The position of the claws in Noasauridae (Dinosauria: Abelisauroidea) and its implications for abelisauroid manus evolution

Abstract

In this note we reassess the position of putative pedal phalanges of some South American noasaurid theropods (Abelisauroidea). Noasaurids were considered as to be distinctive abelisauroids with a peculiar "sickle claw" on the second toe of the foot, convergently developed with that of deinonychosaurians. Among noasaurids, the Argentinean species Noasaurus leali (latest Cretaceous) and Ligabueino andesi (Early Cretaceous) are known from incomplete specimens, including dissarticulated non-ungueal phalanges, and, in N. leali, a claw. A detailed overview of these elements indicates that the supposed raptorial claw of the second pedal digit of N. leali actually belongs to the first or second finger of the manus, and the putative pedal non-ungual phalanges of both genera also pertain to the manus. Thus, the new interpretations of noasaurid pedal morphology blur the distinctions between Noasauridae and Velocisauridae proposed by previous authors. Finally, we suggest, on the basis of phalangeal and metacarpal morphology, that abelisaurids probably lost their manual claws by means of the loss of function of the HOXA11 and HOXD11 genes. Thus Noasauridae differs from Abelisauridae in retaining plesiomorphic long forelimbs with well developed claws, as occurs plesiomorphically in most basal theropods (e. g., Coelophysis).

Full text

RESEARCH PAPER
The position of the claws in Noasauridae (Dinosauria:
Abelisauroidea) and its implications for abelisauroid manus
evolution
Federico L. Agnolin •Pablo Chiarelli
Received: 14 April 2009 / Accepted: 24 October 2009
ÓSpringer-Verlag 2009
Abstract In this note we reassess the position of putative
pedal phalanges of some South American noasaurid thero-
pods (Abelisauroidea). Noasaurids were considered as to be
distinctive abelisauroids with a peculiar ‘‘sickle claw’’ on
the second toe of the foot, convergently developed with that
of deinonychosaurians. Among noasaurids, the Argentinean
species Noasaurus leali (latest Cretaceous) and Ligabueino
andesi (Early Cretaceous) are known from incomplete
specimens, including dissarticulated non-ungueal pha-
langes, and, in N. leali, a claw. A detailed overview of these
elements indicates that the supposed raptorial claw of the
second pedal digit of N. leali actually belongs to the first or
second finger of the manus, and the putative pedal non-
ungual phalanges of both genera also pertain to the manus.
Thus, the new interpretations of noasaurid pedal morphol-
ogy blur the distinctions between Noasauridae and Veloci-
sauridae proposed by previous authors. Finally, we suggest,
on the basis of phalangeal and metacarpal morphology, that
abelisaurids probably lost their manual claws by means of
the loss of function of the HOXA11 and HOXD11 genes.
Thus Noasauridae differs from Abelisauridae in retaining
plesiomorphic long forelimbs with well developed claws, as
occurs plesiomorphically in most basal theropods (e.g.,
Coelophysis).
Keywords Abelisauroidea Noasauridae Phalanges 
Cretaceous Argentina
Kurzfassung In der vorliegenden Studie wird eine Ne-
ubewertung der mutmaßlichen Position der Zehenglieder
su
¨damerikanischer noasaurider Theropoden (Abelisauroidea)
pra
¨sentiert. Noasauride werden als unverwechselbare Abelis-
auroide mit einer besonderen ‘‘Sichelkralle’’ am zweiten Zeh
des Fußes, die konvergent zu der der Deinonychosaurier ent-
stand ist, betrachtet. Die beiden argentinischen Arten Noa-
saurus leali (oberste Kreide) und Ligabueino andesi
(Unterkreide), die beide zu den Noasauriden geho
¨ren, sind nur
durch unvollsta
¨ndigen Exemplaren einschließlich nicht-ung-
ualer Zehenglieder und einer Kralle von N. leali belegt. Eine
ausfu
¨hrliche Untersuchung dieser Elemente deutet darauf hin,
dass die angeblich ra
¨uberische Kralle der zweiten Zehe von
N. leali tatsa
¨chlich zum ersten oder zweiten Finger der Hand
geho
¨rt. Ebenso geho
¨ren auch die nicht-ungualen Zehenglieder
beider Taxa zur Hand. Die unterschiedliche Fußmorphologie,
wie sie von verschiedenen Autoren vorgeschlagen wurde,
beeintra
¨chtigt die Unterscheidung zwischen Noasauriden und
Velocisauriden. Wir vermuten auf Grund der Phalangen- und
Metacarpaliamorphologie, dass der Verlust der Funktion der
HOXA11- und HOXD11-Gene mo
¨glicherweise fu
¨rdasFeh-
len von Fingerkrallen bei den Abelisauriden verantwortlich
ist. Somit unterscheiden sich die Noasauriden von den
Abelisauriden durch den Besitz der plesiomorphen langen
Vorderextremita
¨ten mit gut-entwickelten Krallen, wie es auch
bei basalen Theropoden (z.B. Coelophysis) der Fall ist.
Schlu
¨sselwo
¨rter Abelisauroidea Noasauridae 
Phalangen Kreide Argentinien
F. L. Agnolin (&)
Laboratorio de Anatomı
´a Comparada y Evolucio
´n de los
Vertebrados, Museo Argentino de Ciencias Naturales
‘‘Bernardino Rivadavia’’, Avenida A
´ngel Gallardo 470
(C1405DJR), Buenos Aires, Argentina
e-mail: fedeagnolin@yahoo.com.ar
F. L. Agnolin P. Chiarelli
Fundacio
´n de Historia Natural ‘‘Fe
´lix de Azara’’,
Departamento de Ciencias Naturales y Antropologı
´a,
CEBBAD, Universidad Maimo
´nides, Valentı
´n Virasoro 732
(1405), Buenos Aires, Argentina
e-mail: pablochi@hotmail.com
123
Pala
¨ontol Z
DOI 10.1007/s12542-009-0044-2

Abbreviations
FMNH PR Field Museum of Natural History, Chicago,
USA
MACN-PV-N Coleccio
´n de Paleontologı
´a de Vertebrados,
Provincia de Neuque
´n,MuseoArgentinode
Ciencias Naturales ‘‘Bernardino Rivadavia’’,
Buenos Aires, Argentina
PVL Coleccio
´n de Paleontologı
´a de Vertebrados,
Instituto Miguel Lillo, Tucuma
´n, Argentina
Introduction
Abelisauroidea is a group of very specialized and bizarre
carnivorous dinosaurs that dominated Gondwanan preda-
tory niches along the Late Cretaceous (Bonaparte 1986,
1991a; Sereno et al. 2004). These dinosaurs inhabited ter-
restrial ecosystems together with a large array of endemic
and allochtonous theropods, for example, basal tetanurans
(e.g., Carcharodontosauridae, Megaraptor; Apesteguı
´a
2002; Novas et al. 2005a), and coelurosaurs (e.g., Alv-
arezsauridae; Bonaparte 1991a,b; Novas 1997). At least in
Early and early Late Cretaceous times (Aptian through
Cenomanian), together with the Carcharodontosauridae,
abelisaurids constituted the dominant predatory dinosaurs
of southern continents (Bonaparte 1986).
To date, Abelisauroidea comprises two morphologically
divergent theropod subclades: the mid-sized Abelisauridae
and the small to mid-sized Noasauridae (Bonaparte 1991a),
a phylogenetic arrangement recently confirmed by multiple
cladistic analyses (Wilson et al. 2003; Sereno et al. 2004;
Carrano and Sampson 2008; Canale et al. 2009). However, a
few authors (Bonaparte 1991b; Agnolin et al. 2003) pro-
posed the existence of a third abelisauroid subclade: Ve-
locisauridae Bonaparte 1991b on the basis of the poorly
known genus Velocisaurus. As originally understood, one of
the most striking features of Noasauridae was thought to be
the presence of a raptorial ‘‘sickle claw’’ assigned to the
second pedal digit, morphologically convergent with that of
deinonychosaurs (Bonaparte and Powell 1980; Bonaparte
1996; see Gauthier 1986). In the original description of
Noasaurus leali, Bonaparte and Powell (1980) described a
single non-ungual phalanx as belonging to the second pedal
digit. Subsequently, Bonaparte (1996) described Ligabueino
andesi, and reported the presence of pedal pre-ungual pha-
langes of uncertain anatomical position. Lately, several
isolated ungual and other isolated non-ungual phalanges
attributed to noasaurids were described from the latest
Cretaceous of India and Madagascar (Carrano et al. 2002;
Novas et al. 2004).
In this note we review the morphology of the known
phalanges of Noasaurus leali (El Brete locality, Salta
Province, northwestern Argentina; Lecho Formation,
Maastrichtian, latest Cretaceous; Bonaparte and Powell
1980)andLigabueino andesi (La Amarga locality, Neuque
´n
Province, northern Patagonia, Argentina; La Amarga
Formation, Barremian-early Aptian, Early Cretaceous;
Bonaparte 1996), and we add some new interpretations on
noasaurid and abelisaurid manus morphology.
In this paper we follow the phylogenetic taxonomy of
Wilson et al. (2003) with the modifications introduced by
Carrano and Sampson (2008). We also use the anatomical
terminology employed by Carrano (2007).
Reappraisal of the noasaurid manual and pedal
anatomy
In the following section we discuss the morphology of the
supposed pedal phalanges of both Noasaurus leali, and
Ligabueino andesi. Because the phalanges of these genera
have been described by previous authors this brief over-
view emphasizes traits that allow us to recognize the
manual or pedal nature of noasaurid phalanges.
Ungual phalanges
Recent works have stressed the apomorphic morphology
present in the abelisaurid pedal claws. As observed by
Novas and Bandyopadhyay (2001), these phalanges bear a
well defined ‘‘Y’’-shaped system of lateral grooves, a lat-
eral tuberosity, and a deep ventral concavity, presumably
for flexor tendon attachment (see also Novas et al. 2005b;
Maganuco et al. 2008) that probably replaced in function
the flexor tubercle exhibited by other theropods (Bonaparte
and Powell 1980).
On the other hand, noasaurid pedal unguals are very
poorly known, being represented only by disarticulated
specimens. The first described purported noasaurid pedal
ungual phalanx belonged to Noasaurus leali (Bonaparte
and Powell 1980). It has an unusual shape for a theropod
claw, and was regarded as homoplastic with that of the
dromaeosaurid and troodontid ‘‘sickle claw’’ pedal digit II
(Bonaparte and Powell 1980). Curiously, the Noasaurus
ungual strongly differs from all known abelisauroid pedal
claws (both abelisaurid and noasaurid; e.g. those of
Aucasaurus,Ilokelesia,Majungasaurus, and Masiakasau-
rus; Coria and Salgado 2000; Coria et al. 2002; Carrano
et al. 2002; Carrano 2007) in lacking the ‘‘Y’’-shaped
system of lateral grooves and bump, the large and pro-
truding proximodorsal lip, and also in the extreme curva-
ture of the claw blade (Fig. 1; Novas and Bandyopadhyay
2001; Sampson et al. 2001; Novas et al. 2004,2005b).
These differences were considered in preliminary analyses
(Agnolin et al. 2004; Carrano et al. 2004; Carrano and
F. L. Agnolin, P. Chiarelli
123

Sampson 2008) to be suggestive of the possibility that the
Noasaurus claw actually belongs to the manus rather than
the pes. Furthermore, the Noasaurus claw shows several
traits that are atypical for a theropod pedal claw, but con-
sistent with manual unguals (Fig. 2), as follows:
1 proximal articular surface dorsoventrally tall and
mediolaterally flattened, being deeply concave when
viewed medially or laterally (dorsoventrally extended
and slightly concave in most theropod pedal unguals,
including those of dromaeosaurids and troodontids);
2 proximal articular surface with a sharp median keel
(low-keeled or smooth proximal surface in theropod
pedal claws, especially in those of Abelisauridae);
3 ungual oval cross-section (subtriangular cross-section
in theropod pedal unguals); and
4 strong curvature ungual, describing an arch of nearly
90°along its ventral margin (nearly ventrally flat in
most theropod pedal claws) (Ostrom 1969; Madsen
1976; Gauthier 1986; Currie and Zhao 1993; Russell
and Dong 1993; Novas and Bandyopadhyay 2001;
Novas et al. 2004).
Although traits 2 and 4 may occasionally be present in
the pedal phalanges of some theropod groups (i.e., the
raptorial ‘‘sickle claw’’ of pedal digit II of Dromaeosauri-
dae and Troodontidae; Rauhut and Werner 1995), all four
features are found in the vast majority of (if not all) the-
ropod manual claws. Thus, this combination of features
suggests that the putative pedal ungual of Noasaurus is
actually a manual ungual, as previously indicated by pre-
liminary studies (Agnolin et al. 2004; Carrano et al. 2004;
Carrano and Sampson 2008). Moreover, in addition to the
traits mentioned above the Noasaurus claw clearly differs
from pedal elements previously reported for noasaurids in
lacking a medial deflection of the shaft and in having an
acute and small proximodorsal lip and a sharp ventral edge
(Fig. 2; Carrano et al. 2002; Novas et al. 2004). Based on
its strong curvature and general morphology the Noasaurus
manual claw may belong to the first or second digit,
being rather similar to that of other carnivorous dinosaurs
(e.g., Allosaurus; Madsen 1976).
The only previous mention of a noasaurid manual claw
are two isolated specimens attributed to Masiakasaurus
knopfleri (Carrano et al. 2002). These materials differ from
that of Noasaurus leali in possessing a true flexor tubercle
and in lacking a ventral sulcus. However, this phalanx was
not found in articulation with any definite bone of
Fig. 1 Manual elements of Noasaurus leali (holotype PVL 4061). a
manual ungual of digit II or III, in lateral view; b–d Digit ?III phalanx
in: blateral; cdorsal; and dventral views. CLP, collateral ligamental
pit; HP, hyperextensor pit; LG, lateral groove; MK, median keel;
MVR, median ventral ridge; N, phalangeal neck; PDP, proximodorsal
process; PVP, proximoventral process. Scale bar 1cm
Fig. 2 Left lateral view of manual and pedal claws of selected
theropod dinosaurs. a–d manual claws; e–h pedal claws. aprobable II
or III of Noasaurus leali;bclaw II of Allosaurus fraglis;cclaw II of
Deinonychus antirrhopus;dclaw II of Baryonyx walkeri;eclaw III
of Allosaurus fragilis;fclaw III of Sinraptor dongi;gclaw III of
Deinonychus antirrhopus;hclaw III of unidentified abelisaurid.
(amodified from Bonaparte and Powell 1980;b,emodified from
Madsen 1976;c,gmodified from Ostrom 1969;dmodified from
Charig and Milner 1997;fmodified from Currie and Zhao 1993;
hmodified from Maganuco et al. 2008). Not to scale. LG, lateral
groove; MK, median keel; FT, flexor tubercle; LB, lateral bump
Manual phalanges in Noasauridae
123

Masiakasaurus; accordingly, the assignment of this claw to
that taxon is currently equivocal.
Consequently, the only unquestionable clear evidence of
a noasaurid manual ungual pertains to the holotype of
Noasaurus leali (PVL 4061). Thus, Autapomorphic traits
present on the manual phalanx of Noasaurus, and absent in
remaining theropods include:
1 the presence of a median ventral ridge distal to the
proximo ventral concavity, and
2 a very deep and subtriangular ventral concavity, which
was probably the insertion area of powerful flexor
musculature (Bonaparte and Powell 1980; Novas and
Bandyopadhyay 2001).
The only other ceratosaurian in which manual unguals
have been preserved is in the problematic Limusaurus
(Xu et al. 2009). In Limusaurus the unguals are short, stout,
and small, being mediolaterally expanded on their proxi-
mal ends (Xu et al. 2009), being rather different from that
of Noasaurus, and lacking the proposed autapomorphies of
the latter. However, because manual ungual phalanges are
unknown in other abelisauroids and Ceratosaurus, both
proposed autapomorphies may eventually be shown to
diagnose more inclusive clades within Ceratosauria (sensu
Rauhut 2003). In addition, as proposed by Novas and
Bandyopadhyay (2001), a deep fossa and a reduced flexor
tubercle of the pedal unguals are probably abelisauroid
synapomorphies. As the same morphology is present in
noasaurids, both traits can now be considered as abelisau-
roid synapomorphies.
Furthermore, our reconsideration of the manual unguals
of Noasaurus sheds some light on the validity of the clade
Velocisauridae Bonaparte, 1991. This theropod group was
erected by Bonaparte (1991b) to include the single species
Velocisaurus unicus Bonaparte, 1991 from the Santonian
(Upper Cretaceous) Bajo de la Carpa Formation of Neuque
´n
Province, Argentina. More recently, the Malagasy noa-
saurid theropod Masiakasaurus knopfleri Sampson et al.
(2001), from the Campanian–Maastrichtian Maevarano
Formation, was also included within Velocisauridae
(Agnolin et al. 2003). Agnolin et al. (2003) supported this
hypothesis on the presence of common characters, allowing
these authors to recognize the close phylogenetic rela-
tionship of Noasauridae and Velocisauridae. This view was
subsequently corroborated by Carrano and Sampson (2008)
in a comprehensive cladistic analysis of abelisauroid
interrelationships. Agnolin et al. (2003) distinguished
velocisaurids from noasaurids only on the basis of the
presence of non-raptorial pedal unguals in the former
group, in contrast with the ‘‘sickle claw’’ supposedly
exhibited by Noasaurus. However, following the current
interpretation, Noasaurus pedal unguals are currently
unknown; thus, the raptorial condition of noasaurid pedal
unguals is currently equivocal. Accordingly, the distinction
between noasaurids and velocisaurids presently lacks sup-
port, and we propose to considering Velocisauridae
Bonaparte, 1991 as a junior synonym of Noasauridae
Bonaparte and Powell 1980.
Non-ungual phalanges
Non-ungual phalanges of the pes are well known in the
noasaurids Masiakasaurus and Velocisaurus, and in
noasaurid-like abelisauroids from India (Bonaparte 1991b;
Carrano et al. 2002; Novas et al. 2004). In contrast, in
Noasaurus (Bonaparte and Powell 1980) and Ligabueino
(Bonaparte 1996), only isolated phalanges of the pes have
been described. However, as was indicated briefly in a
previous report, these elements actually may belong to the
manus (Agnolin et al. 2004).
In Noasaurus leali, the only known non-ungual phalanx
is a very short and stout element (Fig. 1b–d). It has a
rectangular-shaped proximal articular surface which bears
a sharp and well defined median keel. This bone also has a
well defined, strikingly narrow, and short proximoventral
process, as also occurs in known abelisaurid manual pha-
langes (e.g., those of Carnotaurus and Aucasaurus;
Bonaparte et al. 1990; Coria et al. 2002), whereas in the
manual phalanges of other theropods this process is absent
or extremely wide, its mediolateral width being subequal to
that of the distal articular surface. The proximodorsal
process is nearly quadrangular in contour, a trait regarded
by previous authors as a noasaurid synapomorphy, because
it is present in both Ligabueino and Noasaurus (Coria and
Salgado 1998). In Noasaurus the constriction between the
proximal and distal articular surfaces is strong; the distal
articular condyle is nearly rounded in lateral view and
exhibits a large, ellipsoidal, collateral ligamental pit on the
dorsal half of each condyle. The distal articular surface
shows a deep median sulcus.
The characters that suggest that the only known non-
ungual phalanx of Noasaurus is a manual element are as
follows:
1 the collateral ligamental pit is located on the dorsal half
on the lateral side of each distal articular condyle in
lateral view;
2 the distal articular condyles are displaced ventrally
relative to the main axis of the shaft in medial and
lateral views;
3 the lateral and medial margins of the distal articular
condyles are oriented subparallel to each other when
viewed dorsally (anterodorsally convergent in most
theropod pedal phalanges); and
F. L. Agnolin, P. Chiarelli
123

4 the proximal surface has an acute median keel that
separates deep articular cotyles (Fig. 3; Ostrom 1969;
Madsen 1976; Currie and Zhao 1993;Xuetal.2002).
Although traits 1 and 2 are occasionally present in pedal
phalanx II-2 of the deinonychosaurian coelurosaurs
(Xu et al. 2002), the combination of features mentioned
above only occurs, as a whole, in theropod manual ele-
ments. Moreover, the non-ungual phalanx of Noasaurus
further differs in several aspects from the pedal phalanx
II-2 of troodontids and dromaeosaurids; the former element
has less dorsally deflected distal condyles, a poorly medi-
olaterally compressed distal end, and a less constricted
phalangeal neck. (Rauhut and Werner 1995).
In the original description of Noasaurus, Bonaparte and
Powell (1980) considered that the ungual and non-ungual
phalanges of the holotype were probably II-2 and I-3I,
respectively. However, in this contribution, we suggest that
these phalanges are not sequential because they do not
articulate properly; in conclusion, they may belong to
different digits (Fig. 1).
Our review of the holotype of Ligabueino andesi
(MACN-PV-N 42; Bonaparte 1996) revealed the presence
of manual phalanx features in a putative pedal phalanx, for
example, distal articular surface and collateral ligamental
pits dorsally displaced and proximal heel narrow and short
(Fig. 4). As a result, we conclude that both the known non-
ungual phalanges of both Noasaurus leali and Ligabueino
andesi belong to the manus rather than the pes.
The precise position of the manual non-ungueal pha-
langes of Ligabueino and Noasaurus within the hand
remains unknown, because the manus is not preserved and
in other noasaurid taxa is unknown. However, the short
and transversely broad condition of the available elements
suggests that they may belong to the third digit. Regrettably,
comparisons between noasaurid and abelisaurid non-ungual
manual phalanges are not especially productive because the
latter clade exhibits a high degree of modification of the
hand (Bonaparte et al. 1990; Coria et al. 2002; Burch and
Carrano 2008). Nevertheless, the manual phalanges of both
Noasauridae (e.g., Ligabueino,Noasaurus; Bonaparte and
Powell 1980; Bonaparte 1996; this paper) and Abelisauri-
dae (e.g., Aucasaurus,Carnotaurus; Bonaparte et al. 1990;
Coria et al. 2002) possess a short and wide proximoventral
process that is unknown in other theropods; consequently,
this character is a probable synapomorphy of
Abelisauroidea.
The robust non-ungual phalanges, large, recurved rap-
torial manual claws, and elongated humerus are features
indicative of the presence of elongated forelimbs in
Noasauridae (Sereno et al.1996; Carrano et al.2002),
clearly contrasting with the reduced arm and manus present
in other ceratosaurians (e.g., Ceratosaurus,Carnotaurus,
Aucasaurus,Majungasaurus; Gilmore 1920; Bonaparte
et al. 1990; Madsen and Welles 2000; Coria et al. 2002;
Burch and Carrano 2008). Elongated forelimbs are also
present in basal saurischians and basal theropods (e.g.,
Herrerasauridae, Coelophysoidea; Rowe and Gauthier
1990; Sereno 1993) and some ceratosaurians (e.g. Elaph-
rosaurus,Limusaurus; Janensch 1920; Xu et al. 2009),
suggesting that the condition exhibited by Noasauridae
may be plesiomorphic within Ceratosauria.
Brief overview of abelisaurid manual morphology
Regrettably, in most ceratosaurians the manus in unknown
or nearly so. The only genera of non-abelisauroid cerato-
saurians in which the manus is known are Ceratosaurus
and Limusaurus (Gilmore 1920; Xu et al. 2009). Although
Fig. 3 Left lateral view of preungueal phalanges of selected thero-
pods. a–c manual elements; d–f pedal elements. adigit ?III phalanx
of Noasaurus leali;bdigit ?III phalanx of Ligabueino andesi;
cphalanx 1 digit III of Allosaurus fragilis.dphalanx 2 digit II of
Deinonychus antirrhopus;ephalanx 3 digit IV of Allosaurus fragilis;
fphalanx 3 digit IV of Velocisaurus unicus.(amodified from
Bonaparte and Powell 1980;bmodified from Bonaparte 1996;
c,emodified from Madsen 1976;dmodified from Ostrom 1969;
fmodified from Bonaparte 1991b). Not to scale
Fig. 4 Ligabueino andesi (holotype MACN PV-N 42). Digit ?III
phalanx in: alateral view; bdorsal view; clateroventral view;
dproximal view. CLP, collateral ligamental pit; HP, hyperextensor
pit; MK, median keel; PDP, proximodorsal process; PVP, proximo-
ventral process. Scale bar 2mm
Manual phalanges in Noasauridae
123

in both genera the manus is reduced, distalmost non-un-
gueal phalanges still retain pulley-like well excavated
distal articular surfaces, and in the case of Limusaurus
manual unguals of digits II and III have been preserved
(Xu et al. 2009). In contrast with the well developed
manual non-ungual phalanges exhibited by other thero-
pods, including noasaurids, the phalanges of abelisaurids
are shortened and reduced (Novas 1991,1992). In abeli-
saurids, manual phalanges are only known in Carnotaurus,
Aucasaurus, and Majungasaurus (Bonaparte et al. 1990;
Coria et al.2002; Burch and Carrano 2008), and all of these
bear several features that are worthy of mention. Several
manual features are apomorphic for abelisaurids, including
the discoidal morphology of proximal phalanges and car-
pals, the absence of hyperextensor pits on the dorsal sur-
face of phalanges and metacarpals, the shallowness of
collateral ligament pits, and the poorly excavated distal
articular surface. These traits indicate, as a whole, the
reduction of flexor and extensor movements in the abeli-
saurid manus (Rauhut 2003). Moreover, the shortened
nature of the distal non-ungual phalanges indicates the lack
of grasping abilities (Sereno 1993) and, perhaps, the total
lack of mobility, as is probably the case for the several
fused phalanges within the manus of Majungasaurus
(Burch and Carrano 2008). A similar morphology may be
observed in the manus of titanosauriform sauropods, in
which phalangeal reduction has been correlated with pae-
domorphic phenomena (Salgado 2003). In the same way,
the extreme forelimb reduction seen in abelisaurids was
interpreted as the consequence of the loss of function of the
HOXA11 and HOXD11 genes (Vargas 2002). The pres-
ence of flat, broad, and rugose distal articular surfaces on
the distalmost non-ungual phalanges of Aucasaurus and
Carnotaurus, together with the reduction of flexor and
extensor pits and overall size, are features that suggest not
only the absence of flexor and extensor movements in the
abelisaurid manus, but also the total absence of unguals. In
Aucasaurus, as was noted by Coria et al. (2002), and in
Majungasaurus, as indicated by Burch and Carrano (2008),
metacarpals I and IV are conical structures that appear to
have carried no phalanges. The same appears to be true for
Carnotaurus, in which the metacarpal IV is a splint-like
structure and metacarpal I is extremely reduced (Bonaparte
et al.1990). Aucasaurus, the only abelisaurid in which non-
ungual phalangeal formula may be clearly established, has
only one phalanx in digit II and two phalanges in digit III,
indicating that these digits bear no unguals. The same
morphology may be also observed in Majungasaurus,
which has a perfectly preserved manus without any sign of
claws (Burch and Carrano 2008). Moreover, in other
abelisaurids, including the beautifully preserved skeletons
of Aucasaurus and Carnotaurus, in which most elements of
the manus are known, there is no evidence of manual
unguals. Large theropod manual ungual phalanges are also
absent in the richly fossiliferous Late Cretaceous forma-
tions of India and Madagascar, from which hundreds of
abelisaurid bones have been recovered, including abundant
pedal unguals (Novas et al.2004; Maganuco et al. 2008).
In summary, we suggest that abelisaurid forelimb reduction
not only included phalangeal shortening and simplification,
but also the loss of manual unguals. The reduction or loss of
manual elements is not an uncommon phenomenon and has
been reported in several tetrapod groups, including squamates
(Greer 1991; Palci and Caldwell 2007) and titanosaurian
sauropod dinosaurs (Salgado et al. 1997).
Conclusions
In this paper we arrive at the following conclusions:
1. The putative ‘‘sickle claw’’ of the second toe of the pes
of Noasaurus, is actually a first or second manual
ungual.
2. In Ligabueino and Noasaurus, the supposed pedal non-
ungual phalanges probably pertain to the manus
(probably digit III), an hypothesis based on several
features exhibited by their proximal and distal articular
surfaces.
3. The presence of an acute and short proximoventral
process may be a diagnostic trait of non-ungual manual
phalanges of Abelisauroidea.
4. The reinterpretation of noasaurid unguals allows us to
propose that the clade Velocisauridae Bonaparte, 1991
be considered a junior synonym of Noasauridae
Bonaparte and Powell 1980.
5. We propose, on the basis of skeletal morphology and
negative evidence, that members of Abelisauridae
probably lacked their manual ungual phalanges.
Additionally, abelisaurids probably lost manual flexor
and extensor movements thus these theropods lacked
grasping abilities.
Acknowledgments We wish to thank Jose
´Bonaparte (Fundacio
´nde
Historia Natural Fe
´lix de Azara) for allowing us to study the
holotypes of Ligabueino andesi and Noasaurus leali, Gabriel
Lio (FHNFA) for some drawings of the holotypical material of
Noasaurus, and Martı
´n Ezcurra (MACN) for photographs of
Ligabueino andesi.
References
Agnolin, F.L., F.E. Novas, and S. Apesteguı
´a. 2003. Velocisaurids in
South America and Madagascar. Ameghiniana 40: 77R.
Agnolin, F.L., S. Apesteguı
´a, and P. Chiarelli. 2004. The end of a
myth: The mysterious ungual claw of Noasaurus leali.Journal
of Vertebrate Paleontology 24: 33A.
F. L. Agnolin, P. Chiarelli
123

Apesteguı
´a, S. 2002. Successional structure in continental tetrapod
faunas from Argentina along the Cretaceous. Boletim do 68
Simpo
´sio sobre o Creta
´ceo do Brasil—28Simposio sobre el
Creta
´cico de Ame
´rica del Sur, Abstracts: 135–141.
Bonaparte, J.F. 1986. History of the terrestrial Cretaceous vertebrates
of Gondwana. Actas 4th Congreso Argentino de Paleontologı
´
ay
Bioestratigrafı
´
a4: 63–95.
Bonaparte, J.F. 1991a. The Gondwanian theropod families Abelisau-
ridae and Noasauridae. Historical Biology 5: 1–25.
Bonaparte, J.F. 1991b. Los vertebrados fo
´siles de la Formacio
´nRı
´o
Colorado, de la ciudad de Neuque
´n y cercanı
´as, Creta
´cico
superior, Argentina. Revista del Museo Argentino de Ciencias
Naturales ‘‘Bernardino Rivadavia 4: 15–123.
Bonaparte, J.F. 1996. Cretaceous tetrapods of Argentina. Mu
¨nchner
Geowissenschaftliche Abhandlungen. Reihe A, Geologie und
Pala
¨ontologie 30: 73–130.
Bonaparte, J.F., F.E. Novas, and R.A. Coria. 1990. Carnotaurus
sastrei Bonaparte, the horned, lightly built carnosaur from the
middle Cretaceous of Patagonia. Contributions in Science of the
Natural History Museum of Los Angeles County 416: 1–42.
Bonaparte, J.F., and J.E. Powell. 1980. A continental assemblage of
tetrapods from the Upper Cretaceous beds of El Brete, northwest-
ern Argentina (Sauropoda-Coelurosauria-Carnosauria-Aves).
Me
´moires de la Socie
´te
´Ge
´ologique de France 139: 19–28.
Burch, S., and M. Carrano. 2008. Abelisaurid forelimb evolution:
New evidence from Majungasaurus crenatissimus (Abelisauri-
dae: Theropoda) from the late Cretaceous of Madagascar.
Journal of Vertebrate Paleontology 28: 58A.
Canale, J.I., C.A. Scanferla, F.L. Agnolin, and F.E. Novas. 2009. New
carnivorous dinosaur from the Late Cretaceous of NW Patagonia
and the evolution of abelisaurid theropods. Naturwissenschaften
96: 409–414.
Carrano, M.T., and S.D. Sampson. 2008. The phylogeny of Cerato-
sauria (Dinosauria: Theropoda). Journal of Systematic Palaeon-
tology 6: 183–236.
Carrano, M.T., S.D. Sampson, and C.A. Forster. 2002. The osteology
of Masiakasaurus knopfleri, a small abelisauroid (Dinosauria:
Theropoda) from the Late Cretaceous of Madagascar. Journal of
Vertebrate Paleontology 22: 510–534.
Carrano, M.T., S.D. Sampson, and M.A.Loewen. 2004. New discoveries
of Masiakasarus knopfleri and the morphology of the Noasauridae
(Dinosauria: Theropoda). Journal of Vertebrate Paleontology
24: 44A.
Carrano, M.T. 2007. The appendicular skeleton of Majungasaurus
crenatissimus (Theropoda: Abelisauridae) from the Late Creta-
ceous of Madagascar. Society of Vertebrate Paleontology,
Memoir 8: 163–179.
Charig, A.J., and A.C. Milner. 1997. Baryonyx walkeri, a fish eating
dinosaur from the Wealden of Surrey. Bulletin of the Natural
History Museum, Geology Series 53: 11–70.
Coria, R.A., L.M. Chiappe, and L. Dingus. 2002. A new close relative
of Carnotaurus satrei Bonaparte 1985 (Theropoda: Abelisauri-
dae) from the Late Cretaceous of Patagonia. Journal of
Vertebrate Paleontology 22: 460–465.
Coria, R.A., and L. Salgado. 1998. A basal Abelisauria, Novas 1992
(Theropoda-Ceratosauria) from the Cretaceous of Patagonia,
Argentina. GAIA 15: 89–102.
Currie, P.J., and X. Zhao. 1993. A new carnosaur (Dinosauria,
Theropoda) from the Jurassic of Xinjiang, People’s Republic of
China. Journal of Canadian Earth Sciences 30: 2037–2081.
Gauthier, J. 1986. Saurischian monophyly and the origin of birds.
Memoirs of the California Academy of Sciences 1: 1–47.
Gilmore, C.W. 1920. Osteology of the carnivorous Dinosauria in the
United States National Museum, with special reference to the
genera Antrodemus (Allosaurus) and Ceratosaurus.Bulletin
United States National Museum 110: 1–154.
Greer, A.E. 1991. Limb reduction in squamates: Identification of the
lineages and discussion of the trends. Journal of Herpetology 25:
166–173.
Janensch,W. 1920. U
¨ber Elaphrosaurus bambergi und die megalosaurier
aus den Tendaguru-Schichten Deutsch-Ostafrikas. Sitzunberichte
Gesselschaft Natursforschende Freunde 1920: 225–235.
Madsen, J.H. 1976. Allosaurus fragilis: A revised osteology. Bulletin
Utah Geological and Mineralogical Survey 109: 1–163.
Madsen, J.H., and S.P. Welles. 2000. Ceratosaurus (Dinosauria.
Theropoda), a revised osteology. Utah Geological Survey
Miscellaneous Publication 2: 1–80.
Maganuco, S., A. Cau, and G. Pasini. 2008. New information on the
abelisaurid pedal elements from the Late Cretaceous of NW
Madagascar (Mahajanga Basin). Atti della Societa Italiana di
Scienze Naturalie del Museo Civico di Storia Naturale di Milano
149: 239–252.
Novas, F.E. 1991. Relaciones filogeneticas de los dinosaurios
teropodos ceratosaurios. Ameghiniana 28: 401–414.
Novas, F.E. 1992. Phylogenetic relationships of the basal dinosaurs,
the Herrerasauridae. Palaeontology 35: 51–62.
Novas, F.E. 1997. South American dinosaurs. In Encyclopedia of
Dinosaurs, ed. P.J. Currie, and K. Padian, 678–689. San Diego:
Academic Press.
Novas, F.E., and S. Bandyopadhyay. 2001. Abelisaurid pedal unguals
from the Late Cretaceous of India. Asociacio
´n Paleontolo
´gica
Argentina, Publicacio
´n Especial 7: 145–149.
Novas, F.E., F.L. Agnolin, and S. Bandyopadhyay. 2004. Cretaceous
theropods from India: A review of specimens described by
Huene and Matley (1933). Revista del Museo Argentino de
Ciencias Naturales 6: 67–103.
Novas, F.E., S. de Valais, P. Vickers Rich, and T. Rich. 2005a. A
large theropod from Patagonia, Argentina, and the evolution of
carcharodontosaurids. Naturwissenchaften 92: 226–230.
Novas, F.E., F. Dalla Vecchia, and D.F. Pais. 2005b. Theropod pedal
unguals from the Late Cretaceous (Cenomanian) of Morocco,
Africa. Revista del Museo Argentino de Ciencias Naturales 7:
167–175.
Ostrom, J.H. 1969. Osteology of Deinonychus antirrhopus,an
unusual theropod from the Lower Cretaceous of Montana.
Bulletin of the Peabody Museum of Natural History 30: 1–165.
Palci, A., and M.W. Caldwell. 2007. Vestigial forelimbs and axial
elongation in a 95 million year old non-snake squamate. Journal
of Vertebrate Paleontology 27: 1–7.
Rauhut, O.W.M., and C. Werner. 1995. First record of the family
Dromaeosauridae (Dinosauria: Theropoda) in the Cretaceous of
Gondwana (Wadi Milk Formation, northern Sudan). Pala
¨onto-
logische Zeitschrift 69: 475–489.
Rauhut, O.W.M. 2003. The interrelationships and evolution of basal
theropod dinosaurs. Special Papers in Palaeontology 69: 1–214.
Rowe, T., and J. Gauthier. 1990. Ceratosauria. In The Dinosauria, ed.
D.B. Weishampel, P. Dodson, and H. Osmo
´lska, 151–168.
Berkeley: University of California Press.
Russell, D.A., and Z.M. Dong. 1993. A nearly complete skeleton of a
new troodontid dinosaur from the Early Cretaceous of the Ordos
Basin, Inner Mongolia, People’s Republic of China. Canadian
Journal of Earth Sciences 30: 2163–2173.
Salgado, L. 2003. Los sauro
´podos de Patagonia: Sistema
´tica,
Evolucio
´n y Paleobiogeografı
´a. In: Colectivo Arqueolo
´gico-
Paleontolo
´gico de Salas, C.A.S. (Ed.):. Actas de las II Jornadas
Internacionales Sobre Paleontologı
´a de Dinosaurios y su
Entorno. Salas de los Infates (Burgos, Espan
˜a). pp 139–168.
Salgado, L., R.A. Coria, and J.O. Calvo. 1997. Evolution of
titanosaurid sauropods. I: Phylogenetic analysis based on the
postcranial evidence. Ameghiniana 34: 3–32.
Sampson, S.D., M.T. Carrano, and C.A. Forster. 2001. A bizarre new
predatory dinosaur from Madagascar. Nature 409: 504–506.
Manual phalanges in Noasauridae
123

Sereno, P.C. 1993. The pectoral girdle and forelimb of the basal
theropod Herrerasaurus ischigualastensis.Journal of Vertebrate
Paleontology 13: 425–450.
Sereno, P.C., D.B. Dutheil, M. Iarochene, H.C.E. Larsson, G.H. Lyon,
P.W. Mgwene, C.A. Sidor, D.J. Varrichio, G.P. Wilson, and
J.A. Wilson. 1996. Predatory dinosaurs from the Sahara and Late
Cretaceous faunal differentiation. Science 272: 986–991.
Sereno, P.C., J.A. Wilson, and J.L. Conrad. 2004. New dinosaurs link
southern landmasses in the Mid-Cretaceous. Proceedings of the
Royal Society of London B 164: 1471–2954.
Vargas, A. 2002. La extrema reduccio
´n del radio y la ulna en la
evolucio
´ndeCarnotaurus sastrei: Posible pe
´rdida de funcio
´nde
los genes HOXA11 y HOXD11. Ameghiniana 39: 17R.
Wilson, J.A., P.C. Sereno, S. Srivastava, D.K. Bhatt, A. Khosla, and A.
Sahni. 2003. A new abelisaurid (Dinosauria, Theropoda) from the
Lameta Formation (Cretaceous, Maastrichtian) of India. Contri-
butions from the Museum of Paleontology of the University of
Michigan 31: 1–42.
Xu, X., M.A. Norell, X.-L. Wang, P.J. Makovicky, and X.-C. Wu.
2002. A basal troodontid from the Early Cretaceous of China.
Nature 415: 780–784.
Xu, X., J.M. Clark, J. Mo, J. Choiniere, C.A. Forster, G.M. Erickson,
D.W.E. Hone, C. Sullivan, D.A. Eberth, S. Nesbitt, Q. Zhao,
R. Herna
´ndez, C.-K. Jia, F.-I. Han, and Y. Guo. 2009. A jurassic
ceratosaur from China helps clarify avian digital homologies.
Nature 459: 940–944.
F. L. Agnolin, P. Chiarelli
123