Archosaur Footprints – Potential for Biochronology of Triassic Continental Sequences

Historically, footprint-bearing localities in eastern and western North America and southern Thuringia and northern Bavaria have played pivotal roles in Triassic archosaur footprint research. In a nearly complete sequence of formations and footprint-horizons in the Moenkopi Group (U.S.A.), Newark Supergroup (U.S.A. and Canada), and Buntsandstein, Muschelkalk, and Keuper groups (Olenekian to Norian-Rhaetian) of central Europe, the principal morphs of archosaur footprints are represented by Rotodactylus, Synaptichnium, Isochirotherium, Brachychirotherium, Chirotherium, Sphingopus, Parachirotherium, Atreipus, and Grallator. Additionally, the temporal and, in principal, the evolutionary “root” is documented by Protochirotherium from the early Olenekian of Hessen. Of utmost significance is the evolutionary succession from Chirotherium to Grallator. Therein, the development of two key features seen in the origin and early evolution of dinosaurs – tridactyl foot-morphology and the bipedal gait – is documented between the Olenekian and Norian. The stratigraphic distribution of these forms and their potential for biochronology yields a biochronological sequence we sketch out as follows: For the Triassic we discriminate six successive biochrons (I-VI). Each biochron is marked by an index taxon ( in bold), a characteristic footprint assemblage, and its stratigraphic distribution. I. Protochirotherium (Synaptichnium), Late Induan-Olenekian. II. Chirotherium, Rotodactylus, Isochirotherium, Synaptichnium (“Brachychirotherium”), Late Olenekian-Anisian. III. Sphingopus–Atreipus–Grallator, Rotodactylus, Isochirotherium, Synaptichnium (“Brachychirotherium”), Late Anisian-Ladinian. IV. Parachirotherium–Atreipus– Grallator, Synaptichnium (“Brachychirotherium”), Late Ladinian. V. Atreipus–Grallator, Brachychirotherium, Carnian-Norian. VI. Grallator–Eubrontes, Brachychirotherium, Norian-Rhaetian. In addition to their biostratigraphic utility, the succession of ichnotaxa and ichno-assemblages also reflects evolutionary developments in foot morphology and in the locomotor apparatus of the Archosauria, a progression thus far incompletely documented by the body fossil record. As a consequence of their limited temporal ranges, and their intercontinental distributions in large quantities, the principal archosaurian ichnotaxa open up additional and innovative possibilities for the biochronology of continental sequences. INTRODUCTION AND BACKGROUND The scientific history of Triassic tetrapod tracks begins with the description and binominal naming of the famous footprints and trackways of Chirotherium barthii and C. sickleri from the Buntsandstein near Hildburghausen in Thuringia, Germany by Sickler (1834) and Kaup (1835). These momentous events were followed, beginning in 1836, by documentation of the extensive discoveries in the “New Red Sandstone” (now the Newark Supergroup) of Connecticut and Massachusetts by E. Hitchcock with the descriptions of important ichnotaxa like Eubrontes and Grallator (Hitchcock, 1845, 1858). The next phase of significance commenced in the middle of the twentieth century with the study of the Early-Middle Triassic tetrapod ichnofauna of the Moenkopi Formation of Arizona by F.E. Peabody. In his pioneering monograph, Peabody (1948) demonstrated for the first time the intercontinental distribution of Chirotherium, with identical ichnospecies (C. barthii and C. sickleri [= C. minus]) present in both the Moenkopi Formation and the Solling Formation of southern Thuringia. Rotodactylus constitutes another remarkable tetrapod ichnotaxon from Arizona described by Peabody that was subsequently recognized preserved in association with C. barthii and C. sicklerii in several strata in the Buntsandstein near Hildburghausen, further supporting the correlation between the two units. Building on the results of Peabody (1948, 1955a, 1956), Haubold (1967, 1971a, b), in his studies of the track assemblages in the Buntsandstein, further substantiated the intercontinental distribution of Rotodactylus. In the following decades, the anatomical interpretations of Chirotherium, Rotodactylus, Eubrontes, Grallator, and other classical Triassic ichnotaxa were shown to parallel the evolutionary sequence of Triassic archosaurs documented by an increasing quantity of skeletal evidence. It was J. Walther (1917) who first proposed a dinosaur-like producer based on the footprint morphology and trackway pattern of Chirotherium. Soergel (1925) likened the C. barthii track maker to Euparkeria, a taxon whose phylogenetic position is presently recognized as close to the base of the hypothesized crown-group Archosauria (Gauthier, 1986; Sereno, 1991). This corresponds well with the geological age and the morphology of C. barthii. Based on these data, and following the excellent reconstruction of Euparkeria presented by Paul (2002), the Chirotherium track maker was reconstructed and displayed in a life-size bronze-sculpture in the Chirotherium Monument that was inaugurated in 2004 in the market-place of Hildburghausen, close to the type locality of Chirotherium barthii (Haubold, 2006). The most common and popular interpretation of Chirotherium barthii and C. sickleri as tracks of various members of the Crurotarsi is not followed here. To the contrary, the Olenekian to early Anisian age of Chirotherium: (1) supports a more general interpretation in the sense stated above, and (2) the morphology of the pes imprints of both C. barthii and C. sickleri include and presage the tridactyl pattern of the later dinosaurs, and of theropods in particular. Support for this interpretation is further substantiated by the development of the two key-features in the evolution of dinosaurs: tridactyly and bipedality, both of which are reflected in a stratigraphic succession of footprint morphs and trackways from Chirotherium, Sphingopus, Parachirotherium and Atreipus through Grallator and Eubrontes (Haubold and Klein, 2000, 2002). For Synaptichnium and “Brachychirotherium,” as well as 121 Isochirotherium in the Middle Triassic and Brachychirotherium (sensu stricto) in the Upper Triassic, a general affinity with the Crurotarsi appears to be realistic. However, this cannot be demonstrated convincingly as is usually supposed, and there is no evidence for any synapomorphies of the group in these footprints.The correlation of footprint morphs, beginning in the Lower Triassic with Protochirotherium, with the archosaurian evolutionary “grades” seems to be well substantiated, in particular by the extraordinarily fine preservation of the latter described by Fichter and Kunz (2004) from the Detfurth Formation of the Middle Buntsandstein of Hessen. The Rotodactylus track maker has been interpreted as a member of the Lagosuchia resp. Dinosauromorpha (Haubold, 1967, 1999; Haubold and Klein, 2002), a conclusion supported by correspondence of the tracks with skeletal anatomy. Of importance is the geological age of the tracks compared to their proposed skeletal correlates: Rotodactylus and Chirotherium, as well as Isochirotherium and Synaptichnium (“Brachychirotherium”), occur as early as the Olenekian-Anisian transition, and prove a diversity that has to be younger than the hypothetical stage of Archosauria, i.e., the beginning of the differentiation of this crown-group. Early-Late Triassic archosaur tracks come from sequences with multiple horizons in localities as far removed today as southern Thuringia and northern Bavaria on one side and Arizona, Connecticut and Massachusetts on the other. In tandem, they play key roles in the biochronological documentation of the evolution of archosaurian foot morphology and locomotion by fossil imprints and tracks. On both continents, sequences with known track horizons range from the Olenekian to the Norian-Rhaetian (Fig. 1). Through intensive research, important correlative locations have been discovered in other regions of Germany, Switzerland, France, Great Britain, Italy and Poland. In North America, the number of occurrences in the Triassic of the Newark Supergroup, for example in Pennsylvania, and in the Chinle Group of Colorado, New Mexico, Utah, Arizona and Texas have increased. Triassic archosaur tracks are known from Lesotho in southern Africa (Molteno and Lower Elliot formations), South America (Argentina), and southern China (Guanling Formation of the Guizhou Province [Lü et al., 2004]). The large number of ichnogenera with archosaurian affinities that have been described so far– roughly 50, even excluding Chirotherium, Rotodactylus, Eubrontes and Grallator – is a further indication that the record is truly extensive. Besides Chirotherium itself, the following ichnogenera are considered chirotherian: Brachychirotherium Beurlen, 1950, Isochirotherium Haubold, 1971, Parachirotherium Kuhn, 1958, Protochirotherium Fichter and Kunz, 2004, Parasynaptichnium Mietto, 1987, Sphingopus Demathieu, 1966, and Synaptichnium Nopcsa, 1923. Furthermore we consider as archosaur tracks forms that have been described under the following names: Aetosauripus Weiss, 1934, Agialopous Branson and Mehl, 1932, Agrestipus Weems, 1987, Anchisauripus Lull, 1904, Atreipus Olsen and Baird, 1986, Banisterobates Fraser and Olsen, 1996, Batrachopus Hitchcock, 1845, Brontozoum Hitchcock, 1847, Coelurosaurichnus v. Huene, 1941, Dahutherium Montenat, 1968, Dinosaurichnium Rehnelt, 1950, Eubrontes Hitchcock, 1845, Evazoum Nicosia and Loi, 2003, Gigandipus Hitchcock, 1856, Grallator Hitchcock, FIGURE 1. Archosaur footprint horizons and track-bearing formations in North America and southern Thuringia/northern Bavaria (Germany). In nearly complete sequences, the intercontinental distributions of characteristic ichnotaxa are used for correlation. Notice conformity of the assemblages at the corresponding levels and differences due to the gap in the North American record between the late Anisian and late Ladinian. 122 1858, Gregaripus Weems, 1987, Otozoum Hitchcock, 1847, Pachysaurichnium Demathieu and Weidmann, 1982, Prorotodactylus Ptaszynski, 2000, Rigalites v. Huene, 1931, Swinnertonichnus, Sarjeant, 1967 and Thecodontichnus v. Huene, 1941. These ichnotaxa pertain to specimens from different parts of the Triassic and their alphabetical listing above does not indicate an evaluation of their validity. A special taxonomic situation surrounds the tracks from the Upper Triassic of Lesotho for which Ellenberger (1972) introduced a wide variety of names like Anatrisauropus, Bosiutrisauropus, Deuterosauropodopus, Deuterotrisauropus, Paratetrasauropus, Paratrisauropus, Pentasauropus, Prototrisauropus, Pseudotetrasauropus, Pseudotrisauropus, Psilotrisauropus, Qemetrisauropus, Sauropodopus, Seakatrisauropus, Tetrasauropus, and Trisauropodiscus. Comparisons of these with ichnotaxa described from elsewhere, and consensus on synonymies, are still in progress; most of his ichnotaxa have not been recognized outside Lesotho and are generally perceived (even in the absence of detailed analyses) as junior synonyms of other, better-known ichnotaxa. Even larger is the number of ichnospecies that have been established within the aforementioned ichnogenera. For Chirotherium alone, the authors counted 50 species names; for all chirotherians, in nearly any combination within the ichnogenera, there are at least 75 ichnospecies. Altogether, about 50 ichnogenera and about 180 ichnospecies have been ascribed to Triassic archosaurs. The status of many ichnotaxa is doubtful; in many cases, they are demonstrably synonyms of well-established taxa. However, this synonymy is evaluated differently, depending on the material and the describing author. Some authors have erected ichnotaxa while ignoring extramorphology, a phenomenon that was recognized as having a misleading influence on tetrapod ichnotaxonomy as long ago as Peabody (1948). For many of the named ichnotaxa, synonymy remains open, and such forms must be considered phantom taxa (sensu Haubold, 1996). Following the very precise guidelines established by Peabody (1955a), ichnospecies and ichnogenera cannot be definitively attributed to a species or genus that is based on body fossils. Body fossil genera are nearly equivalent to ichnospecies. So, the lowest level to which a Triassic archosaur track can be differentiated corresponds with an osteological genus. Consequently, the number of named ichnotaxa cited above would imply a correlative number of osteological ichnogenera. This seems unrealistic – the presently documented ichnological archosaur diversity in the Triassic must be reduced substantially to a smaller number of ichnogenera and ichnospecies. TRIASSIC ARCHOSAUR FOOTPRINTS – THE EVIDENCE Early Early Triassic Tetrapod footprints from this interval come from the Labyrinthodontidae Beds (late Induan) of Wióry (Poland) and from the Dethfurt and Hardegsen formations (early Olenekian) of northern Hessen, Germany (Demathieu and Haubold, 1982; Fuglewicz et al., 1990; Ptaszynski, 2000; Fichter and Kunz, 2004). Essential components of these assemblages are Synaptichnium (Fig. 2A-B) and Protochirotherium, including the type species of the latter, P. wolfhagense Fichter and Kunz, 2004 (Figs. 2C, 3A-B). The status of Synaptichnium and the relationship to Protochirotherium is uncertain (see also below). From the locality in Poland, Brachychirotherium and Isochirotherium were described by Fuglewicz et al. (1990) and Ptaszynski (2000). However, the features exhibited by these specimens, particularly the digit proportions (long pedal digit IV), differ from those listed in the diagnoses for these taxa (Beurlen, 1950; Haubold, 1971b; Karl and Haubold, 1998). We therefore refer this material to Protochirotherium (Fig. 2D-E). Synaptichnium and Protochirotherium indicate a primitive archosaur foot morphology typified by long pedal digits IV and V. Late Early Triassic and Middle Triassic Beginning in the late Olenekian, there is evidence of a broad spectrum of archosaur tracks in the global record, exemplified by Rotodactylus, Synaptichnium, Isochirotherium and Chirotherium, reflecting different evolutionary developments in foot morphology and a biological diversity not thus far documented by the skeletal record. From strata in the “Thüringischer Chirotheriensandstein” (late Olenekian-Anisian) of southern Thuringia, dense concentrations of pentadactyl pes and manus impressions of Rotodactylus are known (Fig. 4A-C; Haubold, 1967, 1971a, b, 1999). A characteristic feature of Rotodactylus is the dominance of digit group II-IV and the extreme posterior position of a small punctiform mark that constitutes the impression of digit V. The digit proportions are I<II<III<IV. Trackways preserve evidence of long strides and a primary, lateral overstep of the manus by the pes, though respective values of overstepping and stride length show high variability. The data indicate cursorial trackmakers that, with our present knowledge, must be attributable to dinosauromorphs comparable in “grade” to lagosuchians FIGURE 2. Characteristic archosaur footprints from the early Early Triassic. A, B, Synaptichnium, Hardegsen Fm., Hessen and Labyrinthodontidae beds, Wiòry, Poland. C, Protochirotherium wolfhagense Fichter and Kunz, 2004, Detfurth Fm., Hessen (holotype at right). D, E, Protochirotherium (Brachychirotherium and Isochirotherium after Ptaszynski, 2000 and Fuglewicz et al., 1990), Labyrinthodontidae beds, Wiòry. Notice digit proportions and long digits IV and V in the pes. A after Demathieu and Haubold (1982); B, D, E after Ptaszynski (2000) and Fuglewicz et al. (1990).