Osteology of the Flightless Patagopteryx deferrariisi from the Late Cretaceous of Patagonia (Argentina)

Full text

L.M. Chiappe, Patagopteryx - Page 1
OSTEOLOGY OF THE FLIGHTLESS PATAGOPTERYX DEFERRARIISI FROM THE LATE
CRETACEOUS OF PATAGONIA (ARGENTINA)
Luis M. Chiappe
Running Head: Patagopteryx

L.M. Chiappe, Patagopteryx - Page 2
INTRODUCTION
The evolution of Mesozoic Gondwanan birds is poorly known. This book attests to the fact that
regardless of the significant contribution of Gondwanan birds to our understanding of early avian history
(Chiappe, 1991, 1996a; Forster et al., 1996, 1998; Clarke and Chiappe, in press) (Chapter 12), the
majority of the available data comes from the northern hemisphere.
Patagopteryx deferrariisi (Alvarenga and Bonaparte, 1992) is a hen-sized, flightless bird (Fig.
13.1) known exclusively from Late Cretaceous beds in the northwestern limits of the Patagonian city of
Neuquén, Argentina (Fig. 13.2). Patagopteryx is important not only because it documents a distinct
lineage of basal birds, but also because, being known from several specimens, it is the best represented
Mesozoic avian from the southern hemisphere.
The discovery of the first specimens of Patagopteryx in 1984-1985 was made in connection with
the expansion of the campus of the Universidad Nacional del Comahue (UNC, Neuquén) and the
resultant clearing and leveling of the hills that form the southwestern margin of the Neuquén river, near
its confluence with the Limay river (Fig. 13.2). Oscar de Ferrariis, the Director of the UNC’s Museo de
Ciencias Naturales at the time, recovered the first specimens and passed them on to José Bonaparte for
study (Alvarenga and Bonaparte, 1992). Additional specimens were later collected by Jorge Calvo and
Leonardo Salgado from the UNC.
The first reference to Patagopteryx was made by Bonaparte (1986), who mentioned the presence
of an “indeterminate family of ratite-like birds” in the Late Cretaceous of Patagonia. The idea of
Patagopteryx as a basal ratite was retained during its formal description (Alvarenga and Bonaparte, 1992;
see also Alvarenga, 1993). Although originally subscribing to the ratite hypothesis (Chiappe, 1987), I
challenged its allocation within Neornithes, and even Ornithurae, in subsequent studies (Chiappe, 1989,
1991). This conclusion was strongly supported by later phylogenetic analyses of basal avians (Chiappe,
1992, 1995a, b, 1996b; Chiappe and Calvo, 1994) (Chapter 20).
Institutional Abbreviations: MACN, Sección Paleontología de Vertebrados, Museo Argentino de
Ciencias Naturales, Buenos Aires (Argentina); MUCPv, Museo de Ciencias Naturales, Universidad
Nacional del Comahue, Neuquén (Argentina).
GEOLOGICAL SETTING
Patagopteryx deferrariisi is known only from the continental beds of the Bajo de la Carpa
Member of the Río Colorado Formation (Neuquén Group), exposed at Boca del Sapo in the northeast
corner of the city of Neuquén (see Chiappe and Calvo, 1994, and references herein cited) (Fig. 13.2).
At Boca del Sapo, the Bajo de la Carpa Member is predominantly composed of medium-to-fine
friable whitish-gray-to-pink sandstones, generally made of clean quartz with very little clay (Chiappe and
Calvo, 1994).
The fauna from Boca del Sapo is composed of a diverse ensemble of small tetrapods (Bonaparte,
1991). Particularily abundant are the crocodilian Notosuchus terrestris (Smith Woodward, 1896;
Gasparini, 1971; Bonaparte, 1991) and the snake Dinilysia patagonica (Smith Woodward, 1901;
Bonaparte, 1991). This fauna also includes the crocodilian Comahuesuchus brachibuccalis (Bonaparte,
1991), the non-avian theropod Velocisaurus unicus (Bonaparte, 1991), the enantiornithine Neuquenornis
volans (Chiappe and Calvo, 1994) (Chapter 11), and Alvarezsaurus calvoi, a close relative of the
Mongolian Mononykus olecranus (Chiappe, 1995a; Novas, 1996) (Chapter 4).
Cazau and Uliana (1973) regarded the depositional environment of the Bajo de la Carpa Member
as a system of braided streams. Opalized trees and large, disarticulated dinosaurian remains were found
in other localities of this member, suggesting transport. In contrast, at Boca del Sapo the fauna is
composed of small and usually articulated tetrapods, and large dinosaurian remains are very rare
(Bonaparte, 1991; Chiappe and Calvo, 1994). The deposits at Boca del Sapo apparently represent a
different depositional environment, one with lower energy than that generally envisioned for the Bajo de
la Carpa Member (Chiappe and Calvo, 1994). Calvo and Heredia (1997) have recently suggested that the

FIGURE 13.1. Skeletal reconstruction of Patagopteryx deferrariisi
(modified from Alvarenga and Bonaparte, 1992). Black areas are
not preserved in any available specimen.

FIGURE 13.2. Map indicating the locality of Boca del Sapo (black arrow in
inset 3) in the city of Neuquén, northwestern Patagonia, Argentina (modified
from Gasparini et al., 1991).

L.M. Chiappe, Patagopteryx - Page 3
sandstones of Boca del Sapo may represent interdune deposits in an eolian depositional regime, although
Clarke et al. (1998) found little basis to support such a claim.
The Río Colorado Formation, along with the entire Neuquén Group, has traditionally been
considered as Late Cretaceous, generally as Campanian (e.g., Cazau and Uliana, 1973; Uliana and
Dellapé, 1981; Danderfer and Vera, 1992; see also Bonaparte, 1991). A Campanian-Maastrichtian
charophyte assemblage was reported for the Anacleto Member (Musacchio, 1993), yet, sequential
stratigraphical analyses in the northern area of the Neuquenian Basin, hinted at an older age for the entire
Río Colorado Formation. Cruz et al. (1989) proposed a Coniacian-Lower Campanian age, while
Legarreta and Gulisano (1989) suggested a Santonian-Campanian age. The pre-Campanian age of the Rio
Colorado Formation has also received indirect support from the recent placement of the basal unit of the
Neuquén Group (Río Limay Formation) as either Cenomanian (Ramos, 1981; Cruz et al., 1989) or
Albian-Cenomanian (Calvo, 1991), as well as from the suggestion that the whole Neuquén Group may be
older than was traditionally thought (Ramos, 1981). Recent magnetostratigraphic analyses of rocks from
the Río Colorado Formation at Auca Mahuevo, a sauropod nesting site some 150 km northwest of the
city of Neuquén (Chiappe et al., 1998), have shed important light on the age of this fossil-rich unit.
Dingus et al. (2000) have provided the first reliable age for the Anacleto Member of the Río Colorado
Formation, constraining it between 83.5 and 79.5 million years ago. This age can reasonably be
extrapolated to the Bajo de la Carpa sandstones of Boca del Sapo, which are therefore considered to be
early to middle Campanian.
SYSTEMATIC PALEONTOLOGY
Aves Linnaeus, 1758
Ornithothoraces Chiappe, 1995
Patagopteryx deferrariisi Alvarenga and Bonaparte, 1992
HolotypeMACN-N-03, partial skeleton including five cervical vertebrae, 11 thoracic vertebrae, a
complete synsacrum, two caudal vertebrae, both humeri and the proximal portion of the radius and ulna,
the shoulder ends of both scapulae and coracoids, the acetabular and postacetabular portions of both ilia,
and portions of the femora and tibiotarsi.
Referred specimensMACN-N-10, five fragments of tarsometatarsi (of at least three individuals) and
pedal phalanges; MACN-N-11, nearly complete skeleton including skull and jaws; MACN-N-14, four
articulated thoracic vertebrae and other vertebral fragments; MUCPv-48, complete hind limb, several
vertebrae, and fragments of the pelvis and skull; MUCPv-207, portions of hind limb, synsacrum, and
several other vertebrae.
Locality and HorizonBoca del Sapo, City of Neuquén, Province of Neuquén, Argentina (Fig. 13.2).
Bajo de la Carpa Member, Río Colorado Formation, Late Cretaceous (Campanian).
DiagnosisPatagopteryx deferrariisi is diagnosed by the following autapomorphies: quadrate fused to
the pterygoid; quadrate pneumatic foramen located laterally; biconvex fifth thoracic vertebra; thoracic
vertebrae 6-11 procoelous, with wide, kidney-shaped centra; synsacrum procoelous; shoulder end of
coracoid formed by a broad, tongue-like caudolateral surface that includes the humeral articular facet and
the area of scapular articulation; acromion of scapula dorsoventrally expanded and key-hole shaped in
shoulder (proximal) view; strong muscular lines in the shaft of the humerus; distal half of shaft of ulna
strongly compressed craniocaudally; minor metacarpal more robust than the major metacarpal; major
metacarpal with a cranioventral, laminar projection; ilium with a prominent iliac crest and a well-
developed caudolateral spine; craniocaudally compressed, strap-like pubis, with its caudal third curved
cranioventrally; paddle-shaped ischium; prominent M. iliofibularis tubercle on the fibula; fibular spine
distally fused to the cranial surface of tibiotarsus; transversely wide tarsometatarsus; pamprodactyl foot.
ANATOMY
Skull and Mandible

L.M. Chiappe, Patagopteryx - Page 4
The skull of Patagopteryx deferrariisi is known only for MACN-N-11 and MUCPv-48. Only the
braincase, caudal portion of the orbits, quadrates, and caudal end of the right pterygoid are preserved
(Figs. 13.3, 13.4).
The skull roof is formed primarily by the frontals, which are separated from one another by a
straight suture. The frontals are flat and they slope rostrally. In MUCPv-48, in the rostralmost preserved
portion of the right frontal, there is a median depression that may have received the caudal end of the
nasal process of the premaxilla. Both MACN-N-11 and MUCPv-48 preserve portions of the endocranial
cavity as a natural cast in some of the areas where the frontals are missing.
The orbits are large and are bound by the frontals, caudodorsally, and the laterosphenoids,
caudoventrally. This latter bone appears to form a short postorbital process, suggesting that Patagopteryx
lacks a postorbital bone.
The caudal portion of the skull roof is formed by the parietals. These are significantly shorter
than the frontals. In MACN-N-11, the frontoparietal contact is not well defined, but in MUCPv-48 there
is a clear, transverse suture between these two bones. The caudal boundary of the parietals is marked by
the strong nuchal crest, which has an inverted V-shape in caudal view that wedges between the parietals
(Figs. 13.3, 13.4).
The squamosal is preserved only in the area of its articulation with the quadrate and prootic, and
along with the latter bone, it caps the otic process of the quadrate (Figs. 13.3B, 13.4A, 13.5). The
articulation of the squamosal to the prootic and its position with respect to other elements indicates that
this bone was incorporated into the braincase, a derived condition contrasting that of other basal avians
(e.g., Archaeopteryx, Mononykus, Confuciusornis, basal enantiornithines). The squamosoprootic
articulation leaves no space for the entrance of the dorsal tympanic recess (Fig. 13.5B). This suggests
that the otic process of the quadrate was not differentiated into two separate condyles (Baumel and
Witmer, 1993), although this cannot be established confidently. Laterally, the squamosal possesses a
short zygomatic process (Figs., 13.3B, 13.4A, 13.5B) that is firmly adhered to the lateral surface of the
quadrate’s otic process. This latter condition recalls the condition of modern ratites (Starck, 1995).
Interestingly, the squamosal also possesses a medial, stout process that abuts the medial surface of the
quadrate, distal to the otic process. The presence of this process along with the closely appressed
zygomatic process must have seriously constrained the quadrate’s kinesis (Starck, 1995).
The prootic is represented by the pila otica (the opisthotic may be taking part in it; see Baumel
and Witmer, 1993) and portions surrounding the columellar recess. The pila otica articulates with the otic
process of the quadrate and connects it with the dorsal margin of the columellar recess (Figs. 13.4D,
13.5B). The fact that the pila otica is quite long suggests that the otic process of the quadrate was not
close to the columellar recess. The latter is a heart-shaped recess with its apex pointing caudodorsally
(Fig. 13.4A). This recess houses the cochlear and vestibular fenestrae, and the entrance to the caudal
tympanic recess. The latter opens dorsal to the vestibular fenestra and it probably connected to a highly
pneumatized area caudodorsally to the columellar recess.
The quadrate is laterally compressed, with both dorsal and ventral ends lying in the same vertical
plane. In lateral view, its caudal margin is strongly notched (Figs., 13.3, 13.4). The rostral margin extends
into a broad, laminar orbital process. The caudal surface is deeply excavated by a dorsoventral furrow
(Figs. 13.4C, 13.5B). The latter starts right at the level of the apex of the lateral notch, and it extends up
to the end of the otic process. This furrow is flanked by dorsoventral bars, the lateral of which is thinner
than the medial. Dorsal to this furrow is the tied contact between the squamosal and prootic bones, both
articulating with the otic process of the quadrate.
The quadrate of Patagopteryx is pneumatic; its hollowed body is exposed in the right element of
MACN-N-11. In contrast to other birds, the entrance of the quadrate diverticulum of the tympanic air sac
is centered on the lateral face of the quadrate (Figs., 13.3, 13.4A). In the left quadrate of MACN-N-11,
immediately dorsal to the pneumatic foramen, there is a small fossa perforating the quadrate. This fossa
is not present on the right element and it most likely represents an artifact.

FIGURE 13.3. Skull and jaw of Patagopteryx deferrariisi (MACN-N-11). A,
right lateral view; B, left lateral view; C, dorsal view. Abbreviations: f, frontal;
fec, fenestra cochlearis; fev, fenestra vestibularis; md, mandible; nuc, nuchal
crest; pnc, pneumatic cavities; pro, prootic; q, quadrate; qjc, quadratojugal
cotyla of the quadrate; qpn, quadrate pneumatic foramen; zpr, zygomatic
process.

FIGURE 13.4. Skull, proatlas, atlas,
and axis of Patagopteryx deferrariisi
(MACN-N-11). A, left lateral view; B,
right lateral view; C, ventral view; D,
occipital view.

L.M. Chiappe, Patagopteryx - Page 5
The mandibular process of the left quadrate of MACN-N-11 preserves the rostral margin of the
quadratojugal cotyla (Fig. 13.3B). Judging from the shape of this portion, the quadratojugal cotyla must
have been large and round. The mandibular articulation is formed by three condyles placed on the
vertices of a triangular surface (Fig. 13.4C). These condyles are ventrally projected, defining a deep,
central depression. The larger, rostral condyle is transversely oriented and it slopes dorsomedially. A
smaller and pointed medial condyle forms the medial vertex of the triangle. The caudal condyle is flat
and less ventrally projected than the others.
The caudal end of the right pterygoid of MACN-N-11 is the only known area of the palate (Fig.
13.5A). The pterygoid, at least on its caudal end, is broad and robust, and it becomes broader rostrally
(Figs. 13.4C, 13.5A). Interestingly, close examination of the pterygoid-quadrate complex does not reveal
even a suture between these two bones. The only conclusion to be drawn from this condition is that the
pterygoid is completely fused to the quadrate, a very unusual condition among birds (Simonetta, 1960).
This condition combined with the presence of a robust, medial process of the squamosal restricting the
quadrate’s movements suggests that the skull of Patagopteryx had limited, if any, capacity for
streptostylic movement (Chiappe, 1996b).
The occiput is nearly vertical and there are no sutures demarcating its component bones. The
cerebellar prominence is feeble (Fig. 13.4D). The foramen magnum and the occipital condyle are covered
by the articulated atlas. The angle formed by the intersection of the planes of the foramen magnum and
the basitemporal plane is slightly larger than 90. This suggests that the longitudinal axis of the skull was
continuous with the main axis of the neck (Saiff, 1983).
To the right of the occipital condyle of MACN-N-11, there are two, very close foramina (joined
by a narrow slit on the left side) forming the exit of the cranial nerve XII (n. hypoglossi) (Figs. 13.4D,
13.5). Ventrolaterally from these foramina, there is another foramen (best observed on the right side of
MACN-N-11) that was probably an additional exit for this nerve. The number of hypoglossal foramina in
birds is variable, but most have three (Webb, 1957). Lateral to the hypoglossal foramina is a large
foramen forming the exit of cranial nerve X (n. vagi) and presumably cranial nerve XI (n.
glossopharyngealis) (Fig. 13.5). On the left side of MACN-N-11, immediately medial to the latter
foramen, are three tiny foramina laid in a triangle. The most medial one appears to be the counterpart of
the one interpreted as an additional exit of cranial nerve XII. The correspondence of the remaining ones
is unclear. Over the left margin of the foramen magnum of MACN-N-11 there is a slit-like foramen that
most likely corresponds to the exit of the external occipital vein (Fig. 13.5); this area has not been
preserved on the right side.
Rostral to the occipital condyle there is a large, deep subcondylar fossa (Figs. 13.4C, 13.5A).
Rostral to this fossa, there is an ample parasphenoidal lamina that is depressed centrally, with its lateral
borders projecting ventrally. Because only the right half of this lamina in known (preserved in MACN-N-
11), it is unclear whether the parasphenoidal lamina was ventrally perforated by a recess (i.e.,
basisphenoidal recess of Currie, 1995), as in the alvarezsaurids Shuvuuia (Chiappe et al., 1998) and
several non-avian theropods (e.g., Currie, 1995). In contrast to many Neornithes (except ratites), the
parasphenoidal lamina projects ventrally well below the level of the occipital condyle. Even though the
basipterygoid processes are not preserved, the presence of a scar on the right pterygoid of MACN-N-11
indicates that these processes were prominent and located caudally, articulating with the pterygoid near
its juncture to the quadrate—such a condition is comparable to that of paleognath birds (Bock, 1963).
The mandible of Patagopteryx is known only from its caudal portion (Fig. 13.6). It is slender and
transversely compressed, except for its robust articular area. The retroarticular process is absent.
Individual bones are not discernible except for a ventral, longitudinal cleft separating the surangular from
the angular (Figs. 13.5A, 13.6C). It is probable that this cleft was housed in a long, caudal projection of
the dentary. These two portions, however, fuse caudal to the beginning of the articular area.
The lateral surface of the mandible is flat. Just rostral to the articular area, the surangular is
perforated by a small, subcircular caudal fenestra (Fig. 13.6A), an apparently primitive condition known

L.M. Chiappe, Patagopteryx - Page 6
for a variety of non-avian theropods (Weishampel et al., 1990). This surangular fenestra is the only
opening perforating the mandible; the presence of an extensive, unperforated area of the surangular
rostral to this fenestra suggests that a rostral mandibular fenestra was absent in Patagopteryx. The medial
surface of the mandible is strongly excavated by an extensive aditus fossa (Baumel and Witmer, 1993).
Caudally, this fossa is bound by a small, medial process.
The quadrate articular fossa bears three cotylae for the articulation with the three condyles of the
quadrate (Fig. 13.6B). The lateral cotyla is large, shallow, and subelliptical. Its main axis is oriented
rostromedially. This cotyla is separated from the medial cotyla by a thick, prominent intercotylar crest.
The medial cotyla is large and it forms a deep basin in between the intercotylar crest, the margin of the
medial process, and the dorsally projected caudal cotyla. Like the lateral cotyla, the main axis of the
medial cotyla is oriented rostromedially. Interestingly, the caudal cotyla crowns a strong, dorsally
projected process that forms the caudal end of the mandible.
Vertebral Column
Cervical vertebrae—The cervical vertebrae are elongate and heterocoelous (Figs. 13.7, 13.8).
Patagopteryx is the most basal bird with fully heterocoelous cervical centra. Pre- and postzygapophyses
are strongly separated from one another. The postzygapophyses bear prominent epipophyses. The
development of the neural spine is variable. Fairly high neural spines are present in the cranial region of
the neck, whereas they are almost absent in the last cervicals. The cervical ribs fuse to the para- and
diapophyses (Figs. 13.7B, D), and all together enclose an ample, round transverse foramen. The cervical
count of Patagopteryx is unknown. Nevertheless, the difference between the epipophysis of the eighth
cervical of MACN-N-11, the last one preserved in articulation, and the first of the five last cervicals
preserved in MACN-N-03 suggests that Patagopteryx had at least 13 cervical vertebrae. Such a number is
comparable to that of many modern avian groups (Verheyen, 1960) and larger than the 9-10 cervicals of
enantiornithines (Chapter 11) and other basal avians (Chiappe et al., 1999).
Despite the retention of some primitive features (e.g., free proatlas, absence of fusion between
the atlantal vertebral arch and centrum), the cervical series of Patagopteryx is essentially comparable to
that of neornithine birds. This is presumably the case from a functional point of view. As in neornithine
birds (see Zusi and Storer, 1969; Zusi and Bentz, 1984; Zusi, 1985), the cervical series of Patagopteryx
can be divided into three functional sections that most likely performed similar movements as those of its
living relatives. Cervicals 3-5 (along with the atlas-axis) and 6-8 constitute sections I and II, respectively,
while the last four cervicals, and possibly the first preserved element of MACN-N-03, comprise section
III. The range of movements of sections I and III of Patagopteryx would have been ventral to the normal
position of the neck, while that of section II would have been dorsal with respect to the neck’s prevalent
position. This morphofunctional complex was also present in Enantiornithes (Chapter 11) and in basal
Ornithurae but absent in Archaeopteryx and non-avian theropods.
ProatlasIn MACN-N-11, between the atlas and the skull, there is a tiny, bony element that is
interpreted here as the proatlas (Figs. 13.4A, 13.7F). This element is preserved only in the left side,
abutting against the lateral surface of the atlantal vertebral arch. In lateral view it has an inverted V-
shape, with its cranial arm projecting ventromedially. Its caudal arm is shorter and more round than the
cranial one, recalling the shape of a postzygapophysis.
The presence of an individualized proatlas in Patagopteryx deferrariisi is startling. In adult extant
birds this element is fused to the skull, and to the best of my knowledge it has not been reported in any
extinct birds. Yet, an individualized proatlas of comparable shape is known for the non-avian theropod
Herrerasaurus (Sereno and Novas, 1993).
AtlasThe atlas has a ring-like shape. Its centrum is small and the neural arch is not fused to it
(Figs. 13.4A-C, 13.7F). The dorsal and dorsolateral portions of the neural arch are wide and laminar. The
neural arch narrows ventrally and ends in a round border that articulates with the centrum.

FIGURE 13.5. Skull and jaw of Patagopteryx deferrariisi (MACN-N-11). A,
ventral view; B, left caudolateral view. Abbreviations: eov, exit for the
external occipital vein; occ, occipital condyle; pro, prootic; psl,
parasphenoidal lamina; pty, pterygoid; q, quadrate; sq, squamosal; X-XII,
cranial nerves.

FIGURE 13.6. Left jaw of Patagopteryx deferrariisi (MACN-N-11). A,
lateral view; B, dorsal view; C, medial view.

L.M. Chiappe, Patagopteryx - Page 7
AxisThe axis is elongate as those cervical vertebrae that follow it (Figs. 13.4A-C). Its centrum
is formed by a single element (Fig. 13.7F). It is laterally compressed and its sides are excavated. The axis
possesses the primitive type of heterocoelous condition present in the last vertebrae of the cervical series
(see below). Unfortunately, both neural spines and zygapophyses are not adequately preserved.
Cervicals 3-5The centra of these vertebrae are heterocoelous. Cranioventrally, between the
parapophysis, the vertebral body has a pronounced depression that receives a ventral projection of the
caudal articular surface of the previous vertebra. Caudal to this fossa, the centrum of the third cervical
narrows ventrally. In the two following vertebrae (C
4
and C
5
), the ventral margin of the centrum becomes
a broader surface with an axial depression bound by slender ridges. Dorsal to these ridges, the vertebral
body of these vertebrae widens remarkably to form a horizontal plane that connects pre- and
postzygapophyses. The zygapophyseal surfaces are ample, elliptic, and broadly separated one from the
other. Of these three vertebrae, only the neural spine of C
5
is preserved. It is broad and well-developed.
In C
3
, and apparently in C
4
and C
5
, the cranial articular surface faces cranioventrally. This
orientation of the cranial articular surfaces combined with the strong development of the cranioventral
fossa of their centra suggests that these three elements belong functionally to section I of the cervical
series of neornithine birds (Zusi and Storer, 1969).
Cervicals 6-8These vertebrae are poorly preserved in MACN-N-11, the only specimen for
which the middle and cranial section of the neck are known. The neural spine of C
6
, as in C
5
, is broad and
well developed. Neural spines are missing in C
7
and C
8
. In C
7
, ventrally, the body is flat—lacking the
slender ridges of C
4
and C
5
. The epipophyses of these vertebrae are well developed. They are laterally
compressed and occupy a central position above the postzygapophyses.
The fact that the cranial articular surfaces of C
7
and C
8
face craniodorsally and that the
cranioventral fossa of their centra is poorly developed suggests that these vertebraeand presumably C
6
as wellcompare functionally to section II of the neck of neornithine birds (Zusi and Storer, 1969).
Caudal cervicalsThe caudal cervicals are known for MACN-N-03, in which the last five
vertebrae are preserved. Because the actual cervical count of Patagopteryx is unknown, the precise
position of these elements within the cervical series remains uncertain. The first three vertebrae behind
the first preserved of these are slightly smaller than the latter, but the last cervical is considerably smaller
(Table 1; Fig. 13.7).
In the last four cervicals, the neural spines are reduced (Figs. 13.7A, D, E). In the first vertebra
preserved in MACN-N-03, the neural spine is broken in a way that suggests that it might have been more
developed. The epipophyses of these five vertebrae are moderately developed (Figs. 13.7B, 13.8A, C),
though considerably less than in the middle cervicals. In none of these vertebrae is there evidence of an
area for the elastic ligament.
The vertebral bodies of the last cervicals are also heterocoelous. The heterocoely of these
vertebrae, however, differs from the typical modern avian condition in that it possesses a lesser degree of
transverse convexity and concavity of the caudal and cranial articular surfaces, respectively.
The bodies are ventrally flat (Figs. 13.7C, G, 13.8B, E), with the exception of the last cervical, in
which the ventral face is sharp and it forms a small ventral process in the cranial portion (Figs. 13.7G,
13.8E). The morphology of this vertebra is transitional between the cervical and thoracic series. It
possesses a ventral process of the centrum, and postzygapophyses that are less separated from one
another, with their articular surfaces facing more laterally.
Thoracic vertebrae—The 11 thoracic vertebrae of Patagopteryx are shorter than the cervicals.
The transverse processes are long. In the cranial portion of the series, the distance between both
prezygapophyses and postzygapophyses decreases caudalwards. This distance reaches a minimum value
in the mid-trunk region, and it increases again in the caudal thoracic vertebrae (Table 1). The vertebral
centra lack pleurocoels (Figs. 13.9, 13.10), and the first seven vertebrae (cranial and mid- thoracic
vertebrae) have ventral processes (Fig. 13.10H). The articular surfaces of the centra are procoelous,

FIGURE 13.7. Cervical vertebrae of Patagopteryx deferrariisi. A,
B, C, first three preserved vertebrae of MACN-N-03 in dorsal
(A), left lateral (B), and ventral (C) view. D, E, G, last two
preserved vertebrae of MACN-N-03 in left lateral (D), dorsal (E),
and ventral (G) view. F, left lateral view of the proatlas, atlas,
and axis of MACN-N-11. Abbreviations: acl, arcocostal lamina;
atc, atlantal centrum; atl, atlantal arch; axi, axis; caf, cranial
articular facet; dto, dorsal torus (epipophysis); occ, occipital articular facet; dto, dorsal torus (epipophysis); occ, occipital
condyle; poz, postzygapophysis; pra, proatlas; prc, carotic
process; prz, prezygapophysis; sca, carotic sulcus; spr, spinal
process; vpr, ventral process.

L.M. Chiappe, Patagopteryx - Page 24
TABLE 1. Measurements (mm) of the vertebral column of Patagopteryx deferrariisi.
MACN-N-03 MACN-N-14 MUCPv-48
Cervical Vertebrae
Total length of centrum
of Cp2* 20.0 — —
Maximum width at the level
of postzygapophyses of Cp2* >14.1 — —
Total length of centrum
of Cp2* 16.0 — —
Maximum width at the level
of postzygapophyses of Cp2* 12.5 — —
Thoracic Vertebrae
Total length of centrum
of D2
13.0 11.6 —
Maximum width at the level
of postzygapophyses of D2 10.3 9.0 —
Total length of centrum
of D5
14.5
>11.5 —
Maximum width at the level
of postzygapophyses of D5
7.0 — —
Total length of centrum
of D10
>12.1 — —
Maximum width at the level
of postzygapophyses of D10 10.1 — —
Synsacrum
Total length 52.6 — —
Width at the level of acetabulum 12.3 — —
Caudal Vertebrae
Maximum length of
proximal caudal** 9.0 — —
Maximum length of
mid-caudal*** — — 10.3
Height of neural spine

L.M. Chiappe, Patagopteryx - Page 25
of mid-caudal*** — — 9.6
*Cp refers to the number of preserved cervicals (e.g., Cp2 is the second preserved cervical).
**Based on the first preserved caudal of MACN-N-03.
***Based on the single mid-caudal preserved in MUCPv-48.
> indicates that the actual value is estimated to be less than 3 mm larger.

FIGURE 13.8. Cervical vertebrae of the caudal portion of the neck of
Patagopteryx deferrariisi (MACN-N-03). A, B, first three preserved
vertebrae in dorsal (A), and ventral (B) view. C, D, E, last two preserved
vertebrae in dorsal (C), right lateral (D), and ventral (E) view.

FIGURE 13.9. Thoracic vertebrae of Patagopteryx deferrariisi
(MACN-N-03). A, C, D2-D4 in left lateral (A), and dorsal (C) view. B, D,
D5-D8 in left lateral (B), and dorsal (D) view. Abbreviations: biv, biconvex
vertebra (D5); fco, costal fovea (parapophysis); prz, prezygapophysis; tp,
transverse process.

FIGURE 13.10. Thoracic vertebrae of Patagopteryx deferrariisi. A, B, D3-D4
of MACN-N-03 in dorsal (A), and right lateral (B) view. C, D, E, D5-D8 of
MACN-N-03 in right dorsolateral (C), right lateroventral (D), and left
dorsolateral (E) view. F, G, H, D2-D5 of MACN-N-14 in dorsal (F), right
lateral (G), and ventral (H) view.

L.M. Chiappe, Patagopteryx - Page 8
biconvex, opisthocoelous, and heterocoelous depending on the position in the thoracic series. The
thoracic vertebrae are free and there is no formation of a notarium.
The large number of thoracic vertebrae of Patagopteryx indicates an early stage in the trend
toward cervicalization of the thoracic series and the incorporation of its elements into the synsacrum.
Thoracic vertebrae (D
1
-D
3
)The first thoracic vertebra has a broad cranioventral depression in
the body and smooth epipophyses over the postzygapophyses; both features are typical of cervical
vertebrae. These characters, however, are combined with thoracic features such as well-developed
transverse processes and ribs articulatednot fusedto the vertebra. The body of this vertebra is
broader cranially than caudally, and its articular surfaces show a primitive degree of heterocoely, as do in
the caudal cervicals.
The bodies of the contiguous two thoracic vertebrae (D
2
and D
3
) are even broader cranially. The
articular surfaces are transitional from a primitive heterocoelous condition to an opisthocoelous or
"pseudopisthocoelous" one. The second thoracic vertebra has a prominent depression in the base of the
transverse process (Fig. 13.9C), between the prezygapophyses and the dorsal process (neural spine). This
depression is less pronounced in the first thoracic vertebra, and absent in the third (Figs. 13.9C, 13.10A).
The parapophyses of these three vertebrae are large and well defined (Figs. 13.9A, 13.10B). In
the first thoracic vertebra, they are located in a lower position on the cranial border of the centrum, while
in the second thoracic vertebra, they have a slightly more dorsal position. In the third thoracic vertebra,
they are located somewhat underneath the transverse process.
Mid- thoracic vertebrae (D
4
-D
7
)In these vertebrae, the cntrum is essentially of the same width
cranially and caudally. On the basis of the large ventral process of the fifth thoracic vertebra of MACN-
N-11, these processes must have also been present in D
4
-D
6
, despite not being preserved. The ventral
process of the seventh thoracic vertebra, however, is poorly developed. The articular surfaces are
opisthocoelous in D
4
, biconvex in D
5
, and procoelous in D
6
and D
7
. The fifth thoracic vertebra (Figs.,
13.9B, 13.10D) is transitional between the procoelous condition of the caudal thoracic vertebrae and the
opisthocoelous-heterocoelous condition of the preceding elements. The parapophyses of D
5
nearly reach
the level of the transverse processes.
In the mid-thoracic vertebrae, the separation between both pre- and postzygapophyses reach a
minimum, increasing from D
7
backwards.
Caudal thoracic vertebrae (D
8
-D
11
)The last four thoracic vertebrae are procoelous as are the
two preceding them (Figs. 13.9B, 13.10D). The bodies of these vertebrae are broad and dorsoventrally
compressed, and kidney-shaped in cranial aspect (Figs. 13.11-13). Ventrally, they lack hypapophyses.
As was mentioned above, the separation between both pre- and postzygapophyses increases with
respect to the mid-thoracic vertebrae (Table 1), although it is not larger here than in the cranial thoracic
vertebrae. The dorsal processes are high and more robust than those of the preceding vertebrae. The
transverse processes are slender and of moderate length with respect to the centrum (Figs. 13.13D-F).
Thoracic vertebrae D
10
and D
11
are situated between the ilia, although they are not coossified to the
synsacral vertebrae (Figs. 13.11, 13.12).
SynsacrumThe synsacrum of Patagopteryx is formed by nine vertebrae (Figs. 13.11, 13.12).
These are completely fused, with no differentiation of either zygapophyses or individual dorsal
processes. As in the caudal thoracic vertebrae, the synsacral bodies are broad and dorsoventrally
compressed (Figs. 13.11B, 13.12B). Caudal to the midpoint, the vertebral centra gradually decrease in
size.The cranial articular surface of the synsacrum is concave while the caudal one is convex. In lateral
view, caudal to the acetabulum, the synsacrum curves slightly ventrally.
The dorsal processes of the synsacral vertebrae are fused, forming a longitudinal crest that
decreases in height caudalwards. Although the cranial part of this crest is not complete in any of the
available specimens, the preserved portion of MACN-N-03 suggests that it was as high as the dorsal
process of the caudal thoracic vertebrae (Fig. 13.12A).

FIGURE 13.11. Caudal thoracic vertebrae, synsacrum, and ilia of
Patagopteryx deferrariisi (MACN-N-03) in dorsal (A), and ventral (B) view.
Pelvis of Patagopteryx deferrariisi (MACN-N-11) in right lateral view (C).
Abbreviations: ace, acetabulum; ant, antitrochanter; bfo, brevis fossa; ctv,
caudal thoracic vertebrae; dic, dorsal iliac crest; ili, ilium; isc, ischium; lpr,
lateral projection of the postacetabular wing; op, obturator process; poz,
postzygapophysis; pub, pubis; syn, synsacrum; tpr, transverse process.

FIGURE 13.12. Caudal thoracic vertebrae, synsacrum, and ilia of
Patagopteryx deferrariisi (MACN-N-03) in dorsal (A), and ventral (B) view.

L.M. Chiappe, Patagopteryx - Page 9
Ventrally, the vertebral bodies of the cranial half define a nearly flat surface grooved
craniocaudally by a shallow broad furrow (Fig. 13.12B) comparable to the ventral sulcus of neornithine
birds (Baumel et al., 1979).
At the level of the acetabulum, the transverse processes of S
3
and S
4
are fused to the medial
surfaces of the ilia. Unfortunately, the remaining transverse processes are missing and it is not possible to
determine whether they were fused to the ilium as well.
Caudal vertebrae—Although in MACN-N-11 there is a minimum of five free caudals following
the synsacrum, the existence of isolated, free middle caudals in MUCPv-48 and MUCPv-207 suggests a
much larger number of free tail vertebrae. Yet, their true number remains unknown.
The caudal vertebrae of Patagopteryx have well-developed, ventrally directed transverse
processes, but lack both pre- and postzygapophyses (Figs. 13.13A-C). In the proximal caudals the neural
arch is high, defining a triangular vertebral canal. In the middle caudals, the neural arch is lower and it
defines a more arched vertebral canal. The proximal caudals are strongly procoelous, although the degree
of their procoelous condition decreases distally. In the middle caudals, as preserved in MUCPv-48 and
MUCPv-207, the proximal articular surface is just slightly concave, while the distal surface is convex.
No haemapophyses have been found in any of the preserved elements of the caudal series.
In the caudal vertebrae preserved in MACN-N-03 and MACN-N-11, the dorsal processes are
missing from their base. The middle caudals of MUCPv-48 and MUCPv-207, however, have very high
(Table 1), laminar dorsal processes (Figs. 13.13B, C), which taper toward their end. The remarkable
development of these processes suggests that the tail of Patagopteryx was long. Unfortunately, the
available material prevents determination about the presence or absence of a pygostyle.
Thoracic Girdle and Sternum
The thoracic girdle seems to comprise only the scapula and coracoid. These two bones articulate
at an angle of approximately 65 (Figs. 13.14B, 13.17M). The glenoid cavity defined by these two bones
is deep and narrow.The presence of an ossified furcula is unlikely. At least in MACN-N-11, in which the
elements of the thoracic girdle were articulated, there are not even vestiges of a furcula. The two
elements interpreted as portions of clavicles by Alvarenga and Bonaparte (1992) are in fact the shoulder
ends of the coracoids.
CoracoidThe coracoid is straight, triangular, and proportionally short (~65% of the scapula;
Table 2; Figs. 13.14, 13.15). The acrocoracoid process is very reduced, and the procoracoid process is
absent (Figs. 13.16A-C, 13.17C, D, G, H). The proximal end is completely different from that of other
birds. The glenoid facet is located at the shoulder end of the coracoid. It is wide and flat, and it slopes
laterally (Figs. 13.16A, 13.17C, D). Its borders are parallel and its lateral margin is round. The glenoid
facet overhangs the lateral border of the coracoid. On the medial border of this facet there is an inflated
area that seems to define the ventral limit of the articular surface with the scapula. This area is barely
distinct from the glenoid facet, however.
The medial surface of the shoulder end of the coracoid is flat (Figs. 13.16B, 13.17G). On the
other hand, the lateral surface is concave. Distal to the shoulder end, the coracoid narrows, and below
the midshaft it widens to form the broad sternal end (Fig. 13.14).
In dorsal view, the coracoidal shaft is flat, without the groove present in the fragment reported by
Alvarenga and Bonaparte (1992) as the sternal portion of the coracoid. The fragment described by these
authors and interpreted as part of the holotype is likely not of Patagopteryx, since no bone with this shape
has been found in any of the known specimens.
The ventral surface of the coracoid is convex. In the shoulder half, a ventral border separates a
medial from a lateral surface, which, combined with the flat dorsal surface, gives a subtriangular cross-
section to the coracoid. In ventral view, the lateral margin is strongly concave, and there is no
supracoracoid nerve foramen or notch (Figs. 13.14, 13.15B). The sternal end is broad and dorsoventrally
compressed.

FIGURE 13.13. Thoracic and caudal vertebrae of Patagopteryx
deferrariisi. A, first two caudal vertebrae of MACN-N-03 in left lateral
view. B, C, middle caudal vertebra of MUCPv-48 in cranial (B), and right
lateral (C) view. D, E, F, caudal thoracic vertebra of MUCPv-48 in dorsal
(D), caudal (E), and ventral (F) view. Abbreviations: caf, caudal articular
facet; crf, cranial articular facet; poz, postzygapophysis; spr, spinal
process; tpr, transverse process.

FIGURE 13.14. Thoracic girdle and sternum of Patagopteryx deferrariisi
(MACN-N-11). A, coracoids and sternum in ventral view; B, right scapula
and coracoid in lateral view. Stipple pattern indicates matrix and
cross-hatch pattern indicates broken or poorly preserved bone.
Abbreviations: acr, acromion; co, coracoid; haf, humeral articular facet; sc,
scapula; stm, sternum.

FIGURE 13.15. Thoracic girdle and sternum of
Patagopteryx deferrariisi (MACN-N-11). A, right
scapula in lateral view, cranial thoracic vertebrae in
dorsal view, left humerus in caudal view; B, coracoids
and sternum in ventral view.

L.M. Chiappe, Patagopteryx - Page 26
TABLE 2. Measurements (mm) of the thoracic girdle of Patagopteryx deferrariisi
MACN-N-11
Coracoid*
Total length 38.0
Maximum width of shoulder end 8.9
Scapula**
Distance from shoulder margin of
humeral articular facies to apex 60.1
Maximum width of blade 7.7
* Based on left element.
**Based on right element.

FIGURE 13.16. Thoracic girdle of Patagopteryx deferrariisi (MACN-N-03). A, B, C,
shoulder end of left coracoid in lateral (A), medial (B), and end (C) view; D, E, F, shoulder
end of right scapula in lateral (D), medial (E), and end (F) view. Abbreviations: acr,
acromion; apr, acrocoracoidal process; cof, coracoidal facet of scapula; haf, humeral
articular; scf, scapular facet of coracoid; tub, tubercle facet.

FIGURE 13.17. Thoracic girdle and humerus of Patagopteryx
deferrariisi. A, B, E, F, shoulder end of right (A, E) and left (B,
F) scapula of MACN-N-03 in lateral (A, F) and medial (B, E)
view; C, D, G, H, shoulder end of left (C, G) and right (D, H)
coracoid of MACN-N-03 in lateral (C, D) and medial (G, H)
view. I, J, K, L, left (I, J) and right (K, L) humerus of
MACN-N-03 in caudal (I, L) and cranial (J, K) view. M, lateral
view of the scapula and humerus of MACN-N-view of the scapula and humerus of MACN-N-11.

L.M. Chiappe, Patagopteryx - Page 10
ScapulaThe scapula is laminar and relatively broad, especially in its distal half (Figs. 13.14B,
13.15A). The glenoid facet is large and subcircular (Figs. 13.16D-E). Above this facet, proximally, there
is a tubercle that topographically corresponds to the coracoidal tubercle of the scapula of neornithine
birds. Ventral to this tubercle, bordering the ventroproximal margin of the glenoid facet there is a flat and
elongate surface situated in an oblique plane with respect to the latter facet (Figs. 13.16D, F). This
surface abuts with another surface located medially to the glenoid facet of the coracoid (see above). The
role of the tubercle mentioned earlier in the scapulocoracoid articulation is uncertain. In neornithine birds
as well as in Ichthyornis, a comparable tubercle (coracoidal tubercle) articulates in a facet of the
coracoid; in the coracoid of Patagopteryx, however, there is no trace of this facet.
The acromion is large (Figs. 13.16D-F, 13.17A, B, E, F). It is remarkably expanded
dorsoventrally, having a small proximal projection. In proximal view, it is separated from the glenoid
cavity by a more slender neck. Interestingly, the presence of a well-developed acromion suggests the
presence of clavicles, although as pointed out above, the completely articulated MACN-N-11 lacks any
sign of these elements.
The scapular blade is sagittally curved, with the ventral border concave (Figs. 13.14B, 13.15A).
Most basal birds have straight scapular blades (e.g., Chiappe et al., 1999; Chapters 9, 11). Patagopteryx
appears to be the most basal bird for which the scapula is sagittally curved; this derived feature is
interpreted as a synapomorphy of Patagopteryx plus Ornithurae (Chiappe, 1996; Chapter 20). the
shoulder half, the width of the shaft increases distally, while in the distal half it decreases toward the
apex. In the midshaft, the dorsal border forms a slender ridge that projects dorsally (Fig. 13.15A). Hence,
the lateral surface of the midshaft is concave.
The scapular apex is not preserved in any of the available specimens; however, the right scapula
of MACN-N-11 preserves the impression of it showing that the apex had a round end (Fig. 13.15A).
SternumThe sternum is thin and slightly convex ventrally (Figs. 13.14A, 13.15B). The cranial
border is round. The articular facet for the coracoids are well separated one from the other, as is typical
for flightless birds. Unfortunately, the median portion is badly preserved, preventing confidence about
the presence or absence of a carina (Figs. 13.14A, 13.15B). The slight convexity of the area near the
median portion, however, suggests that a carina was probably absent. No other information (e.g.,
dimensions, lateral processes) is available considering the state of preservation of the single known
sternum (MACN-N-11).
Thoracic Limb
HumerusThe humerus is robust and longer than the ulna (Table 3; Fig. 13.18). In dorsal view,
it has a sigmoid aspect, with the proximal half convex and the distal half concave toward the cranial face.
It lacks a pneumotricipital fossa and foramen (Figs. 13.17I, 13.18B). As in other non-ornithurine birds,
the humeral head is cranially concave and caudally convex, and it meets dorsally with the deltopectoral
crest (Figs. 13.18C-E). In cranial view, immediately distal to the head and slightly displaced dorsally,
there is a small and shallow circular depression that recalls the condition present in Enantiornithes
(Chapter 11). The deltopectoral crest is robust and extends through the proximal third of the humerus
(Table 3; Figs. 13.17I, J, 13.18). The bicipital area is poorly developed. Unfortunately, the area of the
ventral tuberosity is badly preserved in all available specimens, preventing determination of whether it
was projected caudally.
The humeral shaft is somewhat craniocaudally compressed, and it has a concave ventral border.
Caudally, there are two pronounced ridges (Fig. 13.18B). The most proximal, the strongest, originates on
the dorsal border, distal to the midshaft. This ridge runs proximoventrally until it nears the deltopectoral
crest, where it disappears. The second ridge, situated distal to the latter, also has its origin on the dorsal
border of the humerus, near the distal end. These ridges are interpreted as intermuscular lines separating
different areas of muscular attachment.

L.M. Chiappe, Patagopteryx - Page 27
TABLE 3. Measurements (mm) of the fore limb of Patagopteryx deferrariisi.
MACN-N-03 MACN-N-11
Humerus
Total length 66.3(l) 59.8(r)
Maximum width of distal end 17.2(l) >14.8(r)
Length of pectoral crest 19.7(l) 19.5(r)
Ulna
Total length — >51.5(l)
Maximum width of proximal end 9.6(l) >9.1(l)
Carpometacarpus
Total length — >22.7(r)
Midshaft width — 8.9(r)
(l) and (r) refer to left and right elements, respectively.
> indicates that the actual value is estimated to be less than 3 mm larger.

L.M. Chiappe, Patagopteryx - Page 11
The distal end of the humerus has a width equivalent to the proximal end (Figs. 13.17I, J,
13.18A, B). Unlike more primitive birds (e.g., Archaeopteryx, Enantiornithes), the main axis of the distal
end is oriented in the same plane as that of the proximal end. The dorsal condyle is well developed. It has
an elliptical outline with its major axis oriented 30 to the axis of the shaft (Figs. 13.17J, 13.18A). This
condyle is far from the dorsal margin of the humerus, and is strongly projected distally, in such a way
that it is visible in caudal view. The distal projection of the dorsal condyle is much stronger than the
typical condition in neornithine birds. The ventral condyle is of circular outline, although it is less
bulbous than the lateral one. A well-defined brachial depression is absent, although the M. brachialis
anticus might have originated in a depressed area proximal to the ventral condyle. Both dorsal and
ventral epicondyles are poorly developed (Figs. 13.18A, B). The caudal surface of the distal end is flat,
lacking tricipital sulci. Proximal to the ventral condyle, there is a smooth depression forming a small
olecranon fossa.
UlnaIt is craniocaudally compressed, especially in its distal half, in which it forms a slender
lamina. In the proximal end (Figs. 13.18F-I), the olecranon is well developed and the cotylae well
defined. The dorsal cotyla is located distal to the ventral one, being projected dorsally. The articular
surface of this cotyla is nearly flat. The ventral cotyla is round. Both cotylae are separated by a low
intercotylar area. In proximal view, these cotylae define a concave cranial border that corresponds to the
proximal margin of the radial proximal incision. This well-developed depression is triangular and its
distal vortex ends in a prominent bicipital tubercle. Ventrally, there is no evidence of the impression for
the M. brachialis anticus. The caudal face is flat, and it has no papillae for the insertion of secondary
feathers (Fig. 13.18F). In caudal view, the proximal half of the shaft has a uniform width, while the distal
half broadens distally.
RadiusThe radius is significantly more gracile than the ulna. The shaft has a subcircular cross-
section and a ventral curvature that is increased distally. The proximal end is broad with respect to the
shaft, and in proximal view is subtriangular (Figs. 13.18G, K). The humeral cotyla is subcircular. In
MACN-N-03, there is a projection of the proximal end directed toward the ulnar side, a feature
comparable to the capital tubercle of neornithine birds and Ichthyornis. Distally to the proximal end,
there is a weak bicipital tubercle. This tubercle, however, is in a more distal position with respect to that
in Neornithes.
There is almost no information about the distal end. Only in MACN-N-11 is this end preserved,
although badly. The remarkable torsion of the distal end toward the ulnar side (Figs. 13.18D, E) is
notable, although it might be exaggerated by distortion. In the dorsal portion of the distal end, there is a
condylar structure that is tentatively interpreted as the aponeurosis tubercle, the insertion of the tendon
that fans out the flight feathers of the wrist region (Baumel and Witmer, 1993).
Carpometacarpus and phalangesThe carpometacarpus is very short (Table 3; Figs. 13.18D,
E). The major metacarpal (II) is less robust than the minor one (III) (Figs. 13.18L, M). In its cranial
border, proximal to the midshaft, metacarpal II bears a laminar projection directed cranioventrally. Major
and minor metacarpals abut proximally and distally, forming broad symphyses, and enclosing an
elongated intermetacarpal space. Unfortunately, both distal and proximal carpals are not preserved.
Only the major digit (II) preserves anatomical information. This digit is proportionally long (as
long as the metacarpals), being formed by three phalanges (Fig. 13.18M). The proximal phalanx of this
digit has the greatest depth dorsoventrally. This phalanx is not broad and dorsoventrally compressed as is
the proximal phalanx of the major digit of carinate birds (common ancestor of Ichthyornis and
Neornithes plus all its descendants) (Clarke and Chiappe, in press). The intermediate phalanx is elongate
and cylindrical, also being the longest of the three. The distal phalanx is a claw of triangular shape.
Pelvic Girdle
IliumThe ilia are elongate and broadly separated from each other throughout their length
(Figs. 13.11, 13.12, 13.19).

L.M. Chiappe, Patagopteryx - Page 12
The preacetabular wing, which is slightly longer than the postacetabular (Table 4), is vertical and
laterally compressed (Figs. 13.11C, 13.19). The vertical portion of the preacetabular wing suggests a
poor development of the M. iliofemoralis cranialis (=M. iliotrochantericus caudalis), an important muscle
in the pelvis of extant birds, the function of which is still unclear (Raikow, 1985). In lateral view, the
ventral margin of the preacetabular wing is concave, forming a notch that becomes more pronounced near
the acetabulum. Dorsally, the preacetabular wing continues backwards as a strong crest that passes above
the acetabulum. The relationship between this crest and the other iliac regions suggests its homology
with the iliac crest of neornithine birds.
The acetabulum is large (Table 4), round, and has thick walls (Fig. 13.11C). MACN-N-11 shows
that the acetabulum was not completely perforated but partially obliterated by projections of the ilium,
pubis, and ischium. A partial obliteration of the acetabulum has been reported for hesperornithiforms
(Marsh, 1880; Martin and Tate, 1976) and Archaeopteryx (Martin, 1983, 1991). Caudal to the
acetabulum, there is a well-developed antitrochanter (Table 4). This is below the dorsal margin of the
acetabulum, and its main axis points caudoventrally. This condition resembles that of alvarezsaurids
(Perle et al. 1994) (Chapter 4), but it differs from the typical neornithine (as well as enantiornithine)
condition in which the antitrochanter is above the acetabular margin and its main axis is caudodorsally
oriented.
The postacetabular wing of Patagopteryx is transversally broad (Figs. 13.11, 13.12), and lower
than the preacetabular wing. Dorsally, it bears a prominent caudal iliac crest that continues cranially in
the dorsal border of the preacetabular wing. This crest separates an internal surface, medioventrally
slanted, from an external one sloped lateroventrally. The caudal iliac crest, high and laterally
compressed, has an inflexion point at the level of the caudal third of the acetabulum. In front of this
point, continuing in the preacetabular wing, this crest is slightly concave outwards, while it is convex
behind the inflexion point.
The external surface of the postacetabular wing is large and projected lateroventrally (Figs.
13.11, 13.12). Caudal to the acetabulum, the lateral border of this surface forms a strong notch that ends
in the caudolateral corner of the ilium. This area has a round, laterally projected border. The caudal
margin of the external surface is formed by a thickening that caudally limits a depressed area developed
behind the antitrochanter and lateral to the iliadic crest. Ventrally, both the external and internal surfaces
of the postacetabular wing delimit a broad fossa that originates behind the acetabulum and widens
caudally (Figs. 13.11B, 13.12B)—this fossa strongly resembles the brevis fossa of non-avian theropods.
IschiumThe ischium is an elongate, laterally compressed bone (Figs. 13.11C, 13.19). In its
proximal half, the ventral margin forms a thin obturator process that extends for about 1.5 cm. Caudal
and parallel to this process, on the lateral face, there is a longitudinal groove extending nearly the same
length as the obturator process (Fig. 13.11C). It is likely that this groove provided a larger area for the
insertion of the ischiopubic ligament which, in extant birds, attaches to the obturator process and closes
the obturator foramen. Given the ample notch separating the ischium from the pubis, the obturator
foramen of Patagopteryx must have been much larger than that of ornithurine birds.
The distal half of the ischium is an ample and flat surface that laterally bears a median
longitudinal ridge. Although the distal ends of the ischia are not preserved, it is clear that they did not
form a distal symphysis. In contrast to Enantiornithes and other basal birds, the ischium of Patagopteryx
lacks a proximocaudal process. Hence, if an ilioischiadic fenestra was present, this could not have been
caudally closed by bone.
PubisThe pubis is proximally fused to the ischium. It is a craniocaudally compressed, belt-like
bone of uniform width throughout its length (Figs. 13.11C, 13.19). The craniocaudal width is less than
half of the transverse width. The straight proximal half of the pubis is oriented caudoventrally, defining
an angle of about 45 with respect to the main axis of the synsacrum (Fig. 13.11C). The distal end of the
pubis is curved cranially and slightly medially, in a way that it is oriented cranioventrally. Unlike non-

FIGURE 13.19. Pelvis and caudal series of Patagopteryx deferrariisi
(MACN-N-11) in right lateral (A) and left lateral (B) view.

L.M. Chiappe, Patagopteryx - Page 28
TABLE 4. Measurements (mm) of the pelvic girdle of Patagopteryx deferrariisi.
MACN-N-03 MACN-N-11 MUCPv-48
Ilium
Maximum length of
postacetabular wing 29.9(r) — >28.7(r)
Maximum length of
preacetabular wing — 35.6(l) —
Maximum length of
acetabulum 15.4(r) 14.3(r) —
Maximum length of
antitrochanter 9.7(r) — —
Ischium
Maximum preserved length — 52.5(r) —
Dorsoventral width at the
level of obturator process — 8.6(r) —
Pubis
Total length — >49.6(r) —
(l) and (r) refer to left and right elements, respectively.
> indicates that the actual value is estimated to be less than 3 mm larger.

L.M. Chiappe, Patagopteryx - Page 13
avian theropods and most basal birds, the distal ends of the pubes of Patagopteryx do not fuse to each
other.
Pelvic Limb
FemurThe femur is robust and strongly arched craniocaudally (Figs. 13.20, 13.21). In cranial
view, the medial margin is generally concave, although its central third is flat to slightly convex. In this
view, the lateral margin is virtually flat with only a slight concavity in the central portion. The femur is
hollow and its walls are thick (approximately 1.7 mm). On the femoral shaft, two strong intermuscular
lines are present. The cranial intermuscular line originates above the internal condyle (Fig. 13.20A),
gradually passing toward the external margin. In the central portion, this line becomes more pronounced,
and proximally it weakens until it vanishes distal to the trochanteric crest. Most likely, this line
represents the boundary between the origin of the M. femorotibialis medius and the M. femorotibialis
externus, as occurs in some neornithine birds (Holmes, 1963; McGowan, 1979). The caudal
intermuscular line is a prominent ridge that originates as a projection of the medial margin of the
popliteal fossa and ascends toward the proximal end as a straight line (Figs. 13.20B, 13.21B). Proximal
to the midshaft, this line slopes medially to disappear about 1.5 cm distal to the femoral head. This line is
likely the extensive area of attachment of the M. pubo-ischio-femoralis (=M. adductor longus et brevis)
(McGowan, 1979), a major femoral retractor (Raikow, 1985). It must have also formed the limit of the
caudal extension of the M. femorotibialis externus (Hudson, 1937; Holmes, 1963). Also on the caudal
facelateral to the intermuscular line and somewhat proximal to the midshaftthere is a small foramen
similar to the nutrient foramen of neornithine birds. About 4-5 mm proximal to this foramen, there is an
elliptical tubercle displaced laterally (Fig. 13.21C). This 5-7 mm long tubercle is well-developed in all
known specimens and it probably corresponds to the insertion of the M. iliofemoralis of neornithine birds
(Hudson, 1937; Holmes, 1963; George and Berger, 1966). The central section of the shaft is subcircular,
although the intersection of the two intermuscular lines gives it a more irregular outline.
The femoral head is large, robust, and spherical. It faces medially and it is separated from the
trochanter by a thick, short neck (Figs. 13.20, 13.21). The trochanteric crest is well developed. The
trochanter was not projected proximally; it does not exceed the level of the head. The lateral surface of
the proximal end shows areas for muscular attachment (Fig. 13.20F). The most pronounced of these are
two broad grooves that together form a crescentic line that is cranially convex. It is likely that these were
the areas of insertion of the femoral protractors of the Iliotrochantericus Group (see Rowe [1986] for a
discussion of the different origin of the three muscles composing this group).
In the distal end, the cranial surface is flat, lacking a patellar groove (Figs. 13.20A, 13.21A).
Caudally, there is an ample, triangular, popliteal fossa (Table 5; Figs. 13.20B, 13.21B, C). On the base of
this fossa, in MACN-N-11, there are two small depressions of different size. On the lateral border of the
popliteal fossa, proximal to the external condyle, there is a strong tubercle; on the medial border of this
fossa, there is another tubercle of smaller size. It is probable that the first tubercle served as the area of
attachment of the ligamentary fibers of the ansa M. iliofibularis (Baumel et al., 1979; “biceps loop” of
George and Berger, 1966), that acts as a pulley for the action of the M. iliofibularis (George and Berger,
1966). The other tubercle probably coincides with the insertion of the M. pubo-ischio-femoralis (George
and Berger, 1966). The distal border of the popliteal fossa is limited by a low, transverse ridge that
connects both condyles.
The distal condyles are strongly projected caudally. The medial condyle is flat on its caudalmost
portion but convex distally. The lateral condyle is much larger. It possesses a well-developed tibiofibular
crest that delimits a large, lateral area for the articulation with the fibula.
Tibiotarsus The tibiotarsus is a slender, elongate bone (Figs. 13.22-13.24). The shaft is
straight and hollow, with walls approximately 0.8-1.0 mm thick. The shaft is slightly compressed
craniocaudally (Table 5), especially in its distal portion. The fibular crest extends over the proximal third

L.M. Chiappe, Patagopteryx - Page 14
of the lateral face; it is projected laterally and cranially. This well-developed crest rises 25 mm distal to
the proximal surface, reaching its maximum development near its distal end.
The proximal articular surface of the tibiotarsus is craniocaudally expanded due to the cranial
extension of the single cnemial crest. The medial articular area was more caudally projected than the
lateral one. The cnemial crest is thick and robust throughout its length, decreasing in thickness distally
(Figs. 13.22A, 13.23A, 13.24A). In cranial view, it is laterally twisted. Medially, it forms a slightly
convex, smooth surface that extends over all the proximal end of the tibiotarsus. On this surface,
proximal to the distal end of the cnemial crest and displaced to the caudal border of the tibiotarsus, is a
small tubercle that must have served as an area for muscular attachment. The lateral surface of the
cnemial crest is concave, ending on a ridge in its cranialmost portion.
In the distal end, both astragalus and calcaneum are completely fused to each other and to the
tibia (Figs. 13.22-13.24). The distal condyles are separated from one another by a relatively deep
intercondylar groove. Like in the Late Cretaceous malagasy bird Vorona (Chapter 12), this groove opens
proximally into a circular fossa. A transversal groove projects medially from this fossa and undercuts the
proximal margin of the medial condyle.
Like in other non-ornithurine birds, the medial condyle is much larger and broader than the
lateral (Fig. 13.23A). At the proximal border of the medial condyle, on the medial face, there is a
medially directed prominence. This feature is comparable to the medial ligamentary prominence of
neornithine birds, on which the ligaments of the tibiotarsal-tarsometatarsal articulation are attached
(Baumel et al., 1979). The medial face of the medial condyle has a large, circular depression. The lateral
condyle projects less cranially than the medial, and its proximal border rises less abruptly from the shaft.
On its lateral face there is a small, circular depression (depressio epicondylaris lateralis, Baumel et al.,
1979).
On the medial margin of the distal endproximal to the medial condylethere is an elongated,
proximodistal depression, the development of which varies among different specimens (Figs., 13.22A,
13.23A, 13.24A, B). The medial border of this depression is delimited by a smooth ridge. In MACN-N-
03, this depression forms a groove interpreted by Alvarenga and Bonaparte (1992) as the tendinal groove.
On the cranial surface, and centered with respect to the condyles, there is a deep circular fossa.
Patagopteryx does not have a supratendinal bridge.
FibulaIts proximal end is expanded craniocaudally and laterally compressed. Somewhat distal
to the proximal end, at the level of the proximal half of the fibular crest of the tibiotarsus, there is a large
and robust, laterally directed tubercle (Figs. 13.22D, 13.23, 13.24C, D) for the insertion of the M.
iliofibularis. The position of this tubercle resembles that of the Late Cretaceous malagasy Vorona
(Chapter 12) as well as the bizarre alvarezsaurids (Chiappe et al., 1996) (Chapter 4). Distal to the
iliofibularis tubercle, the fibula narrows, forming a slender bar that progressively becomes craniocaudally
compressed and migrates cranially. The distal end of the fibula is almost laminar and is fused to the
lateral border of the cranial surface of the tibiotarsus (Figs. 13.23C, D, 13.24C, D). Clearly, the fibula fail
to reach the proximal tarsals.
TarsometatarsusThe tarsometatarsus is much shorter than both the femur (~48% of the
femoral length) and the tibiotarsus (~34% of the tibiotarsal length; Table 5). Like in ornithurine birds,
metatarsals II-IV are completely fused with one another and with the distal tarsals (Figs. 13.25, 13.26). In
contrast to these birds, however, longitudinal grooves allow for individualization of each metatarsal and
the medullary cavity of these bones remains individualized (Fig. 13.25G). Metatarsals II-IV are placed on
the same transverse plane (Fig. 13.25G), contrasting with the condition present in neornithine birds, in
which the proximal end of metatarsal III is plantarily displaced with respect to metatarsals II and IV.
Metatarsals II and III are more robust than metatarsal IV, which is slightly slender and more gracile.
Metatarsal III is straight. In cranial view, it has the same thickness throughout its length. In contrast,
metatarsals II and IV narrow distally. Metatarsal I articulates with the medial border of metatarsal II.

FIGURE 13.20. Femur of Patagopteryx deferrariisi.
A, B, C, left femur of MACN-N-11 in cranial (A),
caudal (B), and lateral (C) view; D, E, F, proximal
half of right femur of MACN-N-03 in caudal (D),
medial (E), and lateral (F) view.

FIGURE 13.21. Femur of Patagopteryx deferrariisi. A, B, left femur of
MUCPv-48 in cranial (A) and caudal (B) view; C, left femur of MACN-N-03
in lateral view. Abbreviations: cam, caudal intermuscular line; ftr, fourth
trochanter; pfo, popliteal fossa.

L.M. Chiappe, Patagopteryx - Page 29
TABLE 5. Measurements (mm) of the hind limb of Patagopteryx deferrariisi.
MACN-N-03 MACN-N-11 MUCPv-48
Femur
Total length >98.8(r) >99.1(l) >97.6(l)
Height of popliteal fossa 16.0(r) 16.6(l) —
Lateromedial width at 10.1(r) 10.2(l) 9.8(l)
midshaft
Tibiotarsus
Total length >137.0(l) >136.2(l) >140.0
Length of cnemial crest 21.3(l) — —
Craniocaudal length
of proximal end >23.9(l) — —
Craniocaudal width
at midshaft — 7.2(l) 7.8(l)
Lateromedial width
at midshaft — 9.2(l) 9.5(l)
Maximum width of distal end — 14.5(r) —
Fibula
Total length — — >111.6(l)
Tarsometatarsus
Total length — 50.8(r) 51.0(l)
Maximum width of — >13.5(l) 14.4(l)
proximal end
Craniocaudal width at
midshaft — 5.1(r) 5.1(l)
Lateromedial width at
midshaft — 11.1(l) 10.6(l)
Total length of metatarsal I — 14.8(r) —
(l) and (r) refer to left and right elements, respectively.
> indicates that the actual value is estimated to be less than 3 mm larger.

FIGURE 13.22. Tibiotarsus of Patagopteryx deferrariisi. A, B, C, left
tibiotarsus of MACN-N-03 in cranial (A), medial (B), and lateral (C) view;
D, left tibiotarsus of MACN-N-11 in cranial view.

FIGURE 13.23. Tibiotarsus and fibula of Patagopteryx deferrariisi. A, B, right
tibiotarsus and fibula of MACN-N-11 in cranial (A), and lateral (B) view; C, D, left
tibiotarsus and fibula of MUCPv-48 in cranial (C), and lateral (D) view.

FIGURE 13.24. Tibiotarsus and fibula of Patagopteryx deferrariisi. A, B, left
tibiotarsus of MACN-N-03 in cranial (A), and medial (B) view; C, D, left
tibiotarsus of MUCPv-48 in cranial (C), and cranial (D) view. Abbreviations:
exg, extensor groove; fib, fibula; fic, fibular crest; fis, fibular spine; tubercle
for M. iliofibularis; lco, lateral condyle; lnc, lateral cnemial crest; mco,
medial condyle.

L.M. Chiappe, Patagopteryx - Page 15
The proximal articular surface is well preserved in MACN-N-10 (Figs. 13.25F, 13.26C). It has a
triangular shape with two well-defined, circular cotylae. The medial cotyla is significantly larger that the
lateral one. The area between both cotylae is slightly elevated, and a very weak intercotylar prominence
is present in the cranial border. In the cranial view, the medial cotyla is slightly more elevated than the
lateral one. The latter has an elevated lateral ridge.
The hypotarsus is restricted to the most proximal area. It is simple and poorly developed, without
any tendinal canals or calcaneal ridges. The hypotarsus is primarily developed behind the medial cotyla,
where it is caudally projected. In MACN-N-10, on the proximal surface of this projection, there is a
broken triangular surface that indicates the presence of a proximally projected hypotarsal process.
The tarsometatarsal shaft is straight and strongly compressed dorsoplantarily (Figs. 13.25B,
13.26D). In the mid-portion of MUCPv-48, the thickness is less than 50% of its width (Table 5). The
distal half of the shaft curves slightly cranially, in a way that, in lateral view, the cranial border is slightly
concave.
Cranially, somewhat distal to the proximal end, and on the boundary of metatarsal III and IV,
there is a small foramen perforating the tarsometatarsus (Figs. 13.25D, 13.26A). This foramen is most
likely homologous to the lateral proximal foramen of neornithine birds. At the same level of this
foramen, but between metatarsals II and III, there is a small fossa. The position of this fossa is identical
to that of the medial proximal foramen of Neornithes, although in Patagopteryx this fossa is not
perforated by any foramen.
The distal boundary of the proximal medial fossa is formed by swellings of metatarsals II and III.
This area may represent the area of attachment of M. tibialis cranialis, the main tarsometatarsal flexor
(Hudson, 1937; Raikow, 1985). Distal to these swellings there is a pronounced groove that reaches the
midshaft of the tarsometatarsus. This groove is well developed in MUCPv-48. Parallel to it, between
metatarsals III and IV, there is another longitudinal groove, although it is much less pronounced. It is in
the most proximal portion of this groove that the proximal external foramen is located.
In plantar view, the tarsometatarsus is transversely concave, especially in its proximal two-thirds
(Figs. 13.25C, E, 13.26B). Metatarsals II and IV are well-projected plantarily, defining a broad,
triangular depression, the floor of which is formed by metatarsal III. This depression narrows distally
down to the level of the articular facet of metatarsal I. This depression must have been the area of origin
of several digital muscles (e.g., M. flexor hallucis brevis, M. adductor digiti II, M. abductor digiti IV) as
well as of tendons of the flexor digital musculature (George and Berger, 1966). In the proximolateral
corner of this depression, the plantar opening of the proximal lateral foramen is visible.
Slightly above the distal end of this depression, at the level of the articulation for metatarsal I,
there is a swelling on the plantar face of metatarsal II (this is only slightly visible in MUCPv-48 but well-
developed in MACN-N-11). A similar structure, although less pronounced, is visible in certain
neornithine birds (e.g., Cathartes aura, Fulica rufifrons), despite the fact that in these examples the
swelling is proximal to the articular facet for metatarsal I. At the level of the swelling of metatarsal II, but
on the opposite side of the plantar depression, is the origin of a belt-like groove that is directed
laterodistally. This groove borders the lateral surface of the tarsometatarsus, and it reaches the lateral
fossa of the lateral trochlea. Most likely, this groove corresponds to the mark left by the tendon of M.
abductor digiti IV (George and Berger, 1966). If this interpretation is correct, it suggests a well-
developed M. abductor digiti IV in Patagopteryx, a muscle that in many groups of neornithine birds is
only slightly developed. Metatarsal I is small (Table 5). It articulates by a teardrop-shaped facet to the
medial border of metatarsal II, which is slightly displaced cranially. The proximal end of this facet is
only slightly distal to the midshaft.
The distal end of the tarsometatarsus is slightly narrower than the proximal end. Trochleae for
metatarsals II-IV are nearly in the same transverse plane. The trochlea for metatarsal III is the most
projected distally. Trochleae for metatarsals IV and II are slightly less projected, with the latter being
shorter than the former. Proximal to the trochlear notch of metatarsals III and IV, there is the cranial
opening of the distal vascular foramen (Fig. 13.26A). This foramen originates from a short extensor

FIGURE 13.25. Tarsometatarsus and pes of Patagopteryx
deferrariisi. A, B, C, right tarsometatarsus and pes of
MACN-N-11 in dorsal (A), medial (B), and plantar (C) view; D, E,
left tarsometatarsus and pes of MACN-N-11 in dorsal (D) and
plantar (E) view; F, G, proximal end of left tarsometatarsus of
MACN-N-10 in proximal (F) and distal (cross-section) (G) view;
H, tarsometatarsus and pes of MUCPv-48 in medial view.

FIGURE 13.26. Tarsometatarsus and pes of Patagopteryx deferrariisi. A,
left tarsometatarsus and pes of MACN-N-03 in dorsal view; B, right
tarsometatarsus and pes of MACN-N-11 in plantar view; C, proximal end of
left tarsometatarsus of MACN-N-10 in proximal view; D, left
tarsometatarsus and pes of MUCPv-48 in medial view. Abbreviations: dvf,
distal vascular foramen; hyp, hypotarsus; pvf, proximal vascular foramen;
I-IV, pedal digits I to IV.

L.M. Chiappe, Patagopteryx - Page 16
groove. The distal vascular foramen does not open on the plantar surface, but rather probably opens on
the actual notch between the trochleae; the foramen has not been exposed because of the difficulty of
removing the matrix filling the trochlear notch.
The robust trochlea for metatarsal III is the larger of the three and it projects cranially. This
trochlea bears a well-defined central furrow that is very broad plantarily. The lateral and medial trochlear
notches, which separate the trochlea for metatarsal III from the trochleae for metatarsals IV and II,
respectively, are very narrow. The trochlea for metatarsal II is somewhat more robust than the trochlea
for metatarsal IV. Its medial surface is flat, without a collateral ligamental fossa. This trochlea also lacks
a central furrow. The trochlea for metatarsal I possesses its lateral rim more distally projected than the
medial one. These borders are separated by a broad, shallow central furrow.
Pedal digitsThe foot of Patagopteryx is pamprodactyl (Raikow, 1985), with its four robust,
long digits directed cranially (Figs. 13.25, 13.26). The phalangeal formula is 2-3-4-5-x (see Padian
[1992] for the phalangeal formula designation), as is typical of theropods (including birds). Digit III, the
largest, is as long as the tarsometatarsus. Roughly 30% shorter than this digit is digit II, which is only
slightly longer than digit IV. Digit I, although the shortest, is still robust and well developed, being about
50% the length of digit III.
The ungual phalanges are well developed. Except for digit III, the ungual phalanges are the larger
phalanges of their respective digits. The ungual phalanges of digits I, III, and IV are only slightly curved.
This curvature is more pronounced for the ungual of digit II. Both lateral and medial surfaces bear a
longitudinal groove slightly displaced to the ventral margin. The flexor tubercles are weak.
The first phalanx of digit I, which is slightly shorter than the ungual, possesses on its proximal
end a tongue-like, ventral projection. Centered on its distal end is a broad, circular collateral ligamental
fossa. The proximal phalanx of digit II is very high in its proximal end. The proximal end of the dorsal
margin strongly embraces the trochlea for metatarsal II. Toward the distal end, the dorsal border of this
proximal phalanx diminishes in height. Slightly proximal to the distal articular facet, the dorsal margin is
excavated by a deep depression. A comparable depression is also present in the proximal phalanges of
digits III and IV, as well as in some intermediate phalanges, although it may be located more proximally.
Phalanx 2 of digit II has a short shaft but a long proximal projection of the ventral border. On its dorsal
surface, the fossa described for the proximal phalanx of this digit is not present. Next from this phalanx
is the ungual, which is large and laterally compressed. Like in the Late Cretaceous malagasy Rahonavis
(Forster et al., 1998), the articular surfaces of the phalanges of digit II are well developed. Patagopteryx,
however, does not show the raptorial specializations of the latter bird—its digit II is not more robust than
the others and its claw is not strongly curved. Phalanx 1 of digit III is the largest and more robust of all
the phalanges. Phalanges 2 and 3 of this digit successively diminish in size. The first four phalanges of
digit IV are small. Among these, the proximal and phalanx 4 are the largest. The first three phalanges
also have a depression on their dorsal surface, although in these phalanges this depresion is located in the
midshaft portion. The larger ungual phalanx has a pronounced proximal expansion from the ventral
margin that embraces the precedent phalanx for over half of its size.
PHYLOGENETIC RELATIONSHIPS
As pointed out in the introduction, the notion of a close relationship between Patagopteryx and
extant ratites was stressed in early discussions of this taxon (e.g., Bonaparte, 1986) and was retained in
its formal description (Alvarenga and Bonaparte, 1992; see also Alvarenga, 1993). Yet, as shown by
Chiappe (1995a), the characters used to support such a claim are either plesiomorphic or have been
misinterpreted.
Nessov and Pantheleev (1993) related the Coniacian Kuszholia mengi (Nessov, 1992) from the
Bissekty Formation at Dzhyrakuduk (Uzbekistan), a taxon erected on the basis of two fragmentary
synsacra, to Patagopteryx. These authors were followed by Kurochkin (1995b) who mentioned the
presence of common features between Patagopteryx, Kuszholia, and the Early Cretaceous Chaoyangia

L.M. Chiappe, Patagopteryx - Page 17
from northeastern China (Chapter 7), although the alleged common features were not specified. The
incomplete and poorly preserved nature of the known material of Kuszholia makes comparison to this
taxon difficult—indeed, even its avian nature is difficult to support given the available material. Yet, the
synsacrum of Kuszholia differs from that of Patagopteyx in the presence of pleurocoels and a concave
caudal articular surface.
In another recent paper, Kurochkin (1995a) also supported Alvarenga and Bonaparte’s (1992)
hypothesis of the palaeognath affinity of Patagopteryx, although in this case it is unclear whether
Kuszholia and Chaoyangia were also regarded as closely related to palaeognaths. The characters used by
Kurochkin (1995a) to support such a claim are: (1) absence of a bicipital crest; (2) open ilioischiadic
fenestra; (3) symmetrical lateral and medial ridges of the trochlea of metatarsal III; (4) presence of a
deltopectoral crest that begins at the humeral head, (5) thick postacetabular wing of the ilium; (6)
hypotarsus lacking canals and (7) proximally raised; (8) large cranial iliosacral crest; and (9) proximally
placed medial epicondyle. As pointed out by Kurochkin himself (1995a:Tab. 1), most of these characters,
if not all of them, appear to be primitive—characters 1-6 clearly occur in one or more basal taxa (e.g.,
Archaeopteryx, Confuciusornis, enantiornithines, and non-avian theropods). Furthermore, some of these
characters are difficult to divide into discrete character states. For example, Kurochkin’s “anterior ilio-
sacral crest,” which presumably refers to the dorsal iliac crest (see Baumel and Witmer, 1993), is simply
the dorsal border of the preacetabular wing of the ilium. Thus, differences between “large” and “small”
sizes of this border are rather difficult to distinguish as well as being highly arbitrary.
None of the aforementioned hypotheses were framed into a cladistic analysis and are therefore
difficult to evaluate on parsimonious grounds. Nevertheless, various cladistic analyses have supported
the sister-group relationship between Patagopteryx and Ornithurae (Chiappe, 1992, 1995a, b, 1996b;
Chiappe and Calvo, 1994) (Chapter 20) (Fig. 13.27). The hypothesis of ratite affinities has also been
challenged by recent histological studies documenting the presence of lines of arrested growth (LAGs) in
the compacta of the femora of Patagopteryx, which are typically absent in ornithurine birds, including
ratites (Chinsamy et al., 1994, 1995) (Chapter 18). Consequently, compelling osteological and
histological evidence supports the hypothesis that P. deferrariisi acquired flightlessness independently
from ratites and any other known flightless avian group (Chiappe, 1995a, b, 1996b).
PALEOBIOLOGY
Patagopteryx was a flightless, land bird. Estimations of its body mass based on the diameter of its
femur vary significantly depending on the equation used. Yalden’s (1984) equation (W = 13.25 x
D2.353; D = diameter) provides values of 2,211 and 2,152 grams for specimens MACN-N-11 and
MUCPv-48, respectively. These values are comparable to the mean mass of the male Brown Kiwi
(Apteryx australis; the female is usually heavier) (Dunning, 1993). Yet, Campbell and Marcus (1992)
standard equation (Wlog = - 0.065 + 2.411 x Clog; C = circumference) gives values of 3,672 and 3,334
grams for these two specimens, respectively. These masses exceed the mean value of female Brown
Kiwis given by Dunning (1993), although they are smaller than the maximum value registered by this
author (i.e., 3,850 grams) for females of this species.
Flightlessness has been a common theme in the history of birds. At least 12 avian lineages have
extant flightless representatives and a large variety of extinct and recently extirpated birds are known to
have been, or have been interpreted as, flightless (Raikow, 1985; Livezey, 1995). The large majority of
these are land birds.
Regardless of the group in question and its life-style, flightlessness is usually associated with a
number of distinct morphological attributes (James and Olson, 1983; Diamond, 1981, 1991; Raikow,
1985; Feduccia, 1996). Most evident among these is the reduced fore limb, which is extreme in some
cases, such as hesperornithiforms and certain ratites. In comparison to flighted birds, the fore limb of a
flightless bird typically has a substantially smaller carpometacarpus and ulna-radius, and the relative
proportions between these mid-wing elements and the humerus, and to the entire fore limb, are very
different from those of a flighted counterpart (Livezey, 1990). In those birds in which the fore limb is

FIGURE 13.27. Cladogram depicting the relationship of Patagopteryx
deferrariisi to other basal birds (modified from Chiappe, 1996b).

L.M. Chiappe, Patagopteryx - Page 18
heavily reduced, wing elements bear simple and rudimentary structures. The thoracic girdle also shows
radical differences from that of a flighted bird. Flight muscles form a minor portion of the body mass
(e.g., 0.13% in the North Island Brown Kiwi; see McNab, 1996) and their area of origin is highly
reduced: the sternal carina and the furcula are absent in a variety of flightless birds (e.g., ratites,
Diatryma, moa-nalos). The scapula and coracoid articulate to each other in an obtuse angle, and in certain
cases (e.g., ratites, Diatryma) these bones are fused, forming a scapulocoracoid. Furthermore, in
flightless birdsparticularly the terrestrial onesthe hind limb is robust and proportionately larger than
in a flighted counterpart.
In Patagopteryx, the ratio between maximum length of fore limb and hind limb (1:2.2), and the
proportion between wing elements (e.g., humerus larger than ulna-radius) and of the wing to the rest of
the skeleton, indicates its incapacity to fly. Moreover, Patagopteryx possesses several features present in
flightless birds: small carpometacarpus, widely separated sternal articulations of coracoids, and absence
of furcula and sternal keel. Yet, the fore limb and thoracic girdle of Patagopteryx are considerably less
reduced than those of ratites and other forms with an extremely simplified flight apparatus. For example,
the articulation between the scapula and coracoid forms an acute angle, and these bones are not fused to
each other.
The hind limb and pelvis of Patagopteryx are robust and very well-developed in proportion to the
rest of the skeleton. The hind limb elements exhibit prominent structures for either origin or insertion of
muscles (e.g., strong intermuscular lines, robust cnemial crest, prominent tubercle for M. iliofibularis),
which point at a strong hind limb musculature. This, combined with the reduction of the fore limb,
indicates that the locomotion of Patagopteryx was governed by its hind limbs.
Flightless birds are essentially specialized for two types of locomotion: foot-propelled aquatic
divers, and terrestrial locomotion. Patagopteryx lacks the most obvious specializations of foot-propelled,
aquatic birds such as the compression and elongation of the postacetabular portion of the pelvis and
synsacrum, the proximal development of the cnemial crests, and the lateral compression of the
tarsometatarsus (Raikow, 1985). On the contrary, the general hind limb anatomy of Patagopteryx
approaches that of obligate terrestrial birds: the pedal anatomy is generalized and the foot, although
pamprodactyl, is functionally tridactyl. Undoubtedly, Patagopteryx was an obligate terrestrial bird. Yet,
its hind limb does not show the specializatons for high speed present in other obligate terrestrial birds
such as certain ratites (e.g., Rheidae, Dromaeus, Struthio) and most phorusrhacids (Alvarenga and
Bonaparte, 1992). For example, in Patagopteryx, the tarsometatarsus is significantly shorter than the
tibiotarsus. Also, fast runners possess reduced toes (both in size and number) that minimize the surface of
contact with the substrate (Raikow, 1985). This is not the case in Patagopteryx, in which the toes are
large. This suggests that Patagopteryx was probably unable to achieve the high running speeds of many
flightless birds.
The phylogenetic position of Patagopteryx (Fig. 13.27) strongly suggests that this bird evolved
flightlessness from flighted ancestors. Indeed, a combined set of functional and phylogenetic analyses
indicates that all flightless birds, with the possible exception of the alvarezsaurids, if they were to be
considered avian (Chapter 4), are descendants of flighted ancestors. Secondary flightlessness has often
been viewed within an adaptational context, as a functional modification shaped by a specific force:
natural selection (adaptation is best defined as an apomorphic function—in this case non-function—due
to natural selection; see Coddington, 1988). The occurrence of flightlessness in many different lineages
of birds and in particular, its recurrence in relatively small monophyletic groups (e.g., more than 25% of
rails; see Diamond, 1981; Feduccia, 1996) suggests that in some instances, flightlessness may qualify as
an adaptation (Andersen, 1995). This is also supported by studies on monophyletic lineages of oceanic
islands (e.g., Australasian teals, Hawaiian moa-nalos and geese; see Livezey, 1990; Olson and James,
1991) which have shown a strict correlation between flightlessness and a historical change in
environmental conditions (e.g., colonization of a restricted, insular system). Yet, caution is advised

L.M. Chiappe, Patagopteryx - Page 19
because even in the best scenarios, phylogenetic studies may be insufficient to demonstrate the “truth” of
a given adaptational proposal (Wenzel and Carpenter, 1994).
Attempts to explain the causes and processes that led to secondary flightlessness have focused
more on island lineages (see Olson, 1973; Feduccia, 1980, 1996; Diamond, 1981; James and Olson,
1983; Livezey, 1993; McNab, 1994). It is widely known that most oceanic islands lack aboriginal
mammalian predators, and that most of them are devoid of major geographic barriers or environmental
heterogeneity that complicate the dispersion of their inhabitants (Diamond, 1981). Under these
circumstances, secondary flightlessness has been viewed as a byproduct of energy conservation, reducing
the mass of metabolically expensive thoracic muscles (Olson, 1973; Feduccia, 1980, 1996; Diamond,
1981, 1991; Livezey, 1990, 1992, 1993; McNab, 1994, 1996). McNab (1994, 1996) has shown that in
several instances (e.g., rails and kiwis), flightlessness is associated with low basal metabolic rates
(average rate of energy consumption under standard conditions), linking the evolution of this functional
modification to a reduced rate of energy expenditure. Olson (1973), Feduccia (1980, 1996), Livezey
(1990, 1992, 1993, 1995) and others (e.g., Diamond, 1981, 1991) have suggested that heterochronic
processes (in particular, paedomorphosis) may have been responsible for the acquisition of flightlessness.
In contrast to the case of insular forms, the evolution of flightlessness in continental birds has been
usually linked to an increase in size (e.g., Dromornithidae, Diatryma, Phorusrhacidae) and sometimes
speed (e.g., Struthio, Rhea, Dromaeus). If factors leading to secondary flightlessness in island lineages
are intricate to dissect, those of continental birds appear to be even more complex. Yet, regardless of its
causes, it is possible that elimination of the restrictions imposed by the high metabolic demands of flight
allowed the increase in size (Raikow, 1985). Again, heterochrony has also been claimed as an important
process in the development of flightlessness in continental birds (Feduccia, 1980; James and Olson,
1983). Although neoteny has been highlighted as the specific underlaying process in the heterochronic
evolution of flightlessness, more than one underlaying process (including combinations of paedomorphic
and peramorphic processes) is likely to have been at play (Livezey, 1995).
Interestingly, Patagopteryx appears to be a singular case in the evolution of secondary
flightlessness, not fitting neatly in either of the two models in which flightlessness has often been
accounted. Patagopteryx lived in a continental environment—there are vast deposits of contemporaneous,
continental rocks surrounding Boca del Sapo. Yet, it was neither highly specialized for running nor
achieved a large size, despite being sympatric with a variety of predators—dinilysid snakes, notosuchid
crocodiles, small non-avian theropods such as Velocisaurus may have been potential predators. The
morphology of the wing and thoracic girdle of Patagopteryx fits well with the predictions expected for
paedomorphic skeletal change (see Livezey, 1990, 1993), but such a process theory can be formulated
only if close relatives of the lineage under study are known. Unfortunately, this is not the case in
Patagopteryx. Perhaps, future discoveries of close relatives of this most interesting bird can help us
understand the evolution of flightlessness in this ancient lineage.
ACKNOWLEDGEMENTS
I am grateful to S. Cerruti, S. Copeland, L. Meeker, and A. Rizzetto for preparation of many of
the illustrations. S. Orell and S. Copeland edited the final version of this manuscript. Thanks also to J.
Bonaparte, L. Salgado, and J. Calvo for letting me study material under their care. Support for this
research was provided by grants from the Guggenheim Foundation, the Dinosaur Society, the McKenna
Foundation, and the Consejo Nacional de Investigaciones Científicas y Técnicas (Argentina).

L.M. Chiappe, Patagopteryx - Page 20
LITERATURE CITED
Alvarenga, H. M. F. 1993. A origem das aves seus fósseis; pp. 16–26 in M. A. de Andrade (ed.), A Vida
das Aves. Editora Littera Maciel, Belo Horizonte.
——— and J. F. Bonaparte. 1992. A new flightless land bird from the Cretaceous of Patagonia; pp. 51–
64 in K. E. Campbell (ed.), Papers in Avian Paleontology, Honoring Pierce Brodkorb. Natural
History Museum of Los Angeles County, Science Series 36.
Andersen, N. M. 1995. Cladistic inference and evolutionary scenarios: locomotory structure, function,
and performance in water striders. Cladistics 11:279–295.
Baumel, J. J., A. S. King, A. M. Lucas, J. E. Breazile, and H. E. Evans. 1979. Nomina Anatomica
Avium. Academic Press, London, 637 pp.
——— and L. M. Witmer. 1993. Osteologia; pp. 45–132 in J. J. Baumel, A. S. King, J. E. Breazile, H. E.
Evans, and J. C. Vanden Berge (eds.), Handbook of Avian Anatomy: Nomina Anatomica Avium,
Second edition. Publications of the Nuttal Ornithological Club 23.
Bock, W. J. 1963. The cranial evidence for ratite affinities. Proceedings of the 13
th
International
Ornithological Congress: 39-54.
Bonaparte, J. F. 1986. History of the terrestrial Cretaceous vertebrates of Gondwana. IV Congreso
Argentino de Paleontología y Bioestratigrafía, Actas 2:63–95.
——— 1991. Los vertebrados fósiles de la Formación Río Colorado de Neuquén y cercanias, Cretácico
Superior, Argentina. Revista del Museo Argentino de Ciencias Naturales "Bernardino Rivadavia"
(Paleontología). 4(3):17–123.
Calvo, J. O. 1991. Huellas de dinosaurios en la Formación Río Limay (Albiano-Cenomaniano), Picun
Leufu, Provincia del Neuquén, República Argentina (Ornithischia-Saurischia: Sauropoda-
Theropoda). Ameghiniana 28(3–4):241–258.
Campbell, K. E., Jr., and L. Marcus. 1992. The relationship of hind limb bone dimensions to body weight
in birds; pp. 395–412 in K. E. Campbell (ed.), Papers in Avian Paleontology, Honoring Pierce
Brodkorb. Natural History Museum of Los Angeles County, Science Series 36.
Cazau, L. B., and M. A. Uliana. 1973. El Cretácico superior continental de la cuenca Neuquina. V
Congreso Geológico Argentino 3:131–163.
Chiappe, L. M. 1987. Aves Mesozoicas. Ciencias Naturales (MACN) 1:1–7
——— 1989. A flightless bird from the Late Cretaceous of Patagonia (Argentina). Archosaurian
Articulations 1(10):73–77.
——— 1991. Cretaceous birds of Latin-America. Cretaceous Research 12:55–63.
——— 1992. Osteología y sistemática de Patagopteryx deferrariisi Alvarenga y Bonaparte (Aves) del
Cretácico de Patagonia. Filogenia e historia biogeográfica de las aves cretácicas de América del Sur.
Ph.D. dissertation. Universidad de Buenos Aires, Argentina, 429 pp.
——— 1995a. The first 85 million years of avian evolution. Nature 378:349–355.
——— 1995b. The phylogenetic position of the Cretaceous birds of Argentina: Enantiornithes and
Patagopteryx deferrariisi. Courier Forschungsinstitut-Senckenberg 181:55–63.
——— 1995c. A diversity of early birds. Natural History 104(6):52–55.
——— 1996a. Late Cretaceous birds of southern South America: anatomy and systematics of
Enantiornithes and Patagopteryx deferrariisi; pp. 203–244 in G. Arratia (ed.), Contributions of
Southern South America to Vertebrate Paleontology, Münchner Geowissenschaftliche Abhandlungen
Volume 30.
——— 1996b. Early avian evolution in the southern hemisphere: Fossil record of birds in the Mesozoic
of Gondwana. Memoirs of the Queensland Museum 39:533–556.
———, Ji S., Ji Q., and M. A. Norell. 1999. Anatomy and systematics of the Confuciusornithidae (Aves)
from the late Mesozoic of northeastern China. Bulletin of the American Museum Novitates 242:1–
89.

L.M. Chiappe, Patagopteryx - Page 21
——— and J. O. Calvo. 1994. Neuquenornis volans, a new Upper Cretaceous bird (Enantiornithes:
Avisauridae) from Patagonia, Argentina. Journal of Vertebrate Paleontology 14(2):230–246.
———, Norell, M. A., and J. Clark. 1998. The skull of a new relative of the stem-group bird
Mononykus. Nature 392:275–278.
Chinsamy, A., L. M. Chiappe, and P. Dodson. 1994. Growth rings in Mesozoic birds. Nature 368:196–
197.
———, ———, and ——— 1995. Mesozoic avian bone microstructure: physiological implications.
Paleobiology 21(4):561–574.
Clarke, J., Dingus, L., and L. M. Chiappe. 1998. Review of Upper Cretaceous deposits at Neuquén,
Argentina (Bajo de la Carpa Member, Río Colorado Formation). Journal of Vertebrate Paleontology
18(3):34A.
Coddington, J. A. 1988. Cladistic tests of adaptational hypotheses. Cladistics 4:3–22.
Cruz, C. E., P. Condat, E. Kozlowski, and R. Manceda. 1989. Analisis estratigráfico secuencial del
Grupo Neuquén (Cretácico superior) en el valle del Río Grande, Provincia de Mendoza. I Congreso
Argentino de Hidrocarburos, Actas 2:689–714.
Currie, P. J. 1995. New information on the anatomy and relationships of Dromaeosaurus albertensis
(Dinosauria, Theropoda). Journal of Vertebrate Paleontology 15(3):576–591.
Danderfer, J. C., and P. Vera. 1992. Geología. Boletín del Servicio Geológico Neuquino 1:23–45.
Diamond, J. 1981. Flightlessness and fear of flying in island species. Nature 293:507–508.
——— 1991. Twilight of Hawaiian birds. Nature 353:505–506.
Dingus, L., J. Clarke, G. R. Scott, C. C. Swisher III, L. M. Chiappe, and R. A. Coria. 2000.
Stratigraphy and magnetostratigraphic/faunal constraints for the age of sauropod embryo-bearing
rocks in the Neuquén Group (Late Cretaceous, Neuquén Province, Argentina). American Museum
Novitates 3290:1-11.
Dunning, J. B., Jr. 1993. CRC Handbook of Avian Body Masses. CRC Press, Boca Raton, 371 pp.
Feduccia, A. 1980. The Age of Birds. Harvard University Press, Cambridge, 196 pp.
——— 1996. The Origin and Evolution of Birds. Yale University Press, New Haven, 420 pp.
Forster, C. A., L. M. Chiappe, D. W. Krause, and S. D. Sampson. 1996. The first Cretaceous bird from
Madagascar. Nature 382:532–534.
Gasparini, Z. 1971. Los Notosuchia del Cretácico de América del Sur como un nuevo infraorden de los
Mesosuchia (Crocodilia). Ameghiniana 8(2):83–103.
George, J. C., and A.J . Berger. 1966. Avian myology. Academic Press, New York, 499 pp.
Heredia, S. and J. O. Calvo. 1997. Sedimentitas eólicas en la Formación Río Colorado (Grupo Neuquén)
y su relación con la fauna del Cretácico superior. Ameghiniana 34(1):120.
Holmes, B. E. 1963. Variation in the muscles and nerves of the leg in two genera of Grouse
(Tympanuchus and Pedioectes). University of Kansas Publications, Museum of Natural History
12(9):363–474.
Hudson, G. E. 1937. Studies on the muscles of the pelvic appendage in birds. The American Midland
Naturalist 18(1):1–108.
James, H., and S. L. Olson. 1983. Flightless birds. Natural History 92(9):30–40.
Kurochkin, E. N. 1995a. Morphological differentiation of palaeognathous and neognathous birds.
Courier Forschungsinstitut Senckenberg 181:79–88.
——— 1995b. The assemblage of the Cretaceous birds in Asia; pp. 203–208 in A. Sun and Y. Wang
(eds.), Sixth Symposium on Mesozoic Terrestrial Ecosystems and Biota, Short Papers. China Ocean
Press, Beijing.
Legarreta, L., and C. Gulisano. 1989. Análisis estratigráfico secuencial de la Cuenca Neuquina (Triásico
Superior-Terciario Inferior); pp. 221–243 in G. Chebli and L. Spalletti (eds.), Cuencas Sedimentarias
Argentinas. Serie Correlación Geológica Volume 6.
Livezey, B. C. 1990. Evolutionary morphology of flightlessness in the Auckland Islands teal. Condor
92:639–673.

L.M. Chiappe, Patagopteryx - Page 22
——— 1992. Morphological corollaries and ecological implications of flightlessness in the Kakapo
(Psittaciformes: Strigops habroptilus). Journal of Morphology 213:105–145.
——— 1993. An ecomorphological review of the dodo (Raphus cucullatus) and solitaire (Pezophaps
solitaria), flightless Columbiformes of the Mascarene Islands. Journal of Zoology, London 230:247–
292.
——— 1995. Heterochrony and the evolution of avian flightlessness; pp. 169-193 in K. J. McNamara
(ed.), Evolutionary change and heterochrony. John Wiley and Sons, New York.
Marsh, O. C. 1880. Odontornithes: A monograph on the extinct toothed birds of North America. Rept.
Geol. Expl. 40th Parallel. Government Printing Office, Washington D.C., 201 pp.
Martin, L. D. 1983. The origin and early radiation of birds; pp. 291–338 in A. H. Bush and G. A. Clark,
Jr. (eds.), Perspectives in Ornithology. Cambridge Universty Press, New York.
——— 1991. Mesozoic birds and the origin of birds; pp. 485–539 in H.-P. Scultze and L. Trueb (eds.),
Origins of the Higher Groups of Tetrapods: Controversy and Consensus. Comstock Publishing
Associates, Ithaca.
——— and J. Tate, Jr. 1976. The skeleton of Baptornis advenus (Aves:Hesperornithiformes).
Smithsonian Contributions to Paleobiology 27:35–66.
McGowan, C. 1979. The hind limb musculature of the Brown Kiwi, Apteryx australis mantelli. Journal
of Morphology 160:33–74.
McNab, B. K. 1994. Energy conservation and the evolution of flightlessness in birds. American
Naturalist 144(4):628–642.
——— 1996. Metabolism and temperature regulation of Kiwis (Apterygidae). Auk 113(3):687–692.
Musacchio, E. A. 1993. Use of global time scale in correlating nonmarine Cretaceous rocks in southern
South America. Cretaceous Research 14:113–126.
Nessov, L. A. 1992. Review of localities and remains of Mesozoic and Paleogene birds of the USSR and
the description of new findings. Russian Journal of Ornithology 1(1):7–50.
——— and B. V. Pantheleev. 1993. On the similarity of the Late Cretaceous ornithofauna of South
America and Western Asia; pp. 84–94 in R. L. Potapov (ed.), Ecology and Fauna of the Palaeartic
Birds. Proceedings of the Zoological Institute, Russian Academy of Sciences 252 [in Russian].
Novas, F. E. 1996. Alvarezsauridae, Cretaceous maniraptorans from Patagonia and Mongolia. Memoirs
of the Queensland Museum 39:675–702.
Olson, S. L. 1973. Evolution of the rails of the South Atlantic islands (Aves: Rallidae). Smithsonian
Contributions to Zoology 152:1–53.
——— and H. F. James. 1991. Descriptions of thirty-two new species of birds from the Hawaiian
Islands: Part I. Non-passeriformes. Ornithological Monographs 45:1–88.
Padian, K. 1992. A proposal to standardize tetrapod phalangeal formula designations. Journal of
Vertebrate Paleontology 12(2):260–262.
Perle, A., L. M. Chiappe, R. Barsbold, J. M. Clark, and M. A. Norell. 1994. Skeletal morphology of
Mononykus olecranus (Theropoda: Avialae) from the Late Cretaceous of Mongolia. American
Museum Novitates 3105:1–29.
Raikow, R. J. 1985. Locomotor system; pp. 57–147 in A. S. King and J. McLelland (eds.) Form and
Function in Birds. Academic Press, San Diego.
Ramos, V. A. 1981. Descripción geológica de la Hoja 33c, Los Chihuidos Norte. Boletín del Servicio
Geológico Nacional, Buenos Aires 182:1–103.
Rowe, T. 1986. Homology and evolution of the deep dorsal thigh musculature in birds and other Reptilla.
Journal of Morphology 189:327–346.
Saiff, E. I. 1983. The anatomy of the middle ear region of the Rheas (Aves: Rheiformes, Rheidae).
Historia Natural 3(6):45–55.
Sereno, P. C., and F. E. Novas. 1993. The skull and neck of the basal theropod Herrerasaurus
ischigualastensis. Journal of Vertebrate Paleontology 13(4):451–476.

L.M. Chiappe, Patagopteryx - Page 23
Simonetta, A. 1960. On the mechanical implications of the avian skull and their bearing on the evolution
and classification of birds. Quarterly Review of Biology 35(3):206–220.
Smith Woodward, A. S. 1896. On two Mesozoic crocodilians, Notosuchus (genus novum) and
Cynodontosuchus (gen. nov.) from the red sandstones of the territory of Neuquén (Argentina).
Anales del Museo de La Plata 4:1–20.
——— 1901. On some extinct reptiles from Patagonia of the genera Miolania, Dinilysia and
Genyodectes. Proceedings of the Zoological Society of London 1:169–184.
Starck, M. J. 1995. Comparative anatomy of the external and middle ear of palaeognathous birds.
Advances in Anatomy, Embryology and Cell Biology 131:1–137.
Uliana, M. A., and D. Dellapé. 1981. Estratigrafía y evolución paleoambiental de la sucesión
Maastrichtiano-Eoterciaria del Engolfamiento Neuquino (Patagonia septentrional). Actas VIII
Congreso Geológico Argentino 3:673-711.
Verheyen, R. 1960. Considerations sur la colonne vertebrale des oiseaux (non-passeres). Bulletin Institut
Royal des Sciences Naturelles de Belgique 36(42):1–24.
Webb, M. 1957. The ontogeny of the cranial bones, cranial peripheral and cranial parasympathetic
nerves, together with a study of the visceral muscles of Struthio. Acta Zoologica 38:81–203.
Weishampel, D. B., Dodson, P. and H. Osmólska. 1990. The Dinosauria. University of California Press,
Berkeley, 733 pp.
Wenzel, J. W., and J. M. Carpenter. 1994. Comparing methods: adaptive traits and tests of adaptation;
pp. 79–101 in P. Egelton and R. Bane-Wright (eds.) Phylogenetics and Ecology. Academic Press,
London.
Yalden, D. W. 1984. What size was Archaeopteryx? Zoological Journal Linnean Society 82:177–188.
Zusi, R. L. 1985. Muscles of the trunk and tail in the Noisy Scrub-bird, Atrichornis clamosus and Superb
Lyrebird, Menura novaehollandiae (Passeriformes: Atrichornithidae and Menuridae). Records of the
Australian Museum 37:229–242.
——— and G. D. Bentz. 1984. Myology of the Purple-throated Carib (Eulampis jugularis) and the
hummingbirds (Aves: Trochilidae). Smithsonian Contributions to Zoology 385:1–70.
——— and R. W. Storer. 1969. Osteology and myology of the head and neck of the pied-billed grebes
(Podilymbus). Miscellaneous Publications Museum of Zoology, University of Michigan 139:1–49.