 3, 6).
Four groups (TLP-TLH, TLP-TLF, TLP-MLW and TLP-MLH) of visibly separated
Advanced online publication
clusters of data represent the pygostyle measurements of four investigated clades. Despite
of the ratio values discussed above, cluster distributions of four clades can be more or less
separated as their borders with slight overlaps show (Fig. 7). Therefore, four avian clades can
be, to some extent, differed according to the absolute sizes.
4 Discussion
4.1 Convergent evolution of the “pygostyle”
Fusion or partial fusion of the terminal caudal vertebrae in maniraptorans is observed
in the Therizinosauroidea, Oviraptorosauria and potentially also the Scansoriopterygidae (Xu
et al., 2003; Persons and Currie, 2011a, b; Persons et al., 2013; Gao et al., 2008). However,
morphological differences between these phylogenetically separated taxa indicate these
co-ossified structures cannot be considered equivalent to the avian pygostyle. Outside the
Ornithuromorpha, no group preserves evidence of a tail complex. The rod-like fused caudals
in the primitive Beipiaosaurus completely lack rectrices. Although pennaceous tail feathers
are present in oviraptorosaurs, the absence of a pygostyle lamina or a similar dorsal ridge
makes the development of rectricial bulbs or equivalent structures in this clade unlikely.
Co-ossification, when present, only occurs between the centra and prezygopophyses;
other processes, such as the transverse processes inferred to fuse into the pygostyle base
in ornithuromorphs, are absent in the distal caudal vertebrae of oviraptorosaurs. In the
Oviraptorosauria, the presence of extremely deep hemal arches (chevrons) suggests a massive
caudofemoralis muscle (Persons et al., 2013). Hemal arches are remarkably shortened in
almost all primitive pygostylians and even lost in most living birds (Baumel and Witmer,
Anatomical differences between the tightly associated caudal vertebrae in oviraptorosaurs
and the pygostyle in birds suggest the decrease and fusion of caudal vertebrae in maniraptorans
occurred through different genetic pathways, indicating fusion of the distal most caudals
evolved independently in different maniraptoran lineages (Rashid et al., 2014). To avoid
confusion (Lü and Hou, 2005), fusion in the distalmost caudals of non-avian theropods should
be described as pygostyle-like and a true pygostyle should be regarded as a synapomorphy of
the Pygostylia (Aves).
4.2 Evolution of the tail complex
In the last major study of the evolutionary transition of the avian tail, researchers
considered the pygostyle in the basal bird Confuciusornis to be roughly equivalent to that
of modern birds (Gatesy and Dial, 1996a, b). However, in light of the numerous specimens
uncovered over the past two decades, representing a diverse aviary and a wealth of
morphological data, signicant differences are now clearly observable between the pygostyle
and associated tail feathers of modern birds and that of basal clades (Sapeornithiformes,
Confuciusornithiformes, Enantiornithes and Ornithuromorpha).
Advanced online publication
The sapeornithiform pygostyle is relatively smaller and poorly co-ossified compared
to most other Early Cretaceous pygostylians. Although forming a single element, individual
processes can still be observed in the cranial half of the pygostyle in most specimens; the
adult condition is exemplied in the adult or near adult specimens IVPP V 13275 and V 13276
(Fig. 3) (Zhou and Zhang, 2003b). As a result of the increased caudal fusion, the pygostyle in
confuciusornithiforms have comparatively smoother margins. Compared to sapeornithiforms,
the pygostyle is proportionately longer in confuciusornithiforms and enantiornithines (only
exceptions of the Pengornithidae discussed later). Notably, the bilateral dorsolateral ridges
present in confuciusornithiforms and enantiornithines define an excavated dorsal platform
whereas in ornithuromorphs (including extant birds) the neural spines are fused forming the
blade-like pygostyle lamina so that the surface area of the dorsal margin is negligible. In the
blade-like ornithuromorph pygostyle, the centra fuse forming the narrow pygostyle base,
which is obscured in other pygostylians by the presence of lateral ridges and ventrolateral
processes. The transverse processes are highly reduced in primitive ornithuromorphs, forming
only a series of uneven low ridges on the lateral surface (Fig. 6). However, in some taxa of
living birds (like pheasant and raptors) the transverse processes are inferred to secondarily
form the ventral process, and expanded in wood peckers to form the pygostyle disc (Baumel,
1988) (Fig. 8). These specialized transverse processes are restricted to the cranioventral end
of the pygostyle where present in living birds, differing strongly from the fused transverse
processes observed forming a ridge on the lateral surfaces of the sapeornithiform pygostyle or
the unfused processes on the lateral surface in ornithuromorphs (Fig. 3).
Differences in pygostyle morphology between these four groups co
rrespond to consistent
differences in tail plumage. Sapeornis has been interpreted as having a strongly graded fan
consisting of approximately eight pennaceous feathers (Zheng et al., 2013; Wang X et al.,
2014; Xu et al., 2014), but this taxon has not been studied directly with regards to its rectricial
morphology and only one published specimen preserves this feature in lateral view. Additional
specimens preserved in dorsal or ventral view confirm that a fan-shaped array of rectrices
is present, but they also suggest that the rectrices did not overlap and thus would not have
formed an aerodynamic surface (O’Connor et al., 2016). However, until these specimens can
be further studied tail function remains inconclusive in sapeornithiforms. Tail feathers are not
present in all confuciusornithiform specimens, some of which preserve only rachis-less body
feathers like those found on the neck and other parts of the body, which are also present in
sapeornithiform, enantiornithine and ornithuromorph fossils also preserving rectrices. Other
confuciusornithiform specimens additionally preserve a pair of elongate rachis-dominated
rectrices interpreted as male-specific ornaments (Chiappe et al., 2008). Consistent with
similarities in pygostyle morphology, similar ornamental rachis-dominated rectrices are also
present in some enantiornithines (e.g., Protopteryx, Paraprotopteryx, Dapingfangornis) (Zhang
and Zhou, 2000; Zheng et al., 2007; Li et al., 2006). The only exception is the pengornithid
Chiappeavis (STM 29-11), which has recently been described with a “tail fan” consisting of
Advanced online publication
approximately eight overlapping rectrices (O’Connor et al., 2016). This fan is proportionately
shorter than that observed in Jehol ornithuromorphs. Rectricial fans in early ornithuromorphs
apparently consisted of eight to more than ten overlapping rectrices, comparable to living birds.
Differences in the pygostyle between clades of Early Cretaceous birds, supported by
differences in rectricial morphology, suggest that the complete modern tail complex only
evolved in the ornithuromorph lineage. In neornithines the rectricial bulbs rest on sockets on
either side of the pygostyle, separated medially by the pygostyle lamina (Fig. 8). These bulbs
root the calami of the rectrices forming the tail fan, allowing their position to be precisely
controlled by the rectricial bulb muscle (Baumel, 1988; Gatesy and Dial, 1996b). The nearly
modern morphology of the pygostyle in all Early Cretaceous ornithuromorphs and the common
preservation of a rectricial fan strongly suggest the complete tail complex evolved very early
in this lineage. A few specimens even preserve carbonized imprints surrounding the pygostyle
interpreted as soft tissue traces of the rectricial bulbs (Zhou et al., 2014), supporting inferences
this feature was present. The presence of a fan-shaped array of rectrices and a possible low
pygostyle lamina in sapeornithiforms might suggest the presence of rudimentary rectricial bulbs
in this lineage as well. However, the proportionately greater length (and width) of the pygostyle
indicates that the rectricial bulbs, if present, would have differed from that in ornithuromorphs.
The tail complex can be considered absent in the confuciusornithiforms based on the absence of
a fan-shaped tail in all the hundreds of known specimens; this is supported through differences
in pygostyle morphology, notably the absence of a pygostyle lamina.
A pygostyle lamina is also totally absent in enantiornithines. We consider short body
feathers (apparently rachis-less) were present on the tail in all enantiornithine groups regardless
of rectricial morphology. A majority of enantiornithine specimens preserve a pair of rachis-
Fig. 8 Illustration of extant avian pygostyles in both dorsal view (A–D) and left-lateral view (a–d)
A, a. Cynus cynus, IVPP OV 1721; B, b. Lophura nycthemera, IVPP OV 1863;
C, c. Streptopelia orientalis, IVPP OV 1883; D, d. Buteo buteo, IVPP OV 1888
Light blue: pygostyle lamina, its lateral surfaces are the major attachments for rectricial bulbs and rectricial
bulb muscle, anteriorly up-lateral and dorsal surfaces are attachments for caudal levator muscle;
purple: pygostyle base, its anteriorly low-lateral and ventral surfaces are attachments for caudal depressor
muscle; peach: ventral expansion of pygostyle base for well-developed caudal depressor muscle;
light yellow: dorsal expansion of pygostyle lamina for well-developed caudal levator muscle. Scale bar=10 mm
Advanced online publication
dominated ornamental long rectrices, resembling those in confuciusornithiforms. The “fan
tail” from which longiptergyid Shanweiniao gets its name (O’Connor et al., 2009) has been
reinterpreted as non-aerodynamic, consisting of at least two pairs of rachis-dominated feathers,
similar to the tail in Paraprotopteryx (O’Connor et al., 2016). This argument is partially based
on pygostyle morphology, which agrees with the detailed observations from this study. The tail
fan in pengornithid Chiappeavis (STM 29-11) greatly differs from the rectricial morphology in
other pengornithids, all other members of which possess a pair of rachis-dominated streamer
rectrices (Parapengornis, Eopengornis). However, the pygostyle of Chiappeavis is typical of
pengornithids, although it is proportionately shorter and broader than in Parapengornis. Given
the observed coevolution between pygostyle and tail feathers, similar pygostyle morphologies
are expected to share similar rectrices both within Cretaceous primitive birds and living taxa
(Clarke et al., 2006; Felice and O’Connor, 2014; Felice, 2014), making the two apparently
functionally disparate tail morphologies observed in the Pengornithidae difcult to interpret.
Even if Chiappeavis possessed some structural equivalent to rectricial bulbs, their morphology
would be expected to differ greatly from that of modern birds given the different shape of
the pygostyle in pengornithids. As a result, there may also exist differences in tail function
and ability between the two groups. Disparity in the tail plumage of Chiappeavis compared
to other pengornithids, and differences in pygostyle morphology between Chiappeavis
and ornithuromorphs suggests that if a tail complex was present in Chiappeavis, it evolved
independently in parallel to ornithuromorphs. We suggest that this tail fan even may have been
capable of generating lift but did not have the muscular control of the fan shape associated
with the ornithuromorph tail complex. The tail complex consisting of a blade-like pygostyle,
rectricial bulbs, and aerodynamic rectrices appears to be unique to the Ornithuromorpha as
originally hypothesized by Clarke et al. (2006).
4.3 Coevolution of tail feathers and the pygostyle
Modern birds exhibit a huge diversity of tail feather morphologies. To a lesser extent the
shape of the pygostyle also shows quite a bit of variation. Bird tails can have both signicant
motor and/or display functions. The diversity of tail morphologies in extant birds are
hypothesized to be a product of natural and/or sexual selection (Balmfold et al., 1993; Thomas,
1997; Hedenström, 2002). Recent work has documented the coevolution of the tail feathers
with the pygostyle in living birds (Felice and O’Connor, 2014; Felice, 2014), indicating that
rectrices with specialized functions require structurally modied pygostyles.
In addition to the rectricial bulbs and their muscle, the caudal depressor and caudal
levator (m. levator caudae) muscles also form connections with the pygostyle (Baumel, 1988).
Fleshy part of the pars profunda of the caudal depressor muscle adhere to the ventrolateral
portion of the pygostyle base and the tendon of this muscle attaches along the anterior-ventral
margin of the pygostyle. The tendon of the distal part of the caudal levator muscle (pars
distalis) attaches on the craniodorsal portion of the lateral surface of the pygostyle (Baumel,
Advanced online publication
1988). In certain extant raptor clades, the birds descend on their prey from great height at
great speeds. Such taxa must be capable of generating large braking forces after seizing prey
near the ground. This maneuvering requires a robust caudal depressor muscle. In order to
provide an expanded area for the attachment of this expanded muscle, falconids possess an
expanded tubercle, or in some cases paired accessory sesamoid akelets (Richardson, 1972),
on or near the cranioventral margin of the pygostyle probably (Fig. 8D). Even greater ventral
expansions, in the form of the pygostyle disc (Baumel and Witmer, 1993), are developed in
woodpeckers, and some piciforms. Woodpeckers have evolved stiffened feathers in order
to support their bodies against gravity as an adaptation for vertical climbing (Manegold and
Töpfer, 2013). This posture also requires a strong caudal depressor muscle as the expanded
pygostyle disc suggests. The enantiornithine Parapengornis has been described possessing a
pygostyle that is homomorphic to that of woodpeckers, indicating a similar vertical climbing
behavior in some Mesozoic enantiornithines (Hu et al., 2015). However, in the holotype
of Parapengornis eurycaudatus (IVPP V 18687) the pygostyle is preserved in dorsal view,
preventing observation of any expanded ventral surface (also preserved in dorsal view in the
referred specimen IVPP V 18632). Instead, V 18687 displays a broad, concave dorsal platform
dened by lateral ridges generally similar to those of other pengornithids. The dorsal expansion
present in the pygostyle of Parapengornis and other pengornithids (Hu et al., 2015; Wang X et
al., 2014) is in fact opposite the condition observed in the woodpecker pygostyle, in which it
is the ventral surface that is expanded. The dorsal platform present in pengornithids and other
enantiornithines is more consistent with the presence of an expanded caudal levator muscle,
which attaches dorsally in living birds, rather than an expanded caudal depressor muscle as
in woodpeckers and falconids (Fig. 9). Furthermore, in woodpeckers the specialized stiffened
rectrices of the tail are not especially elongated and are additionally supported by a layer
of tectrices. The tail feathers are proportionately longer in Parapengornis and pennaceous
tectrices appear to be absent. Together, it seems unlikely that these rachis-dominated rectrices
could serve as a prop.
In extant phasianids with ornamental tails, the craniodorsal margin of the pygostyle is
laterally expanded to provide additional surface for the attachment of the enlarged caudal
levator muscle, which acts to elevate the elongated ornamental deck feathers (Gatesy and
Dial, 1993). This craniodorsal expansion is obviously more strongly developed in males,
though it presents in both genders (Fig. 8B). This dorsal platform is reminiscent of that in
confuciusornithiforms and enantiornithines (Figs. 4, 5). The pygostyle of the male peafowl
(Pavo) represents an extreme case: to lift its huge tail (coverts and rectrices), the narrow
body of the pygostyle separates enlarged dorsal and ventral surfaces abnormally expanded
for muscular attachments (“”-shaped in cranial view) (Fig. 9). Male birds living in
wooded environments, like that inferred for the Jehol (Zhou, 2006), often have extravagant
ornaments; the high degree of clutter and reduced visibility in wooded environments allows
the evolution of such features under sexual selection (Endler and Thery, 1996). Paired elongate
Advanced online publication
Fig. 9 Reconstruction of caudal fusion/pygostyle from maniraptorans to birds
Line drawing near the group name shows left lateral view of distalest caudal vertebrae of each clade. Line
drawing in the box shows proximal/cranial view of pygostyle and soft tissue attachments: bone in white, articular
surface is gray dot, caudal levator muscle in red, caudal depressor muscle in blue, rectricial bulb in yellow
A. Sapeornithiformes; B. Confuciusornithiformes; C. Pengornis; D. Parabohaiornis; E. Rapaxavis;
F. Yixianornis; G. Piscivoravis; H. Iteravis; I. Gansus; J. Cynus cynus; K. Lophura nycthemera;
L. Pavo cristatus; M. Streptopelia orientalis; N. Buteo buteo; O. Dendrocopos major; P. Passer montanus. (not scaled)
Advanced online publication
tail feathers are common in enantiornithines and have typically been regarded as ornamental
and sexually dimorphic, as in Confuciusornis. The tail morphology in Parapengornis and
other enantiornithine species is reminiscent of those found in male individuals of many
extant arboreal birds, such as paradise-flycatchers (Terpsiphone). One Early Cretaceous
enantiornithine, the holotype of Feitianius paradisi, preserves a robust pygostyle of the typical
enantiornithine morphology and a tail morphology consisting of at least three distinct feather
morphologies that strongly resembles tail displays present in living sexually dimorphic birds
(O’Connor et al., 2016). This level of extravagance is only observed in extant polygamous
birds. This study suggests that in addition to possessing such sexually specic ornaments (Fig.
10), some Cretaceous birds also engaged in display behaviors as common in extant polygamous
birds, including the birds of paradise. Soft tissue surrounding the pygostyle in the holotype of
Feitianius paradisi was previously identified as including rectricial bulbs (O’Connor et al.,
2016), but we consider that this may not be correct – instead this may include expanded levator
musculature, as hypothesized here (Fig. 9). We conclude that the dorsal platform present in
the confuciusornithiforms and enantiornithine pygostyle may have accommodated a fortied
caudal levator muscle, like that in phasianids. This strongly suggests these basal birds raised
the paired long rachis-dominated deck feathers in some form of display, although whether to
attract females or fend of competitors is unknown.
Fig. 10 Reconstruction of Feinianius paradise by Michael Rothman showing the tail relaxed and with the
levator musculature exed, engaged in a display
A. Perched and non-displaying; B. Perched and displaying
Acknowledgement We thank ZHOU Zhonghe, Ni Xijun, LI Zhiheng and WANG Min for
discussion, ZHENG Xiaoting for accession to collections in STM, ZHOU Shuang and HU Han
for providing photos. We thank M. Rothman for use of his reconstruction of Feitianius. This
research was supported by the National Natural Science Foundation of China (91514302) and
the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB18030501).
Advanced online publication
王 维1,2 邹晶梅1
(1 中国科学院古脊椎动物与古人类研究所,中国科学院脊椎动物演化与人类起源重点实验室 北京 100044)
(2 中国科学院大学 北京 100049)
中图法分类号Q 915.865 文献标识码A 文章编号1000−3118(2017)03−0001−26
Balmford A, Thomas A L R, Jones I L, 1993. Aerodynamics and the evolution of long tails in birds. Nature, 361: 628–
Barsbold R, Currie P J, Myhrvold N P et al., 2000a. A pygostyle from a non-avian theropod. Nature, 403: 155–156
Barsbold R, Osmólska H, Watabe M et al., 2000b. A new oviraptorosaur (Dinosauria, Theropoda) from Mongolia: the rst
dinosaur with a pygostyle. Acta Palaeont Pol, 45(2): 97–106
Baumel J J, 1988. Functional morphology of the tail apparatus of the pigeon (Columba livia). Adv Anat Embryol Cell Biol,
110: 1
Baumel J J, Witmer L M, 1993. Osteologia. In: Baumel J J, King A S, Breazile J E et al. eds. Handbook of Avian Anatomy:
Nomina Anatomica Avium. 2nd ed. Cambridge: Nuttall Ornithological Club. 1–401
Baumel J J, Wilson J A, Bergren D R, 1990. The ventilatory movements of the avian pelvis and tail: function of the muscles
of the tail region of the pigeon (Columba livia). J Exp Biol, 151(1): 263–277
Brusatte S L, Lloyd G T, Wang S C et al., 2014. Gradual assembly of avian body plan culminated in rapid rates of evolution
across the dinosaur-bird transition. Curr Biol, 24(20): 2386–2392
Chiappe L M, 2002. Basal bird phylogeny: problems and solutions. In: Chiappe L M, Witmer L eds. Mesozoic Birds: Above
Advanced online publication
the Heads of Dinosaurs. Berkeley: University of California Press. 448–472
Chiappe L M, Ji S A, Ji Q et al., 1999. Anatomy and systematics of the Confuciusornithidae (Theropoda, Aves) from the
Late Mesozoic of northeastern China. Bull Am Mus Nat Hist, 242: 1–89
Chiappe L M, Lamb J P, Ericson P G P, 2002. New enantiornithine bird from the marine Upper Cretaceous of Alabama. J
Ver Paleont, 22(1): 170–174.
Chiappe L M, Marugán-Lobón J, Ji S A, 2008. Life history of a basal bird: morphometrics of the Early Cretaceous
Confuciusornis. Biol Lett, 4(6): 719–723
Chiappe L M, Zhao B, O’Connor J K et al., 2014. A new specimen of the Early Cretaceous bird Hongshanornis longicresta:
insights into the aerodynamics and diet of a basal ornithuromorph. PeerJ, 2: e234
Clarke J A, Zhou Z H, Zhang F C, 2006. Insight into the evolution of avian ight from a new clade of Early Cretaceous
ornithurines from China and the morphology of Yixianornis grabaui. J Anat, 208(3): 287–308
Dyke G J, Norell M A. 2005. Caudipteryx as a non-avialan theropod rather than a ightless bird. Acta Palaeont Pol, 50(1):
Elzanowski A. 2002. Archaeopterygidae (Upper Jurassic of Germany). In: Chiappe L M, Witmer L eds. Mesozoic Birds:
Above the Heads of Dinosaurs. Berkeley: University of California Press. 129–159
Endler J A, Thery M, 1996. Interacting effects of lek placement, display behavior, ambient light, and color patterns in three
neotropical forest-dwelling birds. Am Nat, 148(3): 421–452
Feduccia A, Tordoff H B, 1979. Feathers of Archaeopteryx - asymmetric vanes indicate aerodynamic function. Science, 203:
Felice R N, 2014. Coevolution of caudal skeleton and tail feathers in birds. J Morphol, 275(12): 1431–1440
Felice R N, O’Connor P M, 2014. Ecology and caudal skeletal morphology in birds: the convergent evolution of pygostyle
shape in underwater foraging taxa. PloS ONE, 9(2): e89737
Fisher H I, 1959. Some functions of the rectrices and their coverts in the landing of pigeons. Wilson Bull, 71(3): 267–
Foth C, Tischlinger H, Rauhut O W, 2014. New specimen of Archaeopteryx provides insights into the evolution of
pennaceous feathers. Nature, 511: 79–82
Gatesy S M, Dial K P, 1993. Tail muscle activity patterns in walking and ying pigeons (Columba livia). J Exp Biol, 176(1):
Gatesy S M, Dial K P, 1996a. Locomotor modules and the evolution of avian ight. Evolution, 50: 331–340
Gatesy S M, Dial K P, 1996b. From frond to fan: Archaeopteryx and the evolution of short-tailed birds. Evolution, 50: 2037–
Gao C L, Chiappe L M, Meng Q J et al., 2008. A new basal lineage of Early Cretaceous birds from China and its
implications on the evolution of the avian tail. Palaeontology, 51(4): 775–791
Gao C L, Chiappe L M, Zhang F J, et al., 2012. A subadult specimen of the Early Cretaceous bird Sapeornis chaoyangensis
and a taxonomic reassessment of sapeornithids. J Vert Paleont, 32(5): 1103–1112.
Haeckel E, 1883. The History of Creation. London: Kegan Paul. 1–402
He T, Wang X L, Zhou Z H, 2008. A new genus and species of caudipterid dinosaur from the Lower Cretaceous Jiufotang
Formation of western Liaoning, China. Vert PalAsiat, 46: 178–189
Hedenström A, 2002. Aerodynamics, evolution and ecology of avian ight. Trends Ecol Evol, 17(9): 415–422
Advanced online publication
Heers A M, Dial K P, Tobalske B W, 2014. From baby birds to feathered dinosaurs: incipient wings and the evolution of
ight. Paleobiology, 40(3): 459–476
Hone D W E, 2012. Variation in the tail length of non-avian dinosaurs. J Vert Paleont, 32(5): 1082–1089
Hu D, Xu X, Hou L et al., 2011. A new Enantiornithine bird from the Lower Cretaceous of Western Liaoning, China. J Vert
Paleont, 31(1):154–161
Hu H, O’Connor J K, Zhou Z H, 2015. A new species of Pengornithidae (Aves: Enantiornithes) from the Lower
Cretaceous of China suggests a specialized scansorial habitat previously unknown in Early birds. PloS ONE, 10(6):
Jarvis E D, Mirarab S, Aberer A J et al, 2014. Whole-genome analyses resolve early branches in the tree of life of modern
birds. Science, 346: 1320–1331
Jetz W, Thomas G H, Joy J B et al., 2012. The global diversity of birds in space and time. Nature, 491: 444–448
Ji Q, Currie P J, Norell M A et al., 1998. Two feathered dinosaurs from northeastern China. Nature, 393: 753–761
Li L, Duan Y, Hu D Y et al., 2006. New eoenantiornithid bird from the Early Cretaceous Jiufotang Formation of western
Liaoning, China. Acta Geol Sin, 80(1): 38–41
Liu D, Zhou Z H, Zhang Y G, 2012. Mass estimate and evolutionary trend in Chinese Mesozoic fossil birds. Vert PalAsiat,
50: 39–52
Lü J C, Hou L H, 2005. A possible long-tailed bird with a pygostyle from the Late Mesozoic Yixian Formation, western
Liaoning, China. Acta Geol Sin, 79(1): 7–10
Manegold A, Töpfer T, 2013. The systematic position of Hemicircus and the stepwise evolution of adaptations for drilling,
tapping and climbing up in true woodpeckers (Picinae, Picidae). J Zool Syst Evol Res, 51(1): 72–82
Mayr G, Pohl B, Hartman S et al., 2007. The tenth skeletal specimen of Archaeopteryx. Zool J Linn Soc, 149(1): 97–
Martin L D, Zhou Z H, Hou L H et al., 1998. Confuciusornis sanctus compared to Archaeopteryx lithographica.
Naturwissenschaften, 85(6): 286–289
O’Connor J K, Zhou Z H, 2013. A redescription of Chaoyangia beishanensis (Aves) and a comprehensive phylogeny of
Mesozoic birds. J Syst Palaeont, 11(7): 889–906
O’Connor J K, Sullivan C, 2014. Reinterpretation of the Early Cretaceous maniraptoran (Dinosauria: Theropoda) Zhongornis
haoae as a scansoriopterygid-like non-avian, and morphological resemblances between scansoriopterygids and basal
oviraptorosaurs. Vert PalAsiat, 52: 3–30
O’Connor J K, Wang X R, Chiappe L M et al., 2009. Phylogenetic support for a specialized clade of Cretaceous
enantiornithine birds with information from a new species. J Vert Paleont, 29(1): 188–204
O’Connor J K, Chiappe L M, Gao C et al., 2011. Anatomy of the Early Cretaceous enantiornithine bird Rapaxavis pani. Acta
Palaeontol Pol, 56(3): 463–475
O’Connor J K, Sun C, Xu X et al., 2012. A new species of Jeholornis with complete caudal integument. Hist Biol, 24(1):
O’Connor J K, Wang X L, Sullivan C et al, 2013. Unique caudal plumage of Jeholornis and complex tail evolution in early
birds. Proc Natl Acad Sci USA, 110: 17404–17408
O’Connor J K, Li D Q, Lamanna M C et al., 2015. A new Early Cretaceous enantiornithine (Aves, Ornithothoraces) from
northwestern China with elaborate tail ornamentation. J Vert Paleont, 36(1): e1054035
Advanced online publication
O’Connor J K, Wang X L, Zheng X T et al., 2016. An Enantiornithine with a fan-shaped tail, and the evolution of the
rectricial complex in early birds. Curr Biol, 26(1): 1–6
Pan Y, Sha J, Zhou Z et al., 2013. The Jehol Biota: definition and distribution of exceptionally preserved relicts of a
continental Early Cretaceous ecosystem. Cretaceous Res, 44: 30–38
Pennycuick C J, 1968. Power requirements for horizontal ight in the pigeon Columba livia. J Exp Biol, 49: 527–555
Persons W S, Currie P J, 2011a. The tail of Tyrannosaurus: reassessing the size and locomotive importance non-avian
theropods. Anat Rec, 294(1): 119–131
Persons W S, Currie P J, 2011b. Dinosaur speed demon: the caudal musculature of Carnotaurus sastrei and implications for
the evolution of South American abelisaurids. PloS ONE, 6(10): e25763
Persons W S, Currie P J, Norell M A, 2014. Oviraptorosaur tail forms and functions. Acta Palaeont Pol, 59(3): 553–
Pittman M, Gatesy S M, Upchurch P et al., 2013. Shake a tail feather: the evolution of the theropod tail into a stiff
aerodynamic surface. PloS ONE, 8(5): e63115
Pu H Y, Chang H L, Lü J C et al., 2013. A new juvenile specimen of Sapeornis (Pygostylia: Aves) from the Lower
Cretaceous of Northeast China and allometric scaling of this basal bird. Paleontol Res, 17(1): 27–38.
Rashid D J, Chapman S C, Larsson H C E et al., 2014. From dinosaurs to birds: a tail of evolution. Evo Devo, 5(1): 25
Richardson F, 1972. Accessory pygostyle bones of falconidae. Condor, 74(3): 350–351
Spedding G R, Rayner J M V, Pennycuick C J, 1984. Momentum and energy in the wake of a pigeon (Columba livia) in
slow ight. J Exp Biol, 111(1): 81–102
Thomas A L R, 1993. On the aerodynamics of birds’ tails. Phil Trans B Bio, 340(1294): 361–380
Thomas A L R, 1997. On the tails of birds. Bioscience, 4: 215–225
Thomas A L R, Balmford A, 1995. How natural selection shapes birds’ tails. Am Nat, 146: 848–868
Tucker V A, 1992. Pitching equilibrium, wing span and tail span in a gliding Harris’ Hawk, Parabuteo unicinctus. J Exp
Biol, 165(1): 21–41
Wang M, Zhou Z H, O’Connor J K et al., 2014. A new diverse enantiornithine family (Bohaiornithidae fam. nov.) from the
Lower Cretaceous of China with information from two new species. Vert PalAsiat, 52: 31–76
Wang M, Zheng X T, O’Connor J K et al., 2015. The oldest record of ornithuromorpha from the early cretaceous of China.
Nat Commun, 6: 1–9
Wang X L, O’Connor J K, Zheng X T et al., 2014. Insights into the evolution of rachis dominated tail feathers from a new
basal enantiornithine (Aves: Ornithothoraces). Bio J Linn Soc, 113(3): 805–819
Xu X, Chen Y N, Wang X L et al., 2003. Pygostyle-like Structure from Beipiaosaurus (Theropoda, Therizinosauroidea) from
the Lower Cretaceous Yixian Formation of Liaoning, China. Acta Geol Sin, 77(3): 294–298
Xu X, Zhou Z H, Dudley R et al., 2014. An integrative approach to understanding bird origins. Science, 346: 1253293
You H L, Lamanna M C, Harris J D et al., 2006. A nearly modern amphibious bird from the Early Cretaceous of
northwestern China. Science, 312: 1640–1643
Yuan C X, 2005. Restudy on Sapeornis from the Lower Cretaceous of Yixian County, Liaoning. Ph. D thesis. Beijing:
Paleotology and Stratigraphy, China University of Geosciences. 1–165
Zelditch M L, Swiderski D L, Sheets H D, 2012. Geometric Morphometrics for Biologists: a Primer. San Diego: Elsevier
Academic Press. 1–436
Advanced online publication
Zhang F C, Zhou Z H, 2000. A primitive enantiornithine bird and the origin of feathers. Science, 290: 1955–1959
Zhang F C, Ericson P G, Zhou Z H, 2004. Description of a new enantiornithine bird from the Early Cretaceous of Hebei,
northern China Can J Earth Sci. 41: 1097–1107
Zhang F C, Zhou Z H, Benton M J, 2008a. A primitive confuciusornithid bird from China and its implications for early avian
ight. Sci China Ser Sci, 5(5): 625–639
Zhang F C, Zhou Z H, Xu X et al., 2008b. A bizarre Jurassic maniraptoran from China with elongate ribbon-like feathers.
Nature, 455: 1105–1108
Zhang Z, Chiappe L M, Han G et al., 2013. A large bird from the Early Cretaceous of China: new information on the skull of
enantiornithines. J Vert Paleont, 33(5): 1176–1189
Zheng X T, Zhang Z H, Hou L H, 2007. A new enantiornitine bird with four long rectrices from the Early Cretaceous of
northern Hebei, China. Acta Geol Sin, 81(5): 703–708
Zheng X T, Zhou Z H, Wang X L et al., 2013. Hind wings in basal birds and the evolution of leg feathers. Science, 339:
Zheng X T, O’Connor J K, Huchzermeyer F et al., 2014. New specimens of Yanornis indicate a piscivorous diet and modern
alimentary canal. Plos One, 9(4): e95036
Zhou S, O’Connor J K, Zhou Z H, 2012. A new basal beaked ornithurine bird from the lower cretaceous of western
Liaoning, China. Vert PalAsiat, 50: 9–24
Zhou S, Zhou Z H, O’Connor J K, 2013. A new piscivorous ornithuromorph from the Jehol Biota. Hist Biol, 26(5): 608–
Zhou S, O’Connor J K, Wang M, 2014. A new species from an ornithuromorph (Aves: Ornithothoraces) dominated locality
of the Jehol Biota. Chin Sci Bull, 59(36): 5366–5378
Zhou Z H, 2004. The origin and early evolution of birds: discoveries, disputes, and perspectives from fossil evidence.
Naturwissenschaften, 91(10): 455–471
Zhou Z H, 2006. Evolutional radiation of the Jehol Biota: chronological and ecological perspectives. Geol J, 41(34): 377–
Zhou Z H, 2014. The Jehol Biota, an Early Cretaceous terrestrial Lagerstätte: new discoveries and implications. Natl Sci
Rev, 1(4): 543–559
Zhou Z H, Zhang F C, 2001. Two new ornithurine birds from the Early Cretaceous of western Liaoning, China. Chin Sci
Bull, 46(15): 1258–1264
Zhou Z H, Zhang F C, 2002a. A long-tailed, seed-eating bird from the Early Cretaceous of China. Nature, 418: 405–
Zhou Z H, Zhang F C, 2002b. Largest bird from the Early Cretaceous and its implications for the earliest avian ecological
diversication. Naturwissenschaften, 89(1): 34–38
Zhou Z H, Zhang F C, 2003a. Jeholornis compared to Archaeopteryx, with a new understanding of the earliest avian
evolution. Naturwissenschaften, 90(5): 220–225
Zhou Z H, Zhang F C, 2003b. Anatomy of the primitive bird Sapeornis chaoyangensis from the Early Cretaceous of
Liaoning, China. Can J Earth Sci, 40(5): 731–747
Zhou Z H, Clarke J, Zhang F C, 2008. Insight into diversity, body size and morphological evolution from the largest Early
Cretaceous enantiornithine bird. J Anat, 212(5): 565–577
Advanced online publication
... Much fossil evidence suggests that the feathers on the hindlimbs of Deinonychus and early birds had aerodynamic functions, strongly supporting the idea that leg feathers were an important factor in the origin of bird flight [25,38,40]. Confuciusornithid birds do not have a fan-shaped tail, but their pygostyle exhibits dorsolateral ridges and proximoventral processes, which allow for more muscle attachments to aid in control [48][49]. This may imply that the elongated central tail feathers in confuciusornithids also had some aerodynamic function rather than simply resulting from sexual dimorphism. ...
Full-text available
Eastern Asia can be regarded as a dynamic environment during the Early Cretaceous in which the whole biota was complex and diverse, so the differentiation of early birds was also pronounced in this period. Confuciusornithids only uncovered from Eastern Asia, are an important early bird group for revealing the origin and early evolution of avian flight. However, results on confuciusornithid flight capabilities mainly come from qualitative analysis or speculation and analogism because of insufficient information from fossils, limited analogues in living birds, and no rigorous biomechanical testing. As an essential and popular tool in quantitative simulation and analysis, computational fluid dynamics (CFD) has major advantages over other techniques and has been well developed in recent years to solve general morphological-functional problems in palaeontolog-ical research. CFD is beneficial for studying the flight capability of confuciusornithids and will provide new quantitative evidence for the origin and evolution of early bird flight. In addition, we expect that interdisciplinary, exploratory, and tentative work will soon be applied in other similar studies, which will allow us to develop a quantitative method that can be applied effectively in functional-morphological analyses.
... The caudals are boxy with short transverse processes, small zygopophyses and large neural canals. The pygostyle is most complete in the counterslab where it is exposed in right dorsolateral view and shares with other ornithuromorphs the typical caudally tapered, ploughshare-shaped condition that also characterizes most neornithines [17]. ...
Full-text available
STM35-3 from the Yixian Formation is the only Early Cretaceous ornithuromorph preserving direct evidence of granivory. The crop contains numerous seeds and the preservation of gastroliths presumably within the ventriculus indicates this diet was paired with the presence of a gastric mill as in living granivorous birds. STM35-3 was originally referred to Hongshanornis longicresta, member of a diverse clade of small, basal ornithuromorphs with elongate hindlimbs known as the Hongshanornithidae. Hindlimb proportions suggest that hongshanornithids were wading birds, an ecological inference somewhat in conflict with direct evidence suggesting Hongshanornis fed on seeds. However, close inspection of STM35-3 reveals that the specimen represents a new species not closely related to hongshanornithids, distinguished by large forelimbs that exceed the length of the hindlimbs, robust and narrow coracoids, and a delicate edentulous rostrum. By contrast, all hongshanornithids have hindlimbs that far exceed the length of the forelimbs, coracoids with wide sternal margins, and small teeth throughout the upper and lower jaws. Reinterpretation of this new taxon, Eogranivora edentulata gen. et sp. nov, helps to clarify trophic driven patterns of tooth loss within the Ornithuromorpha. Apparent loss of the hallux may represent the first such occurrence in a Mesozoic bird and suggests a highly terrestrial lifestyle.
... On this basis it is possible to infer that in paravians the M. ilio-ischiocaudalis ran ventrally along the tail, probably within this ventral longitudinal trough (Fig. 11E). This location of the M. ilio-ischiocaudalis is coincident with the M. depressor caudae, which is ventrally located along the pygostyle among pygostylian birds ( Fig. 11G; Wang and O'Connor, 2017). The homology of these two tail muscles require a specific study, which is out the scope of the present contribution. ...
Tail anatomy of unenlagiids is nearly unknown. The aim of the present contribution is to describe in detail the tail of the unenlagiid Buitreraptor gonzalezorum, and to compare it with Archaeopteryx and other paravians. A detailed specimen overview yielded a large number of similarities between Archaeopteryx and Buitreraptor, including long postzygapophyses and lateral laminae and concavities at mid-vertebral centra. The differences in caudal vertebrae morphology along the tail indicate three differentiable functional regions. This contrasts with the presence of only two in other basal paravians, including dromaeosaurids, such as Deinonychus. This contribution sheds light on paravian tail evolution, and provides new data to on the changes in tail morphology that occurred along the theropod line to modern birds.
... The shape of the pygostyle in P. campestris suggests that this species may have had elaborate tail feathers. The bone is flexed dorsally at an approximately 45° angle (figure 9j) and the lamina have a large surface area for muscular attachment, rather like the pygostyle of extant silver pheasant Lophura nycthemera[57], which has a long trailing tail. By contrast, the pygostyle of Al. lathami, which is noted for its short, vertically fanned tail (the name Alectura = cock-tail[58]), is rather straight and lacks deep sulci on the laminae, and L. ocellata, which has a short tail with a downward posture, has a pygostyle that is laterally compressed and points slightly downwards. ...
Full-text available
Megapodes are unusual galliform birds that use passive heat sources to incubate their eggs. Evolutionary relationships of extant megapode taxa have become clearer with the advent of molecular analyses, but the systematics of large, extinct forms (Progura gallinacea, Progura naracoortensis) from the late Cenozoic of Australia has been a source of confusion. It was recently suggested that the two species of Progura were synonymous, and that this taxon dwarfed into the extant malleefowl Leipoa ocellata in the Late Pleistocene. Here, we review previously described fossils along with newly discovered material from several localities, and present a substantial taxonomic revision. We show that P. gallinacea and P. naracoortensis are generically distinct, describe two new species of megapode from the Thylacoleo Caves of south-central Australia, and a new genus from Curramulka Quarry in southern Australia. We also show that L. ocellata was contemporaneous with larger species. Our phylogenetic analysis places four extinct taxa in a derived clade with the extant Australo-Papuan brush-turkeys Talegalla fuscirostris, L. ocellata, Alectura lathami and Aepypodius bruijnii. Therefore, diversity of brush-turkeys halved during the Quaternary, matching extinction rates of scrubfowl in the Pacific. Unlike extant brush-turkeys, all the extinct taxa appear to have been burrow-nesters.
Full-text available
A bizarre coelurosaurian theropod Fukuivenator paradoxus is known only from the holotype specimen preserving majority of the skeleton from the Kitadani Dinosaur Quarry of the Lower Cretaceous Kitadani Formation, Tetori Group, Fukui, Japan. With aids of computed tomography techniques, a re-examination of the holotype specimen reveals additional features of Fukuivenator which was unobservable in the original description, such as the presence of parietals and a quadrate, and the fusion of the posteriormost caudal vertebrae. The thorough description in this study results in the emendation of diagnosis including the retraction of the large promaxillary fenestra subequal in size to maxillary fenestra, and the addition of the large maxillary fenestra expanded well dorsally above the suprantral strut. Expansion of morphological information elaborates the phylogenetic dataset, resulting in locating Fukuivenator as an unambiguous member of Maniraptora at the basalmost position of Therizinosauria. This phylogenetic position of Fukuivenator is supported by several therizinosaurian synapomorphies such as the subotic recess on the braincase, 11 cervical vertebrae some of which having two pneumatic foramina, and distal articular condyles on the anterior surface of the humerus. Among numerous diagnostic features, eight characters shared with some non-maniraptoran coelurosaurs and five shared with different clades within Maniraptora, highlighting the notably mosaic condition of Fukuivenator proposed in the original description. The combination of characters for herbivorous and carnivorous diets suggests the omnivory of Fukuivenator, projecting the dietary shift in the earliest evolutionary stage of Therizinosauria. Also, the large olfactory ratio revealed by the revised brain endocast highlights the unusually high olfactory acuity further developed than the plesiomorphic condition, implying that the acute sense of smell might be a characteristic of therizinosaurian theropods.
Full-text available
New and important pennaraptoran specimens continue to be discovered on a regular basis. Yet, with these discoveries the number of viable phylogenetic hypotheses has increased, including ones that challenge the traditional grouping of dromaeosaurids and troodontids within a monophyletic Deinonychosauria. This chapter will cover recent efforts to address prevailing phylogenetic uncertainties and controversies, both between and within key clades, including deinonychosaurian monophyly, the phylogenetic position of anchiornithines and scansoriopterygids, and the interrelationships of enantiornithines. While recent discoveries mainly from Asia have created much of the latest uncertainty and controversy, new material, particularly from Asia, promises to rather fittingly address these issues. Further curatorship of long-standing phylogenetic datasets and more prevalent use of extended analytical protocols will be essential to meeting this challenge, especially for groups whose boundaries have been blurred. As it becomes increasingly difficult to study all fossil materials, owing to their growing numbers and ever disparate locations, broader use of digital fossils and online character databases for character coding is acutely needed to ensure that errors arising from remote, rather than firsthand, scoring are reduced as far as possible, particularly at this time of rapid data accumulation.
Protopteryx fengningensis is from the 130.7 Ma Huajiying Formation making it one of the oldest known enantiornithines. Contributing to its significance, this taxon is also commonly resolved as the basal-most enantiornithine in phylogenetic analyses. Protopteryx preserves several unusual morphologies that are otherwise absent in the Enantiornithes but common in the Ornithuromorpha such as the procoracoid and lateral processes on the coracoid and proximally convex humeral head. Thus, the morphology of this taxon hints at the morphology of the ornithothoracine common ancestor. Here we supplement existing data with information from a new specimen as well as new morphological data from the holotype and paratype. The new specimen preserves gaps in the right wing suggestive of a sequential molt. The presence of two gaps suggests that, unlike neornithines, primaries and secondaries molted simultaneously. This represents an intermediate condition between Microraptor, in which several feathers are growing simultaneously and sequentially, and modern birds with sequential molts, in which a single feather is replaced at a time. A single patch of feathers was sampled revealing preserved eumelanosomes, indicating that at least part of the remiges was darkly colored.
Full-text available
We provide a complete description of the skeletal anatomy of the holotype of Chiappeavis magnapremaxillo, the first enantiornithine to preserve a rectricial fan, suggesting that possibly rectricial bulbs were present in basal members of this clade. Notably, Chiappeavis preserves a primitive palatal morphology in which the vomers reach the premaxillae similar to Archaeopteryx but unlike the condition in the Late Cretaceous enantiornithine Gobipteryx. If rectricial bulbs were present, pengornithid pygostyle morphology suggests they were minimally developed. We estimate the lift generated by the tail fan preserved in this specimen and compare it to the tail fans preserved in other Early Cretaceous birds. Aerodynamic models indicate the tail of Chiappeavis produced less lift than that of sympatric ornithuromorphs. This information provides a possible explanation for the absence of widespread aerodynamic tail morphologies in the Enantiornithes.
Full-text available
To better determine the history of modern birds, we performed a genome-scale phylogenetic analysis of 48 species representing all orders of Neoaves using phylogenomic methods created to handle genome-scale data. We recovered a highly resolved tree that confirms previously controversial sister or close relationships. We identified the first divergence in Neoaves, two groups we named Passerea and Columbea, representing independent lineages of diverse and convergently evolved land and water bird species. Among Passerea, we infer the common ancestor of core landbirds to have been an apex predator and confirm independent gains of vocal learning. Among Columbea, we identify pigeons and flamingoes as belonging to sister clades. Even with whole genomes, some of the earliest branches in Neoaves proved challenging to resolve, which was best explained by massive protein-coding sequence convergence and high levels of incomplete lineage sorting that occurred during a rapid radiation after the Cretaceous-Paleogene mass extinction event about 66 million years ago.
Full-text available
Recent coelurosaurian discoveries have greatly enriched our knowledge of the dinosaur-bird transition, but all reported taxa close to this transition are from relatively well-known coelurosaurian groups^1-3^. Here we report a new basal avialan, Epidexipteryx hui gen. et sp. nov., from the Middle-Late Jurassic of Inner Mongolia, China. This new species is characterized by an unexpected combination of characters seen in several different theropod groups, particularly the Oviraptorosauria. Phylogenetic analysis shows it to be the sister taxon to Epidendrosaurus^4,5^, forming a new clade at the base of Avialae^6^. Epidexipteryx also possesses two pairs of elongate ribbon-like tail feathers (ETFs), and its limbs lack contour feathers for flight. This finding shows that a member of the avialan lineage experimented with integumentary ornamentation as early as the Middle-Late Jurassic, and provides further evidence relating to this important aspect of the transition from non-avian theropods to birds.
Full-text available
We provide a detailed description of a well-preserved enantiornithine specimen (GSGM-05-CM-004) from the Lower Cretaceous (Aptian) Xiagou Formation of northwestern Gansu Province, China, for which we erect the new taxon Feitianius paradisi, gen. et sp. nov. This specimen has a distinctive pelvic morphology and can be further distinguished from all other Mesozoic birds by a unique caudal plumage formed by multiple rectricial morphotypes. This newly documented tail morphology reveals a previously unrecognized level of complexity in the plumage of basal birds. This complex tail-feather morphology has a parallel in extant sexually dimorphic birds in which the males have the most altered tails; thus, we identify this specimen as male. Ornamental tail morphologies, such as the novel tail plumage described here, dominate Enantiornithes. This reinforces hypotheses that sexual selection was a major driving force in the evolution of basal bird plumage. SUPPLEMENTAL DATA—Supplemental materials are available for this article for free at Citation for this article: O’Connor, J. K., D.-Q. Li, M. C. Lamanna, M. Wang, J. D. Harris, J. Atterholt, and H.-L. You. 2015. A new Early Cretaceous enantiornithine (Aves, Ornithothoraces) from northwestern China with elaborate tail ornamentation. Journal of Vertebrate Paleontology. DOI: 10.1080/02724634.2015.1054035. 2015
Full-text available
We report on a new species of ornithuromorph bird, Iteravis huchzermeyeri gen. et sp. nov., from the previously unreported Sihedang locality of the Lower Cretaceous Yixian Formation, the oldest ornithuromorph bearing deposit in the world. Unlike most other Cretaceous localities, specimens from this new quarry are largely referable to Ornithuromorpha, similar to the Lower Cretaceous Aptian Xiagou Formation in Gansu Province. Also similar to the Xiagou avifauna, the fauna at Sihedang is largely dominated by a single taxon (described here). Differences in faunal dominance may suggest the Sihedang records a unique ecological habitat. This may also explain the dominance of Gansus in the younger Xiagou Formation locality and suggests that previous hypotheses regarding the shift in dominance between Enantiornithes and Ornithuromorpha need to be reassessed in terms of potential ecological biases due to limited sampling. Furthermore, the recognition of an ornithuromorph dominated locality in the Sihedang significantly weakens the signal of such an inferred trend. Compared to most Jehol birds, the new specimen is relatively better preserved in three dimensions revealing morphological details of the skeleton, as well as preserves feather impressions including a rectricial morphology previously unknown among Mesozoic birds. © 2014, Science China Press and Springer-Verlag Berlin Heidelberg.
Full-text available
A description of Nomingia gobiensis gen. et sp. n., the first known dinosaur with a pygostyle, the structure known so far only in birds, is presented. The specimen comes from the Late Cretaceous strata at Bugin Tsav, Trans-Altai Gobi, Mongolia. N. gobiensis is assigned within the Oviraptorosauria based on the following characters: pneumatized caudal vertebrae, posteriorly concave ischium, and deep cervicodorsal hypapophyses. This specimen has been previously partially described without being formally named (Barsbold et al. 2000).
Full-text available
Although more than half of the evolution of birds occurred during the Mesozoic Era our understanding of this long history focused on the spectacular specimens of the Late Jurassic Archaeopteryx lithographica and the more derived Late Cretaceous hesperornithiforms and ichthyornithiforms for over a century of paleontological research. In the last decade, however, a tremendous burst of new evidence—perhaps unparalleled in the field of vertebrate paleontology—has been uncovered. Indeed, the number of species of early birds described during the 1990s nearly tripled the number of taxa discovered during the previous 130 years elapsed since the discovery of Archaeopteryx in the mid-1800s.
Full-text available
We describe a new enantiornithine bird, Parapengornis eurycaudatus gen. et sp. nov. from the Lower Cretaceous Jiufotang Formation of Liaoning, China. Although morphologically similar to previously described pengornithids Pengornis houi, Pengornis IVPP V18632, and Eopengornis martini, morphological differences indicate it represents a new taxon of the Pengornithidae. Based on new information from this specimen we reassign IVPP V18632 to Parapengornis sp. The well preserved pygostyle of the new specimen elucidates the morphology of this element for the clade, which is unique in pengornithids among Mesozoic birds. Similarities with modern scansores such as woodpeckers may indicate a specialized vertical climbing and clinging behavior that has not previously been inferred for early birds. The new specimen preserves a pair of fully pennaceous rachis-dominated feathers like those in the holotype of Eopengornis martini; together with the unique morphology of the pygostyle, this discovery lends evidence to early hypotheses that rachis-dominated feathers may have had a functional significance. This discovery adds to the diversity of ecological niches occupied by enantiornithines and if correct reveals are remarkable amount of locomotive differentiation among Enantiornithes.
A new species of enantiornithine bird from the Lower Cretaceous Yixian Formation of northeastern China is reported. The new taxon, Shanweiniao cooperorum, possesses several enantiornithine synapomorphies as well as the elongate rostral morphology (rostrum equal to or exceeding 60% the total length of the skull) of the Chinese early Cretaceous enantiornithines, Longipteryx chaoyangensis and Longirostravis hani. The discovery of this new specimen highlights the existence of a diverse clade of trophically specialized enantiornithines, Longipterygidae, for which we present phylogenetic support in a new comprehensive cladistic analysis of Mesozoic birds. Shanweiniao provides new information on the anatomy of longipterygids, and preserves a rectricial morphology previously unknown to enantior-nithines, with at least four tail feathers closely arranged. This supports the hypothesis that enantiornithines were strong fliers and adds to the diversity of known tail morphologies of these Cretaceous birds.
The most basal avians Archaeopteryx and Jeholornis have elongate reptilian tails. However, all other birds (Pygostylia) have an abbreviated tail that ends in a fused element called the pygostyle. In extant birds, this is typically associated with a fleshy structure called the rectricial bulb that secures the tail feathers (rectrices) [1]. The bulbi rectricium muscle controls the spread of the rectrices during flight. This ability to manipulate tail shape greatly increases flight function [2, 3]. The Jehol avifauna preserves the earliest known pygostylians and a diversity of rectrices. However, no fossil directly elucidates this important skeletal transition. Differences in plumage and pygostyle morphology between clades of Early Cretaceous birds led to the hypothesis that rectricial bulbs co-evolved with the plough-shaped pygostyle of the Ornithuromorpha [4]. A newly discovered pengornithid, Chiappeavis magnapremaxillo gen. et sp. nov., preserves strong evidence that enantiornithines possessed aerodynamic rectricial fans. The consistent co-occurrence of short pygostyle morphology with clear aerodynamic tail fans in the Ornithuromorpha, the Sapeornithiformes, and now the Pengornithidae strongly supports inferences that these features co-evolved with the rectricial bulbs as a "rectricial complex." Most parsimoniously, rectricial bulbs are plesiomorphic to Pygostylia and were lost in confuciusornithiforms and some enantiornithines, although morphological differences suggest three independent origins.
Reconstructing the tree of life requires deciphering major evolutionary transformations and the functional capacities of fossils with "transitional" morphologies. Some of the most iconic, well-studied fossils with transitional features are theropod dinosaurs, whose skeletons and feathered forelimbs record the origin and evolution of bird flight. However, in spite of over a century of discussion, the functions of forelimb feathers during the evolution of flight remain enigmatic. Both aerodynamic and non-aerodynamic roles have been proposed, but few of the form-function relationships assumed by these scenarios have been tested. Here, we use the developing wings of a typical extant ground bird (Chukar Partridge) as possible analogues/homologues of historical wing forms to provide the first empirical evaluation of aerodynamic potential in flapping theropod "protovvings." Immature ground birds with underdeveloped, rudimentary wings generate useful aerodynamic forces for a variety of locomotor tasks. Feather development in these birds resembles feather evolution in theropod dinosaurs, and reveals a predictable relationship between wing morphology and aerodynamic performance that can be used to infer performance in extinct theropods. By spinning an ontogenetic series of spread-wing preparations on a rotating propeller apparatus across a range of flow conditions and measuring aerodynamic force, we explored how changes in wing size, feather structure, and angular velocity might have affected aerodynamic performance in dinosaurs choosing to flap their incipient wings. At slow angular velocities, wings produced aerodynamic forces similar in magnitude to those produced by immature birds during behaviors like wing-assisted incline running. At fast angular velocities, wings produced forces sufficient to support body weight during flight. These findings provide a quantitative, biologically relevant bracket for theropod performance and suggest that protowings could have provided useful aerodynamic function early in maniraptoran history, with improvements in aerodynamic performance attending the evolution of larger wings, more effective feather morphologies, and faster angular velocities.