PALEVO 1631-0683 Elsevier S1631-0683(10)00043-6 10.1016/j.crpv.2010.05.001 Research article General palaeontology (Palaeoichnology) Paléontologie générale/General Palaeontology (Paléoichnologie/Palaeoichnology) Evidence of an Early Triassic age (Olenekian) in Argana Basin (High Atlas, Morocco) based on new chirotherioid traces Évidence de l’âge Trias inférieur (Olénékien) dans le Bassin d’Argana (Haut Atlas, Maroc) basée sur des traces chirothérioïdes Tourani Abdelilah a Benaouiss Naima a Gand Georges b Bourquin Sylvie c Jalil Nour-Eddine a Broutin Jean jean.broutin@snv.jussieu.fr d Battail Bernard e Germain Damien e Khaldoune Fatima a Sebban Soumaya a Steyer Jean-Sébastien e Vacant Renaud e Département de géologie, faculté des sciences Semlalia, BP 2390, Marrakech 40000, Maroc Biogeosciences UMR 5561, centre des sciences de la Terre, de la vie et de l’environnement, 6, boulevard Gabriel, université de Bourgogne, Dijon, France CNRS/Insu, UMR 6118 géosciences Rennes, université de Rennes 1, campus de Beaulieu, 35042 Rennes cedex, France UMR 7207, CNRS, université Pierre-et-Marie-Curie, centre de recherches sur la paléobiodiversité et les paléoenvironnements, MNHN CP48, 57, rue Cuvier, 75231 Paris cedex 05, France UMR 7207, CNRS, Muséum national d’histoire naturelle, centre de recherches sur la paléobiodiversité et les paléoenvironnements, MNHN CP48, 57, rue Cuvier, 75231 Paris cedex 05, France 9 5 S1631-0683(10)X0005-7 201 208 © 2010 Académie des sciences. Published by Elsevier B.V. All rights reserved. 2010 Académie des sciences Full (PDF)

New chirotherioid traces (Synaptichnium, Chirotherium, Brachychirotherium, Isochirotherium), are described in the Argana Basin (High Atlas of Morocco). Seeing that these ichnotaxa are frequent in the Triassic, their occurrence in outcrops formerly mapped as Permian (T2 Member) has required detailed sedimentological and paleontological studies of the fossiliferous site. These studies clearly show that the ichnite-bearing strata belong actually to the T3 Member of the “regional Triassic”, i.e. lower member of the Timezgadiwine Formation, the age of which was, in fact, unknown up to now. The description of these ichnospecies and their statistical comparison with those of other Early and Middle Triassic areas, suggest an Olenekian age for this footprint site, and consequently for the T3. The trackmakers were Archosauriformes, some of which had autopodia less evolved than those of Anisian age. With Lepidosauria, they lived in a floodplain close to alluvial-fans.

Des traces chirothérioïdes (Synaptichnium, Chirotherium, Brachychirotherium, Isochirotherium) ont été découvertes dans le Bassin d’Argana (Haut Atlas, Maroc). Ces ichnotaxa étant fréquemment représentés dans le Trias, leur présence dans des affleurements antérieurement cartographiés comme Permien (attribués au Membre T2) a nécessité une étude sédimentologique et paléontologique approfondie du site fossilifere. Il en resulte que ces traces sont, en fait, localisées dans le Membre T3 du « Trias régional », membre inférieur de la Formation de Timezgadiwine, qui n’était pas encore réellement daté. Après la description ichnospécifique des ichnites et leur comparaison statistique avec celles du Trias inférieur et moyen d’autres régions, un âge Olénékien est proposé pour ce site à empreintes et, par conséquent, pour le Membre T3. Les auteurs de ces traces furent des Archosauriformes dont certains avaient des autopodes moins évolués que ceux des formes de l’Anisien. Avec les Lepidosauria, ils ont vécu dans une plaine d’inondation en bordure de cônes alluviaux.

 Morocco, Argana Basin, Early Triassic, Olenekian, Chirotherioid footprints, Archosauriformes, Lepidosauria Maroc, Bassin d’Argana, Trias inférieur, Olénékien, Chirothérioïde, Traces de pas, Archosauriformes, Lepidosauria presented Presented by Philippe Taquet Introduction

The stratigraphic study of the Permian and Triassic series of the Argana basin was made by Roch (Roch, 1930), and their cartography carried out by Tixeront (Tixeront, 1973 and Tixeront, 1974). Within this framework, several sedimentological works, quoted in Medina et al. (Medina et al., 2000 and Medina et al., 2001), Tourani et al. (Tourani et al., 2000) and Hofmann et al. (Hofmann et al., 2000), were elaborated among which that of Jones who discovered the first footprints (Rhynchosauroides) in the Argana Basin, on Permian outcrops, south-east of Timezgadiwine (Jones, 1975). Later, others were found by Dutuit (Dutuit, 1976) in the Triassic T5 Member of this same area, and by Beauchamp (in Biron (Biron, 1982)) in the Triassic of the Ourika valley (High Atlas). All these tracks, studied by Biron and Dutuit (Biron and Dutuit, 1981), are currently stored in the collections of the Faculty of Sciences Semlalia (Department of Geology) and the Muséum national d’histoire naturelle (Paris).

Recently, since the 2000s, the first two authors of this work found other footprint sites and gathered new traces during their sedimentological studies of the Argana Basin and neighbouring Permian and Triassic Formations of High Atlas Mountains (Fig. 1A). One of these track places, the Iggui Aouglef mapped as Permian by Tixeront (Tixeront, 1974) yielded Triassic chirotherioid footprints. Thus, this discrepancy justified a new detailed cartographic, sedimentological and paleontological study of these Permian-Triassic beds. The aim of this paper is:

to replace them within their geological and stratigraphic contexts;

to describe the tracks themselves;

to propose a chronological attribution based on a comparison with European Permian-Triassic series.

Geological and stratigraphic context of the tracksite

Because of numerous uses of the lithostratigraphic term Formation and Member, these will be replaced by the abbreviations Fm and Mb.

The track site of Iggui Aouglef is located about 50 km to the south of Imin’ Tanout, 8 km SSW of Timezgadiwine, and 2 km south of Irerhi village, along the Assif Aït Messaoud (Fig. 1B, star). According to the geological map, it is located in the terminal part of the unit T2 (Tourbihine Mb), the upper member of the Ikakern Fm (Tixeront, 1973 and Tixeront, 1974) dated as Late Permian on the basis of vertebrate remains (Jalil and Dutuit, 1996 and Jalil and Janvier, 2005). However, the precise sedimentological study of this footprint site made by the first two authors showed that it rather belongs to the basal Triassic and especially to the unit T3 (Tanameurt Mb) of the Timezgadiwine Fm (Fig. 1C).The footprint level is about 2.5 m above the Permian-Triassic boundary.

In Iggui Aouglef locality, the Triassic succession is divided into the unit T3 and the overlying unit T4 (Agelgal Mb) (Fig. 1C). The Tanameurt conglomerates, 11 m thick, overlaid unconformably the Late Permian T2 unit (Fig. 1C). Their first meter consists of laterally discontinuous beds up to 40 cm thick of crudely stratified matrix-supported breccias deposited by debris flows. This unit is followed by sheet-like beds composed of matrix-supported conglomerates. They are interpreted as sheet-flood and debris flow deposits in alluvial-fan setting. Carbonate nodules or continuous carbonate beds are frequent features of this part of T3. They are interpreted as pedogenic carbonate accumulations in semi-arid paleosols. Moreover, interbedded light brown to light red sheet sandstone and red to chocolate sandy mudstone, with some mudcracks, are frequent. The very coarse- to medium-grained sandstone facies dominate. They are composed of horizontal laminated, subordinate ripple laminated, or massive layers. This facies association is interpreted as overbank deposits laid down under lower flow regime, in middle alluvial fan setting. They contain the chirotherioid tracks described here.

The upper part of the T3 section consists of conglomerates composed of low-angle trough cross-stratification and planar cross-stratification grading upward to horizontal lamination. This association of facies characterizes bedload deposits of gravelly braided river system.

 Palaeontological study

The footprint site is located at the base of a little cliff (Fig. 2O). Traces are convex hyporeliefs which appear, on 2.25 m2, at the basal surface of a conglomeratic sandstone forming an overhang. The original trace fillings were various, made of gravels to silts, explaining the poor quality of several of them. Among these traces were distinguished about twenty footprints which may fit in the Nopsca (Nopcsa, 1923) crocodiloid group and more precisely in that of Chirotherioid as defined by Lessertisseur (Lessertisseur, 1955) or in the Chirotheriidae ichnofamily (Abel, 1935). Some others are lacertoid and close to Rhynchosauroides Maidwell, 1911 (Maidwell, 1911).

 Description of the footprints

The following abbreviations will be P and M = pes and manus traces often named only pes or manus, CPM = associated pes-manus; I, II, III, IV, V: digits numbering; L and W = footprint length and width; LD and WD = I–IV part length and width; LP and LM = pes and manus length, Q = cross-axis angle in degrees = angle between the metapodial-phalangeal axis and the long axis of the footprint (see Fig. 3B); L/W, LD/WD, L/LM, L/LD, III/II, III/IV, II/I: various characters ratios often statistically tested; 1- 10 FM = collection numbering.

Unlike the Rhynchosauroides track site already mentioned (Jones, 1975) in the Permian of the Argana Basin, most of the traces observed in the Iggui Aouglef site are chirotherioid in aspect. This term indicates pentadactyl, heteropod, quadruped tracks of which digits I–IV form a distal part more or less separated from the digit V. This later is reduced or absent, and extended by a large metapod trace. It constitutes the posterior part of the footprint, which often has a lateral posterior outer position.

As a whole, this morphological organisation characterizes the ichnogenera Chirotherium Kaup, 1835 (Kaup, 1835), Brachychirotherium Beurlen, 1950 (Beurlen, 1950) in which the pes digit III is the longest whereas it is IV for Synaptichnium Nopsca, 1923 (Nopcsa, 1923), and Isochirotherium Haubold, 1971 (Haubold, 1971b) characterized by subequal II and III.

 The manus-pes couple 3 FM (<xref rid="fig2" ref-type="fig">Figs. 2</xref>K<italic>,</italic> 3D)

The pes trace dimensions are as follows: L × W = 136 × 48 mm; LD × WD = 80 × 48 mm; I–IV lengths successively equal to 38, 56, 69, 77 mm; cross-axis angle Q = 68°; angle III–V = 40°. The whole is therefore lacertoid in aspect with toes close to each other (angle I–IV = 15°). They are ended by rather strong claws directed towards the footprint inside. An oblong metatarsal mark without the digit V is only present at the back of the I–IV digital part.

The manus is located to 97 mm ahead of the pes. The digit IV is not visible. It seems to be torn off, or its trace was erased by the later passage of another animal which made a print covering partially the preceding footprint. Digits are ended by short claws. II and III are rather broad and the first (I) is present only by the claw mark; recorded dimensions are: L × W = 65 × ? 40 mm; II and III are subparallel, with III = 47 mm, II = 41 mm; digito-metacarpal pad V = 37 mm; Q = 75°, III–V angIe = 50°.

 The 2 FM trace (<xref rid="fig2" ref-type="fig">Figs. 2</xref>H, 3E)

Longer than broad, it is a pes trace of which the digital lengths decrease from IV to I digits. Q is low: 60° and the angle III–V = 40°. L × W must be close to 135 × 70 mm.

 The trackway n° 1 (5, 6, 8 FM traces) (<xref rid="fig2" ref-type="fig">Fig. 2</xref> and <xref rid="fig3" ref-type="fig">Fig. 3</xref>)

It is made of two consecutive couples P1–M1 and P2–M2 (Figs. 2A, C, D), and an incomplete pes P4 (Fig. 2F); P3–M3 seems to be destroyed or removed. The pes traces show a plantigrad support (Figs. 2C, D, F, L and 3F, G, I). The digits I–IV are broad, squeezed and only a little separated (angle I–IV = 22°). I and IV are right whereas II and III are bent laterally to the outside of the trackway. They are ended by rather discrete claws. That of III is clear and constant. The IV is short on P2 and P4, not very printed on P1. The digit II is present in P4, but not preserved in P1.

The I–IV part is complete only in P1 where it appears paraxonic, broader than long, with LD × WD = 79 × 93 mm. It is extended backwards by the quadrangular toe V, without claw, which comes from a broad triangular to ovoid sole representing the metatarsal pad print.

Based on P1, L × W = 170 × 91 mm and digital lengths, without claws, are as follows: I = 65, II = 75, III = 70, IV = 59 mm (II > III > I > IV). This order is the same for P2 and P4. The II is thus the longest, a few millimeters longer than the III, and the IV is the smallest. The length of V is respectively 39, 28 and 11 mm for P1, P2 and P4. The pes length mean is 161 mm. The P1 trace is the best preserved; its surface is covered by transversal or oblique furrows. Those of the internal border partly correspond to a slight autopodium skid. The others are cutaneous folds and round-scale band traces which are quite visible on the toe I and a part of the II (Figs. 2L, 3F, E, J).

The M1 and M2 traces are located at 156 and 180 mm ahead of their respective pes. They show a digitigrad support. On M1, the I–IV part is as long as broad (LD × WD = 62 × 54 mm). The digits IV–II appear there broad, right, a little divergent from each other (angle IV–II = 9°), clawed and enough strong, unlike the I which is thin and not very visible because of it is stuck against the II. The digital lengths are: I = ? 29, II = 45, III = 57, IV = 50 mm; III is thus the longest with III > IV > II > I. The V trace is not distinguished from those of metacarpals. This V whole is located back to the I–IV part. It is 43 mm long and the angle III–V is 55°. M1 is as long as broad (L × W = 77 × 74 mm). The M2 trace shows the same organisation and dimensions as M1: L = 82 mm; LD × WD = 62 × 60 mm; I = ? 35, II = 48, III = 57, IV = 49 mm. The little, thin digit I is not clearly marked. V is the partial metapodium trace, not clearly readable because of a print superposition. Angle II–IV = 12°.

 The 1, 4, 7, 9, 10 FM traces (<xref rid="fig2" ref-type="fig">Fig. 2</xref> and <xref rid="fig3" ref-type="fig">Fig. 3</xref>)

These footprints seem isolated on the surface. Among the best preserved, we distinguished two CPM: 9 and 1 FM (Figs. 2M, J, 3K, L) and two pes traces: 10 and 7 FM (Figs. 2I, G, 3B, C). The four pes length mean equal 113 mm. These traces are typically chirotherioid in aspect as those of the trackway n° 1. But there the I–IV toes, ended by short claws and are more spread (angle II–III = 31° against 15°). As a rule, also the digit V is much shorter than its neighbours (length mean = 20 mm). It is extended inside by metatarsal pads expanded strongly behind the I–IV part, especially for 1 and 7 FM. The manus is located ahead of the pes, outside for 9 FM (Figs. 2M, 3K) and just in front of for 1 FM (Figs. 2J, 3L). The digit V and associated metacarpal pads traces of 1 FM were covered by the pes during the moving. The small 4 FM print could be that of a pes but there is no trackway allowing to know it with certainty (Figs. 2E, 3H).

 Ichnotaxonomy The 3 FM manus-pes couple: <italic>Synaptichnium</italic> cf. <italic>pseudosuchoides</italic>

Several chirotherioid ichnospecies were mentioned in the Lower and Middle Triassic of the USA (lower part of the Moenkopi Fm (Peabody, 1948)) and Europe where they were found in various formations: Keuper Sandstone of England (Haubold, 1971b), Middle and Upper Buntsandstein of Germany (Detfurth, (Fichter and Kunz, 2004 and Kunz and Fichter, 2000), Hardegsen, Solling and Röt Fms (Demathieu and Haubold, 1982, Demathieu and Leitz, 1982, Fichter, 1995, Haubold, 1967, Haubold, 1971a, Haubold, 1971b and Haubold, 1983), and Poland (Labyrinthodontid beds, (Fuglewicz et al., 1990 and Ptaszyński, 2000)).

A comparison of the 11 FM footprints with the various ichnospecies described in the formations quoted above, shows that they more resemble the ichnospecies Synaptichnium pseudosuchoides Nopsca, 1923. The type comes from the English Lower Triassic, and similar forms have been described in the Solling Folge and the Lower Röt (Demathieu and Haubold, 1982 and Demathieu and Leitz, 1982). So the Synaptichnium ichnogenus is known now in Argana Basin, but only from a CPM whose the pes is very narrow (LD/WD = 1,67).

 The 2 FM trace: cf. <italic>Synaptichnium</italic>

We report with doubt to this ichnogenus the 2 FM pes which is not entirely visible in situ (Figs. 2H, 3E). Nevertheless the decreasing of the IV–I digital lengths and the low values of Q (= 60°) and that of the angle III–V (= 40°), plead for this generic attribution.

 The trackway n° 1 (5, 6, 8 FM traces) FM traces: <italic>Brachychirotherium</italic> sp.

In addition to Synaptichnium, the Chirotherium, Isochirotherium and Brachychirotherium ichnogenera are also present in the Early Triassic palichnofauna. Few years ago, the Chirotheriidae ichnotaxa grew with Protochirotherium wolfhagense Fichter and Kunz 2004 (Fichter and Kunz, 2004), described in the Detfurth Fm (Middle Bunter, Germany).

Although the trackway n°1 footprints show II–III digital lengths virtually subequal, they cannot be assigned to Isochirotherium because of the manus size. Indeed, the ratio (LP/LM = 2.2) of these footprints differs significantly from that of Isochirotherium which is included between 2.6 and 3.7, but fits with those of Chirotherium (1.9–2.7) and Brachychirotherium (1.4–3.1) (calculus by authors). Based on the LP/LM ratio mean equal to 2.2, the trackway traces can thus be regarded as Chirotherium or Brachychirotherium. But in respect to the distal tip of the broad subparallel II–IV toes, which are rounded in shape and ended by narrow claws, we ascribe them to Brachychirotherium. Moreover these traces of the Argana Basin have similar measurements with B. hauboldi (Ptaszyński 1990 in Fuglewicz et al., 1990 and Ptaszyński, 2000) from the Labyrinthodontid Beds of Poland. In each case, the pes prints are directed towards outside of the trackway, by nearly the same value, respectively 17° and 20° for B. hauboldi and those of the trackway. The pace angulation is variable in B. hauboldi: 157° (Ptaszyński, 2000), 120° to 65° (Fuglewicz et al., 1990), and 110° for the trackway n°1 of the Argana Basin. But whereas the longest toe of the trackway n°1 is the II (sample n = 3), it is the III in B. hauboldi where, for n = 11, III/II mean = 1.17 (1.12–1.22) with confidence interval at 5% level, and III/IV mean = 1.19 (1.16–1.22).

The pes of the trackway n°1 was also compared to that of Protochirotherium wolfhagense. At first sight, they are rather close by a LD/WD ratio equal to 0.93 and 0.97. Thus the I–IV part appears a little broader for P. wolfhagense. Nevertheless the latter differs from the trackway n° 1, by its claws stronger and directed inwards (towards inside of the trackway), and by the absence of round digital ends. It is the reason why P. wolfhagense is closer to Chirotherium than to Brachychirotherium. But with the III toe hardly longer than the IV, its autopodium pes has a rather archaic organisation.

Brachychirotherium hauboldi (Ptaszyński, 2000) of which the pedal I–IV part is almost as broad as long, was then statistically compared with ichnospecies sharing this characteristic, namely B. gallicum Willruth, 1917 (Willruth, 1917), B. pachydactylum Demathieu and Gand, 1973 (Middle Triassic, France) (Demathieu and Gand, 1973), I. soergeli (Haubold, 1967) and C. sickleri Kaup, 1835 (Solling Folge, Early Triassic, Germany) (Haubold, 1967 and Haubold, 1971b). The means and variances tested at 5% level showed that B. hauboldi shares similarities with B. gallicum Willruth, 1917, concerning the ratios values of III/IV, II/I and the Q angle, but they differ for L/W, LD/WD, L/LD. With B. pachydactylum Demathieu and Gand, 1973 of the French Middle Triassic (Demathieu and Gand, 1973), there is only the Q value (= 67°) which is no significant. Using the measurements made by Haubold (Haubold, 1971a), the III/II, LD/WD, L/LD values of B. hauboldi and C. barthii Kaup, 1835, are also no significant. Except LD/WD, all the other ratio values are significantly different between I. soergeli (Haubold 1967) (Haubold, 1967) and B. hauboldi, the latter being fully different of C. sickleri Kaup, 1835 for all the ratios.

In the absence of sufficient measurement series for P. wolfhagense and the trackway n° 1 traces, we also executed a statistical comparison of the outlines of the I–IV part between the pes P1 of the trackway n°1, 3 pes of P. wolfhagense, and 11 specimens B. hauboldi. For this, the complex discrete Fourier analysis clarified in Navaro et al. (Navarro et al., 2004) was used. The Dommergues program (Dommergues, 2001) was selected. Then 200 points, 11 and 20 pairs of harmonics were chosen in order to reconstitute the outlines. They were successively restored at 94 and 97% values. The results were analysed by a CPA (Components Principal Analysis) and visualized by a group of points traces in the two graphs Fl/F2 and F2/F3; the factor F representing the factorial weight.

From the two Fl/F2 graphs, P1 of the trackway and one P. wolfhagense pes are outside of the B. hauboldi cloud, two others being on the limit, and similar by their shape with two C. hauboldi traces. In the two Fl/F3 graphs, the three P. wolfhagense points and the P1 of Argana are external. Thus, there is a certain morphological similarity between two P. wolfhagense samples and the B. hauboldi whole (= population traces) inferred from the Fl/F2 graphs, but none between the Argana traces and the two preceding ichnospecies. Thus, it does not seem appropriate to have synonymise B. hauboldi with Protochirotherium as made Klein and Haubold (Klein and Haubold, 2007).

 The 1, 7 FM traces: <italic>Isochirotherium</italic> cf <italic>gierlinskii</italic>; 9, 10 FM: <italic>Chirotherium barthii</italic>

The 1 and 7 FM pes prints have subequal digits II–III, with a slight prevalence for the III, a characteristic favourable to its attribution to Isochirotherium. This hypothesis is also corroborated by the LP/LM value = 2.7, but the manus is not quite complete. Based on the II–IV lengths, which exceed faintly those of II and IV in the CPM 1 FM, it appears that these last tracks are neither Isochirotherium archaeum Demathieu and Haubold, 1982 (Demathieu and Haubold, 1982) from Hardegsen Fm (Germany, Early Triassic) nor I. soergeli, I. hessbergense, I. jenense and I. herculis from the Solling Fm (Germany, Early Triassic) (Haubold, 1967, Haubold, 1971a, Haubold, 1971b and Haubold, 1983). On the other hand, except for the IV length, smaller for the two Argana traces, the length ratios I/III, taken two by two, are close to those of I. gierlinskii from the Labyrinthodontid Beds of Poland. This species also has important posterior digito-metatarsal pads (Fuglewicz et al., 1990 and Ptaszyński, 2000).

Following a mean and variance statistical comparison, the 9 and 10 FM pes (Fig. 3B, K) have more common points with Chirotherium barthii than with C. sickleri. Eight tested characters (Q and I–IV angles, L/W, LD/WD, L/LD, II/I, III/IV, and III/II ratios) are similar to those of C. barthii whereas three (Q, L/W and III/I) differ from those of C. sickleri.

 Palaeontological, palaeoenvironmental and stratigraphic results

The osteological interpretation of Synaptichnium, Chirotherium, Isochirotherium and Brachychirotherium has long been discussed (Demathieu, 1970, Demathieu and Haubold, 1974, Haubold, 1971b and Peabody, 1948). This led, in the 1970s, to a general agreement that ascribes them to Archosauria Pseudosuchia, or Archosauriformes Crurotarsi in cladistic nomenclature (Sereno, 1991). In such a phylogenetic concept, Synaptichnium would have been made by a basal Archosauriforme such as Euparkeria, and the other ichnogenera by different undetermined Crurotarsi. From the autopodium structure inferred from the pes traces of the trackway n° 1, it seems that the author of these Argana footprints was an archaic Crurotarsi. By using various methods quoted in Gand et al. (Gand et al., 2007), the analysis of the trackway characters and measurements allow to reconstitute a pentadactyl quadruped animal, approximately 2 m long, with forelimbs shorter than the posterior ones, a measured (from the trackway) and calculated trunk length about 50 cm, and a barycentre position located between 1/3 and 1/4 from the acetabulum (Demathieu, 1970). The pace angle value (110°) is indicative of semi-erected members. Nevertheless, this stance could be occasional as for the B. hauboldi trackmakers. Aside these Crurotarsi, Lepidosauria were also present as indicated by the Rhynchosauroides footprints.

The sedimentologic characters of the beds bearing the footprints level are indicative of an alluvial plain under the influence of alluvial fan inputs. As it was suggested, this environment was therefore inhabited by a tetrapod community dominated by Archosaurian Crurotarsi. The measured orientation of their traces in the Iggui Aouglef site indicates that animals walked west-east in both directions.

Until now the unit T3 (Tanameurt Mb) of the Timezgadiwine Fm was only tentatively referred to the Triassic. It overlies unconformably the unit T2 (Tourbihine Mb) of the Ikakern Fm, assigned to the Upper Permian thanks to vertebrate remains (Jalil and Janvier, 2005). It is overlain by the unit T4 (Aglegal Mb), dated as Middle Triassic by charophytes and ostracodes (Medina et al., 2001) and as Anisian by Heylerosaurine Cyclotosauridae remains (Jalil et al., 2009). The tetrapod footprint assemblage studied here provides new element for biostratigraphic dating of the T3 unit.

Any of the ichnospecies described in the Middle Triassic of France (Anisian-Ladinian “Grès inférieurs” Fm) (Demathieu, 1970 and Gand et al., 2007)) and Germany (Röt Fm (Haubold, 1967, Haubold, 1971a and Haubold, 1971b) dated Lower Anisian (Bourquin et al., 2006)) have been identified in the present study. Moreover, the closeness of the Argana ichnospecies to the German Synaptichnium cf. pseudosuchoides and Chirotherium barthii (Solling Fm., Upper Olenekian (Bourquin et al., 2006)), and to the Polish Isochirotherium cf Gierlinskii (Labyrinthodontid beds, Upper Olenekian (Bourquin et al., 2006 and Niedźwiedzki and Ptaszyński, 2007)) is clearly indicative of an Early Triassic age of this Moroccan ichnofauna, probably Upper Olenekian (Spathian). An Early Triassic age, possibly later than mid-Olenekian, could also be deduced from the Brachychirotherium sp. trackway, which suggests Crurotarsi with autopodia less advanced than those known in the Anisian.

 Conclusions

The Iggui Aouglef ichnites are represented by Rhynchosauroides sp. and chirotherioid footprints: Synaptichnium cf pseudosuchoides, Synaptichnium sp., Brachychirotherium sp., and Isochirotherium cf gierlinskii. They have been identified above all at the generic level, except Chirotherium which is clearly present by the ichnospecies C. barthii. Animals were Archosauriformes Crurotarsi, and Lepidosauria represented by Rhynchosauroides. On the footprint surface, it could be shown that all these animals moved in east–west opposite directions. From the sedimentological characters of the fossiliferous site, we deduced that the palaeoenvironment they run through was a floodplain periodically invaded by braided channels during wet periods. These last alternated with dryness phases during which vertisols could develop.

According to the closeness of the aforesaid footprints with Lower Triassic ichnospecies, and the relative archaic pes bony structure of the Brachychirotherium sp. trackmaker, an Upper Olenekian (Spathian) age can be proposed for the footprint level from the T3 unit of Argana Basin.

 Acknowledgements

This study was led in the framework of collaborative research work entitled “La limite Permien-Trias en milieu continental : crise de sédimentation et/ou crise d’extinction biologique en masse ? Paléoenvironnements de dépôts et modalités de fossilisation dans les bassins permo-triasiques du Haut Atlas Marocain”. Financial support was provided by the MNHN-PPF04 project “Biodiversité actuelle et fossile. Crises, stress, restaurations et panchronisme : le message systématique”. We are deeply grateful to P. Janvier for his great interest and encouragement to our study and we thank very much the two reviewers for their useful remarks and suggestions.

 Abel O. Vorzeitliche Lebensspuren 1935 Fischer Verlag Iena 644 p Beurlen K. Neue Fährtenfunde aus der Fränkischen Trias N. Jahrb. Geol. Palaont. 1950 308–320 Biron P.E. Le Permo-Trias de la région de l’Ourika (Haut Atlas de Marrakech, Maroc), sédimentologie tectonique et métallogénie, thèse de 3e cycle 1982 Institut Dolomieu Grenoble 170 p Biron P.E. Dutuit J.M. Figurations sédimentaires et traces d’activité au sol dans le Trias de la formation d’Argana et de l’Ourika (Maroc) Bull. Mus. Natl. Hist. Nat. C 3 1981 399–427 Bourquin S. Peron S. Durand M. Lower Triassic sequence stratigraphy of the western part of the Germanic Basin (west of Black Forest): fluvial system evolution through time and space Sediment. Geol. 186 2006 187–211 Demathieu G. Les empreintes de pas de vertébrés du Trias de la bordure nord-est du Massif central Cahiers de Paléontologie 1970 Éditions CNRS Paris 291 p Demathieu G. Gand G. Deux espèces ichnologiques nouvelles des Grès à empreintes du Trias du plateau d’Antully Bull. Soc. Hist. Nat. Autun 67 1973 11–26 Demathieu G. Haubold H. Evolution und Lebensgemeinschaft terrestrischer Tetrapodenden nach ihren Fährten in der Trias Freiberg. Forschungsh. C 298 1974 51–71 Demathieu G. Haubold H. Reptilfährten aus dem Mittleren Buntsandstein von Hessen (BRD) Halle. Jahrb. Geowiss. 7 1982 97–110 Demathieu G. Leitz F. Wirbeltier-Fährten aus dem Röt von Kronach (Trias Nordost-Bayern) Mitt. Bayer. Staatssaml. Palaont. Hist. Geol. 22 1982 63–89 Dommergues C.H. 2001. CDFT complex discrete Fourier transform (Matlab package), version 2.7. Dijon: Biogeosciences-Dijon, UMR CNRS 5561, Dijon. Dutuit L.-M. Introduction à l’étude paléontologique du Trias continental marocain. Description des premiers Stégocephales recueillis dans le couloir d’Argana (Atlas Occidental) Mem. Mus. Natl. Hist. Nat, Paris C 36 1976 253 p Fichter J., 1995. Ein bemerkenswerter Fund von Hans Penndorf (1879–1960), Philippia, 7/2, 91–108. Fichter J. Kunz R. New genus and species of chirotheroid tracks in the Detfurth-Formation (Middle Bunter, Lower Triassic of Central Germany) Germany Ichnos 11 2004 183–193 Fuglewicz R. Ptaszyński T. Rdzanek K. Lower Triassic footprints from the Świętokrzyskie, (Holy Cross) Mountains, Poland Acta Palaeont. Polonica 35, 3–4 1990 109–164 Gand G. Demathieu G. Montenat C. Les traces de pas d’Amphibiens, de dinosaures et autres reptiles du Mésozoique français: inventaire et interprétations Palaeovertebrata 35,1–4 2007 149 p Haubold H. Eine Pseudosuchier-Fährtenfauna aus dem Buntsandstein Südthüringens Halle. Jahrb. Mitteldt. Erdgesch. 8 1967 12–48 Haubold H. Die Tetrapodenfährten des Buntsandsteins in der Deutschen Demokratischen Republik und in Westdeutschland und ihre Äquivalente in der gesamten Trias Palaont. Abh. A IV 3 1971 395–660 Haubold H. Ichnia Amphibiorum et Reptiliorum fossilium Kuhn O. Handbuch der Paläoherpetologie 18 1971 Encyclopedia of Paleoherpetology Stuttgart 123 p. Haubold H. Archosaur evidence in the Buntsandstein (Lower Triassic) Acta Palaeont. Polonica 28 1983 123–132 Hofmann A. Tourani A. Gaupp R. Cyclicity of Triassic to Lower Jurassic continental red beds of the Argana Valley Morocco: implications for palaeoclimate and basin evolution Palaeogeogr. Palaeoclimatol. Palaeoecol. 161 2000 229–266 Jalil N.-E. Dutuit J.-M. Permian captorhinid reptiles from the Argana formation Morocco. Palaeontol. 39 1996 907–918 Jalil N.-E. Janvier P. Les paréiasaures (Amniota, Parareptilia) du Permien supérieur du Bassin d’Argana, Maroc Geodiversitas 27 2005 35–132 Jalil N.-E. Janvier P. Steyer J.-S. A new cyclotosaurid (Amphibia, Temnospondyl) from the Triassic of Argana basin (High Atlas mountains, Morocco); biostratigraphy implications Abstr. 1st Intern. Congr. North African Vertebr. Paleont., NAVEP l, Marrakech 2009 36–37 Jones D.F. Stratigraphy, environments of deposition, petrology, age, and provenance of the basal redbeds of the Argana Valley, western High Atlas Mountains, Morocco, Ms Thesis, New Mexico, Inst Mining Technol. 1975 148 p Kaup J.J. Thier-Fährten von Hildburghausen: Chirotherium oder Chirosaurus N. Jahrb. Min. Geogn. Geol. Petref. 1835 327–328 Klein H. Haubold H. Archosaur footprints, potential for biochronology of Triassic continental sequences Lucas S.G. Spielmann J.A. The Global Triassic Bull. New Mexico Mus. Nat. Hist. Sci. 41 2007 120–130 Kunz R., Fichter J., Die Wolfhager Saurierfahrten. Schr. Vereins Regionalmus. Wolfhagen, Reihe Forschungen 9, 2000, 17 p. Lessertisseur J. Traces d’activité animale et leur signification paléobiologique Mém. Soc. géol. France 74 1955 150 p Maidwell F.T. Note on footprints from the Keuper of Runcorn Hill Proc. Liverpool Geol. Soc. 11 1911 140–152 Medina F. Tourani A. Benaouiss N. Jalil N. Le Permien supérieur et le Trias du bassin d’Argana Livret guide 1e journée, excursion de l’AGP-GMPT 2000 17 p Medina F. Vachard D. Colin J.-P. Ouarhache D.D. Ahmamou M. Charophytes et ostracodes du niveau carbonaté de Taourirt Imzilen (Membre d’Aglegal, Trias d’Argana): implications stratigraphiques Bull. Inst. Sci., Rabat, Sect. Sci. Terre 23 2001 21–26 Navarro N. Zatarin X. Montuire S. Effects of morphometric descriptor changes on statistical classification and morphospaces BioI. J. Linn. Soc. 2004 243–260 Niedźwiedzki G. Ptaszyński T. Large Chirotheriidae tracks in the Early Triassic of Wióry, Holy Cross Mountains, Poland Acta Geol. Polonica 57 2007 325–342 Nopcsa F. Die Familien der Reptilien Fortschr. Geol. Paläont. Berlin 2 1923 1–210 Peabody F.E. Reptile and Amphibian trackways from the Lower Triassic Moenkopi Formation of Arizona and Utah Univ. Calif. Publ. Bull. Dept. Geol. 27,8 1948 295–468 Ptaszyński T. Lower Triassic vertebrate footprints from Wióry, Holy Cross Mountains Poland. Acta Palaeont. Polonica 45,2 2000 151–194 Roch E. Études géologiques dans la région méridionale du Maroc occidental Notes Mem. Serv. Min. Carte Geol. Maroc 9 1930 542 p Sereno P.C. Basal archosaurs: Phylogenetic relationships and functionnal implications J. Vertebr. Paleont. Memoir 2, Suppl. 114 1991 53 p Tixeront M. Lithostratigraphie et minéralisation cuprifères et uranifères stratiformes, syngénétiques et familières des formations détritiques permo-triasiques du Couloir d’Argana (Haut Atlas occidental, Maroc) Notes Mém. Serv. Geol. Maroc 33, 249 1973 147–177 Tixeront M. Carte géologique et minéralogique du Couloir d’Argana 1/100 000, Notes et Mém. Serv. Geol. Maroc 205 1974 Tourani A. Lund J.J. Benaouiss N. Gaupp R. Stratigraphy of Triassic syn-rift deposition in western Morocco Zbl. Geol. Paläont. 9–10 2000 1193–1215 Willruth K. Die Fährten von Chirotherium Z. Naturwiss 86 1917 395–444

The Argana Basin. A: in northern Morocco, B: geological map, modified from Tixeront (Tixeront, 1974); C: stratigraphy of Permian-Triassic-Liassic red series, modified from Tourani et al. (Tourani et al., 2000).

Le bassin d’Argana. A : localisation au Nord du Maroc ; B : carte géologique d’après Tixeront (Tixeront, 1974) modifié ; C : stratigraphie des séries rouges permienne, triasique et liasique, d’après Tourani et al. (Tourani et al., 2000) modifié.

Chirotherioid traces (Argana Basin, T3 unit). A: trackway n° 1 (5-6 FM), Brachychirotherium. sp.; B: 11 FM, unnamed; C, D, F: footprints of trackway n°1, respectively: 5 FM = CP1M1, 6 FM = CP2M2, 8 FM = P4; E: 4 FM, unnamed; G: 7 FM, Isochirotherium cf. gierlinskii; H: 2 FM, cf. Synaptichnium; I: 10 FM, Chirotherium barthii; J: 1 FM, Isochirotherium cf gierlinski; K: 3 FM, CPM Synaptichnium cf. pseudosuchoides; L: pes 5 FM = P1, Brachychirotherium sp; M: 9 FM, C. barthii, possible CPM; N: possible relation between 4 and 2 FM; O: Fossiliferous site (F); P: Footprint level (F). Scale-bar = 1 cm for B-L.

A : piste n° 1 (5-6 FM), Brachychirotherium. sp. ; B : 11 FM, non nommé ; C, D, F : traces de la piste n°1, respectivement : 5 FM = CP1M1, 6 FM = CP2M2, 8 FM = P4 ; E : 4 FM, non nommé ; G : 7 FM, 1.cf. gierlinskii ; H : 2 FM, cf. Synaptichnium ; I : 10 FM, C. barthii ; J : 1 FM, 1. cf. gierlinskii ; K : 3 FM, CPM Synaptichnium cf. pseudosuchoides ; L : 5 FM = P1, Brachychirotherium sp ; M : 9 FM, C. barthii, possible CPM ; N : possible relation entre 4 et 2 FM ; O : site fossilifère (F) ; P : Niveau à traces (F). Barre d’échelle = 1 cm pour B-L.

Chirotherioid traces (Argana Basin, T3 unit). A: trackway n° 1 (5, 6, 8 FM), Brachychirotherium sp.; B: 10 FM, Chirotherium barthii; C: 7 FM, Isochirotherium cf. gierlinskii; D: 3 FM, CPM Synaptichnium cf pseudosuchoides; E: 2 FM, cf. Synaptichnium; F: 5 FM = CP1M1, Brachychirotherium sp., e = scales; G: 8 FM = P4, Brachychirotherium sp.; H: 4 FM, chirotherioid trace; I: 6 FM = CP2M2, Brachychirotherium sp.; J: detail of the scale traces of the digit I in fig. F; K: 9 FM, possible CPM, Chirotherium barthii; L: 1 FM, CPM, Isochirotherium cf. gierlinskii. Scale-bar = 1 cm, except for A and J; ? = doubtful link.

Empreintes chirothérioïdes. A : piste n°1 (5, 6, 8 FM), Brachychirotherium sp. ; B : 10 FM, Chirotherium barthii ; C : 7 FM, Isochirotherium cf. gierlinskii ; D : 3 FM, CPM, Synaptichnium cf pseudosuchoides ; E : 2 FM, cf. Synaptichnium ; F : 5 FM = CP1M1, Brachychirotherium. sp., e = écailles ; G : 8 FM = P4, Brachychirotherium sp. ; H : 4 FM, trace chirothérioïde ; I : 6 FM = CP2M2, Brachychirotherium sp ; J : traces d’écailles e du doigt I de fig. F ; K : 9 FM, CPM possible, Chirotherium barthii ; L : 1 FM, CPM, Isochirotherium cf. gierlinskii. Barre d’échelle = 1 cm, sauf pour A et J ; ? = liaison douteuse.