Number 515 23 August 2007 u , UcS-k , “fOf^ Contributions IN Science The Xochixtlapilco Dinosaur Ichnofauna, Middle Jurassic of Oaxaca, Southeastern Mexico: Description and Paleontologic Significance Ismael FerrusquIa-Villafranca, Victor Manuel Bravo-Cuevas, and Eduardo Jimenez-Hid algo Natural of Los Angeles County Serial Publications OL THE Natural History Museum ol Los Angeles County Scientific Publications Committee John Heyning, Deputy Director for Research and Collections John M. Elarris, Committee Chairman Brian V. Brown Joel W. Martin Xiaoming Wang K. Victoria Brown, Managing Editor The scientific publications of the Natural History Museum of Los Angeles County have been issued at irregular in- tervals in three major series; the issues in each series are numbered individually, and numbers run consecutively, re- gardless of the subject matter. • Contributions in Science, a miscellaneous series of tech- nical papers describing original research in the life and earth sciences. • Science Bulletin, a miscellaneous series of monographs describing original research in the life and earth sciences. This series was discontinued in 1978 with the issue of Numbers 29 and 30; monographs are now published by the Museum in Contributions in Science. • Science Series, long articles and collections of papers on natural history topics. Copies of this publication are available through the Scholarly Publications Office at 213/763-3330 or by vis- iting our website at for a PDF file version. Natural History Museum OF Los Angeles County 900 Exposition Boulevard Los Angeles, Calieornia 90007 Printed at Allen Press, Inc., Lawrence, Kansas ISSN 0459-8113 The Xochixtlapilco Dinosaur Ichnofauna, Middle Jurassic of Oaxaca, Southeastern Mexico: Description and Paleontologic Significance Ismael Ferrusquia-Villafranca,^ Victor Manuel Bravo-Cuevas,^ and Eduardo Jimenez-Hidalgo^ ABSTRACT.. The Xochixtlapilco Dinosaur Ichnofauna was recovered from fine-grained, red phyllarenitic strata of the Middle Jurassic Tecocoyunca Group partim, which was laid down in a coastal lagoon, and dated as Early Bajocian-Early Bathonian on the basis of ammonites. The site lies in the Oaxacan Mixteca, southeastern Mexico. The ichnofauna chiefly consists of small footprints, whose makers are referred to as a “basal coelurosaur” (Morphotype A tracks); an undescribed sauropod taxon, probably of family rank (Morphotype C tracks); and an ankylopollexian ornithopod (Morphotype D tracks). There is also a single large footprint made by an Pallosaurid carnosaur (Morphotype B track). This small but relatively diverse ichnofauna is the southernmost record of Jurassic dinosaurs in North America, and adds a new^ fauna to the meager record of dinosaurs in Middle America. During the Jurassic the Mixteca territory (—Mixteca Terrane) was one of several small continental- crust blocks laid down in the inter-American/African space, as Pangea became disassembled. Ecologically such a scenario corresponded to an isolated setting where limited space and resources might have induced selective pressures toward small size, especially to the primary consumers and associated predators; it also shielded the island fauna from competition and exchange with neighboring continental faunas. Nonetheless, the Xochixtlapilco dinosaur fauna shows a closer biogeographic/ phylogenetic resemblance to the North American faunas than to the South American or African faunas; how^eve-r, the meaning of this fact can not be fully ascertained at present because of the Xochixtlapilco fauna’s small size. RESUMEN. La Dinosauricnofauna Xochixtlapilco fue recolectada de estratos rojos, finogranudos fiiareniticos del Grupo Tecocoyunca partim, depositados en una laguna costera, y fechados como bajocianos tempranos-batonianos tempranos, con base en amonitas. El sitio se encuentra en la Mixteca Oaxaquena, sureste de Mexico. La icnofauna principalmente consiste de huellas podiales pequehas, cuyos autores son referibles a un “celurosaurio basal” (huellas del Morfotypo A), a un taxon no descrito de sauropodo, probablemente de rango de familia (huellas del Morfotipo C), y a un ornitopodo ankyiopolexiano (huellas del Morfotipo D); tambien hay una sola huella grande, hecha por un carnosaurio Paliosaurido (huella del Morfotypo B). Esta pequena, pero relativamente diversa icnofauna es ei registro mas austral de dinosaurios jurasicos en Norteamerica, y agrega una nueva icnofauna al escaso registro de dinosaurios en Mesoamerica. Los modelos de tectonica de placas sobre la evolucion geologica/tectonica de Mesoamerica, presentan al territorio Mixteco (— Terreno Mixteco) durante el Jurasico, como uno de los pequehos bloques de corteza continental dispuestos en el espacio interamericano-africano, a medida que Pangea se desmembraba. Ecologicamente tal escenario correspondia a un marco isleho, donde lo limitado del espacio y recursos, pudieron haber inducido presiones de seleccion hacia tamaho pequeho, especialmente en los consumidores primarios y en los depredadores asociados a ellos; dicho escenario protegio a la fauna isleha de competencia e intercambio con faunas continentales vecinas. Sin embargo, la dinosaurofauna Xochixtlapilco muestra un mayor parecido biogeografico/filogenetico con la fauna norteamericana que con la sudamericana o la africana; al presente; sin embargo el significado de este hecho no puede establecerse aun, debido al pequeho tamaho de la fauna Xochixtlapilco. INTRODU CTION somewhat better known than those of Asia and Africa. The Jurassic assemblages have a high The Mesozoic tetrapod faunas of temperate percentage of shared taxa, which in turn suggests North and South America and Europe are land connections that allowed migration of taxa 1. Institute de Geologia, Universidad Nacional Autonoma de Mexico, Ciudad Universitaria, Coyoacan, Mexico, D.F., C.P. 04510; Research Associate Vertebrate Paleontology, Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, California 90007 USA 2. Museo de Paleontologia, Centro de Investigaciones Biologicas, Universidad Autonoma del Estado de Hidalgo, Ciudad Universitaria, Pachuca, Hidalgo Mexico C.P. 42188 3. Universidad del Mar - Campus Puerto Escondido-Oaxaca, Puerto Escondido, Oaxaca Mexico C.P. 71980 Contributions in Science, Number 515, pp. 1-40 Natural History Museum of Los Angeles County, 2007 2 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna and/or geographic extension of faunas. However, the detailed Jurassic paleogeography and the tetrapod faunas of southern North America and of the middle American/Caribbean region are very poorly known. The Mexico record is based mainly on Late Cretaceous faunas north of the Trans- Mexican Volcanic Belt (TMVB). Location of the chief continental Mesozoic tetrapod faunas of Mexico are plotted in Figure 1 and listed in Table 1. To properly evaluate the faunistic relationships between North and South America, it is critical to have a large database on the Mexican Mesozoic, particularly the Jurassic tetrapod record. An added benefit of this database is that it poses objective constraints on the paleogeographic and tectonic models proposed to depict the complex geologic evolution of the middle American- Mexican region during the Mesozoic. In this paper, the Xochixtlapilco Dinosaur Ichnofauna, Middle Jurassic of Oaxaca, southeastern Mexico (Figure 1 (Ferrusquia-Villafranca et al., 1995)), a small, but important, footprint assemblage from a locality south of the TMVB, is described in full, and its regional significance discussed, as a con- tribution to increase the meager dinosaur data- base of the Mexican-middle American region. Further, it is essential to monograph this ichno- fauna now, before it becomes irretrievably lost to erosion within a few years because of the frailness of the footprint-bearing rock strata. A preliminary study of the site was conducted in 1981 by Mr. Oscar Comas, then teaching in the Facultad de Ciencias, Universidad Nacional Autonoma de Mexico (UNAM) (Comas and Applegate, 1982). Visiting the site when the footprints were less weathered, he made plaster casts of several footprints and deposited them in the Museo de Paleontologia of that school; these casts were lent to the senior author by Dr. Pedro Garcia-Barrera, Curator of the Museo. Mr. Comas also furnished two tracing-paper silhou- ettes of footprints (IGM-9309 and IGM-9310), as well as measurements of some prints. All the material is deposited in the Coleccion Nacional de Paleontologia, Instituto de Geologia, UNAM (acronym IGM used for formal cataloguing). The general patterns of vertebrate locomotion and its functional anatomical design generate a limited number of basic configurational patterns reflected in the dinosaur footprints. Therefore, similarity of footprint morphology discloses only a generalized degree of taxonomic affinity, usually no lower than family level. The taxonomic value of footprints is real but limited, because both biological and nonbiological factors intro- duce variables that affect footprint morphology. Among the biological factors, the most significant one is the pedal/manual anatomy of the individual that made the prints. In turn, this anatomy is modified by individual variation (sex, age, size, etc.), and locomotory and behavioral aspects (standing, resting, stalking, walking or running, being solitary or gregarious, etc.). The nonbiological factors mainly pertain to the nature and to the later geological history of the strata bearing the footprints. The nature of the stratum determines the precision and quality of the prints. The diagenesis and later history of the track-bearing strata also influence the print quality. Such considerations lead us to take a conservative approach and limit our taxonomic identifications to the family level at most. MATERIAL, METHODS, AND TERMINOLOGY The footprints lie on a hill slope formed by steeply tilted red sandstone strata with few clear expo- sures. The footprint-bearing strata were in most places too friable to attempt removal or casting of prints, so there is no direct material record of them. During visits to exposed areas by the senior author, each individual print was numbered and measured. To record the relative position and spacing of the prints, they were carefully de- lineated in full size with indelible ink markers on transparent plastic sheets placed over the outcrop. Three sequences were recorded: IGM-7958, a 4- m-wide and 6-m-long sheet with 34 footprint silhouettes from the main outcrop; IGM-7960, a 1-m-wide and 1.5-m-long sheet with three footprint silhouettes from a smaller outcrop located 20 m east of the main one; IGM-3006, a 0.8-m by 0.8-m square sheet with a single footprint silhouette from another small outcrop located about 80 m east from the main one. From the latter a plaster cast was made (IGM-3006). In an effort to detect the orientation of the prints when they were made, the plastic sheet was properly oriented both topographically and struc- turally. The structural restoration of the strata to the original horizontal position allows one to establish the orientation of the prints with respect to the present-day North Pole. This information could then be transferred to the Middle Jurassic North Pole position inferred from paleomagnetic data of the Mixteca region (cf. Caballero-Miran- da et al., 1991). Epirelief/hyporelief casts pairs were made from the epirelief cast suite of the Facultad de Ciencias, UNAM (IGM-9303/-9304, IGM-7428/-9305, IGM-7430/-9306, IGM-9311/-9312, IGM-9313/ -9314, and IGM-7425/-9315). A photographic record was made of the casts and plastic sheets. The description of each track, as well as the shape and numerical parameters used, follows Peabody (1948), Sarjeant (1975), Ferrusquia-Villafranca et al. (1978), and Lockley (1991b). We adhere to the usage of Thulborn (1989:41) for the following terms: small theropods are those whose average footprint length is less than 25 cm, large theropods are those whose average footprint length is more than 25 cm; small ornithopods are Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 3 Figure 1 Index map of Mexico showing the location of the chief continental Mesozoic tetrapod faunas (1-14), the TMVB, and the Mixteca Terrane (MT). The state of Oaxaca (southeastern Mexico) is blown up to show the position of the study area (HJ, Huajuapan de Leon Area, Mixteca Alta region). Faunas: Early Jurassic, Tamaulipas State: 1, Huizachal Fauna. Middle Jurassic, Oaxaca State: 2, Xochixtiapilco Dinosaur Ichnofauna. ?Middle/Late Jurassic, Puebla State: 3, Otlaltepec single occurrence. Late Jurassic, Michoacan State: 4, Chuta Dinosaur Ichnofauna. Early Cretaceous, Puebla State: 5, San Martin Atexcai Dinosaur Ichnofauna; Late Cretaceous, Baja California State: 6, El Rosario Fauna. Sonora State: 7, CabuOona Faunuie. Chihuahua State; 8, Ojinaga single occurrence. 9, Altares faunule. Coahuila State: 10, El Pelliliai Fauna. 11, Rincon Colorado Fauna. 12, Sabinas Dinosaur Ichnofauna. 13, Patau single occurrence. Puebla State: 14, Mitepec Dinosaur Ichnofauna. Michoacan State: 15, El Aguaje Dinosaur Ichnofauna. 16, Tiquicheo single occurrence those whose average footprint length is less than 25 cm, large ornithopods are those whose average footprint length is more than 25 cm. It should be noted that the qualifiers large(er) and small(er) are sometimes used as comparative terms and related to the known or inferred adult size of the taxon under consideration; this usage is clearly deduced from the context. Other terms and measurements used, as well as the corresponding abbreviations, are illustrated and spelled out in Figure 2. The nomenclatural treatment adopted in this paper is based on the concept and usage of morphotypes and morphic varieties. Sarjeant (1992:304) also recommends that the names of ichnotaxa should be based upon ichnofossil 4 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna Table 1 Major Mesozoic tetrapod faunas of Mexico Period State Fauna Early Jurassic Tamaulipas Huizachal Fauna (Clark et al., 1994) Middle Jurassic Oaxaca Xochixtlapilco Dinosaur Ichnofauna (Comas and Applegate, 1982; Ferrusquia-Villafranca et al., 1995, 1996, this study). ?Middle/Late Jurassic Puebla Otlaltepec (Applegate and Ferrusquia, unpublished information) Late Jurassic Michoacan Chuta Dinosaur Ichnofauna (Ferrusquia-Villafranca et al., 1978; Tilton et al., 1996) Early Cretaceous Puebla San Martin Atexcal Dinosaur Ichnofauna (Rodriguez-de la Rosa et al., 2004) Late Cretaceous Baja California El Rosario Eauna (Morris, 1967, 1972, 1973) Sonora Cabullona Eaunule (Lucas and Gonzalez-Leon, 1990, 1993; Taliaferro, 1993) Chihuahua Ojinaga single occurrence (Eerrusquia-Villafranca, unpublished information) Altares Eaunule (Andrade-Ramos et al., 2002) Coahuila El Pellilal Eauna (Rodriguez-de la Rosa and Cevallos-Ferriz, 1998, Rodriguez-de la Rosa, 2003) Rincon Colorado Eaunule (Kirkland et al., 2000) Sabinas Dinosaur Ichnofauna (E. Jimenez-Hidalgo, unpublished information) Palau single occurrence (Eerrusquia-Villafranca, unpublished information) Puebla Mitepec Dinosaur Ichnofauna (Eerrusquia-Villafranca et ah, 1993) Michoacan El Aguaje Dinosaur Ichnofauna (Ortiz-Mendieta et al., 2000) Tiquicheo single occurrence (Benammi et al., 2004) morphology, not on the presumed affinities of the trace maker. Much the same is advocated by Lockley (1989:441) and by Farlow et al. (1989:385), who strongly recommend that formal names should be given only to dinosaur tracks preserved well enough to provide significant information about the pedal structure of their maker. Thus, given the small number of available ichnites and their moderate to poor preservation, we have chosen not to assign formal ichnogeneric names to them but rather to refer to them as morphotypes and morphic varieties. However, in an effort to tie morphotypes to known ichnogen- era, we discuss the relevant ichnotaxonomic record. A morphotype is the fundamental configuration of an ichnite that allows it to be differentiated from other such configurations and constitutes an informal taxonomic category; in this paper, however, morphotypes are given a capital letter designation. Morphotype makers can confidently be linked to family categories of the Linnean system. Morphic varieties (MV) document vari- ability within a given morphotype. In this paper, morphic varieties are designated with the corre- sponding morphotype capital letter, followed by a particular lowercase letter, thus producing a unique two-letter combination. These categories are also informal. The possible Linnean taxonom- ic identification of the morphotype maker is discussed; the assignment made, usually a category of family or higher rank, is the most parsimonious estimate on the basis of the morphological information, known biogeochronologic ranges. and geographic distribution of the taxa involved. Inevitably, given the current extensive usage of cladistic analysis of skeletal remains to character- ize and work out the phylogenetic relationships of generic and higher-rank dinosaur taxonomic categories, some of the names referred to in this study are now cladistically defined, but still remain Linnean taxonomic categories, and are named accordingly. We use and refer to such categories as taxonomic/nomenclatorial “han- dles” to convey the possible taxonomic position of the morphotype maker, and certainly no cladistic analysis of bone remains is implied. GEOLOGICAL SETTING AND PALEOGEOGRAPHIC CONSIDERATIONS GEOLOGIC SYNOPSIS OF THE HUAJUAPAn DE LEON AREA, OAXACA The dinosaur footprint locality lies in the Mixteca Alta Oaxaquena, Municipality of Huajuapan de Leon, northwestern Oaxaca State, southeastern Mexico, 17°42'-17°50'N Lat. and 97°45'- 7°52'W Long. (Figure 2). The area includes nearly 160 sq. km of rugged, complexly deformed territory, where Paleozoic to Quaternary units crop out (Figures 3 and 4). The Paleozoic Acatlan Complex in the Mixteca region is unconformably overlain by the Jurassic System, which largely consists of a 1,700- to 2,500-m-thick, sedimenta- ry, continental-to-marine sequence; its current lithostratigraphic differentiation (Erben, 1956a and b, Cortes-Obregon et al., 1957) into two Contributions in Science, Number 515 Ferrusqma et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 5 Figure 2 Descriptive track and trackway terminology used (modified from Peabody (1948), Sarjeant (1975), Ferrusqma-Villafranca et al., (1978), and Lockley (1991b)). A1-A5: track shapes of Al, rhomboidal; A2, rounded; A3, oval; A4-A5, hyperelongated; the interdigital notch is V-shaped in Al and A3 and U-shaped in A2; B1-B4: track measurements of Bl, theropod footprint; B2-B3, sauropod pedal and manual prints respectively; B4, ornithopod footprint. ABBREVIATIONS: Da, divergence angle; dl, apparent digit length; Dm, dorsal margin (of manual print); dw, width of digit at its base; fl, antero-posterior footprint (= pedal print) length; fw, transverse footprint (= pedal print) width; ia, interdigit angle (between stated digits); ml, antero-posterior length of manual print; mw, transverse width of manual print; pm, palmar margin (of manual print); C trackway measurements; PA, pace angulation; sa, step angle; si, step length; STl, stride length groups (the Early Jurassic Consuelo Group and the Middle to early Late Jurassic Tecocoyunca Group) and eight formations needs a thorough revision (not attempted here) because the recog- nition of the formations beyond their type areas is uncertain or altogether impossible due to their lithic resemblance and complex structural de- formation. In the study area (Figures 3 and 4), the footprint bearing strata belong to the Tecocoyunca Group partim (i.e., an undifferentiated lithostratigraphic unit of group rank, where only some of the formations that make it up are present). This group is a fine to medium-grained, phyllarenitic clastic body, largely laid down in a transitional environment, and corresponds to a huge delta complex. About 18 km S 8°W of the dinosaur footprint locality (see Figure 3), in the Diquiyu-Rio Santa Catarina ‘subarea’, Taberna Formation strata (a largely clastic, shallow marine and transitional unit of the Tecocoyunca Group) located above the dinosaur footprint-bearing beds contain ammonites such as Duashnoceras (lore si Burckhardt, 1927, Subcollina lucretia d’Orbigny, 1847 (Sandoval and Westermann, 1986), Strigo- ceras (Lyroxinites) cf. S. (L.) kellumi Imlay, 1961, 6 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna HUAJUAPAN DE LEON AREA, OAXACA Knrrt 17 42’ Ttt Kmt : 7 Tmi QsQal TmpTml Tmp Ttt Kmt * ~_t JmEA' Pma Jmt Pma 0' =>ma Acatlan Complex Jmt Tecocoyunca Group Partim Argillo-Calcareous unit Kmt Teposcolula 00 Tamazulapan -r Limestone | " | Conglomerate | Pyroclastic unit Tmi La Quebrada L .I Lavic L0J‘ '■ Figure 3 Geologic map and structural section of the Huajuapan de Leon Area, Mixteca Alta, Oaxaca, and southeastern Mexico Lissoceras cf. L, oolithicum d’Orbigny, 1845 (Sandoval and Westermann, 1986), Oppelia sub- radiata Westermann, 1983, Parastrenoceras zapo- tecum Ochoterena, 1963, and Stephanosphinctes buitroni Sandoval and Westermann, 1986 (Erben, 1956a; Sandoval and Westermann, 1986), that indicate an Early Bajocian-Early Bathonian age. In the study area, poorly preserved plant remains referred to as Otozamites sp. (Silva-Pineda, 1984) are present; their geochronological range is con- gruent with the Middle Jurassic age assignment obtained from the ammonites. The structure of the Tecocoyunca Group partim in the area includes two north-northwest-south-southeast trending. complexly and tightly folded anticlinoria, separat- ed by a narrow synclinorium (Figure 3). Normal faulting further complicates the structures. The dinosaur footprint-bearing strata are located near the western edge of the eastern anticlinorium, where a thin basaltic andesite sill intrudes the sequence. The textural and primary structural features of these strata, such as laminar to thin bedding, ripple marks, immature fine grain sand- stones, siltstones, and claystones, indicate that they were deposited very probably on the shores of a coastal lagoon. The Tecocoyunca Group partim is overlain by a Late Jurassic marine unit, in turn overlain by the Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 7 IND. ui V \ + + + + OQO Oo°oQO ^Oo°0°c?o° ® O fN O On^ OSo OOq O^Ooo /|v V V/ V V V V/ V \ry\f V/ ',r vy V V- V \/ V V V V/ V \/ V V fmlV V V V/ V/ V/ V V/ . + + + + + + + 4- + + Tmp+ r~T I', I' i‘, I 1 r 1 I I I. ri nzx T~ri T~71 III r^T "T 1 "T" r ^ 40 o o CNI 6 UO O O CSJ I o LO o o CsJ I o LO o o o o o CD O to CNJ o o o QUATERNARY DEPOSITS: Alluvium and soil. Tml, UNNAMED LAVA UNIT: Andesitic flows. Tmp, UNNAMED PYROCLASTIC UNIT: Thin to medium bedded, diverse- ly textured and welded silicic tuffs and lapilli-tuffs. Tmi, LA QUEBRADA SILL: Basaltic andesite. TAMAZULAPAN CONGLOMERATE: Thickly bedded calclithite. TEPOSCOLULA LIMESTONE: Thickly bedded biomicrite and biopel- micrite; partly recristalized. UNNAMED ARGILLO-CALCAREOUS UNITS: Quartz-bearing calcareous argillite and siltstone. TECOCOYUNCA GROUP Partim: Red, thin to medium bedded phylla- renitic silstone, sandstone and con- glomerate. It bears dinosaur foot- prints. ACATLAN COMPLEX: High grade metamorphites such as biotite schist and pyroxene granulite (green stone). * in meters Figure 4 Generalized lithostratigraphic column of the Huajuapan de Leon Area, Mixteca Alta, Oaxaca, and southeastern Mexico 8 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna Albian Teposcolula Limestone, which is overlain by a Cenozoic continental sequence (Figures 3 and 4). PALEOGEOGRAPHIC CONSIDERATIONS ON THE MIXTECA REGION The Mixteca is part of the southeastern Mexico- middle American-Caribbean region, whose geo- logic evolution during the Mesozoic is far from well known. Several competing models of tectonic and/or paleogeographic evolution have been pro- posed (see Anderson and Schmidt, 1983, for a review of early models). Figure 5 depicts six such models proposed (between 1988 and 2004) for the Jurassic. Notice that all of them include several small continental-crust blocks lying in the space left between North America, South Amer- ica, and Africa as Pangea became disassembled. Paleogeographically they corresponded to islands. However, there are not enough data to positively constraint the sea/land boundary for any of them during any particular time interval (nor for the adjacent continents either). Later, some such blocks accreted to North America, generating southeastern Mexico-Central America, or formed the Greater Antilles. The Mixteca is one or part of one of these blocks. There have been three studies on the paleogeo- graphic and tectonic evolution of the Mixteca region (which is one, or part of one such blocks) during the Middle Jurassic, both based on paleo- magnetic data. Caballero-Miranda et al. (1991: 206-209) considered two paleogeographic hy- potheses. In one the Mixteca was located in the Southern Hemisphere, perhaps by the Central Andes (~20°S Lat.). This hypothesis was proposed by Westermann et al. (1984), Taylor et al. (1984), and Sandoval and Westermann (1986) to explain the close affinities of the Mixtecan and Andean ammonite faunas and implies at least a 37° north- ward rotation (i.e., about 4,500 km) from a start- ing position by present-day northern Chile. In the second hypothesis the Mixteca was located in the Northern Hemisphere, around today’s west-central Sinaloa (about 20°N Lat. and 108°W Long.), as advocated by Scotese et al., (1979), Anderson and Schmidt (1983), and Urrutia-Fucugauchi (1984), among others, who postulated a left lateral displacement related to the Sonora-Mojave and TMVB megashears. This hypothesis implies an oblique (southeastward, — 1,000 km) rotation from a starting position located 8° farther north and 10° farther west of the present-day location. However, the data to support this hypothesis (Caballero-Miranda et al., 1991:table 2) locate the Middle Jurassic Refer- ence Paleomagnetic Pole at 61°N Lat. and 116°E Long., whereas the rotated Middle Jurassic Paleomagnetic Pole’s location varies from 53° to 62°N Lat. and 144° to 165°E Long., thus cal- ling for a much greater rotation (ca., 15° to 30°) than the one proposed; nonetheless, Caballero- Miranda et al. preferred the second hypothesis. Bohnel (1999) proposed a third hypothesis, namely a 25° southward post-Bajocian translation of the Mixteca from a starting position in northern South America across northeastern North America (i.e., by present-day New York, then located just south of the Paleoequator (Figure 5.4)). In these hypotheses, the Mixteca region is regarded as an island located close to west-central South America (first hypothesis), close to south- western North America (second hypothesis), or close to northeastern North America/northern South America (third hypothesis). The largely undifferentiated Pangeatic nature of the Middle Jurassic dinosaur fauna as a whole (Weishampel, 1990; Russell and Bonaparte, 1997; Sues, 1997a) lends no support to any of these hypotheses. However, the island component of all three hypotheses is consistent with the ecologic and geographic scenarios proposed for the Xochixtla- pilco ichnofauna towards the end of this article. SYSTEMATICS This ichnofauna is named for the village of Santa Maria Xochixtlapilco, which lies 4.5 km south- east of the locality by Highway 49, segment Huajuapan de Leon-San hdarcos Arteaga (Fig- ure 3). The locality lies on an east-west trending, — 80-m-long, steep slope formed by brick red, laminar, phyllarenitic silty-clayey strata dipping 62° south-southeast (Figure 6); the average height of the slope is 8 m, and the footprints occur on bedding planes within a strata thickness no greater than 45 mm, hence they are regarded as contemporaneous. Close to the western end of the slope, there is a well-exposed bedding plane covering 16 m^, where 33 footprints of small theropods and sauropods are present (Figures 6 and 7). About 20 m east of this main exposure, there is a much smaller one, where only three small ornithopod footprints (and a digit impres- sion) are discernible (Figure 7). Farther east, close to the end of the slope, on its lower part, there is a single, large, well-preserved theropod print. The footprints of the latter exposures and some of the main outcrop are faintly outlined and shallow, which suggests that they are erosional remnants of true prints or underprints. Order Saurischia Seeley, 1888 Suborder Theropoda Marsh, 1881 Coelurosauria, Huene, 1914, sensu Holz et ah, 2004 “Basal Coelurosauria” sensu Holz et al., 2004 Morphotype A, Morphic Varieties Aa to Ak (Figures 8.1-8.4, Tables 2-3) DESCRIPTION. Small, rhomboidal to oval tridactyl footprints; with short to moderately long Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 9 SYMBOLS \ Convergent margin Spreading zone Suture zone \ Transform fault Volcanic I . 1 Addition- * arc I I al land Figure 5 Paleogeographic and paleotectonic hypothetical reconstructions of the North American-South American- African region during the Jurassic, after 1, Ross and Scotese, 1988; 2, Smith et al., 1994; 3, Golanka et al., 1996; 4, Bohnel, 1999; 5, Dickinson and Lawton, 2001; 6, Elias-Herrera, 2004 (all are slightly modified); notice that only some continental-crust blocks are labeled; the Mixteca Terrane Block is shown in solid black. ABBREVIATIONS: AF, Africa; APO, Ancestral Pacific Ocean; ch, Chortis Block; ds, Del Sur (i.e., Sierra Madre del Sur) Block; NA, North America; SA, South America; y, Yucatan Block and robust digits separated by V-shaped notches, digit III is the longest and the others are subequal; divergence angle (between digits II-IV) typically of 54° to 78°, angle interdigits II-III of 39°-42°, and angle interdigits III-IV of ~20°-39°; estimat- ed hip height of 0.50 m to 0.70 m; other numerical parameters are presented in Tables 2 and 3. Sixty percent of the recorded footprints are hyperelongated, seemingly combining the dactylar/plantar impression with a large meta- tarsal impression into a single print, which has a humanlike footprint appearance because the pedal/metatarsal margin is diffuse. Eleven mor- phic varieties are recognized for this morpho- type. Morphic Variety Aa REFERRED MATERIAL. Plaster and plastic epirelief/hyporelief cast pairs numbers IGM-9303, IGM-9304 and IGM-7428, IGM-9305 (Figures 8.1AalP, 8.1AalN, and 8.1Aa2P, 8.1Aa2N); both correspond to left footprints. 10 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna Table 2 Measurements of footprints assigned to Morphotype A. ABBREVIATIONS AND SYMBOLS: aiII~III, angle between interdigits ll-III; ailll-FV, angle between interdigits DA, divergence angle; dl, digit length; dw, width at digit base; e, estimated; fl, antero-posterior footprint length; fw, transverse footprint width; f#, number of footprints recorded in IGM-7958; I, indeterminate; H, hip height; L, left; MV, morphic variety; Mw, apparent maximum footprint transverse width; R, right; TPl, apparent maximum antero-posterior footprint length; (4), (6), casts made from these footprints in 1981. (A), morphometric ratio method: h = 4.5 fl (Thulborn, 1990:251, Equation 8.2); (B), allometric equation method: h = 3.49 (1.5 fl)^ (Thulborn, 1990:255, Equation 8.15. Linear measurements H ai ai dl dw dl dw dl dw f# Side MV TPl Mw fl fw (A) (B) DA II-III III-IV II II III III IV IV _ L Aal _ — 130 119 585 693 78° 39° 39° 65 24 72 25 58 24 - L Aa2 - - 100 83 450 533 56° 42° 14° 30 22 52 28 30 17 - L Ab - - 120 77 540 640 45° 15° 20° 32 18 50 22 35 17 - R Ac 130 80 110 80 495 587 54° 30° 24° 44 16 60 13 13 - R Adi 210 84e - - - 60° 30° 30° 44 24 63 28 63e 21 - R Ad2 125 53e -- - 56° 33° 23° 23 19 33 18 32 19 (6) L Ae 195e 78e - ~ - - - _ 18 44e 13 42e 14e (4) L Af 200e 85e - - - ~ - - - 45e 20 68 23 18 - L Ag 205 78 122 74 9 L Ah 260 130 125 124 15 L Al 150 133e - ~ - - - - - - - - - 31 R Ai 295 122 128 126e 2 L Aj 295 126 - 121 10 I A) - 110 - 110 - - - - - - _ - - - 16 R Aj 315 125 _ 123 29 L Aj 328 130 - 129e 1 3 4 5 6 12 13 22 23 24 25 Ak Ak Ak Ak Ak Ak Ak Ak Ak Ak Ak 215 260 265 200 270 300 230 205 256 256 294 110 140 120 140 135 123 112 110 115 143 143 DESCRIPTION. They are small, rhomboidai to oval, have well-developed, moderately long digit impressions, with acute apices, which suggests the presence of narrow, elongated claws, digit III is the longest, a divergence angle of 56°-78°, angle interdigits II-III of 39°-42° and angle interdigits III-IV of 14°-35°; one of the footprints is slightly larger, with straight digits; the other has curved, shorter digits (particularly II and IV). The plantar portion of the footprint is little developed; no pads and heel impressions are discernible. Morphic Variety Ab REFERRED MATERIAL. Plaster and plastic epirelief/hyporelief cast pair number IGM-7430 and IGM-9306 (Figures S.lAbP and S.lAbN); it corresponds to a left footprint. DESCRIPTION. This morphic variety is also small, with well-developed, acute-tipped digit impressions, which suggests the presence of claws, but differs from MV Aa in having narrower, nearly straight digits, with their tips slightly bent; the divergence angle of 45° is much smaller than in MV Aa, and so are the interdigit angles (15° between II and III, and 20° between III and IV). Another difference is that the plantar region is larger, but no heel or pad impressions are discernible. Morphic Variety Ac REEERRED MATERIAL. Plaster and plastic epirelief/hyporelief cast pair number IGM-9307 and IGM-9306 (Figures 8.2AcP and 8.2AcN); it corresponds to a right footprint. DESCRIPTION. This morphic variety is broad- ly similar to MV Aa (divergence angle of 54°), differing from it in being smaller, with rather thick, rounded-tipped digit impressions, and in possessing a long and narrow, posteriorly di- rected projection, interpreted as a metatarsal impression. Morphic Variety Ad REEERRED MATERIAL. IGM-9309 and IGM-9310, two silhouettes outlined in 1981 over tracing paper on a now-eroded part of the lower trackway (see Figure 7, and in “General Discus- sion of the Morphotype,” Morphotype Assign- Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 11 Figure 6 Photograph of the main outcrop of the Tecocoyunca Group partim, where most of the dinosaur tracks are exposed; black stripes are asphalt stains. This outcrop is what remains of a much larger bedding plane exposure that extended —80 m to the east (right-hand side of the picture). The site is a road cut in Highway 49, located 6.3 km nearly due south of Huajuapan de Leon, Oaxaca. The square delimits the space illustrated in Figure 7; the bar on the lower part measures 1 m ment section, number 2), Figures 8.2Adl and 8.2Ad2; both correspond to right footprints. DESCRIPTION. Both footprints are very similar, but one is 50% shorter and —37% narrower than the other; the digits are moderately long and have a broad base and acute tips, such as those of MVs Aa and Ab; digit II is shorter that digit IV, which is nearly as long as digit III; the divergence angle is —58°; the angle between digits II and III is slightly greater than that between digits III and FV. The plantar region is broad and shows a posteriorly directed projection, inter- preted as a metatarsal impression. Morphic Variety Ae REFERRED MATERIAL. Plaster and plastic epirelief/hyporelief cast pair number IGM-9311 and IGM-9312 (Figures 8.2AeP and 8.2AeN); it corresponds to a left print. DESCRIPTION. This morphic variety has a hyperelongated outline and shows two distinct regions: the dactylar/plantar region makes up the anterior third, it is wider but shallower than the rest of the footprint; the digit impressions are nearly parallel, that of digit II is longer and wider than those of digits III and IV, which are successively shorter; the plantar portion is small, not well defined, shows no pads or heel impres- sions. The remaining two-thirds are narrower and deeper and correspond to a large metatarsal impression; no discontinuity separates the plantar from the metatarsal impressions, so that the whole track has a human footprint-like appearance. Morphic Variety Af REFERRED MATERIAL. Plaster and plastic epirelief/hyporelief cast pair number IGM-9313 and IGM-9314 (Figures 8.3AfP and 8.3AfN); it corresponds to a left footprint. DESCRIPTION. This morphic variety is —15% longer and wider than MV Ae, but otherwise similar; it differs from it in having the digit II impression much shorter and that of digit III much longer, that is, it has the opposite condition that of MV Ae. MV Af is shallower, and has a wider and shorter metatarsal region (which makes up the posterior half of the footprint) than MV Ae. Morphic Variety Ag REFERRED MATERIAL. Plaster and plastic epirelief/hyporelief cast pair number IGM-7425 and IGM-9315 (Figures 8.3AgP and 8.3 AgN); it corresponds to a left footprint. DESCRIPTION. This morphic variety is deep, hyperelongated, and relatively narrow; the dacty- lar/plantar region makes up —60% of the whole print; the digit impressions are long and wide 12 ■ Contributions in Science, Number 515 Ferrusqma et al.: Southeastern Mexico Dinosaur Ichnofauna Figure 7 Sketch plane of the main outcrop of the Tecocoyunca Group partim depicted on Figure 6, showing the spatial distribution of the dinosaur footprints. The inset, upper right, corresponds to another (smaller) outcrop of the same track-bearing strata, located 20 m east of the main one. The capital-lowercase letter labels on each track indicate the morphotype (A, C, or D) and morphic variety to which it belongs (see text for description). The arrows denote the antero-posterior axes of the prints, the arrow heads point to the anterior end. The numbers inside the tracks correspond to those registered on the plastic sheets IGM-7958 and 7960. The N-S line represents the present-day north-south direction, after the structural restoration of the strata to their original horizontal position. The unnumbered footprint dark silhouettes located on the lower and right-hand-side margins correspond to plaster casts made in 1981, from tracks set on now-eroded parts (indicated by the dark arrows) of the main outcrop. The letters in parentheses (Ae) and (Af), refer to plaster casts also made in 1981, which still remain, but are much deteriorated. The narrow X and Z rhomboids correspond to trackways. (especially so that of digit III), the angle between digits II-III is slightly narrower than that between digits III and IV; the plantar portion is small, showing no discernible pad or heel. The meta- tarsal region of the print is narrow, and joins the dactylar/plantar region at an oblique ( — 160°) angle, as if the digit/plantar region would have been laterally displaced some 20° off the straight metatarsal-plantar/digit III axis. Morphic Variety Ah REFERRED MATERIAL. IGM-7958, plastic sheet, impression number 9, which corresponds to a left footprint (Figure 8.3Ah). DESCRIPTION. In this morphic variety, the dactylar/plantar region is wider than the meta- tarsal one; it has three short digit impressions on the anterior margin; that of digit III is longer and wider. The recess or notch between digits II and III is much narrower than that between digits III and IV. The posterior half of the print corre- sponds to the metatarsal impression. Morphic Variety Ai REFERRED MATERIAL. IGM-7958, plastic sheet, impression numbers 15 (left, incomplete, only the anterior half remains) and 31 (right, partly overprinting a sauropod footprint. Figure 8.3Ai). DESCRIPTION. This morphic variety is similar to MV Ah, but the dactylar/plantar region is less well defined, and has on the anterior margin three short, uneven digit impressions: that of digit II is wide and subround, the one of digit III is the longest, its tip is acute, and that of digit IV is barely visible. The posterior half of the footprint corresponds to the metatarsal impression. Morphic Variety Aj REFFERED MATERIAL. IGM-7958, plastic sheet, impression numbers 2 (left), 10 (indetermi- Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 13 Figures 8. 1-8.4 Morphotype A, footprint assemblage originated by trackmakers referred to the “Basal Coelurosauria” sensu Holtz et al., 2004, scale bars = 5 cm. 8.1, morphic varieties Aa and Ab; AalP and AalN, photographs of IGM-9303, epirelief (P) and IGM-9304, hyporelief (N) casts of a left footprint; Aa2P and Aa2N, photographs of IGM-7428, epirelief (P) and IGM 9305, hyporelief (N) casts of a left footprint; AbP and AbN, photographs of IGM-7430, epirelief (P) and IGM-9308, hyporelief (N) casts of a left footprint 14 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna Figure 8.2 Morphic varieties Ac, Ad, and Ae; AcP and AcN, photographs of IGM-9307, epirelief (P) and IGM-9308, hyporelief (N) casts of a right footprint; Adi and Ad2, computer drawings of IGM-9309 and IGM-9310 respectively, silhouettes of right footprints taken directly from the main outcrop in Oaxaca; AeP and AeN, photographs of IGM 9311, epirelief (P) and IGM-9312, hyporelief (N) casts of a left footprint Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 15 Figure 8.3 Morphic varieties Af, Ag, and Ah; AfP and AfN, photographs of IGM-9313, epirelief (P) and IGM-9314, hyporelief (N) casts of a left footprint; AgP and AgN, photographs of IGM-7425, epirelief (P) and IGM-9315, hyporelief (N) casts of a left footprint; Ah and Ai, computer drawings of selected footprint silhouettes (f# 9, left and f# 31, right respectively) from IGM-7958, plastic sheet outline record of footprints exposed on the main outcrop; (f# = footprint number) 16 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna nate, nearly complete), 16 (right, Figure 8.4Aj), and 29 (left). DESCRIPTION. This morphic variety resem- bles MV Ai, differing from it in having a shallow constriction on the internal and external margins of the footprint, as well as a large anterior lobe (interpreted as the joint digits II and III impres- sions), separated by a shallow recess from a smaller lobe (interpreted as the digit IV impression). The posterior half of the footprint corresponds to the metatarsal impression. Morphic Variety Ak REFERRED MATERIAL. IGM-7958, plastic sheet, impression numbers 1, 3, 5, 13, 22, and 24 (right ones), impression numbers 4, 6, 12, 23, and 25 (left ones; Figure 8.4Ak). DESCRIPTION. About 60% of the Morphotype A footprints belong to this morphic variety. Although the most frequent, it is the least typical of all, with an ellipsoidal hyperelongated outline, and no digit prints. It is interpreted that this odd, humanlike track shape corresponds to the combined dactylar/plantar and metatarsal impressions, as discussed in “General Discussion of the Morpho- type,” Morphotype Assignment section, number 2. MV Ak includes the footprints that form the two trackways present in the main outcrop (cf. Figure 7 and Table 3). Trackway X lies in the lower right part of the outcrop. It consists of four footprints directed S 55°W with respect to the present-day North, the motion was from north- west to southeast. The stride length is 92 cm, the pace angle is 135°, and the step angle varies between 18° and 27°. Trackway Z lies in the lower left part of the outcrop, being nearly vertical. It is directed N 60°E, the motion was from southwest to northeast. The stride length varies from 92 cm to 103 cm, the pace angle ranges from 135° to 145°, and the step angle is similar to that of Trackway X. GENERAL DISCUSSION OF THE MORPHOTYPE Morphotype Assignment The 20 footprints assigned to this morphotype show considerable diversity in shape, and less so in size, but in a grading fashion between end-size (smallest/largest) and end-shape (typical theropod/ atypical theropod), which suggests that the track makers belonged to the same population, being just a small sample of it. Elsewhere, sets of footprints showing different morphologies (age-, gender-, individual variation-, preservation-, or deforma- Figure 8.4 Morphic varieties Aj and Ak; computer drawings of selected footprint silhouettes (f# 16, right and f# 12, indefinite respectively) from IGM 7958, plastic sheet outline record of footprints exposed on the main outcrop; (f# = footprint number) Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 17 Table 3 Measurements of trackways ABBREVIA- TIONS AND SYMBOLS: PA, pace angulation; Sa, step angle; SI, step length; STl, stride length; Xf# = footprint numbers of trackway X; Zf# = footprint numbers of trackway Z. Linear measurements in cm Measurements Xf# Data Zf# Data SI 1-2 56 22-23 50 SI 2-3 40 23-24 56 SI 3-4 58 24-25 42 STl 1-3 89 22-24 98 STl 2-4 90 23-25 92 PA 2 135° 23 145° PA 3 135° 24 140° Sa 1 18° 22 18° Sa 2 27° 23 18° Sa 4 18° 25 25° tion-related), have already been confidently as- signed to a single theropod taxon (Breithaupt et ah, 2003, 2004). The size spectrum (Tables 2-3) includes ‘little’ individuals interpreted as juveniles which as expected, are less numerous than the ‘big’ ones interpreted as adults. It should be noted though, that the actual footprint length might have been at least 25% longer because the plantar region of the prints seems to be not fully preserved, thus appearing less well developed than typical small theropod footprints (contrast Figures 8.1 and 8.2 with those of Lull, 1953:figs. 37-39; Lockley, 1991b:fig. 3.6; Thulborn, 1990:figs. 6.8- 6.9). If morphotype tracks are indeed incomplete, their makers were larger and taller than implied by the dimensions reported in Table 2. The morphic spectrum involves prints that grade from the typical tridactyl theropod footprints (MV Aa) to the atypical, hyperelongated, digitless, human footprint-like tracks (MV Ak), which makes up 60% of the entire sample. The gradational footprint shape changes illustrated by MV Aa through MV Ak were accomplished by incorporat- ing more of the metatarsal impression to the footprint, and by the concomitant reduction and eventual deletion of the dactylar impressions. Because of the great departure from the typical theropod track morphology, MVs Af to Ak are in fact extramorphological variants. There could be three alternate reasons for this mode of locomotion: 1. The trackmaker was crouched, stalking prey in such a way that the weight was borne by the metatarsal, leaving a deep impressions of this bone, whereas the digits remained retracted, barely (if at all) touching the ground. However, digit retraction is unlikely, because the theropod foot structure allows a great deal of contraction, but very little or no retraction. Further, the presence in the same bedding plane of two differently oriented and directed trackways (made with this kind of footprints, see Figure 6), would indicate concur- rent prey stalking, which seems unlikely. 2. The track maker had adopted a semiplanti- grade gait to provide support while walking over slippery and/or soft, unconsolidated ground. The resulting prints would be deep, have a well- marked metatarsal impression, but sediment collapse around the footprint margin would have selectively destroyed small structures such as the digit impressions. These features are present to a varying degree in MV Ac to MV Ak, thus lending credence to this hypothesis. This second interpretation calls for the odd configuration of MV Ae through MV Ak to have resulted from a particular activity of the theropod track maker, rather than reflecting an unusual foot structure. Further, the configuration of each variety reflects the firm to slippery conditions of the substrate, as well as the hardness/softness of the sediment. Finally, the weathering and erosion after exposure reduce the footprint quality, producing changes in their shape and size and eventually deleting them. (Eight years after their discovery, no footprints of the MV Aa through MV Ag remained.) 3. The track maker could be walking on substrata of different firmness, which is inversely related to their degree of wetness, as shown by Gatesy et al. (1999). They studied a suite of Late Triassic small theropod tracks from Greenland. The suite displays a morphic spectrum that varies from typical tridactyl tracks to atypical, hyper- elongated tracks with a large metatarsal impres- sion and a digit I impression much longer than in typical tracks, thus producing a shape not unlike that of MV Ag (cf. Gatesy et al., 1999.:figs. le-f; Figures 8.3AgP and 8.3AgN). They interpreted the spectrum as being made by theropods belonging to the same taxon but walking upon grounds of different firmness. Experimental stud- ies of turkey and helmeted guinea fowl walking on substrata differing in firmness and wetness show that the track shape changes from typical to atypical as the wetness increases, thus supporting their interpretation. Comparison between Gatesy et al., 1999:figs. la-f, and this work. Figures 7.1AaP-7.1AaN to 7.4Ak, shows in the latter a much greater shape diversity, both in toe and metatarsal impressions. Such diversity could not be readily explained by differences in leg/foot stance, penetration in the sediment substratum, and retraction, which is the mechanism that explains shape differences in the experimental study. Because of this. Alternative 3, although possible, seems less likely an explanation than Alternative 2. Ichnological Assessment: Introductory Remarks The features displayed by MVs Aa to Ad are reminiscent of those of small theropods, as characterized by Lull (1953), Thulborn and Wade (1984), Thulborn (1990), and others, who set forth the following characters as diagnostic of small theropod footprints: (a) shape oval, sub- elongated to elongated, so that the width is 70% to 75% of the antero-posterior length; (b) general small size, with an antero-posterior length usually 18 ■ Contributions in Science, Number 515 Ferrusqma et al.: Southeastern Mexico Dinosaur Ichnofauna Figure 9A-C Ichnological comparison of Morphotype A with Shensipus. Prints are adjusted to the same size to ease comparisons. Scale bar = 5 cm. 9A, IGM-7428, an epirelief cast of a left pedal print assigned to MV Aa; 9B, Shensipus tungchuanensis Young (1960) from the Mid- dle Jurassic of China, virtual left print, (after Zhen et ah, 1989:fig. 19.4E; point line added to better delineate the ichnite); 9C, outline of print depicted in 9B to enhance the shape no greater than 20 cm; (c) mesaxonic tridactyl, with clawed digits directed anteriorly; (d) digits II and IV shorter, subequal, relatively broader than digit III, which is the longest, diverging nearly symmetrically from it; the total divergence angle (between digits II and IV) usually does not exceed 50°; (e) step angle between 150° and 180°, which indicates a small pedal divergence angle, and that the gait was that of narrow-hipped individuals. The estimated hip height (—0.6 m) is that of small individuals. The hyperelongated, human footprint-like ap- pearance of MVs Ae to Af, admittedly uncommon, has already been documented in theropods from several sites across the world: North America, Early Jurassic: (Connecticut Valley (Lull, 1953)), Cretaceous: (Utah (Strevell, 1940), Texas (Kuban, 1989; Pittman, 1989)); China, Late Jurassic: (Sichuan Province (Zhen et al., 1989)); Europe, Cretaceous: (Spain (Brancas et al., 1979)); North Africa, Middle Jurassic: (Morocco (Ambroggi and Lapparent, 1954)); Australia, Cretaceous: (Queensland (Thulborn and Wade, 1984)); South America, Late Jurassic/Early Cretaceous: (Brazil, Paraiba (Leonard!, 1994)). Kuban (1986, 1989) reviewed possible explanations, and concluded that most frequently, this elongated footprint shape involves a metatarsal impression, made while the theropod track maker adopted a planti- grade or semiplantigrade gait. The description and discussion of MVs Aa to Ak as integrating a morphic spectrum strongly support Kuban’s conclusion. Morphotype A footprints then are thus interpreted as those of small theropods. Ichnogeneric Summary Review Lormally described Jurassic (and/or Late Triassic) ichnogenera attributed to small theropods (i.e., ‘coelurosaurs’ in a precladistic sense) are rather numerous and geographically quite widespread; among the better known are Anchisauripus Lull 1904, Atreipus Olsen and Baird 1986, Coelur- osaurichnus Huene 1941, Delatorreichnus Casa- miquela 1964, Grallator Hitchcock 1858, Laiyangpus Young 1960, Otouphepus Cushman 1904, Paracoelurosaurichnus Zhen, Zhen, and Rao 1986, Sarmientichnus Casamiquela 1964, Schizograllator Zhen, Zhen, and Rao, Seleichnus Hitchcock 1858, Shensipus Young 1966, Steno- nyx Lull 1953, Taupezia Delair 1962, and Wild- eichnus Casamiquela 1964. It should be noted that the detailed stratigraphic study of the Newark supergroup of New England conducted by Olsen (1980) and Olsen and Baird (1986), has led to the geochronologic reassignment to the Early Jurassic of several ichnogenera such as Anchisauripus, Grallator, Otouphepus, Seleich- nus, and Stenonyx, originally described from the Connecticut Valley by Lull, (1953) and assigned by him to the Late Triassic. We go along with this change, but keep open to the possibility that some such ichnogenera might be also of Late Triassic age. With this in mind, we proceed to compare Morphotype A with the ichnogenera mentioned in the preceding paragraph. Anchisauripus (Late Triassic and possibly Early Jurassic of eastern North America (Lull, 1953; Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 19 Olsen, 1980; Olsen and Baird, 1986), Early Jurassic of Brazil (Leonard!, 1994)) resembles and co-occurs with Grallator in eastern North America; because of this some people regard it as a synonym of the latter. The taxonomic review of this problem lies beyond the scope of this paper; here, for the sake of completeness, we regard both as independent ichnogenera. Morphotype A tracks are —20% to 30% antero-posteriorly shorter, but much wider (—60% and more) than those of Anchisauripus; have less robust, subequal toes with poorly de- veloped pads; and a much greater divergence angle. In contrast, Anchisauripus shows rather narrow tracks with robust, unequal toes (digit III is much longer than II and IV); well-developed pads; and a small divergence angle (—48°). Morphotype A tracks fall in the size range of those of Atreipus (Late Triassic and probably Early Jurassic of eastern North America (Olsen and Baird, 1986); and Late Triassic of Europe (Ger- many: Olsen and Baird, 1986)). According to Olsen and Baird (1986), Atreipus includes Anchisauripus, Coelurosaurichnus, and some species of Grallator; they are probably right, but this is no place to attempt a formal revision of these taxa, so for the sake of completeness, we treat them as independent ichnogenera. Morphotype A tracks differ from those of Atreipus, being wider and relatively shorter (length: width ratio = 1.0:0.65 to 1.0:0.90, vs. 1.0:0.50 to 1.0:0.57 in Atreipus), have a poorly developed plantar region, and moderately robust, subequal toes, which display a large divergence angle (frequently —60°, whereas in Atreipus di- vergence angle is commonly 28° to 32°). Morphotype A tracks are —15% to 35% longer and two times wider than those of Coelurosaur- richnus (Middle Jurassic of France (Kuhn, 1958; Demathieu, 1989)), and also differ from them in having less robust, subequal toes; in Coelurosaur- ichnus digit III is much longer than the others. Morphotype A tracks are —40% to 50% smaller than those of Delatorreichnus (Late Jurassic of Argentina (Casamiquela, 1964)), and differ also from them in having less robust toes and a shorter plantar region. Morphotype A tracks fall in the size range of Grallator (Late Triassic and probably Early Jurassic of eastern North America (Lull, 1953; Olsen, 1980; Olsen and Baird, 1986), Early Jurassic of China (Young, 1960; Zhen et ah, 1986); Early Cretaceous of Brazil (Leonard!, 1994); Grallator includes Dilophosaurus (Early Jurassic of South Africa (Ellenberger, 1972) and possibly of Arizona (Irby, 1996)). Morphotype A differs from them in having subequal, rather robust toes, which display a large divergence angle; by contrast, in Grallator digit III is much longer than II and IV, pads are well developed, and the divergence angle is small (—45°), thus having the configuration of a narrow track (i.e., much longer than wide). Morphotype A tracks are at least five times larger than those of Laiyangpus (Late Jurassic of China (Young, I960)), which in addition show delicate, subparallel, acuminated digits, thus being quite different from those of Morphotype A. Morphotype A tracks fall in the size range of those of Otouphepus (Late Triassic and probably Early Jurassic of eastern North America (Lull, 1953; Olsen and Baird, 1986)), but are propor- tionally wider and have less robust, subequal toes, which display a large divergence angle. In contrast, Otouphepus tracks have very robust toes, particularly so digit III, which is much longer than the other digits; the divergence angle is small (—35°), thus shaping a narrow track. Morphotype A tracks are nearly half as large as those of Paracoelurosaurichnus (Early Jurassic of China (Zhen et ah, 1986)), which show narrow, delicate digits, small divergence angle (—45°), and digit III much longer that II and IV; hence these tracks are very different from those of Morphotype A. Morphotype A tracks are slightly longer, but two to three times wider than the only known footprint of Sarmientichnus (Middle Jurassic of Argentina (Casamiquela, 1964)), which shows only one well-developed toe print (digit II?), and thus its maker was interpreted by Casamiquela as functionally didactylous; hence Sarmientichnus is very different from this morphotype. Morphotype A tracks are at least half as large as those of Schizograllator (Early Jurassic of China (Zhen et ah, 1986)), which show moderately robust, unequal toes (digit III is much longer than II and IV), with well-developed pads, thus con- trasting in shape with Morphotype A tracks, whose toes are subequal. Morphotype A tracks are —27% to 38% longer and three to four times wider than those of Seleneichnus (Late Triassic and probably Early Jurassic of eastern North America (Lull, 1953; Olsen and Baird, 1986)), which show an odd ellipsoidal shape, largely made by the plantar region, with three very unequal toe impressions on the anterior margin (that of digit III is by far the longest). The track maker is interpreted as functionally didactylous (Lull, 1953). In contrast, Morphotype A tracks are tridactylar, with poorly developed plantar region, and moderately robust, subequal toes. Morphotype A tracks are —25% larger than those of Shensipus (Middle Jurassic of China (Young, 1966; Haubold, 1971, 1984; Zhen et ah, 1983, 1989)), but show a close overall resemblance, particularly so with MV Aa2 (Fig- ures 8.1 and 9); both have subequal, distally tapering, slightly curved toes, seemingly devoid of pads, and a poorly developed plantar region; their divergence angle is similar. Yet, we have decided to regard Morphotype A and Shensipus as different, but related, ichnotaxa, on the basis of the size difference, the scant material basis (which pre- cludes assessing intraspecific variation), and the enormous geographic separation between south- eastern Mexico and China. Morphotype A tracks are —3.6 to 4.3 times larger than those of Stenonyx (Late Triassic and 20 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna probably Early Jurassic of eastern North America (Lull, 1953; Olsen and Baird, 1986)), which have a very small plantar region and short, robust unequal toes (digit III is much longer than II and IV) with well-developed pads. Morphotype A tracks have a better developed plantar region, and subequal toes. Morphotype A tracks are —15% to 25% smaller than those of Taupezia (Middle Jurassic of England (Delair, 1962)), and have very different morphological features. Taupezia tracks are quite reminiscent of those experimentally generated by extant birds walking on very wet sediments (Gatesy et ah, 1999); such prints are similar to those of Late Triassic small theropods from eastern Greenland. So, all that can be said about Taupezia tracks is that their maker was a small theropod. Morphotype A tracks are —34% to 44% larger than those of Wildeichnus (Middle Jurassic of Argentina (Casamiquela, 1964)), which show a moderate to well-developed plantar region, and very unequal, delicate (i.e., narrow and long) toes, so that digit III is much longer than II and IV. These tracks then are very different from those of Morphotype A, which have a poorly developed plantar region, and moderately robust, subequal toes. Summing up then, it could be said that although Morphotype A tracks show some re- semblance to the Middle Jurassic ichnogenus Shensipus (from China), the scant material basis, size differences, and great geographic separation make it unadvisable to formally refer Morphotype A to this ichnogenus. Possible Correspondence with Linnean Taxonom- ic Categories By middle Early Jurassic time, Syntarsus (Late Triassic of South Africa (Raath, 1969); Kayenta Em., Arizona, (Rowe, 1989; Glut, 1997) was the only known survivor of the coelophysoid cerato- saurs {sensu Holtz, 1994, and Tikosky and Rowe, 2004), which represent the first extensive thero- pod radiation (Rowe and Gauthier, 1990; Rowe et al., 1997; Tikosky and Rowe, 2004). Syntarsus was a small to medium-size, gracile, digitigrade coelophysoid; its feet (described and figured by Raath, 1969, and Glut, 1997:873) were function- ally tridactylar, with a dactylar length —120 mm, which falls in the length range of Morphotype A, but with a width of —60 mm, thus 20% to 45% narrower than the latter, and would have pro- duced narrow tracks with a small divergence angle, quite unlike that of Morphotype A. Sub- sequent ceratosaurs (Neoceratosauria Novas 1991, sensu Tikosky and Rowe, 2004) are Late Jurassic to Late Cretaceous theropods of medium to large size (cf. Gilmore, 1920), hence much larger than the morphotype trackmaker. The Tetanurae {sensu Gauthier, 1986) constitute the second major radiation of theropod dinosaurs, ranging from the Middle Jurassic to the Late Cretaceous; basal tetanurans {sensu Holz et al.. 2004) of Middle Jurassic age include megalosaur- ids, chiefly from Europe (Glut, 1997, 2000, 2002, 2003; Holz et al., 2004), all medium-size to large theropods; the small Proceratosaurus from England (Huene, 1926); a coelurosar {sensu Holz et al., 2004) of uncertain position, whose feet are un- known, so no comparison is possible with Mor- photype A; and several genera of uncertain position (cf. Holtz et al., 2004:table 4.1), mostly of medium to large size, whose foot morphology is unknown or undescribed. Coelurosaurs extensively diversified in the Late Jurassic-Cretaceous, developing small, medium, and mostly large-size theropods. The Morphotype A trackmaker is roughly the size of the North American Late Jurassic Ornitholestes (cf. Osborn, 1903:fig. 1; Ostrom, 1978, 1980), differ- ing from it in having rather thick digits (cf. Eigures 8.1.Aa, 8.1.Ab, and 8.2.Ac) and in being geologically older. Coelurus, coeval and sympatric with Ornitholestes, is of similar size (cf. Marsh, 1884; Ostrom, 1980:fig. 2; Miles et al., 1998), but its precise foot structure remains undescribed. These reasons preclude assigning the Morphotype A trackmaker to either Coelurus or Ornitholestes. Summing up, because of their greater recorded Middle Jurassic diversity and pedal morphology, it seems probable that the Morphotype A track- maker represents a taxon that could have belonged to basal coelurosaurs {sensu Holz et al., 2004). Geographic Distribution and Geological Age Bajocian-Early Bathonian ‘coelurosaur’ (— small theropod) footprints are known from South America (Brazil and Argentina (Casamiquela, 1964; Leonard!, 1989, 1994)); Australia (Weishampel, 1990); Europe (southeastern Eng- land (Delair, 1962; Lockley and Meyer, 2000) Erance (Kuhn, 1958; Demathieu, 1989)); China (Young, 1966; Haubold, 1971, 1984; Zhen et al., 1983; 1989); and western North America (Carmel Eormation, eastern Utah (Lockley et al., 1998; Hamblin et al., 2000) Lower Sundance Eormation, Wyoming (Breithaupt et al., 2003) and Oaxaca, southeastern Mexico (this study)). The footprints from Oaxaca extend the record —3,000 km south- ward, making it the southernmost Middle Jurassic sample for the Northern Hemisphere. Suborder Theropoda Marsh, 1881 Tetanurae Gauthier, 1986 Avetheropoda Paul, 1988 Carnosauria (Huene, 1914), sensu Holtz et al., 2004 Allurosauroidea Currie et Zhao, 1993, sensu Holtz et al., 2004 Allosauridae Marsh, 1879 PAllosauridae Morphotype B (Eigure 10, Table 4) DESCRIPTION. Large subrounded tridactyl pedal print, digits II and IV short and thick, less Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 21 Figure 10 Morphotype B, single footprint originated by a track maker referred to PAllosauridae; BN, outcrop photograph of a large left footprint; outlined to enhance visibility; BP, photograph of IGM-7959, epirelief cast of this footprint; BS, a computer drawing of the same footprint from a blow up of the photograph, and of IGM-3006, plastic sheet silhouette record of the footprint outlined in the outcrop, scale bar =10 cm 22 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna Table 4 Measurements of the left footprint assigned to Morphotype B ABBREVIATIONS as in Table 2. (A), morphometric ratio method: h = 4.9 fl (Thulborn, 1990:251, Equation 8.3); (B), allometric equation meth- od: h = 3.14 fl‘ (Thulborn, 1990:254, Equation 8.11); (C), allometric equation method: h = 8.6 fl ° (Thulborn, 1990:254, Equation 8.10). Linear measurements in mm Measurements Data Fl 391 fw 340 H(A) 1916 H(B) 1397 H(C) 2855 dill 72 dwII 96 dl III 180 dwIII 116 dllV 60 dwIV 100 DA 70° aill-III 30° ailll-IV 40° than half the length of digit III, which is curved inward; a well-developed lateral notch reaches inside the interdigital notch between digits III and IV, so that the digit IV stands out from the rest of the print. The pads are barely discernible: there is one in digits II and IV, and two in digit III; the plantar pad seems to be four-lobed; no traces of the claws are distinguishable. The total divergence angle is 70°; the estimated hip height is —1.91 m. It should be noted that digit III shows a marked curvature that makes its tip diverge 30° from the antero-posterior digit axis; such curvature is much greater than in other large theropod footprints (cf. Lockley and Meyer, 2000:146, fig. 6.11). Whether this curvature resulted from preservational distortion, an anatomical anomaly, or a normal structure can not be resolved at present. REFERRED MATERIAL. IGM-7959, epirelief plaster cast of a single left footprint located in the smaller outcrop, 1 m from its eastern margin (Figure lOBP); IGM-3006, plastic sheet with the silhouette of this footprint directly outlined in the outcrop (Figures lOBN and lOBS). GENERAL DISCUSSION OF THE MORPHOTYPE Ichnological Assessment: Introductory Remarks The footprint closely resembles that of typical theropods in having a significantly longer digit III, and a deep lateral notch that makes digit IV fully stand out from the rest of the foot (Lull, 1953; Thulborn, 1990); its size fits that of large theropods, that is, carnosaurs (cf. Haubold, 1971, 1984; Lockley and Hunt, 1995; Lockley and Meyer, 2000). Ichnogeneric Summary Review Named Early and Middle Jurassic large theropod ichnogenera are few and far apart; among the better known are Changpeipus Young, 1960, Dilophosauripus Ellenberger, 1970, Eubrontes Hitchcock, 1845, Kayentapus Welles, 1971, Mega- losauripus Lockley, Meyer, and dos Santos, 1996, and Youngichnus Zhen, Zhen, and Rao, 1986. Given that the formal nomenclatorial and taxo- nomic status of some mentioned ichnogenera is not settled, before comparing Morphotype B to them, some comments are in order. Olsen (1980) and Pittman (1992) have discarded ichnotaxa based on insufficient material and/or unsatisfactory descrip- tions, and have synonymized ichnotaxa that share the character states already recognized in another ichnotaxon, for instance Grallator includes as junior synonyms Eubrontes, Changpeipus, Kayen- tapus, Megalosauripus, and Youngichnus. Lockley and Hunt (1995) recognized merit in Olsen’s and Pittman’s approach, but regard Eubrontes as a valid genus diagnostically different from Grallator; further, they synonymized Kayen- tapus to Eubrontes and Dilophosauripus to Grallator, and cited them as examples of ‘pro- vincial taxonomy’, that is, assigning new names to tracks from a localized area, where suitable names already exist. Lockley and Hunt were aware of the problematic status of Megalosauripus, but chose to regard it as a valid ichnogenus, and used it as the basis of intercontinental correlation (Lockley and Hunt, 1995). We concur with the ideas expressed above, but for the sake of completeness, we shall compare Morphotype B with the ichnogenera mentioned, except Dilopho- sauripus and Kayentapus, following Lockley and Hunt (1995). The Morphotype B track is —10% to 25% larger than those of Changpeipus (Early and Middle Jurassic of China (Young, I960)), and differs from them in having shorter, stouter toes; a wider plantar (—metatarsal) region; and a much greater divergence angle (70° vs. 45° in Chang- peipus). The Morphotype B track is —15% to 45% larger than those of Eubrontes (Late Triassic and probably Early Jurassic of eastern North America (Lull, 1953; Olsen, 1980; Olsen and Baird, 1986), southern United States (Lockley and Hunt, 1995); it includes Kayentapus southern United States (Welles, 1971; Lockley et al., 1995)), and also differs from them in having shorter and stouter toes, with larger and less numerous pads, and a larger plantar region; the divergence angle is larger than that of most Eubrontes tracks, but some Eubrontes tracks show a divergence angle as large or larger than that of Morphotype B. It appears that the maker of the Morphotype B track was somewhat heavier and/or more plantigrade than the maker of Eubrontes tracks. The Morphotype B track is —10% longer but 18% wider than the tracks of Gigandipus Hitch- Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 23 cock, 1855 (Late Triassic and probably Early Jurassic of eastern North America (Lull, 1953; Haubold, 1971, 1986; Olsen and Baird, 1986). Olsen and Baird (1986:64) regard Gigandipus (PAnchisauripus) milfordensis as the type species of their new ichnogenus Atreipus, but do not state whether Gigandipus remains valid or invalid; other authors (cf. Haubold, 1971, 1986) deem it valid, and here we follow their opinion. The Morphotype B track also differs from Gigandipus tracks in having shorter toes, longer and curved digit III, greater divergence angle (70° vs. 50° in Gigandipus), and the lack of hallux. The Mor- photype B track is much larger (3.5 times longer and five times wider) than tracks of Hyphepus Hitchcock 1858 (Late Triassic of eastern North America (Lull, 1953)) and further differs from them in having longer and stouter toes, curved digit III, shorter plantar region, and a greater divergence angle (70° vs. 45° in Hyphepus). The Morphotype B track is —13% to 48% smaller than those of Megalosauripus [sensu Lockley et ah, 1986, 1996b; Middle Jurassic of western United States (Lockley et ah, 1996b), England (Lockley and Meyer, 2000); Late Jurassic of Europe (Nopcsa, 1923; Haubold, 1971), western Asia (Lockley et ah, 1996b); Early Cretaceous of Australia (Colbert and Merrilees, 1967), and Uzbekistan (Gabuniya and Khurbatov, 1988)), and differs from them in having shorter, stouter toes with a greater divergence angle (70° vs. 55° to 60° in Megalosauripus), more curved digit III, larger and less numerous pads, and a larger plantar region. The Morphotype B track is —30% longer and — 45%^wider than those of Youngichnus (Early Jurassic of China, Zhen et ah, 1986), and differs from them in having shorter and stouter toes, a greater divergence angle (70° vs. 40° in Youngichnus), more curved digit III, narrower plantar region, and better developed pads. Summing up, Morphotype B is diagnosti- cally different from the ichnogenera discussed above; in fact this morphotype could be the basis of a new ichnogenus; however we refrain from formally proposing it, because of the scarce available material. Possible Correspondence with Linnean Taxonomic Categories Large Middle Jurassic theropods are represented by the Megalosauridae Huxley 1869, sensu Holtz et ah, 2004 (chiefly from western Europe (Holtz et ah, 2004:table 1), the Carnosauria Huene 1920, sensu Holtz et ah, 2004 (from China (Dong, 1992; Zhao and Currie, 1993), and Antarctica (Hammer and Hickerson, 1994), un- fortunately no published description of their feet is available) and a few tetanurans of uncertain position (cf. Holtz et ah, 2004:table 4.1). In spite of its complex taxonomic/nomenclatorial history (Glut, 1997), Megalosaurus Huxley, 1869 is the best known megalosaurid (Padian, 1997): it reached 7 to 8 m long; it was tridactylous and fully digitigrade, with feet long and narrow; footprints attributed to it are 640 mm long and 210 mm wide (Lapparent and Zbyszewski, 1957). A similar foot structure is seen in other mega- losaurids (e.g. Piatnizkysaurus Bonaparte, 1979). In contrast, Morphotype B is much wider, and its track maker would have had a foot structure closer to that of allosauroids sensu Currie and Zhao, 1993, for example the eponymous Allo- saurus, who had robust hind limbs with digits II- IV evenly spaced (Glut, 1997:107). Although allosauroids are known from the Late Jurassic of North America and the Cretaceous of South America, North Africa, and North America (Holz et ah, 2004), and megalosaurids were fairly common in the Middle Jurassic (Padian, 1997), we deem it more probable that the track maker of Morphotype B, because of its closer inferred foot structure, may have been an early member of the Allosauroidea {sensu Holtz et ah, 2004), not yet represented by bone remains. Support for this hypothesis stems from the discovery of sauropod bone remains (Buffeteaut et ah, 2000) from an earlier age than previously known (cf. McIntosh, 1990; Glut, 1997), but suspected on the basis of ichnological evidence (Lockley et ah, 2001). Geologic Age and Geographic Distribution Large Jurassic theropod tracks, although not numerous are known nearly worldwide: North America, Late Triassic and possibly Early Juras- sic: New England region (Lull, 1953; Haubold, 1971, 1986; Olsen, 1980; Olsen and Baird, 1986); Early Jurassic: Arizona (Kayenta Formation, Navajo Sandstone (Welles, 1971; Thulborn, 1990)), Utah (Moenave Formation (Miller et ah, 1989)); Middle Jurassic: Colorado and Utah (Summerville Formation and Entrada Sandstone of the Moab Megatracksite (Lockley, 1991a; Lockley and Hunt, 1995; Lockley et ah, 1996a)); Late Jurassic: Mexico and Colorado (Playa Azul, Michoacan, Mexico (Ferrusquia et ah, 1978), Morrison Formation (Lockley et ah, 1986)). Europe, Middle Jurassic: United Kingdom (Oxfordshire, southern England (Lockley and Meyer, 2000), Hebrides Isles of Scotland (An- drews and Hudson, 1984)), Portugal (near Fatima (Dos Santos et ah, 1994)); France and Germany, Middle and Late Triassic (Nopcsa, 1923; Hau- bold, 1971); China, Middle Jurassic: Sichuan (probably Xiashaximiao Formation (Young, I960)); western Asia: Late Jurassic, Turmeki- stan/Uzbekistan region (Lockley et ah, 1996b); and South America: Brazil, Late Triassic or Early Jurassic: Parana Basin, southern Brazil (Botucato Formation (Leonardi, 1994)); Middle or Late Jurassic: Chile (Tarapaca Province (Galli and Dingman, 1965)). The Xochixtlapilco find in southeastern Mexico adds a sixth site to the meager record of Middle Jurassic large theropod footprints worldwide. 24 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna Figure 11 Morphotype C, footprint assemblage originated by track makers referred to an undescribed family of Eusauropoda Upchurch, 1995; scale bar = 5 cm; Cl (f# 26, left), Cm (f# 8, right), and Cn (f# 19, left), pedal prints; Co (f# 28, left), manual print; all are computer drawings of selected tracks silhouettes from IGM-7958, plastic sheet outline record of tracks exposed on the main outcrop, assigned to MVs Cl, Cm, Cn, and Co respectively, (f# = footprint number) Suborder Sauropodomorpha Huene, 1932 Infraorder Sauropoda Marsh, 1878 Eusauropoda Upchurch, 1995 Family Undescribed Morphotype C, Morphic Varieties Cl-Co (Figure 11, Table 5) DESCRIPTION. Small, oval to subrounded pedal prints (most fall in the 16- to 19-cm antero-posterior length range), with a wide plan- tar region, and short, antero-laterally directed digits. The only manual print is antero-posterioriy much shorter than the pedal prints; its outline is such that the dorsal (‘external’) margin is convex. whereas the palmar margin is concave; one of the side margins is concave too, and the opposite is nearly straight. The estimated hip height ranges between 56 and 116 cm; other measures and numerical parameters are given in Table 5. Morphic Variety Cl REFERRED MATERIAL. IGM-7958, plastic sheet, pedal impressions numbers 20, 26 (left. Figure llCl), and 27 (indeterminate). DESCRIPTION. The footprints of this morphic variety are the most clearly sauropodal ones, and from them the morphotype characterization was made. It should be noted though, that in the Contributions in Science, Number 515 Ferrusqma et aL: Southeastern Mexico Dinosaur Ichnofauna ■ 25 Table 5 Measurements (in mm) of pedal and manual prints assigned to Morphotype C ABBREVIATIONS as in Table 2. (A), morphometric ratio method: h = 5.9 fl (Thulborn, 1989:42); (B), morphometric ratio method: h = 4.0 fl (Alexander, 1976:129) F# Side MV fl Fw H(A) H(B) 20 L Cl 189 145 1115 756 26 L Cl 197 153 1162 788 27 I Cl 160 187 944 640 7 R? Cm 166 133 979 664 8 R Cm 221 106 1304 884 11 U Cm 176 112 1038 704 17 u Cn 160 90 944 640 18 I Cn 129 98 761 516 19 L Cn 182 146 1073 728 21 R? Cn 180 143 1062 720 30 L? Cn 172 138 1014 688 32 L? Cn 140 120 826 560 33 I Cn 140e 110 826 560 28 L Co 91 123 — — footprint 26, where the dactylar impressions are best preserved, digit I has a rounded tip, whereas digits III and IV have slightly acute tips; digit II shows a small prominence that might be a claw mark; digit V is not preserved. DISCUSSION. The sauropod pes was gravi- portal, with reduced digits protected at the tip by ungual claws of diminishing size from digit I to III (McIntosh, 1990); this basic anatomical structure is reflected in the footprints. The MV Cl footprint differs from the typical sauropod one in having the clawlike mark only in digit II; whether it corresponded to an anatomical structure or it is an artifact can not be ascertained because of the limited available sample. Morphic Variety Cm REFERRED MATERIAL. IGM-7958, plastic sheet pedal impression numbers 7, 8 (right. Figure llCm), and 11 (?left). DESCRIPTION. The footprints of this variety are oval, slightly longer than those of MV Cl, and show minor lobes on the lateral margin that might correspond to toe marks. DISCUSSION. The antero-posterior elongation of this morphic variety is an uncommon feature for sauropod footprints. Rather than representing a structural peculiarity, such elongation might be the result of asymmetrical printing between the external and internal halves of the print, whereby the half receiving relatively greater pressure would be more deeply imprinted than the other; in underprints the better defined half would usually be wider than the other, thus producing a virtual elongation that is not present in the true footprint. Larger sauropod footprints with this general shape have already been reported in the literature (cf. Pittman and Gillette, 1989). Morphic Variety Cn REFERRED MATERIAL. IGM-7958, plastic sheet, pedal impression numbers 17, 19, 30, and 32 (?left), 21 (Pright), 18 and 33 (indeterminate). Figure llCn. DESCRIPTION. Half of the sauropod foot- prints belong to this morphic variety, it is the least typical for the lack of toe impressions, so that the plantar region remains. DISCUSSION. Poorly preserved, ovoid to subrounded footprints with no digit marks have already been reported in the literature, and interpreted as being made by sauropods (cf. Thulborn, 1990). Morphic Variety Co REFERRED MATERIAL. IGM-7958, plastic sheet, number 28 corresponds to a manual impression. Figure 11 Go. DESCRIPTION. Only one of the 15 sauropod footprints is referred to this morphic variety. Because it was described in the characterization of Morphotype C, there is no need to repeat its description here. DISCUSSION. The manual print is located very close to footprint 27 (MV Cn); both prints have their greatest axes parallel, however such axis is transverse to the antero-posterior axis in the manus print, whereas it largely corresponds to the antero-posterior axis in the podial prints. This spatial relationship suggests that the track maker of footprint 27 also produced the manual print 28. If so, the manus contacted the ground in a position that was not outwardly rotated relative to the sagittal plane of the individual (in a ground sloth fashion), which is unlike the transverse or oblique manus ground contact, usual among sauropods (Farlow et ah, 1989; Thulborn, 1989; Pittman, 1992). The pedahmanual print ratio recorded in the Xochixtlapilco assemblage (14:1) is far from the expected 1:1 ratio of sauropods. Several possible explanations include the following: 1. Differential preservation favoring pedal over manual prints might be the cause; however, the 26 ■ Contributions in Science, Number 515 Ferrusqma et al.: Southeastern Mexico Dinosaur Ichnofauna print-bearing strata show no compositional or sedimentary-structural differences that could sup- port such an interpretation. 2a. The track maker had a specialized body structure and/or gait that placed less weight to the fore limbs than to the hind limbs, hence the manual prints would have been less deep, and would have had a lesser chance of preservation. The small number of prints and the lack of skeletal remains make it impossible to test this hypothesis, which in any case is contrary to the tendency for manus to be overrepresented in many samples (Lockley et ah, 1994). 2b. The footprints are actually underprints made as in (2a). The shallowness of the prints argues in favor of this hypothesis, but it is still open to the same objections as (2a). 3. The footprints were made subaqueously, that is, by sauropods walking/paddling over ground covered by shallow water. The lack of striations or drag marks on the prints is inconsistent with this hypothesis (cf. McAllister, 1989). 4. Overprinting obliterated the manual prints (cf. Lockley, 1991b:216; Paul, 1991). Overprint- ing commonly occurs in graviportal tetrapods where the length of the hind and fore limbs is similar, the distance between them is close or equal to such length, and the pes is larger than the manus; thus the hind and fore limbs may partly or fully overlap resulting in erasure of the front footprints. The lack of appendicular skeletal material and/or trackways does not allow one to test this hypothesis. Full overprinting is not common (see Farlow et ah, 1989, and Lockley et ah, 1986, for well documented instances and further observations), so that most trackways show both front and hind footprints nearly equally represented (cf. Lockley and Hunt, 1995; Lockley and Meyer, 2000). 5. The track maker was a facultative biped. Again, the lack of skeletal material and/or track- ways make it impossible to test this hypothesis. Alexander (1985) and Bakker (1993) have pro- posed that some sauropods were able to adopt a tripod-based standing/sitting position, support- ing their weight on the hind limbs and the tail. The lack of tail impressions in the Xochixtlapilco assemblage argues against this explanation. Summing up, the available evidence allows no positive choice from the possible explanations discussed; however, it seems probable that the observed manuahpodial print ratio in the Xochix- tlapilco assemblage represents some sort of a preservational artifact. GENERAL DISCUSSION OE THE MORPHOTYPE Morphotype Assignment The sauropod footprints referred to this morpho- type are similar in shape, size (most fall in the 16- to 19-cm antero-posterior length range), and parameters (both configurational and numerical. Table 5); these facts indicate that quite probably, the track makers belonged to the same popula- tion. Ichnological Assessment: Introductory Remarks The limited number, moderate to poor preserva- tion, and small size of the Morphotype C prints render difficult the identification of the track maker. Thulborn (1990) has diagnosed the sauropod footprints as (a) ovoidal to subround, wider than long; (b) usually large (antero-poste- rior length range: 20 cm to 100 cm, commonly 30 to 60 cm), however, undoubted small sauropod footprints have been reported, such as those of the Jindong Formation, from the Cretaceous of Korea (Lim et ah, 1989, 1994); (c) pentadactyl, with digits oriented antero-laterally, (MV Cl shows at least three digits so oriented); (d) having a well- developed plantar pad (the shallowness of the Xochixtlapilco prints suggests that they are either underprints where the pad impression is not preserved, or that most of the footprint thickness has been eroded); and (e) having a step angle of 120°-140° (the lack of trackways does not allow us to detect this character). Sauropod manual prints are (f) about half as large as the corre- sponding pedal prints, (g) transversely much wider than antero-posterior long, (h) semicircular to horseshoe-shaped, (i) lacking digital marks, (j) convex at the anterior margin and concave at the palmar one. As shown, the Morphotype C prints have most of these features, which allows one to refer the track maker to the Sauropoda. Ichnogeneric Summary Review There are few named sauropod ichnogenera, and most belong to the Cretaceous; for example, Breviparopus Dutuit and Ouazzou, 1980 (Early Cretaceous of the Gulf Coast (Farlow et al., 1989; Pittman, 1992)), Rotundichnus Hendricks, 1981 (Early Cretaceous of Germany (Hendricks, 1981)), and Koreanosauripus Kim, 1986 (Late Cretaceous of Korea (Kim, 1986)); among the named Jurassic ichnogenera are Breviparopus (Late Jurassic of Morocco (Dutuit and Ouazzou, 1980)) and Gigantosauropus Mensink and Mertmann, 1984 (Late Jurassic of Spain (Mensink and Mertmann, 1984)); all are much larger than Morphotype C, and need no further consideration. Three small named ichnogenera attributed to sauropods merit discussion: Agrestipus (Late Tri- assic of Virginia, eastern United States (Weems, 1987)), Tetrasauropus Ellenberger, 1970 (Late Triassic of western United States (Lockley et al., 2001)), and Hamanosauripus Kim, 1986 (Late Cretaceous of Korea (Kim, 1986)). Morphotype C differs from Agrestipus Weems, 1987 in having ovoid rather than trapezoidal, posteriorly nar- rower pedal prints, with well-discernible toes located on the antero-lateral margin (in Agrestipus no toes are discernible, instead very faint promi- Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 27 nences are present on the anterior margin). The pedal prints of Agrestipus (length, 150 mm; width, 110 mm) fall in the range of those of Morphotype C. The absence of manual prints is attributed to bipedality or to full overprinting (Weems, 1987). If Agrestipus is indeed a sauropod, Morphotype C could be regarded as a surviving member of a small, hitherto unrecognized sauropod lineage. The second ichnogenus is Tetrasauropus, long interpreted as a prosauropod from the Late Triassic Chinle Group (upper part) of Colorado and New Mexico, but recently reinterpreted by Lockley et al., (2001) as a true sauropod, on the basis of track morphology (the podial print shape closely resem- bles that of Brontopodus bairdii Farlow, Pittman, and Hawthorne, 1989 from the Cretaceous of the Gulf Coast, but its size is much smaller), and the discovery of Late Triassic sauropod skeletal re- mains (Buffetaut et al., 2000). Tetrasauropus might be congeneric to Tetrapodus/Tetrapodosaurus from the Late Triassic of South Africa (Ellenberger, 1972, 1974); however, in the latter, size is larger (pes length range: >-^440 mm vs. 200-300 mm in Tetrasauropus-^ Lockley et al., 2001:185), and the toes curve inwardly rather than outwardly, as in true sauropods. Morphotype C approaches the lower end of Tetrasauropus size range, but differs from it in these features: (a) ovoid rather than subtrapezoi- dal podial prints, (b) antero-laterally rather than anteriorly directed toes, (c) less-developed toes, and (d) antero-posteriorly longer manual prints, with no digit impressions (they are shorter, more curved, and show well-developed digits in Tetra- sauropus). These reasons show that Morphotype C and Tetrasauropus are diagnostically different. Both ichnogenera and Agrestipus seem to repre- sent a sauropod lineage distinctly characterized by small size that thrived during the Late Triassic- Early Jurassic; its survival in the Middle Jurassic of southeastern Mexico (as represented by Mor- photype C), probably was due to the peculiar ecologic/geographic setting there, as discussed below (see “Geographic, Ecologic and Biogeo- graphic Considerations of the Xochixtlapilco” section). Further, the presence of small sauropods in the Late Jurassic of Colorado, western North America (Lockley et al., 1986) lends support to the hypothesis on the post-Early Jurassic survival of this small sauropod lineage. Hamanosauripus (Late Cretaceous Jingdong Formation, of Korea (Kim, 1986; Lim et al., 1995)) podial tracks are 330 mm long and —200 mm wide, ellipsoid, with the anterior margin less curved than the posterior, and bearing three clawed toes, where the inner one is larger than the others. The manual prints are ovoid and show no digit impressions. Morphotype C is — 33% smaller than Hamanosauripus., differing from it in morphology (e.g., antero-laterally directed, clawless toes), geologic age, and geo- graphic location, thus ruling out any close relationships between their track makers. Further, Farlow et al. (1989) have questioned the validity of Hamanosauripus. Also from the Jingdong Formation, Lim et al., (1989) figured but did not describe or name very small, undoubted sauropod tracks (length 185 to 195 mm), which differ in size and shape from those of Hamanosauripus, but closely resemble in shape those of Brontopodus bairdi from the Gulf Coast (cf. Lim et al., 1989:fig. 35.4 and Farlow et al., 1989:fig. 42.3). Hence by the Late Cretaceous, at least two kinds of small sauropods lived in Korea, and left footprints in the Jingdong Formation. Possible Correspondence with Linnean Taxonomic Categories By Middle Jurassic time, members of the first sauropod radiation (e.g. Datosaurus Dong and Tang, 1984 and Klamelisaurus Zhao, 1993, of China (Dong and Tang, 1984; Dong, 1992)) coexisted with members of the neosauropod radiation then underway (McIntosh, 1990, Ser- eno, 1999; Upchurch et al., 2004), such as Omeisaurus Young, 1939 and Shunosaurus Dong and Tang, 1984 of China (Young, 1939; Dong et al., 1983), Cetiosaurus Owen, 1841 of England (McIntosh, 1990; Glut, 1997) and Lapparento- saurus Bonaparte, 1986 of Madagascar (Bona- parte, 1986), all of which were far too large to include small sauropods as the trackmaker of Morphotype C. Hence it could be possible that the Morphotype C maker belongs to a hitherto unrecognized and undescribed suprageneric taxon of truly small sauropods, such as the track makers discussed above. Geographic Distribution and Geologic Age Middle Jurassic sauropod track published records are few and far apart: Australia: (Queensland (Molnar, 1991, seemingly a questionable re- cord)); Europe: England (White Limestone For- mation in Oxfordshire (Lockley and Meyer, 2000)), France (Dept, de I’Indre (Farlow, 1993)), Portugal (Pedrera do Galhina site near Fatima (Dos Santos et al., 1994)); and North America: United States (New Mexico, Summer- ville Formation (Lockley et al., 1994; Lucas and Heckert, 2000)) and Mexico (Oaxaca (this re- port)). The Summerville tracks are much larger than those from Oaxaca. It should be noted that the age and stratigraphic position of the Summerville Formation seem unsettled (cf. Gillette, 1996a and b for a review of the problem); however, according to S. Lucas (personal communication, Jan. 2005), the Sum- merville Formation includes strata of latest Mid- dle Jurassic age and earliest Late Jurassic age, so it may be that the Summerville tracks are actually of Late Jurassic age. In any case, the Mexican tracks are older than the Summerville ones, and may be the oldest Jurassic record of sauropods in North America; in addition, they extend —2,500 km 28 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna Figure 12 Morphotype D, small footprint assemblage originated by track makers referred to Pankylopollexian ornithopods sensu Norman, 2004; DO, photograph of the small outcrop where footprints assigned to this morphotype are exposed; ruler indicates 10 cm; Footprint MVs Dp and Dq: Dp (f# 34, right), Dql (f# 36, right), and Dq2 (f# 35, indefinite), computer drawings of footprint silhouettes from IGM-7960, plastic sheet outline record of footprints exposed on the small outcrop depicted in DO; scale bar = 5 cm. (f# = footprint number) southward the record of Middle Jurassic saur- opods in this subcontinent. Order Ornithischia Seeley, 1888 Suborder Cerapoda Sereno, 1986 Ornithopoda Marsh, 1881 Iguanodontia Sereno, 1986 “Basal Iguanodontia” sensu Norman, 2004 PAnkylopollexia Sereno, 1986 Morphotype D, Morphic Varieties Dp-Dq (Figure 12, Table 6) DESCRIPTION. Small, rounded to ovoid footprints with short, wide, round-tipped digits. Digit III is the longest, the other two are subequai, with a total divergence angle close to 60°, and an estimated hip height of —0.43 to 0.84 m; other measurements on Table 6. These footprints are exposed on the same bedding plane as those of the main outcrop located —20 m east. Morphic Variety Dp REFERRED MATERIAL. IGM-7960, plastic sheet, footprint number 34 (right. Figure 12Dp). DESCRIPTION. This print is the best preserved and provided the morphotypic characters. It should be noted that the internal posterior margin was not clearly preserved. Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 29 Table 6 Measurements of footprints assigned to Mor- photype D ABBREVIATIONS: f#, footprint number recorded on IGM 7960; others as in Table 1. (A), Morphometric ratio method: h = 4.8 fl (Thulborn, 1989:251, Equation 8.4); (B), Allometric equation method: h = 3.97 fl* * (Thulborn, 1990:254, Equation 8.12). Linear measurements in mm Atributes f#34 f#35 F# 36 Side R I R MV Dp Dq2 Dql Fl 175 110 lOOe fw 172 150 133 H(A) 840 528 480 H(B) 750 471 428 dill 45 — 15 dwII 55 — 37 dim 69 42 32 dwIII 55 58 65 dllV 50 — 33 dwIV 58 — 44 DA 57° — 82° ai II-III 28° — 35° ai III-IV 29° — 47° DISCUSSION. See the morphotype’s “General Discussion” below. Morphic Variety Dq REFERRED MATERIAL. IGM-7960, plastic sheet, footprint numbers 36 (right, Figure 12Dql) and 35 (indeterminate. Figure 12Dq2). DESCRIPTION. Short prints whh widely based short digits, having a total divergence angle of 80° to 100°. DISCUSSION. The shape of these footprints differs from that of MV Dp in being relatively shorter and wider, as well as in having shorter and wider digits; however, it should be noted that these footprints are less well defined than footprint 34 (the only MV Dp); this in turn suggests that deficient printing may account for the shape differences, particularly so for the virtual lack of the digit II and IV impressions in footprint 35. The latter and footprint 36 are actually smaller than footprint 34 (their width is 13%-22% smaller than that of footprint 34), this size difference may represent sex, individual, or age variation. How- ever, it seems most probable that the smaller footprints were made by juvenile individuals. GENERAL DISCUSSION OF THE MORPHOTYPE Morphotypic Assignment In spite of the differences mentioned above, these footprints show an overall similar morphology, which indicates that their track makers belonged to the same population, and are therefore assigned to the same morphotype. Ichnological Assessment: Introductory Remark The size and morphology of the footprints assigned to Morphotype D are those of small ornithopods (Thulborn, 1990; Lockley, 1991b, Leonard!, 1994). Thulborn (1990) characterized such footprints as being tridactylar, mesaxonic, with digits II and IV subequal, slightly divergent; digit IV is smaller, 20 to 25 cm anterior-posterior length range and a total divergence angle around 60°. Other criteria include tracks as wide as (or wider than) long, toes without claws or with little-developed claws, lack of hallux impression, lack of plantar notch (cf. Haubold, 1971, 1984). It should be noted though, that still all these criteria are not universally followed (cf. Olsen and Baird, 1986). Although the preservation of Morphotype D tracks is moderate to poor, they show most of the features listed here as characteristic of those attributed to small ornithopods, namely being nearly as wide (or wider than) as long, tridactylar, mesaxonic, with subequal digits II and IV, digit IV smaller, absence of claws, and lack of plantar notch. Thus they could not be extramorphological variants of small theropods, that is, tracks whose shape and diagnostic features greatly depart from those of the typical or characteristic track morphology attributed or known to belong to a given track maker (concept proposed by S. Lucas, personal communication, July 2005); rather, Morphotype D tracks are clearly eo- morphic (i.e., preserved well enough to allow detection of their shape and other diagnostic features; it is the antonym of the previous concept), and readily attributed to small ornitho- pods. Ichnogeneric Summary Review Formally named Jurassic ornithopod ichnogenera are not numerous; among the better-known ones are Anomoepus Hitchcock, 1848, Dinehichnus Lockley, dos Santos, Ramalho, and Galopin, 1993, Gyrotrisauropus Ellenberger, 1972, Gyp- sichnites Stenberg, 1932, Iguanodon Mantell, 1825, Jialingpus Zhen, Li, and Zhen, 1983, Pseudotrisauropus Ellenberger, 1972, and Si- noichnites Khun, 1958. Perhaps more than in any other group, the morphological and size diversity displayed by the footprint record has made it quite difficult to characterize and relate taxa. For instance, within the named ichnogenera, there are forms such as Anomoepus and Jialing- pus, characterized by small size and narrow toes, thus resembling Grallator; they contrast with forms such as Gypsichnites, Gyrotrisauropus, and Iguanodon, characterized by medium to large size, and medium to broad toes, with pointed to rounded tips. Another complicating matter is the unsettled formal taxonomic and nomenclatorial status of some ichnogenera, so before making comparisons, a few comments are put forward. Several species of Grallator have been trans- ferred to Anomoepus (cf. Haubold, 1971). Pitt- 30 ■ Contributions in Science, Number 515 Ferrusqma et al.: Southeastern Mexico Dinosaur Ichnofauna Figure 13A-C Ichnological comparison of Morphotype D with Sinoichnites and Gyrotrisauropus. Prints are adjusted to the same size to ease comparisons; scale bar = 5 cm; A, footprint #34 of IGM-7960, a right pedal print assigned to MV Dp; B, Sinoichnites youngi Kuhn, 1958 from the Middle Jurassic of China. (Redrawn from Haubold, 1971:fig. 54.10; and Zhen et ah, 1989:fig. 19. 2E); C, Gyrotrisauropus Ellenberger, 1972 from the Early Jurassic of South Africa. (Redrawn from Thulborn, 1990:fig. 6.33b) man (1992) proposed a different course: Grallator should include as junior synonyms Anomoepus and Gypsichnites, because they share most character states of Grallator. Further, given the close resemblance and geologic age bemeen Anomoepus and Jialingpus, there is a real possi- bility that both taxa be congeneric, hence Jialingpus could be a junior synonym of Anom- oepus. Hopiichnus Early Jurassic of Arizona (Welles, 1971) is another example of the ‘pro- vincial taxonomy’ approach, and is considered a junior synonym of Anomoepus by Lockley and Hunt (1995:122). At any rate, the resolution of these and related problems lies beyond the scope of the present paper, so in the following compar- ison of Morphotype D with the named ichnogen- era, we shall make comments to express our position on some problematic taxa. Morphotype D tracks are slightly larger than those of the nearly ubiquitous Anomoepus (Late Triassic of eastern North America (Lull, 1953; Olsen, 1980; Olsen and Baird, 1986), southwest- ern North America (Texas and New Mexico, Murry, 1986; Clark and Fastovsky, 1986), Early Jurassic of eastern North America (Olsen and Sues, 1986), and Europe and South Africa (Haubold, 1971, 1986)). Olsen and Baird (1986) synonymized Moyenisauripus (Early Jurassic of South Africa (Ellenberger, 1972)) with Anomoe- pus on the basis of size and strong overall resemblance; this change has been accepted (cf. Haubold, 1986). Lockley and Hunt (1995) have synonymized Hopiichnus (Early Jurassic of Ar- izona, Welles, 1971) to Anomoepus also on the basis of size and strong overall resemblance. Both moves are followed here. Jialingpus (Early Juras- sic of China, Zhen et ah, 1983) is also very similar in size and morphology to Anomoepus (cf. Zhen et al., 1989:figs. D and E, the only difference between these ichnogenera is the shape of the metatarsal impression); therefore we include Jialingpus within Anomoepus. Morphotype D tracks also differ from Anomoepus tracks in being nearly as wide as (or wider than) long, with short, stout toes displaying a large diver- gence angle; the pads are poorly (if at all) developed, and the plantar region is wider; the metapodial impression frequently present in Anomoepus tracks is absent in Morphotype D. Morphotype D tracks are about the same size that those of Apatichnus (Early Jurassic of eastern North America (Eull, 1953)), but differ from them in being much wider, with short, stout toes, which display a greater divergence angle. Morphotype D tracks are —30% larger than those of Atreipus (Early Jurassic of eastern North America (Olsen and Baird, 1986)). According to these authors, Atreipus includes species of Gigandipus (same age and provenance, Bock, 1952), and of fAnchisaur- ipus (same age and provenance, Bock, 1952). Here we have partly followed their interpretation (see our ichnogeneric assessment of Morphotype A). Morphotype D tracks differ from Atreipus tracks in being wider (width may be greater than length), with short, stouter toes, which display much a greater divergence angle. It should be noted that Atreipus has also been interpreted as made by a theropod, and as such it was discussed in Morphotype A. Morphotype D tracks are less than half as long as those of Gypsichnites (Early Jurassic of South Africa (Ellenberger, 1972)), but differ from them in being wider (length:width ratio 175:172 mm vs. 450:290 mm in Gypsichnites) and in having short, stout, round-tipped toes which display a greater divergence angle; further, digit III in Morphotype D is straight and shorter than in Gypsichnites, which is longer and slightly curved. Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 31 Morphotype D tracks are less than half as long as those of Gyrotrisauropus (Early Jurassic of South Africa (Sternberg, 1932; Ellenberger, 1972; Thul- born, 1990)), but the shape shows some re- semblance; both are wide tracks with short, stout, round-tipped toes, which display a large diver- gence angle (Figure 13). Unlike Gyrotrisauropus, Morphotype D tracks show no pads. Morphotype D tracks are less than half as large as those of 'Iguanodon' (Fate Jurassic of England (Sarjeant, 1974)), and also differ from them in having short, stout, and round-tipped toes. Morphotype D tracks are 40% smaller than those of Pseudotrisauropus (Early Jurassic of South Africa (Ellenberger, 1972)), and differ from them in having toes with a wider basal part that tapers distally, whereas in Pseudotrisauropus, the wider part is subdistal; also in Morphotype D, the interdigit II-III cleft is much shallower than in Pseudotrisauropus. Morphotype D tracks are -^40% to 50% smaller than those of Sinoichnites (Fate Jurassic of China (Kuhn, 1958; Young, 1960; Haubold, 1971, 1984; Zhen et al., 1983, 1989)), but show an overall resemblance (Figure 13) in shape (wide tracks with short, stout toes, shallow interdigit cleaves) and divergence angle (large). Morphotype D tracks significantly differ in size and shape from the putative ornithopod ichnogenera Yangtzepus (early Fate Jurassic of China (Young, 1960; Kuhn, 1963; Flaubold, 1971, 1984)) and Youngichnus (Early Jurassic of China, (Zhen et al., 1989)), also interpreted as a theropod track (Zhen et al., 1989)). In conclusion, Morphotype D significantly differs in size and morphological features from the ichnogenera discussed, although it shows some shape resemblance to Gyrotrisauropus, and more so to Sinoichnites-, however, the limited material basis of Morphotype D, plus the size and geologic age differences, as well as the enormous geographic separation between this morphotype and both Gyrotrisauropus and Sinoichnites, lead us to regard it as not congeneric with either ichnotaxon. Possible Correspondence with Linnean Taxonomic Categories By Middle Jurassic time, basal ornithopods (Euornithopoda Sereno, 1986, Hypsilophodonti- dae included (cf. Sues, 1997a and b; Sues and Norman, 1990)) have succeeded the Fate Trias- sic-Early Jurassic heterodontodaurid (Weisham- pel and Witmer, 1990; Smith, 1997) and basal thyreophoran ornithischians (Weishampel, 1990, 2004; Dong, 1992; Glut, 1997); known only from China, they include Yandusaurus He, 1979, Gongbusaurus Dong et al., 1983, and Agilisaurus Peng, 1990 (and 1992); they were gracile, small, cursorial, tridactylar digitigrade dinosaurs; their feet had long, slender metatarsals; long, delicate digits; and pointed to rounded unguals; digits II and IV slightly diverged from digit III. This foot structure would have produced long and narrow tracks quite different from Morphotype D. By default, the possibility that the Morphotype D track maker was an iguanodont merits consid- eration. As previously discussed, this morphotype closely resembles tracks commonly attributed to iguanodonts (cf. Thulborn, 1990; Eockley and Meyer, 2000). Further, in size Morphotype D corresponds to tracks that could have been made by a medium-size iguanodont, like the Late Jurassic North American Dryosaurus (cf. Galton, 1981; Glut, 1997; Ryan, 1997); in shape, because of its short, blunt, divergent digits, Morphotype D approaches the condition seen in tracks attributed to much larger iguanodonts, like the Early Cretaceous Iguanodon (Norman and Weisham- pel, 1990; Sarjeant et al., 1998; Eockley and Meyer, 2000); the well-known ankylopollexian Camptosaurus from the Late Jurassic of North America and Europe (Glut, 1997; Norman, 2004) has already ponderous feet with short, stocky digits (cf. Glut, 1997:247), that could have produced tracks not unlike Morphotype D, save that digit IV would be less divergent. Under these circumstances, it appears more parsi- monious to hypothesize that the Morphotype D track maker probably was a Middle Jurassic early ankylopollexian iguanodont, as yet un- recorded in bone. The discovery of Late Triassic sauropod bone remains (Buffeteaut et al., 2000), long suspected on the basis of tracks, and lends support to this possibility (cf. Eockley et al., 2001). Geographic Distribution and Geologic Age Jurassic small ornithopod footprint published records are rather scarce: Early Jurassic: North America (Portland Formation of the Connecticut Valley (Lull, 1953; Olsen, 1980; Olsen and Sues, 1986); Kayenta Formation of southwestern Unit- ed States (Welles, 1971; Eockley and Hunt, 1995)), Europe: Poland (Swietokrzyskie Moun- tains (Karaszewski, 1969)), Germany (southern region (Haubold, 1971, 1984)), South Africa: Lesotho (Upper Elliot Formation (Sternberg, 1932; Ellenberger, 1972, 1974; Thulborn and Wade, 1984)), and South America: Brazil (Sao Paulo, Botucatu Formation (Leonardi, 1994)). Middle Jurassic: Europe: Scotland (Leak Shale (Andrews and Hudson, 1984)). Late Jurassic: North America: Mexico (Michoacan (hypsilophodontid tracks, Ferrusquia-Villafranca et al., 1978)), Europe: England (southern region (Sarjeant, 1974)), China (largely the Sichuan Province (Kuhn, 1958, 1963; Young, 1960; Haubold, 1971, 1984; Zhen et al., 1983, 1989)). The tracks from Oaxaca, southeastern Mexico extend —3,000 km southward the record of Middle Jurassic small ornithopods in North America. 32 ■ Contributions in Science, Number 515 Ferrusquia et at.: Southeastern Mexico Dinosaur Ichnofauna GEOGRAPHIC, ECOLOGICAL, AND BIOGEOGRAPHIC CONSIDERATIONS OF THE XOCHIXTLAPILCO DINOSAUR ICHNOFAUNA AND ITS ENVIRONMENT Recent models of the Mesozoic geologic and tectonic evolution of the southeastern Mexico- middle American region (Figure 5), portray the Mixteca territory (also known as the Mixteca Terrane), as one of the several small continental- crust blocks set in the widening space between North America, Africa, and South America, as Pangea became disassembled. Paleogeographical- ly, such blocks would have been islands; however, there are not sufficient data to constraint the sea/ land boundary of any one block during the Jurassic and most of the Cretaceous. Regardless of the model, the Mixteca territory would have been a Middle Jurassic island, probably of small size, still lying close to North America, South America, and Africa (Figure 5). The diversity of this ichnofauna, given the reduced number of tracks and the small outcrop area where they occur, is indeed noteworthy. The fact that three of the four track makers were small dinosaurs, two herbivorous (one sauropod and an ankylopollexian ornithopod) and one carnivorous (a “basal coelurosaur” theropod), appears to be not merely coincidental, and calls for an expla- nation; we offer as such this speculation: The Xochixtlapilco dinosaurs belonged to a communi- ty set in a restricted and/or isolated scenario, where limited space and resources would have induced selective pressures toward small size, particularly to the primary consumers (i.e., the herbivores), and to their associated predators (the “basal coelurosaur”). Larger predators such as the allosauroid recorded by the Morphotype B footprint, could survive in a setting like this, having much lower population densities than the small dinosaurs; hence their representation in the ichnofauna would be lesser than that of the small forms. The paleogeographic island scenario pro- posed above, although conjectural, would provide the environmental setting required for this eco- logical hypothesis. In addition such a scenario would be consistent with the idea that the fauna was shielded from competition and exchange with continental faunas, thus promoting its endemic condition and peculiar physiognomy. Finally, given the supposed location of the Mixteca block during the Middle Jurassic (Fig- ure 5), one would expect some overall biogeo- graphic/phylogenetic resemblance of the Xochix- tlapilco dinosaur fauna with coeval faunas from North America, South America, and Africa. To test this hypothesis. Table 7 was prepared; it is a compilation of pertinent taxonomic/distribu- tional data for these and other continents. It discloses that the Xochixtlapilco fauna shows greater resemblance to the North American fauna and to the Northern Hemisphere Chinese and Western European faunas than to the Southern Hemisphere, Gondwanic South American, Afri- can, and Australian faunas. However, the Xo- chixtlapilco fauna is too small to objectively assess the validity of this resemblance. SUMMARY AND CONCLUSIONS 1. The Xochixtlapilco Dinosaur Ichnofauna was recovered from steep outcrops of thinly bedded, red, phyllarenitic, fine-grained sand- stone and shaley siltstone belonging to the Tecocoyunca Group partim, which was laid down in a tropical coastal lagoon, and dated as Early Bajocian-Early Bathonian on the basis of ammonites. The site lies in the Oaxacan Mixteca, southeastern Mexico. 2. The ichnofauna mainly consists of small footprints, whose makers are referred to a “basal coelurosar” (Morphotype A tracks), an undescribed eusauropod taxon, probably of family rank (Morphotype C tracks), and an ankylopollexian ornithopod (Morphotype D tracks); there is also a single large footprint, made by an Pallosaurid carnosaur. The scant material record of this fauna makes notewor- thy its relatively high diversity. 3. The Xochixtlapilco ichnofauna is the south- ernmost record of Jurassic dinosaurs in North America, and adds a new fauna to the meager record of dinosaurs in Middle America. 4. Middle American plate tectonics models of geologic/tectonic evolution portray the Mix- teca territory (—Mixteca Terrane), for the Jurassic, as one of the several small, continen- tal-crust blocks laid in the inter-American/ African space as Pangea became disassembled. 5. Ecologically, this paleogeographic scenario would have been an isolated setting, where limited space and resources might have imposed selective pressures toward small size, particularly to the primary consumers and associated predators. Such a setting would have shielded the island fauna from competition and exchange with neighboring continental faunas, thus promoting its en- demic condition and identity. 6. Nonetheless, the Middle Jurassic Xochixtla- pilco dinosaur fauna shows a closer bio- geographic/phylogenetic resemblance to the North American fauna than to the South American or African ones; however, the meaning of this fact can not be fully assessed at present, because of the Xochixtlapilco fauna’s small size. ACKNOWLEDGMENTS The Institute de Geologia, UNAM, the Direccion General de Asuntos del Personal Academico de la Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna ■ 33 I S S s -rt S .2 >. 3 3 c/5 3 2 2-^ s A 13 s ^ ^ ,4^ (j 2 cj _0 3 g 3 1 -o 2 g o « o 2, g- 3 O. W [— ' ^ 2 S J .2 o I- CJ ex ^ o 3 "3 « V- 'U 3 y >.3 C ^=: ^ qj U c5 c=3 S 2 .3 C« *J 3 E(« C3 cn 0«! 2 -X ■C u d o 7N 3 G m -3 cn 33 O « O a 2 3^ >030 — OJ 3 -3 33 *-. 3 T3 ■« 3 a 2 ^ P o ^ o ^ :s ^ 0/ W c/5 ^ , %1 o • S .2'^^ , c o iJ -=! "2 -C . y y ^ 5° 2 ■- 2 :. 2 2-2 ^ ^ ,3 2, 3 I 3 o OJ 2 3, t! u .y -o C« o 3 o. ;C O c/5 «J 3 ^ 3 03 '3 I * "O tr 3 - 3 ■3 3 I §-2 &| ^ S J M I (5 OJ I—) c/5 3 c« W OJ 3 -o - .y -5 ^ y 2 2 3 2 ^ s OJ 2 _3 33 OJ ^ T3 r~ OJ “ c« .y -o S o 3 Oh y o ~o 2 a c« O ^ Vh 3 OJ S-S ^ ^ OJ ,3;^ OJ tsjD y 1-2 2 u -2 o ^ o'© 33 2 o a 2^ -i'Q P ::§ w y .s 111 IM ^ o ^ Ji ‘P Cl( _v • =r *3 GD ..^ l-i Wh c/5 ^ 2 •C o 2 3 Oh 3 3 2 ■M c/5 < .2 •- -y £ 2 1:sili‘, 03 O O H a w X !3 — ^ tN !5 -2 3y_y::g03o33 i| S -£ 2 3 cxW 39 3 c/5 3 3 3 -a 33 3 o 3 O- O 2 S 3 jj 3 _2 OJ <1 OJ _y £05 CJ Wh 3 JS 2-S .. § 2 o 3 OJ 2 2 _0 Oj -*-* ^ S? “3 ■i'sl I rt cn 03 ^ 1-1 y/! j- O ^ y ^ 2 5 < OJ £3 I< n*. 3 C OJ H as ’S « 3 2’§ ^ 3 Oh 33 I— 4 O O Ji O Tracks Middle Jurassic/MphC: Middle and Late — Middle Jurassic/ Late Jurassic/large Middle Jurassic/ small tracks: undescribed Jurassic/small medium to sauropod tracks medium to large sauropod family sauropod tracks; large sauropod Middle Jurassic/ sauropod tracks Late Triassic/small tracks medium to large sauropod tracks sauropod tracks 34 ■ Contributions in Science, Number 515 Ferrusquia et al.: Southeastern Mexico Dinosaur Ichnofauna G ^ fif) ^3 (U QJ Dh o u G !U Oh .S s u u t; c w e '55 d o O Oh o U O rrl ■« C> « -o d O s-H Vh d ' ’ . bJO 1 ^ S i W -O ' ^ ^ 2 1 "o G S ! a o -2 H 2^ T3 IH 2 d i-H T3 O 0-1 d O bJO I ^ 2 ■55 C/) O 2:2^ d d o I , H rrt O' — > c« ■fi ^ T1 •3 ^ ^ "o ^ 2 Oh o -d :h^-3 2 .y o O bJD' X s tl 5j t! w u _ « Oh:t: d3 -d « -o -§ 2 -g Oh's Oh-2 Oh o^'d-2 'S 2f— I 2 w < Q o hJ o jj 2 a d o>^ S CO ■7 ^ " 2 |2|J sill C o ^ d O T ■ 3"§ >— >_o