SciELO - Scientific Electronic Library Online

 
vol.45 número2Revisión del género Stegotherium Ameghino, 1887 (Mammalia, Xenarthra, Dasypodidae)Vertebrados cretácicos de la Formación Loncoche en Calmu-Co, Mendoza, Argentina índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Articulo

Indicadores

  • No hay articulos citadosCitado por SciELO

Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google

Bookmark


Ameghiniana

versión On-line ISSN 1851-8044

Ameghiniana v.45 n.2 Buenos Aires abr./jun. 2008

 

Hindlimb musculature and function of Neuquensaurus australis (Sauropoda: Titanosauria)

Alejandro Otero1 and Sergio F. Vizcaíno2

1CONICET. Museo de Geología y Paleontología (Universidad Nacional del Comahue), Buenos Aires 1400, Neuquén, 8300 Neuquén, Argentina. alexandros.otero@gmail.com
2CONICET. División Paleontología Vertebrados, Museo de La Plata, paseo del bosque s/n, 1900 La Plata, Argentina. vizcaino@museo.fcnym.unlp.edu.ar

Abstract. We present the reconstruction of the hindlimb musculature of the titanosaur sauropod Neuquensaurus australis (Lydekker), integrating data from crocodilians and birds as a phylogenetic bracket. First, we evaluate sites of origin and insertions in N. australis by comparison with extant archosaurs as well as with previous works on dinosaur soft anatomy, and reconstruct principal muscles of the leg, such as Triceps femoris group, M. iliofibularis, Mm. adductores femores, Mm. gastrocnemii, M. ischiotrochantericus and Mm. caudofemorales. Second, we infer the probable function of each muscle comparing with our bracket taxa and with living analogs, principally mammals, because they are thought to be the best model to understand non-avian dinosaur locomotion. This analysis suggests that N. australis has developed larger moment arms in Mm. iliotibiales and M. iliofibularis than in other non-Saltasaurinae sauropods, in relation to the well developed preacetabular and postacetabular ilium; a large Mm. femorotibiales, due to the mediolateral development of the femur; and a large M. iliofibularis due to the large fibular lateral trochanter. We also interpret functional attributes of sauropod wide-gauged trackmakers, in which the extension-flexion and adduction actions are improved with respect to narrow-gauged sauropods.

Resumen. Musculatura y función del miembro posterior de Neuquensaurus australis (Sauropoda: Titanosauria). Se presenta la reconstrucción de la musculatura del miembro posterior del saurópodo titanosaurio Neuquensaurus australis (Lydekker). Para ello, se integró información de cocodrilos y aves, tomadas como marco filogenético. En primera instancia, se localizaron los sitios de origen e inserción de la musculatura en N. australis por comparación con arcosaurios actuales y con trabajos previos sobre anatomía blanda en dinosaurios. Se reconstruyeron los principales músculos de la pata, como el grupo Triceps femoris, M. iliofibularis, Mm. adductores femores, Mm. gastrocnemii, M. ischiotrochantericus y Mm. caudofemorales. A continuación, se infirió la probable función de cada músculo comparando con homólogos y análogos actuales, principalmente mamíferos, ya que estos constituyen un buen modelo para comprender la locomoción en dinosaurios no avianos. Del análisis se concluye que N. australis habría desarrollado mayores brazos de momento en los músculos iliotibiales e iliofibularis que en saurópodos no saltasaurinos, en relación a la extensión de los procesos pre-y postacetabulares del ilión. También habría desarrollado un gran femorotibialis, debido al notable desarrollo mediolateral del fémur, y un gran iliofibularis, debido al prominente trocánter lateral de la fíbula. Por último, se interpretaron atributos funcionales de saurópodos con huella ancha, en los cuales la acción de extensión-flexión es más importante respecto de la de los saurópodos con huellas angostas.

Key words. Archosauria; Extant Phylogenetic Bracket; Soft tissue; Functional morphology.

Palabras clave. Arcosauria; Soporte Filogenético Viviente; Tejidos blandos; Morfología funcional.

Introduction

Neuquensaurus australis (Lydekker) is among the most derived titanosaur sauropods; belonging to the clade Saltasaurinae (sensu Powell, 1992).This group of sauropods is characterized by smaller size in comparison with other sauropods and features of the appendicular skeleton, such as a well developed femoral lateral bulge, a preacetabular lobe of the ilium that is nearly horizontal and laterally projected, and a well developed fibular lateral trochanter (McIntosh, 1990; Upchurch, 1995; Salgado et al., 1997; Wilson and Sereno, 1998; Salgado, 2000; Salgado and Azpilicueta, 2000; Apesteguía, 2002; Wilson, 2002; Powell, 2003). These features suggest muscular arrangements, and hence locomotor habits, different from those inferred in other titanosaur sauropods (see Wilson and Carrano, 1999).
Inferring soft tissues in extinct forms is not an easy task, especially in non-avian dinosaurs (particularly sauropods) which have achieved many differences in their locomotor apparatus (Salgado et al., 1997; Wilson and Sereno, 1998; Upchurch, 2004)
with respect to dinosaurs in the line of birds (Gatesy, 1990; Hutchinson, 2001a, 2001b; Carrano and Hutchinson, 2002; Hutchinson et al., 2005). For this reason, reconstruction of soft tissues (such as musculature) in extinct forms requires integration of paleontological and neontological data as well as an explicit methodology (Bryant and Seymour, 1990; Witmer, 1995, 1997).
Archosauria is a monophyletic group that includes forms with living representatives, such as crocodilians and birds, and extinct forms, such as pterosaurs and non- avian dinosaurs (Benton, 2004). Attempts to infer homologous structures in extinct forms, for example, non-avian dinosaurs, necessarily imply the study of extant ones; especially if we wish to reconstruct nonpreserved attributes, such as soft tissue. Application of explicit, outgroup-based phylogenetic methodology in several recent studies (Gatesy, 1990, 1995; Carrano, 1998, 2000; Hutchinson, 2001a, 2001b; Carrano and Hutchinson, 2002), has improved the understanding of soft tissue features in extinct forms. As noted by Carrano and Hutchinson (2002; but also see Witmer, 1995, 1997), parsimony requires reconstruction of the primary soft tissue with a minimum number of inferred differences from extant forms.
Non-avian dinosaurs and birds have in common some derived morphological features in their appendicular skeleton (e.g. parasagittal limb posture and medially directed femoral head) (Chiappe, 1995; Benton, 2004) although they differ in many osteological features related to their locomotor behavior (Carrano, 1998, 2000). For example, the neornithine pelvis and femora are strikingly modified with respect to the dinosaurian ancestral condition, with even greater differences if we consider associated soft tissues (Romer, 1923b, 1927; Galton, 1969; Chiappe, 1995; Gatesy, 1995; Hutchinson, 2001a, 2001b). Non-avian dinosaurs have many differences in the musculoskeletal system in relation to birds that suggest changes in limb orientation, posture, and function in relation to locomotor habits and their use of the substrate (Carrano, 1998, 2000). Interpretation of appendicular characters to evaluate locomotory behavior requires data on bone shape and inferences about its relationship with associated musculature, which for extinct animals, must be achieved by comparison with extant species (Lauder, 1995; Hutchinson and Gatesy, 2000; Carrano, 2001; Hutchinson, 2004). In light of this, what extant group of vertebrates provides the most appropriate model to understand function in non-avian dinosaurs?
Previous studies about appendicular musculature in dinosaurs and derived forms include Romer's classical works on dinosaurs and birds (1923a, 1927), and on saurischian dinosaurs (1923b). Galton (1969) and Coombs (1979), in turn, restored pelvic and hindlimb musculature of ornithischian dinosaurs (i. e. Hypsilophodon and ankylosaurs, respectively). Avian dinosaurs were studied by several authors (Gatesy, 1990, 1995; Hutchinson, 2001a, 2001b; Carrano and Hutchinson, 2002; Hutchinson et al., 2005). Few contributions described pelvic and hindlimb musculature of sauropod dinosaurs (Borsuk-Bialynicka, 1977; Wilhite, 2003) and no attention has been paid to the reconstruction of the appendicular musculature in South American sauropods. Understanding the relationship between the appendicular skeleton and its associated musculature allows predictions to be made about aspects of function in relation to the biological role of the characters implied in locomotion (Hutchinson and Gatesy, 2000; Hutchinson, 2004). For example, Wilson and Carrano (1999) argued that the skeletal morphology is responsible for "wide-gauge" ichnotype in titanosaurid sauropods. But, what about the musculature associated? How might musculature be related to the modifications of the skeleton?
Recently discovered specimens of Neuquensaurus australis allow us to determine the areas of origin and insertion of several pelvic and hindlimb muscles and to evaluate its muscle function in relation to locomotor behavior. The relationship between the morphology of the appendicular skeleton, its associated musculature, and locomotor habits is well documented in living crocodilians and lizards (Blob, 2000; Blob and Biewener, 2001), and birds (Gatesy, 1995; Hutchinson and Gatesy, 2000) as well as in living mammals (Biewener, 1983; Alexander and Pond, 1992).
In this work, we assume that muscles that have the same rough positions and connections in extant archosaurs (crocodilians and birds) can reasonably be inferred to have had the same features in extinct dinosaurs and confidently placed into a musculoskeletal model (see Materials and methods). We reconstruct principal muscles groups of the leg as Triceps femoris group, M. iliofibularis, Mm. adductores femores, M. ischiotrochantericus, Mm. gastrocnemii and Mm. caudofemorales. The function of each muscle is inferred using the "form-function correlation approach" noted by Radinsky (1987; see Materials and methods), by comparison with extant relatives (crocodilians and bids) and with living analogs (mammals). This reconstruction represents the first attempt to interpret appendicular muscle arrangements and function in South American sauropods.

Materials and methods

Acronyms

MCS: Museo Regional Cinco Saltos, Cinco Saltos, Argentina.
MLP-Av: Museo de La Plata, Rancho de Ávila collection, La Plata, Argentina.
MLP-CS: Museo de La Plata, Cinco Saltos collection, La Plata, Argentina.
MLP-Ly: Museo de La Plata, Lydekker's collection, La Plata, Argentina.
PVL: Colección de Paleontología de Vertebrados de la Fundación-Instituto "Miguel Lillo", Tucumán, Argentina.

Material. The specimens of Neuquensaurus australis analyzed were: one sacrum articulated to both ilia (MCS-5/16); several incomplete ilia (MLP-CS 1056, MLP-CS 1057, MLP-CS 1258, MLP-CS 1298, MLP-Ly 17); four fragments of pubes (MLP-CS 1102, MLP-CS 1304, MLP-CS 1294, MLP-Ly 14); one left ischium (MCS- 5/24); three femora (MLP-CS 1101, MLP-CS 1121, MCS-9); four tibiae (MLP-CS 1093, MLP-CS 1123, MLP-CS 1103, MCS-6); two fibulae (MLP-CS 1198, MLP-Ly 127); several proximal haemal arches (MCS-5/30-32).

We also included specimens assigned to Neuquensaurus robustus (Huene 1929) nomen dubium because differences between the latter and N. australis are considered to be simply individual variation or owing to sexual dimorphism (Powell, 2003; Salgado et al., 2005). The material referred to N. robustus was: an incomplete left ilium (MLP-Av 2069); an incomplete right pubis (MLP-Av 2066); one left femur (MLP-CS 1480); two tibiae (MLP-CS 1264, MLP-CS 1303); two fibulae (MLP-CS 1265, MLP-Av 2060).
All specimens (Lydekker's material and recent discovery) were collected from the iron bridge which crosses the Río Neuquén, close to the city of Neuquén, Neuquén Province, Cinco Saltos, Cerro Policía and Lake Pellegrini, Río Negro Province (Bajo de la Carpa Formation and lower member of Allen Formation, Río Colorado Subgroup, Neuquén Group, Senonian, Campanian?, Early Maastrichtian).

Phylogenetic control of muscular inferences

The anatomical framework was provided by detailed observations of extant archosaurs. For soft tissue data two specimens of Caiman latirostris were dissected. The ostelogical data were provided by four cocodrilians and eleven Neornithes housed in collections of Museo de La Plata and Instituto-Fundación"Miguel Lillo".
The Extant Phylogenetic Bracket (Witmer, 1995, 1997), provided by crocodilians and birds has been confirmed to be the best approach for soft tissue reconstruction (Witmer, 1997; Hutchinson and Gatesy, 2000; Hutchinson, 2001a, 2001b; Carrano and Hutchinson, 2002). It allows formulation of hypotheses about the soft-tissue relations in extinct taxa that may be tested by reference to the known osteological correlates of the soft tissues in fossil taxa enclosed by the bracket, constraining speculation to a minimum (Witmer, 1995).
Origin and insertion sites for Neuquensaurus were inferred separately using the EPB method. The degree of speculation in reconstructions was quantified using Witmer's "levels of inference" for soft tissue reconstruction, with Level I less speculative and Level III more speculative. If both brackets (crocodilians and birds) have the osteological correlate for a specific origin or insertion of a given muscle, the soft tissue reconstruction is a Level I inference (the soft-tissue assessment is said to be decisive and positive). Lack of the feature in one outgroup would require a level II inference (soft-tissue assessment equivocal), and if both outgroups lack the feature a Level III inference would be required (soft-tissue decisive and negative). According to Witmer (1995), when inferences require additional speculation there is a hierarchy of inference parallel to Levels of Inference outlined above: (I´) decisive and positive, (II´), equivocal and (III´) decisive and negative (table 1).

Table 1. Muscles and function inferred present in Neuquensaurus / músculos y funciones inferidas presentes en Neuquensaurus.

Muscle nomenclature

Muscle nomenclature used herein follows Romer (1923a), Rowe (1986), Hutchinson (2001a, 2001b) and Carrano and Hutchinson (2002). We based muscle comparisons and homology hypotheses on previous studies of anatomy of crocodilians and birds (Romer, 1923a, 1923b; Vanden Berge, 1982; Rowe, 1986; Vanden Verge and Zweers, 1993; Hutchinson, 2001a, 2001b; Wilhite, 2003). We also referred to previous works on sauropod dinosaurs including Opisthocoelicaudia skarzynskii (Borsuk-Bialynicka, 1977), North American sauropods (Camarasaurus, Diplodocus and Apatosaurus) (Wilhite, 2003) and other dinosaurs as Hypsilophodon (Galton, 1969), ankylosaurs (Coombs, 1979) and Tyrannosaurus rex (Carrano and Hutchinson, 2002). We do not reconstruct muscles whose inference implies high speculation (Deep dorsal group, Flexor cruris Group and Mm. puboischiofemorales externii). We infer the probable function of each muscle by comparison with crocodiles and birds, but principally with living mammals for which detailed anatomical and functional studies have been made by veterinarians (Sisson, 1982) and because of their similarities in locomotor habits with non-avian dinosaurs (Carrano, 1998, 2000).

Form, function and living analogs

In this paper, the probable function of each muscle is evaluated using the "form-function correlation approach" (Radinsky, 1987), which assumes that a close relation exists between form and function, so that the latter can be predicted from the former. The observed relationship between form and function in extant taxa then can be applied to extinct taxa of interest. Homologous structures between extant and extinct forms are supposed to have homologous functions, allowing the function of the structure in the fossil taxon to be inferred (but also see Lauder, 1995). In the absence of suitable homologies, the tendency has been to argue for function based on biological analogues (Radinsky, 1987). In this work, we also infer muscular function comparing with mammals, despite their phylogenetic distance from dinosaurs, because they share a parasagittal limb posture, relatively slender femoral dimensions, and similar limb segment proportions, providing a better model for understanding muscle function in nonavian dinosaurs (Carrano, 1997, 1998, 1999, 2001).

Results

Areas of origin and insertion and probable function of hindlimb muscles of Neuquensaurus australis.

We present detailed anatomical observations to interpret osteological data and associated soft tissues directly from bracket taxa (crocodilians and birds) using the EPB approach (Witmer, 1995, 1997) and placed into a phylogenetic framework (figure 1) (Brochu, 2001; Benton, 2004). Comparisons with previous studies are given separately (see Discussion). The homology scheme is given in table 2.


Figure 1. Phylogenetic framework used in this study. Based on Brochu (2001) and Benton (2004) / marco filogenético utilizado en el presente estudio. Basado en Brochu (2001) y Benton (2004).

Table 2. Muscle homologies of the hindlimb in extant archosaurs / homologías musculares del miembro posterior de arcosaurios vivientes. Following / según Romer (1923a), Rowe (1986), Hutchinson (2001a,2001b) and/y Carrano and/y Hutchinson (2002).

Triceps femoris group

Mm. iliotibiales (IT). In extant archosaurs, Mm. iliotibiales is a large and superficial muscle comprises of three separate heads, from the dorsolateral edge of the ilium, M. iliotibialis 1-3 from anterior to posterior. Neornithes keeps the tripartite divisions as in crocodilians, although the relative development of each head shifts, with the anterior (M. iliotibialis cranialis/ IC) and posterior one (M. iliotibialis lateralis/IL) showing greatest development (Carrano and Hutchinson, 2002). In crocodilians and birds the three separated heads converge with M. ambiens and Mm. femorotibiales, generating the common knee extensor tendon that inserts on the cnemial crest of the tibia, via the patellar ligament in crocodilians, or the patella in birds (Romer, 1923a; Carrano and Hutchinson, 2002).
It is not possible to discriminate the origin of the individual heads of this muscle on the ilium in sauropod dinosaurs (see Borsuk-Bialynicka, 1977, fig. 17 B). In Neuquensaurus, the anterior portion of that muscle probably has been well developed, as in Neornithes, due to the great development of the preacetabular lobe. There is no evidence of patella in sauropods, so it probably inserted directly on the well developed cnemial crest (figures 2 and 3).


Figure 2. Hindlimb musculature of / musculatura del miembro posterior de Neuquensaurus australis. Right femur of / fémur derecho de N. australis in / en vistas 1, anterior / anterior; 2, medial / medial; and / y 3, posterior view / vista posterior; 4, right tibia of / tibia derecha de N. australis in anteromedial view / en vista anteromedial. Left fibula of / Fíbula izquierda de N. robustus in 5, anterior view and 6, lateral view / en vistas 5, anterior y 6, lateral. Cc, cnemial crest / cresta cnemial; gtr, greater trochanter / trocánter mayor; lb, lateral bulge / comba lateral; lic, linea intermuscularis cranialis / línea intermuscular craneal; ltr, lateral trochanter / trocánter lateral; ts, trochanteric shelf / repisa trocantérica; 4tr, fourth trochanter / cuarto trocánter. The extent of the FMTE and FMTI are ambiguous as is the extent of the insertion of ADD1+2. Muscle abbreviations are in table 1 / la extensión de los músculos FMTE y FMTI y las inserciones de los ADD1+2 son ambiguas. Las abreviaturas de los músculos se presentan en la tabla 1. Scale bar / escala: 15 cm.

Figure 3. Hindlimb bones of / huesos del miembro posterior de Neuquensaurus. Right femur / fémur derecho (MCS-9) in / en 1, anterior / vista anterior; 2, medial / vista medial; and 3, posterior views / vista posterior. 4, anteromedial view of right tibia / vista anteromedial de la tibia derecha (MCS-6). 5, lateral view of left fibula / vista lateral de la fíbula izquierda (MLP-CS- 1265). Scale bar / escala : 15 cm.

Function. In extant archosaurs, M. iliotibialis extends the knee joint, abducts and flexes the leg, and also extends the hip (Hutchinson, personal communication). An interesting aspect of Neuquensaurus is the nearly horizontal and laterally projected preacetabular lobe of the ilium (figure 4) (Salgado et al., 1997; Powell, 2003). This suggests that the line of action of the cranial portion of this muscle has had principally an anterior-posterior component, improving the extension of the leg.

Figure 4. Sacrum and ilia of some sauropods showing the relative development of the preacetabular process of the ilium / sacro e iliones de algunos saurópodos mostrando el desarrollo relativo de los procesos preacetabulares del íleon. 1, Neuquensaurus (ventral view / vista ventral)); 2, Saltasaurus (dorsal view / vista dorsal); 3, Opisthocoelicaudia (dorsal view / vista dorsal); 4, Camarasaurus (dorsal view / vista dorsal). S1, first sacral vertebrae / primera vértebra sacra; S7, last sacral vertebrae / última vértebra sacra. 2, 3 and 4 redrawn from Salgado et al. / redibujado de Salgado et al. , 1997. Scale bar / escala: 30 cm.

M. iliofibularis (ILFB). In extant archosaurs, this muscle originates fleshly from the lateral surface of the postacetabular lobe of the ilium. In crocodilians, it originates ventral to the Mm. iliotibiales, between IT2 and IT3. In Neornithes, it originates from the postacetabular iliac crest, under M. iliotibialis lateralis. It inserts on a tubercle on the anterolateral proximal fibular shaft and also on the lateral head of the M. gastrocnemius. In Neuquensaurus, as well as in Opisthocoelicaudia (Borsuk-Bialynicka, 1977), the ILFB originates from the postacetabular ilium. The ILFB inserts onto the fibular lateral tuberosity, a prominent bump on the proximolateral fibula (figures 2 and 3). This structure is a characteristic of Neuquensaurus and more developed in this species than in other titanosaurs (Powell, 2003). We do not reconstruct the secondary tendinous attachment to the origin of M. gastrocnemius because of the lack of osteological data (Level I´ of inference).
Function. In extant archosaurs, M. iliofibularis flexes the knee joint, extends and abducts the hip (Hutchinson, personal communication). Neuquensaurus has a well developed postacetabular lobe of the ilium, much more than other sauropods (figure 4). Besides, in MCS-5, both preacetabular and postacetabular processes of the ilium are outwardly expanded (Salgado et al., 2005). Then, the line of action of the muscle might be anterior-posterior. In addition, this muscle abducts the hip, as in extant archosaurs.
Mm. femorotibiales (FMT). This muscle presents two heads in crocodilians (M. femorotibialis externus/ FMTE and internus/FMTI). In Neornithes, the external head corresponds to M. femorotibialis lateralis (FMTL) and the internal to M. femorotibialis medialis (FMTM) and intermedius (FMTI). In crocodilians both heads are separated by M. iliofemoralis. The M. femorotibialis externus originates from the anterior surface along the femoral shaft, while M. femorotibialis internus originates from the posterior surface of the shaft, immediately besides the insertion of M. iliofemoralis. This muscle has a fleshy origin, although three intermuscular lines are related to the limits of the origin of this muscle and represent an ancestral condition for archosaurs (Hutchinson, 2001b): linea intermuscularis cranialis, caudalis and lateralis, which run along the anterior, posterior, and lateral surfaces, in the femur respectively (for details see Hutchinson, 2001b). Neuquensaurus presents a structure on the anterior surface of the femur which corresponds to the linea intermuscularis cranialis. It runs along the long axis of the shaft, from the proximal edge of the medial condyle to the base of the greater trochanter (figures 2 and 3). Both heads of Mm. femorotibiales would have originated from the anterior surface of the femoral shaft, separated by the linea intermuscularis cranialis, and inserted onto anterolateral cnemial crest, together with Mm. iliotibiales and M. ambiens, as in extant archosaurs (figure 2). The anterior intermuscular line separates both heads of Mm. femorotibiales. Neuquensaurus probably has a well developed Mm. femorotibiales due to the extensive lateromedial development of the femur, as in other titanosaurs (Salgado et al., 1997; Wilson and Sereno, 1998).
Function. In extant archosaurs, Mm. femorotibiales are a knee extensor muscle group. They extend the femorotibial joint. In mammals, the Quadriceps femoris
group, with similar origin and insertion, fulfills the same action (Sisson, 1982). In Neornithes (Feduccia 1982), prosauropods (Cooper, 1981; Galton, 1990) and basal sauropods such as Vulcanodon (Cooper, 1984; Wilson and Sereno, 1998) the cnemial crest is cranially orientated, and in Eusauropoda it is laterally projected (Wilson and Sereno, 1998) shifting laterally the line of action of the muscle. In Neuquensaurus, the function may be similar to that in extant archosaurs, even though the former has not patella. As aforementioned, Neuquensaurus has well developed fibular condyles of the femur (figure 5), so it is possible that this sauropod could achieve an effect analogous to that of patella.

Figure 5. Relative development of the mediolateral diameter of the femur and femoral distal condyles in several sauropods / desarrollo relativo del diámetro mediolateral y cóndilos distales del fémur en distintos saurópodos. 1, Barapasaurus tagorei; 2, Dicraeosaurus; 3, Camarasaurus; 4, Brachiosaurus brancai; 5, Opisthocoelicaudia skarzynskii; 6, Neuquensaurus australis. Black arrow shows the fibular (lateral) condyle; white arrow shows the tibial (medial) condyle; m, minimum transverse width. Redrawn from Salgado et al. / la flecha negra indica el cóndilo fibular (lateral); la flecha blanca indica el cóndilo tibial (medial); m, ancho transverso mínimo. Modificado de Salgado et al. (1997).

M. ambiens (AMB). In crocodilians, the AMB has a double head, although in the rest of Reptilia (including birds), there is a single one. In non-crocodilian reptiles, the single head of the AMB originates from the pubic tubercle (ambiens process), anteroventral to the acetabulum, near the pubo-iliac joint. In crocodilians, the forward head originates from the anterior union of the ilium and the preacetabular ramus of the ischium. Near the knee joins the external head of M. gastrocnemius. The posterior head originates from the ventral portion of the pubis and joins Mm. iliotibiales (Romer, 1923a; see also Wilhite, 2003). In Neornithes, the single head, equivalent to the posterior head of crocodilians, originates from the pectineal process (pubic spine sensu Vanden Berge, 1982) and inserts onto the cnemial crest, together with muscles of the Triceps femoris group, as in crocodilians (Romer, 1923a; Carrano and Hutchinson, 2002 fig. 3 E).
In Neuquensaurus, M. ambiens likely originated from the pubic tubercle. As in extant archosaurs, it inserted onto the cnemial crest (figure 2). It is not possible to identify the accessory insertions to the M. gastrocnemius; then we assume a Level I´ inference. Wilhite (2003) does not mention the insertion onto the cnemial crest. Besides an insertion onto the M. gastrocnemius, he proposes part of the insertion onto the proximoanterior surface of the fibula. This last could be possible because in Neuquensaurus that region of the fibula has vertical grooves that could represent sites of attachment.
Function. This muscle has different actions, depending of the group. In crocodilans, it extends the femorotibial joint; in Neornithes, it contributes to the extension of the tibiotarsus and adducts the leg. The mammalian homologue, the M. pectineous, adducts the leg and flexes the hip joint (Sisson, 1982). Due to the
parasagittal limb posture and similar origins and insertion areas in Neuquensaurus, as in birds and mammals, this muscle could present a combined action, extending the femorotibial joint, and adducting the leg.

Mm. Adductores Femores

M. adductor femoris (ADD). In extant archosaurs this muscle presents two heads, which originate from the lateral surface of the ischium and insert on the posterior surface of the femur. In crocodilians, M. adductor femoris 1 (M. puboischiofemoralis pars medialis/ PIFM of birds) originates fleshly from the anterolateral surface of the ischium, associated with the obturator process (Hutchinson and Gatesy, 2000; Hutchinson, 2001a) and separated from ADD2 by the M. puboischiofemoralis externus 3 (PIFE3). The M. adductor femoris 2 (M. puboischiofemoralis pars lateralis/PIFL of birds) originates from the posterolateral ischium. In Neornithes, PIFM and PIFL originates from a topologically similar position, although they are not separated by PIFE3. In extant archosaurs they inserts on the posterior shaft of the femur. Hutchinson (2001b) proposes a lateral insertion for extant archosaurs, between the linea intermuscularis caudalis (mentioned above) and the adductor ridge, distal to the fourth trochanter (medially) (see also Ballman, 1969; Baumel and Witmer, 1993). In Neuquensaurus, M. adductor femoris originated fleshly from the external border of the ischium, as in crocodilians and birds (Romer, 1923a; Wilhite, 2003), and a similar condition is present in Opisthocoelicaudia (Borsuk-Bialynicka, 1977). It inserted onto the posterior surface of the femur (figure 2).
Function. The ADD is the principal femoral adductor of crocodilians. In birds, this muscle also helps with the extension of the femur. In mammals, there are several muscles that help with the adduction of the leg. Nevertheless, M. adductor itself could be analogous to the archosaurian M. adductor femoris, because they have similar origins and insertions. Those muscles adduct the leg, extend the hip joint and medially rotate the femur (Sisson, 1982; Vanden Berge, 1982).
In the Patagonian sauropod, the ADD could have a combined action that includes adduction, medial rotation and retraction of the femora.

Mm. Caudofemorales

M. caudofemoralis brevis (CFB). In crocodilians, the CFB originates from the last sacral vertebrae, and from mediolateral surface of postacetabular ilium as well as from the first caudal. The avian homologue (M. caudofemoralis pars pelvica/CFP) originates from the lateral surface of the postacetabular ilium, behind the M. ischiofemoralis origin and beneath M. iliofibularis origin. This leads to a shift in the origin of the muscle, from a medial to a lateral surface. In Tyrannosaurus rex, the CFB originates from the brevis fossa, a transitional structure located on the medial surface of the ilium (Carrano and Hutchinson, 2002). The M. caudofemoralis brevis inserts proximomedial on the femur, proximal to the fourth trochanter. In Neuquensaurus, the CFB probably originated from the postacetabular ilium, although it is not possible to recognize its origin precisely because sauropods lack a brevis shelf and fossa. The fourth trochanter of the femur lacks any differential site of insertion, so we assume the existence of a common tendon with M. caudofemoralis longus (CFL) (see also Wilhite, 2003) (figures 2 and 3).
Function. This muscle aids CFL in femoral retraction.

M. caudofemoralis longus (CFL). In crocodilians, this muscle originates from the sides and ventral surface of the transverse processes of caudal vertebrae 3-15 (Romer, 1923a; Galton, 1969, but see Discussion). In Alligator, Wilhite (2003) places the origin of the CFL from the ventral surface of proximal caudal centra as well as on lateral surface of the first 13 chevrons. It inserts on the fourth trochanter of the femur, distal to the insertion of CFB (Galton, 1969; Hutchinson, 2001b). A secondary CFL tendon joins M. gastrocnemius and inserts on the knee region, close to the fibula (Romer, 1923a) or onto the distal femur and proximal tibia (Wilhite, 2003). In birds the fourth trochanter reduces substantially (Gatesy, 1990; Hutchinson, 2001b); this is carried to the extreme in Neornithes, where the M. caudofemoralis pars caudalis (CFC), homologous to the crocodilian M. caudofemoralis longus, is reduced or lost.
In some titanosaurian sauropods (e.g. Titanosaurus araukanicus), Salgado and García (2002) observed certain modifications on proximal caudals that affect the inclination and development of lateral sides and the width of the ventral sides. This involves a lateral inclination of the lateral sides of the first 7-8 caudal vertebrae, determining a narrow ventral surface. Then, a secondary lateral side arises from subsequent vertebrae. This variation has a direct correlation with the distribution and extent of M. caudofemoralis longus (Salgado and García, op. cit., figures 1 and 2). In this sense, this muscle originates from the primary lateral sides of the proximal caudals, under the transverse processes, up to that vertebrae in which observes the last trace of the primary side.
In Neuquensaurus, there are two interesting patterns. First, in the proximal caudals the replacement of primary sides for secondary sides can be seen. Second, anterior chevrons have a well developed antero-posterior expansion, determining a broad lateral surface (figure 6). As seen in Caiman, this area could represent the site of origin of CFL. It inserts on the femoral fourth trochanter, together with CFB (figure 2).


Figure 6. Anterior haemal arches of Neuquensaurus australis (MCS-5/32) in 1, anterior and 2, lateral view / arcos hemales anteriores de Neuquensaurus australis (MCS-5/32) en 1, vista anterior y 2, vista lateral. Scale / escala: 6 cm.

Function. M. caudofemoralis longus is the principal retractor of the femur, and also contributes to the long axis rotation and adduction of the leg, as in extant archosaurs.

Mm. Gastrocnemii

M. gastrocnemius (G). It's the most important crural muscle. In birds, is formed commonly by three heads (M. gastrocnemius lateralis/GL; M. gastrocnemius medialis/GM and M. gastrocnemius internus/GI); and crocodilians have two (M. gastrocnemius externus/ GE and M. gastrocnemius internus/GI). The lateral (external) head originates from the posterior distal femur, near the base of the lateral condyle. The internal head originates from the medial surface of the proximal tibia. Both heads inserted via tendon onto the plantar surface of the pes (Carrano and Hutchinson, 2002). In Neuquensaurus there is not a double attachment in the femur, so we assume a unique site of origin, as in crocodilians (GE). Muscle gastrocnemius internus originates from the medial surface of proximal tibia, as in extant archosaurs (figure 2). The insertions onto the pes are not discussed here.
Function. In birds extends the tarsus-metatarsus and flexes the fingers. In mammals the M. gastrocnemius flexes the femorotibial joint. In Neuquensaurus the action should be similar to that of mammalians because in the Patagonian specimen, the hindlimb resembles to that of mammalians than to the Neornithes.
Mm. Ischiotrochantericus (ISTR). The ISTR is a small muscle that, in crocodilians, originates from the posteromedial surface of the ischium. In Neornithes, its homologue, the M. ischiofemoralis (ISF), arises from the lateral surface of that bone, under the M. iliofibularis, as well as the ilio-ischiadic membrane. The ISTR inserts on posterolateral proximal femur, distal to the greater trochanter (Hutchinson, 2001b).
Novas (1996) and Hutchinson (2001b) propose an insertion on the caudal portion of the trochanteric shelf, a posterodistally oriented sigmoid process on the lateroproximal surface of the femur. The ISTR is prominent in Dinosauriformes. Ornitischians and saurischians present a derived condition in which this muscle appears reduced or even absent (Novas, 1996).
It is not possible to reconstruct with confidence the precise origin of M. ischiotrochantericus in Neuquensaurus because of the lack of any osteological evidence. However, we assume the existence of M. ischiotrochantericus in the specimen MCS-5/24 due to the presence in crocodilians and birds (Level I´ inference). The femur of Neuquensaurus presents a slight sigmoid structure, vertically oriented, that could represent the trochanteric shelf. In that case, M. ischiotrochantericus would insert there (figure 2 and 3). Following Hutchinson (2001b), that feature is conserved throughout archosaurian evolution and differences correspond to character states.

Function. In birds, the M. ischiofemoralis laterally rotates the femur, generating the "lateral displacement" of the body, and contributing to the extension of the femur (Vanden Berge, 1982). The mammalian M. gemellus, which has a similar origin and insertion that the ISTR of extant archosaurs, shares the same function. It is assumed that M. ischiotrochantericus of Neuquensaurus has same the action than of birds and mammals.

Discussion

Previous reconstructions of the hindlimb musculature of sauropod dinosaurs, focused predominantly on comparisons with crocodilians (Romer, 1923a; Borsuk-Bialynicka, 1977; Wilhite, 2003). Our study highlights the significance of integrating both crocodilians and birds as a phylogenetic framework in order to identify osteological correlates of muscle attachments, and thus allowing associated soft tissues to be mapped with confidence. Below, we will discuss muscle homologies on the basis of previous studies on extant and extinct archosaurs, evaluating the confidence of each reconstruction. Second, we will analyze the inferred function of each muscle by comparison with non-saltasaurinae sauropods, and evaluate the musculoskeletal and biomechanical correlates of the sauropods "wide-gauge" trackmakers.

Muscle homologies

Several difficulties appear when trying to identify osteological correlates and interpret the soft tissues of any extinct taxa. This is because many muscles present in Crocodylia show shifted areas of origin and insertion in birds, reducing or even disappearing due to differences in locomotor habits (Romer, 1923a; Hutchinson, 2001a, 2001b; Carrano and Hutchinson, 2002).
For example, M. femorotibiales has a fleshy origin in extant archosaurs and thus there is not a direct osteological correlate (e.g. grooves or ridges) for this muscle in Neuquensaurus; although the origin of this muscle is associated with the linea intermuscularis cranialis, which is present in both extant bracket taxa as well as in Neuquensaurus. Therefore, it is possible to reconstruct the origin of both heads of this muscle with confidence. This pattern is also seen in other dinosaurs such as Hypsilophodon (Galton, 1969) and Tyrannosaurus (Carrano and Hutchinson, 2002); although such divisions are not indicated on ankylosaurs (Coombs, 1979). Neither Wilhite (2003) nor
Borsuk-Bialynicka (1977) establishes separations for the heads of this muscle.
Borsuk-Bialynicka (1977, fig. 17 B) proposes the lateral trochanter of the fibula as the origin site for the M. flexor digitorus longus, whereas she places M. iliofibularis insertion on the posterolateral surface of the fibula, an opinion also shared by Wilson and Sereno (1998). We do not have reasons to support here this interpretation, since in extant archosaurs, the M. iliofibularis inserts on a tubercle (lateral trochanter) placed on the anterolateral proximal fibular shaft (Carrano and Hutchinson, 2002; Wilhite, 2003). Brochu (2003) also places M. iliofibularis insertion scar in a similar area in Tyrannosaurus. In Neuquensaurus, we place M. iliofibularis insertion scar onto the fibular lateral trochanter (figures 2 and 3), a prominent bump on the proximolateral fibula, an equivalent surface as that of extant archosaurs. Galton (1969) and Coombs (1979) besides found the insertion scar onto the lateral surface of proximal fibula, a similar position to that seen in Neuquensaurus.
Another muscle that deserves consideration is M. adductor femoris. As Hutchinson and Gatesy (2000) and Hutchinson (2001b) note, the linea intermuscularis caudalis and the adductor ridge are linked to the insertion of this muscle in extant archosaurs. In Neuquensaurus, as well as in other sauropods (see Borsuk-Bialynicka, 1977; Wilhite, 2003) there is no possible to identify the double heads of this muscle, due to the lack of any osteological correlates, maybe because that muscle could have a fleshy attachment. Therefore, in the patagonian genus the M. adductor femoris is a Level I´ inference.
In crocodilians, M. caudofemoralis brevis originates from the medial surface of postacetabular ilium, and in Neornithes, from the lateral surface of the postacetabular ilium. In Tyrannosaurus rex, the CFB originates from the brevis fossa, which is placed on the medial surface of the ilium (Hutchinson, 2001b). As noted by Carrano and Hutchinson (2002), this marks a shift of the origin in the line of birds, from a medial (in crocodilians) to a lateral (Neornithes) position, showing a transitional state in Tyrannosaurus. The ilium of Neuquensaurus lacks any feature associated to that muscle; however, not reconstructing M. caudofemoralis brevis in Neuquensaurus would be a Level III´ Inference and thus, requires more speculation because crocodilians and birds have the muscle. So we infer the presence of that muscle in Neuquensaurus (Level I´ inference).
The precise origin of M. caufofemoralis longus is also a subject of discussion in extant as well as in extinct archosaurs (Romer, 1923a; Galton, 1969; Borsuk- Bialynicka, 1977; Carrano and Hutchinson, 2002; Wilhite, 2003). There are two principal difficulties in reconstructing the origin of this muscle in fossils. First, there is not agreement on where to place its area of origin. In crocodilians, Romer (1923a) and Galton (1969) place it from the sides and ventral surface of proximal caudal vertebrae, while Wilhite (2003) argued that the M. caudofemoralis longus originates from
the ventral surface of proximal caudals and the first 13 chevrons, based on his dissection of appendicular musculature of Alligator mississipiensis. Second, in previous reconstructions of this muscle in dinosaurs (Borsuk-Bialynicka, 1977; Coombs, 1979) the place for the origin of this muscle is not clearly shown, because any of these authors specify the surfaces of attachment on the vertebrae. In Neuquensaurus, combining the interpretation of Salgado and García (2002) and Wilhite (2003), we assume the origin of the M. caudofemoralis longus with high confidence from the primary lateral sides of proximal caudal vertebrae as well as from the lateral surfaces of proximal chevrons. Respect to the insertion of the CFL, we do not reconstruct the secondary tendon of this muscle because, although crocodilians have it, it was lost in the line to birds (Hutchinson, 2001b) Reconstruction of this muscle in Neuquensaurus is an inference of Level II´.
Finally, M. ischiotrochantericus has not been mentioned in previous studies on sauropod pelvic musculature (Wilhite, 2003; Borsuk-Bialynicka, 1977). However, not reconstructing the origin of this muscle in Neuquensaurus would be very speculative (Level III´ inference). In archosaurs, there is a wide consensus in placing the insertion of M. ischiotrochantericus onto the posterolateral femur, near the greater trochanter (Romer, 1923a; Galton, 1969; Coombs, 1979; Hutchinson, 2001b; Carrano and Hutchinson, 2002; Wilhite, 2003), although few authors specify its origin in dinosaurs (Hutchinson, 2001b; Carrano and Hutchinson, 2002). The osteological correlate of this muscle is the trochanteric shelf, a sigmoid process placed on the lateroproximal surface of the femur, present in Dinosauriomorpha and reduced among ornithischians (Novas, 1992; Hutchinson, 2001b). Neuquensaurus has a structure behind the greater trochanter that corresponds to the trochanteric shelf. Therefore M. ischiotrochantericus would insert there.

Muscle function

We infer the function of each muscle by comparison with extant archosaurs and mammals. In terms of locomotory posture, extant crocodilians are considered to be intermediate between sprawlers and erect, adopting a semi-erect posture, where the body is held half-way between sprawling and erect grades during locomotion (Reilly and Elias, 1998). Birds and mammals have a parasagittal limb posture, in which the limbs are beneath the body, adopting a fully upright posture. Such differences in limb posture are correlated with differences in the musculoskeletal system, limb kinematics and muscle function (Blob, 2000, 2001; Carrano, 2000). For that reason, caution is warranted when the function of an extinct animal, such as a dinosaur, is inferred by comparison with living forms with different locomotor habits, such as crocodilians and birds (see Lauder, 1995). For that reason, mammals provide an additional source of comparison by analogy in our study.
Figure 7 shows vectors and moment arm of some hindlimb muscles of Neuquensaurus. The principal differences between the muscle function of Neuquensaurus and non- Saltasaurinae sauropods are related to the moment arms of the inferred lines of action of the described muscles. The extent of the preacetabular and postacetabular iliac blades in Neuquensaurus (figure 4) determines greater moment arms, increasing the force action. The muscles involved are the cranial portion of the Mm. iliotibiales and M. iliofibularis. In Neuquensaurus, the preacetabular and postacetabular process of the ilium are well developed, determining a more distal attachment of the cranial portions of that muscles, increasing the moment arm about the hip joint for those portions, and also improving the extension action for the IT and ILFB. It would produce more force for the muscle but less velocity (Clair, 1982).

Figure 7. Lines of actions and moment arms of some hindlimb muscles of Neuquensaurus. The lines of action of each muscle (heavy arrows) run approximately between the midpoints of origin and insertion, with the arrowhead directed at the origin. Moment arms are shown by lighter lines. The axes of femoral and tibial rotation are indicated by empty circles. Muscle abbreviations are given in table 1 / líneas de acción y brazos de momento de algunos músculos del miembro posterior de Neuquensaurus. Las líneas de acción de cada músculo (flechas gruesas) corren aproximadamente entre los puntos medios de origen e inserción, con la punta de la flecha dirigida hacia el origen. Los brazos de momento se indican líneas más delgadas. Los ejes de rotación del fémur y la tibia se indican con círculos vacíos. Las abreviaturas de los músculos se indican en la tabla 1.

The medial and posterior portions of Mm. iliotibiales show lower moment arms respect the hip joint in the antero-posterior direction (figure 7). The latter could be because the lines of action of those portions of the muscle are acting principally in a mediolateral direction, improving the abduction-adduction action.
There is no evidence of patella in sauropod dinosaurs. That structure acts as a pivot of the muscle,
increasing the moment arm about the knee joint. However, saltasauriniae sauropods femora have distal lateral (fibular) condyles well developed and expanded at the knee joint (figure 5). It is possible that this sauropod could achieve an effect analogous to that of patella (see also Salgado et al., 1997, figure 10; Wilson and Carrano, 1999). The action of Mm. femorotibiales in Neuquensaurus should be similar with respect to extant archosaurs despite of the lack of patella.
M. ambiens shows two interesting patterns in our specimen. It present lower moment arm about the knee join but higher ones respect the hip joint. It suggests that M. ambiens could achieve more importance adducting the leg than extending the femorotibial joint (figure 7). Mm. adductores femores also shows one of the highest moment arms about the hip joint.
In Neuquensaurus, Mm. femorotibiales and Mm. gastrocnemmii have lower moment arms in relation to the others muscles. However, both muscles have relative high moment arms, in relation to other non- Saltasaurinae sauropods (e.g. Barapasaurus, Dicraeosaurus, Camarasaurus) because of the expanded distal condyles of the femur that increases the moment of the force about the knee joint (figures 5 and 7).

Musculoskeletal patterns in wide-gauge trackmakers: morpho-functional implications

As Wilson and Carrano (1999) notes, sauropod trackways fall into two major ichnotypes: "widegauge" and "narrow-gauge", referring to the relative distance of the prints from the trackway midline (see also Farlow, 1992; Lockley et al., 1994; Moratalla et al. 1994; Wilhite, 2003). Wide-gauge trackmakers (e.g. Saltaraurinae sauropods) have developed several aspects of pelvic and hindlimb morphology such as a wide pelvic girdle, an angled femoral axis and an eccentric femoral midshaft cross-section (Wilson and Carrano, op. cit.). These particularities of the hindlimb have a direct correlate with the posture and limb orientation, but also with the arrangements of the associated musculature and hence the biomechanics of locomotion.
First, the wide pelvic girdle in Neuquensaurus and, in general, in all titanosaurids, is a result of the nearly horizontal and laterally projected preacetabular process of the ilium (figure 4). The anterior portion of the Mm. iliotibiales is involved in that skeletal change, shifting its origin far and laterally from the hip joint, producing: 1) greater moment arms, improving the
force action respect to a "narrow-gauged" sauropods, such as diplodocids, in which the preacetabular blades of the ilium are not outwardly oriented; and 2) an increasing on the antero-posterior component of the line of action of the muscle, improving the extension action. In a femur angled laterally outward from the acetabulum, produced by "beveled" distal condyles, the lines of action of the adductor musculature act in different ways than in narrow-gauged sauropods, in which the femora are straight shafted (figure 8). In wide-gauged trackmakers, such as Neuquensaurus, the legs are not just beneath the body, but laterally outward from the hip joint; then, the line of action of the M. adductor femoris has a principal mediolateral component, improving the adduction action, in relation to narrow-gauged sauropods. Finally, the eccentricity of the femoral midshaft cross-section is a result of the mediolateral ("bending") forces acting on the femora, determining a morphology in which the mediolateral diameter is great that the anteroposterior (figure 7) (see also Wilson and Carrano, 1999; Carrano, 2001). It determines a broad surface of the anterior femur for the origin of M. femorotibialis, a muscle that contributes together with the M. iliotibialis in the extension of the leg.


Figure 8. Femoral morphology within Sauropoda. Right femora in anterior view of 1, diplodocid (sensu Wilson, 2002); 2, titanosauriform (sensu Salgado et al., 1997) and 3, Saltasaurinae (sensu Powell, 1992). Ellipses at right of each femur are reconstructed cross-sections. Solid lines centered at base of femora mark the vertical and horizontal; solid line at proximal end of femur represents the vertical distance between the femoral head (hip) and the center of distal condyles (knee). Dashed line in 3 passes through the long axis of the femur. Modified from Wilson and Carrano (1999) / Morfología femoral dentro de Sauropoda. Fémures derechos en vista anterior de 1, diplodocoideo (según Wilson, 2002); 2, titanosauriforme (según Salgado et al., 1997) y 3, saltasaurino (según Powell, 1992). Las elipses del lado derecho de los fémures representan la sección media. Las líneas sólidas centradas en la base de los fémures marcan la vertical y la horizontal; la línea sólida en la parte proximal del fémur representa la distancia vertical entre la cabeza femoral (cadera) y el centro de los cóndilos distales (rodilla). La línea punteada en 3 pasa a través del eje mayor del fémur. Modificado de Wilson y Carrano (1999).

Conclusions

In this study we evaluate the probable origin, insertions, and actions of hindlimb musculature of Neuquensaurus, on the basis of comparisons with extant archosaurs (both crocodilians and birds). The methodology utilized (EPB) allows us to find osteological correlates to establish homologies and thus reconstruct muscles, reducing speculation. Our analysis reveals several differences in the origin of the muscles, their insertions, their extents and their actions, with respect to extant archosaurs and with previous studies on other sauropods. This study also clarifies certain ambiguities related to the origins and insertions of some controversial muscles (e.g. M. iliofibularis, M. ischiotrochantericus and M. caudofemoralis longus).
Functional analysis of pelvic and hindlimb musculature reveals several differences with respect to extant taxa and narrow-gauged sauropods, related primarily to morphological variations in the pelvis and femora. That is the cases of the Triceps femoris group and M. iliofibularis, whose origins on the preacetabular and postacetabular lobes of the ilium, respectively, increases the extension-flexion action; and Mm. adductores femores, which has achieved a principal mediolateral component in wide-gauged trackmakers.

Acknowledgements

We are grateful to L. Salgado, R. Fariña and J.A. Wilson, whose critical reviews improved the manuscript. We also would like to thank L. "el gordo" Pérez, J. "Lucy" Canale and "Juje" Haluza for useful comments on an earlier draft of this paper. To J. Noriega and C. Piña (CICYTTP) for their generous help and for bringing us the Caiman specimens. To M. Reguero, J. Williams, E. Etcheverry, J. Powell and I. Cerda for permitting us access to the collections under their care.

References

1. Alexander, R.McN and Pond, C. 1992. Locomotion and bone strength of the white rhinoceros, Ceratotherium simum. Journal of Zoology 227: 63-69.        [ Links ]

2. Apesteguía, S. 2002. Successional structure in continental tetrapod faunas from Argentina along the Cretaceous. Boletim do 6º simpósio sobre o Cretáceo do Brasil/2º Simposio sobre el Cretácico de América del Sur (Rio Claro), pp. 135-141.        [ Links ]

3. Ballman, P. 1969. Les oiseaux miocènes de la Grive-Saint-Alban (Isère). Geobios 2: 157-204.        [ Links ]

4. Baumel, J.J. and Witmer, L.M. 1993. Osteologia: 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. 2º ed. Cambridge, Massachussets. Publications of the Nutall Ornithological Club 23: 45-132.        [ Links ]

5. Benton, M.J. 2004. Origin and relationships of Dinosauria. In: D.B. Weishampel, P. Dodson and H. Osmólska (eds.), The Dinosauria, University of California Press, Berkeley, pp. 7-19.        [ Links ]

6. Biewener, A.A. 1983. Locomotory stresses in the limb bones of two small mammals: the ground squirrel and chipmunk. Journal of Experimental Biology 103: 131-154.        [ Links ]

7. Blob, R.W. 2000. Interespecific scaling of the hindlimb skeleton in lizards, crocodilians, felids and canids: does limb bone shape correlate with limb posture? Journal of Zoology 250: 507-531.        [ Links ]

8. Blob, R.W. 2001. Evolution of hindlimb posture in nonmammalian therapsids: biomechanical tests of paleontological hypoteses. Paleobiology 27: 14-38.        [ Links ]

9. Blob, R.W. and Biewener, A.A. 2001. Mechanics of limb bone loading during terrestrial locomotion in the green iguana (Iguana iguana) and American alligator (Alligator mississippiensis). The Journal of Experimental Biology 204: 1099-1122.        [ Links ]

10. Borsuk-Bialynicka, M. 1977. A new camarasaurid sauropod Opisthocoelicaudia skarzynskii, gen. n., sp. n. from the Upper Cretaceous of Mongolia. Palaeontologia Polonica 37: 1-64.        [ Links ]

11. Brochu, C.A. 2001. Progress and future directions in archosaur phylogenetics. Journal of Paleontology 75: 1185-1201.        [ Links ]

12. Brochu, C.A. 2003. Osteology of Tyrannosaurus rex: Insights from a Nearly Complete Skeleton and High -Resolution Computed Tomographic Analysis of the Skull. Society of Vertebrate Paleontology Memoir 7:1-138.        [ Links ]

13. Bryant, H. and Seymour, K.L. 1990. Observation and comments on the reliability of muscle reconstruction in fossil vertebrates. Journal of Morphology 206: 109-117.        [ Links ]

14. Carrano, M.T. 1997. Morphological indicators of foot posture in mammals: a statistical and biomechanical analysis. Zoological Journal of the Linnean Society 121: 77-104.        [ Links ]

15. Carrano, M.T. 1998. Locomotion in non-avian dinosaurs: integrating data from hindlimb kinematics, in vivo strains, and bone morphology. Paleobiology 24: 450-469.        [ Links ]

16. Carrano, M.T. 1999. What, if anything, is a cursor? Categories versus continua for determining locomotor habit in mammals and dinosaurs. Zoological Journal of London 247: 29-42.        [ Links ]

17. Carrano, M.T. 2000. Homoplasy and the evolution of dinosaur locomotion. Paleobiology 26: 489-512.        [ Links ]

18. Carrano, M. T. 2001. Implications of limb bone scaling, curvature and eccentricity in mammals and non-avian dinosaurs. Journal of Zoology 254: 41-55.        [ Links ]

19. Carrano, M.T. and Hutchinson, J.R. 2002. Pelvic and Hindlimb musculature of Tyrannosaurus rex (Dinosauria: Theropoda). Journal of Morphology 253:207-228.        [ Links ]

20. Chiappe, L.M. 1995. The first 85 million years of avian evolution. Nature 378: 349-355.        [ Links ]

21. Clair, L.E. St. 1982. Miología general. In: S. Sisson y J.D. Grossman, (eds.), Anatomía de los animales domésticos, Vol. 1, Salvat, Barcelona, pp. 45-55.        [ Links ]

22. Coombs, W.P. 1979. Osteology and myology of the hindlimb in the Ankylosauria (Reptilia, Ornithischia). Journal of Paleontology 53: 666-684.        [ Links ]

23. Cooper, M.R.A. 1981. The prosauropod dinosaur Massospondylus carinatus Owen from Zimbawe: its biology, mode of life and phylogenetic significance. Occasional Papers of the National Museums and Monuments of Rhodesia B, Natural Sciences 6: 689-840.        [ Links ]

24. Cooper, M.R.A. 1984. A reassesment of Vulcanodon karibaensis Raath (Dinosauria: Saurischia) and the origin of the Sauropoda. Palaeontologia Africana 25: 203-231.        [ Links ]

25. Farlow, J.O. 1992. Sauropod tracks and trackmakers: integrating the ichnological and skeletal records. Zubia 10: 89-138.        [ Links ]

26. Feduccia, A. 1982. Osteología de las aves. In: S. Sisson y J.D. Grossman (eds.), Anatomía de los animales domésticos, Vol. 2, Salvat, Barcelona, pp. 1960-1973.        [ Links ]

27. Galton, P.M. 1969. The pelvic musculature of the dinosaur Hypsilophodon (Reptilia: Ornithischia). Postilla 131. p 64.        [ Links ]

28. Galton, P.M. 1990. Basal Sauropodomorpha-Prosauropoda. In: D.B. Weishampel, P. Dodson and H. Osmólska (eds), The Dinosauria, University of California Press, Berkeley, pp. 320-344.        [ Links ]

29. Gatesy, S.M. 1990. Caudofemoral musculature and the evolution of theropod locomotion. Paleobiology 16: 170-186.        [ Links ]

30. Gatesy, S.M. 1995. Functional evolution of the hindlimb and tail from basal theropods to birds. In: J.J. Thomason (ed.), Functional morphology in vertebrate paleontology, Cambridge University Press, Cambridge, pp. 219-234.        [ Links ]

31. Huene, F. 1929. Los Saurisquios y Ornitisquios del Cretácico Argentino. Museo de la Plata, Anales 3: 194 pp.        [ Links ]

32. Hutchinson, J.R. 2001a. The evolution of pelvic osteology and soft tissues on the line to extant birds (Neornithes). Zoological Journal of the Linnean Society 131: 123-168.        [ Links ]

33. Hutchinson, J.R. 2001b. The evolution of femoral osteology and soft tissues on the line to extant birds (Neornithes). Zoological Journal of the Linnean Society 131: 169-197.        [ Links ]

34. Hutchinson, J.R. 2004. Biomechanical Modeling and Sensitivity Analysis of Bipedal Hability. II. Extinct Taxa. Journal of Morphology 262: 441-461.        [ Links ]

35. Hutchinson, J.R. and Gatesy, S.M. 2000. Adductors, abductors, and the evolution of archosaur locomotion. Paleobiology 26: 734-751.        [ Links ]

36. Hutchinson, J.R, Anderson, F.C., Blemker, S.S. and Delp, S.L. 2005. Analysis of hindlimb muscle moment arms in Tyrannosaurus rex using a three dimensional musculoskeletal computer model: implications for stance, gait and speed. Paleobiology 31: 676-701.        [ Links ]

37. Lauder, G.V. 1995. On the inference of function from structure. In: J.J. Thomason (ed.), Functional morphology in vertebrate paleontology, Cambridge University Press, Cambridge, pp. 1-18.        [ Links ]

38. Lockley, M.G, Farlow, J.O. and Meyer, C.A. 1994. Brontopodus and Parabrontopodus ichnogen. nov. and the significance of wideand narrow-gauge sauropod trackways. Gaia 10: 135-145.        [ Links ]

39. Lydekker, R. 1893. The dinosaurs of Patagonia, Anales Museo de La Plata, Vol. 2.        [ Links ]

40. McIntosh, J.S. 1990. Sauropoda. In: D.B. Weishampel, P. Dodson and H. Osmólska (eds.), The Dinosauria, University of California Press, Berkeley, pp. 345-401.        [ Links ]

41. Moratalla, J.J., García-Mondéjar, J., Santos, V.F., Lockley, M.G., Sanz, J.L. and Jiménez, S. 1994. Sauropod trackways from the Lower Cretaceous of Spain. Gaia 10: 75-84.        [ Links ]

42. Novas, F.E. 1992. Phylogenetic relationships of the basal dinosaurs, the Herrerasauridae. Paleontology 35: 51-62.        [ Links ]

43. Novas, F.E. 1996. Dinosaur monophyly. Journal of Vertebrate Paleontology 16: 723-741.        [ Links ]

44. Powell, J.E. 1992. Osteología de Saltasaurus loricatus (Sauropoda- Titanosauridae) del Cretácico Superior del Noroeste Argentino. In: J.L. Sanz y A.D. Buscalioni (eds.), Los Dinosaurios y su entorno biótico. 2º Curso de Paleontología de Cuencas-Instituto 'Juan De Valdés'Cuencas, Actas pp. 165-230.        [ Links ]

45. Powell, J.E. 2003. [Revision of South American Titanosaurid dinosaur: palaeobiological, palaeobiogeographical and phylogenetic aspects. PhD. Dissertation, Records of the Queen Victoria Museum 111, Launceston, 173 pp.].        [ Links ]

46. Radinsky, L.B. 1987. The evolution of vertebrate design. The University of Chicago Press, Chicago, 188 pp.        [ Links ]

47. Reilly, S.M., and Elias, J. A. 1998. Locomotion in Alligator mississipiensis: kinematic effects of speed and posture and their relevance to the sprawling-to-erect paradigm. The Journal of Experimental Biology 201: 2559-2574.        [ Links ]

48. Romer, A.S. 1923a. Crocodilian pelvic muscles and their avian and reptilian homologues. Bulletin of the American Museum of Natural History 58: 533-552.        [ Links ]

49. Romer, A.S. 1923b. The pelvic musculature of saurischian dinosaurs. Bulletin of the American Museum of Natural History 58: 605-617.        [ Links ]

50. Romer, A.S. 1927. The pelvic musculature of ornithischian dinosaurs. Acta Zool, Stockholm 8: 225-275.        [ Links ]

51. Rowe, T. 1986. Homology and evolution of the deep dorsal thigh musculature in birds and other Reptilia. Journal of Morphology 189: 327-346.        [ Links ]

52. Salgado, L. 2000. [Evolución y paleobiología de los saurópodos Titanosauridae. Tesis Doctoral, Universidad Nacional de La Plata, La Plata, Argentina, 300 pp. Unpublished.].        [ Links ]

53. Salgado, L., Coria, R.A. and Calvo, J.O. 1997. Evolution of Titanosaurid Sauropods. I: Phylogenetic análisis base on the postcranial evidence, Ameghiniana 34: 3-32.        [ Links ]

54. Salgado, L. and Azpilicueta C. 2000. Un nuevo saltasaurino (Sauropoda, Titanosauridae) de la provincia de Río Negro (Formación Allen, Cretácico Superior), Patagonia, Argentina, Ameghiniana 37: 259-264.        [ Links ]

55. Salgado, L. y García, R. 2002. Variación morfológica en la secuencia de vértebras caudales de algunos saurópodos titanosaurios. Revista Española de Paleontología 17: 211-216.        [ Links ]

56. Salgado, L., Apesteguía, S. and Heredia, S.E. 2005. A new specimen of Neuquensaurus australis, a Late Cretaceous Saltasaurinae titanosaur from North Patagonia. Journal of Vertebrate Paleontology 25: 623-634.        [ Links ]

57. Sisson, S. 1982. Miología de los equinos. In: S.Sisson y J.D. Grossman (eds.), Anatomía de los animales domésticos, Vol. 1, Salvat, Barcelona, pp. 423-508.        [ Links ]

58. Upchurch, P. 1995. The evolutionary history of sauropod dinosaurs. Philosophical Transactions of the Royal Society of London, B 349: 365-390.        [ Links ]

59. Upchurch, P. 2004. Sauropoda. In: D.B. Weishampel, P. Dodson and H. Osmólska (eds.), The Dinosauria, University of California Press, Berkeley. Pp. 259-322.        [ Links ]

60. Vanden Berge, J.C. 1982. Miología de las aves. In: S. Sisson y J.D. Grossman (eds.), Anatomía de los animales domésticos, Vol. 2, Salvat, Barcelona, pp. 1973-2022.        [ Links ]

61. Vanden Berge, J.C. and Zweers, G.A. 1993. Myologia. In: J.J. Baumel, A.S. King, J.E. Breazile, H.E. Evans and J.C. Vanden Verge (eds.), Handbook of Avian Anatomy: Nomina Anatomica Avium. Second Edition, Publications of the Nutall Ornithological Club 23: 189-250.        [ Links ]

62. Wilhite, R. 2003. [Biomechanical reconstruction of the appendicular skeleton in three North American Jurassic Sauropods. Ph. D. dissertation. Louisiana State University, Baton Rouge, 198 pp.} Unpublished.].        [ Links ]

63. Wilson, J.A. 2002. Sauropod dinosaur phylogeny: critique and cladistic analysis. Zoological Journal of the Linnean Society 136: 217-276.        [ Links ]

64. Wilson, J.A. and Carrano, M.T.1999. Titanosaurs and the origin of "wide-gauge" trackways: a biomechanical and systematic perspective on sauropod locomotion. Paleobiology 25: 252-267.        [ Links ]

65. Wilson, J.A. and Sereno, P.C. 1998. Early evolution and higherlevel phylogeny of sauropod dinosaurs. Society of Vertebrate Paleontology Memoir 5:1-68.        [ Links ]

66. Witmer, L.M. 1995. The extant phylogenetic bracket and the importance of reconstructing soft tissues in fossils. In: J.J. Thomason (ed.), Functional morphology in vertebrate paleontology, Cambridge University Press, Cambridge, pp. 19-33.        [ Links ]

67. Witmer, L.M. 1997. The evolution of the antorbital cavity in archosaurs: a study in soft-tissue reconstruction in the fossil record with an analysis of the function of pneumaticity. Society of Vertebrate Paleontology Memoir 3: 1-73.        [ Links ]

Recibido: 30 de enero de 2007.
Aceptado: 9 de mayo de 2008.