Middle Triassic Chirotherid Trackways on Earthquake Influenced Intertidal

Cent. Eur. J. Geosci. • 4(3) • 2012 • 495-529 DOI: 10.2478/s13533-011-0080-9

Central European Journal of Geosciences

Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

Review Article

Cajus G. Diedrich∗

Private Research Institute Paleologic, Nansenstr. 8, Germany, www.paleologic.de

Received 30 April 2012; accepted 16 July 2012

Abstract: Chirotherid footprints of Synaptichnium, Chirotherium and Isochirotherium appeared during the late Early (Aegean) to early Late (Carnian) Triassic in central Europe. These taxa are partly revised herein, using both perfect and variably preserved tracks, and very long trackways from an upper Pelsonian intertidal-flat megatracksite of the Germanic Basin coast Pelsonian (Karlstadt Formation). The global Middle Triassic distribution of those footprints suggests seasonal migrations across Pangaea of possible archosauriform reptile trackmakers, such as Euparkeria, Ticinosuchus, Arizonasaurus and Batrachotomus, caused by horseshoe-crab mass migrations into tidal-flat beach reproductive zones in the Germanic Basin. Such seasonal migrations may even suggest a Pangaea-wide food-chain reaction, possibly including the mobilization of fish, marine and terrestrial reptiles, and of which situation the Germanic Basin intertidal-flats is a globally unique example. Keywords: Chirotherid trackways • archosauriform trackmakers • intertidal megatracksite • Middle Triassic • horseshoe-crab reproduction zones • intertidal environments • migration behaviour on Pangaea and coasts © Versita sp. z o.o.

1. Introduction tory of discovery of ‘Chirotherium’ and its various tracks has been described in detail by Swinton [12]. Pieces of the main historical negative sandstone slab from Hild- burghausen (Fig.2A) are today distributed among several The oldest known record of reptile footprints, from Hild- European museums [13]. burghausen, in Thuringia, central Germany, (Fig. 1), be- The same suite of chirotherid tracks have been reported came famous following the discovery of a large slab with from similar-aged layers in the Upper Buntsandstein tracks of the ‘hand animal’ Chirotherium Kaup, 1835 [1], (Roet Formation, Aegean, upper Lower Triassic), near and other isolated footprints. These tracks are from the Jena, in Saxony, central Germany [14], and more recently Roet Formation (uppermost Upper Bunter, Aegean, upper from Middle Muschelkalk (Pelsonian, Middle Triassic) in- Lower Triassic). The discovery of this tracksite prompted a tertidal biolaminites, and intercalated seismically shocked wave of publications by a number of well-known palaeon- layers at Bernburg in central Germany ([15], Fig.2), tologists during the 19th century (e.g. [2–11]). The his- the latter of which forms the focus of this paper. The trackbeds from Hildburghausen, Jena, and other sites in ∗E-mail: [email protected] central Germany [16] are terrestrial ﬂuvial sandstones

495 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

Figure 1. A. Previously described terrestrial tracksites from the Triassic in Germany (Lower Triassic tracksites mainly from Haubold [89]; Bernburg Lower and Middle Triassic tracksites from Diedrich [117]). B. Chirotherid and other tracksites on intertidal ﬂats of the Middle Triassic (uppermost Upper Bunter and Lower to Middle Muschelkalk, Aegean to Illyrian) in Germany (from Diedrich [117]). C. Chirotherid track documentation at the new Middle Triassic tracksite at Bernburg, central Germany. D. Photographic and AUTOCAD documentation of large trackway surfaces at Bernburg. E. Excavation site in the active Solvay Quarry in 2008. F. Slickensided arenite of track-bed 8 (cf. Fig. 2D) overlaying the mud-cracked biolaminite surfaces of track-bed 7 (cf. Fig. 2C), seen during the excavations in 2010.

overlaying the track horizons in the clay beds of the up- across much of the world. Unfortunately, confusing ich- permost Roet Formation. These trackbeds contain three notaxonomy, used in different countries, within different different chirotherid track ichnogenera — Synaptichnium sediment types, and by many scientists with different doc- Nopcsa 1923, Isochirotherium Haubold, 1971 and Chi- umentation techniques to describe variant states of foot- rotherium Kaup, 1835 — as well as other small footprints print preservation, single tracks, or incomplete footprints such as Rhynchosauroides Maidwell 1911, Rotodactylus and trackways, has resulted in the creation of dozens Peabody, 1948, and Capitosauroides Haubold, 1971 [17– of Middle Triassic ichnospecies. The resulting long list 20]. of names, ‘outline sketches’, and chirotherid track de- Chirotherid footprints have also been described from con- scriptions often serves to confuse rather than solve ques- temporaneous upper Lower to Middle Triassic layers

496 Cajus G. Diedrich

Figure 2. A. Stratigraphy of the 110 m thick, Middle Triassic Lower to basal Middle Muschelkalk (Bithynian–Pelsonian) marine carbonate series of Bernburg, with high resolution stratigraphy of the Karlstadt Formation. B. Section showing the boundary between the Jena and Karlstadt Formations, with Wellenkalk, dolomitic platy limestones, and biolaminite carbonates as main facies types. C. Mini-sequences with originally growing, autochthonous, centimetres-thick biolaminites covered by 1–4 cm thick arenites. D. Mud-cracked biolaminites of the most important chirotherid track layer, track-bed 7. E. Slickensided and mud-cracked arenite (tsunamite/tempestite) of track-bed 8, which overlies the chirotherid trackway-bearing biolaminte track-bed 7.

497 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

Figure 3. Distribution of megatracksite beds in intertidal marine carbonates of the Germanic Basin (compiled from Diedrich [72, 73, 84, 86, 117] and new results).

498 Cajus G. Diedrich

Figure 5. Human trackway experiments conducted in the lower intertidal to lower sabkha coastal environments of the Al Dabb’iya Peninsula, west of Abu Dhabi (UAE). A. Upper- most upper intertidal to lower sabkha, with typical surface structures of partly dried and brownish biomats. B. Tracks after several days in the gypsum-dominated lower sabkha, which were ‘fossilized’ early by a thin halite crust. C. Footprints in the uppermost upper intertidal, preserved on smaller, wrinkled biomats. D. Trackway reaching from the lower to lowermost upper intertidal environments. E. Foot- prints, only a few centimetres deep, in the lower intertidal, stabilized carbonate mush, into which mangrove leaves were washed after one flooding; this flooding did not de- Figure 4. Chirotherid trackway preservation types from Bernburg, stroy the tracks at all. F. Deeply printed footprints on the showing the differences between the two main sedi- polygonally cracked biomats, as a result of underlying soft ment surfaces of the Karlstadt Formation (basal Mid- carbonate mush. Many leaves were also flooded into the dle Muschelkalk). A. Faint poorly preserved chirotherid tracks here, although the tracks are quite sharp in outline trackway and trampling area on a 2 cm thick tsuna- due to the damaged, 5 mm thick, biomats. G. Feet sink- mite/tempestite layer with parallel seismic slickensides, ing slowly through the leathery microbial mats. H. Cross track-bed 8 (part of trackway 11; cf. Fig. 7). B. Detailed, section, with the upper 15 cm a thick, organic-rich biolami- well-preserved, chirotherid trackway on millimetre-thick nite microbial mat layer, below which is a soft carbonate mud-cracked biolaminites of track-bed 7 (part of track- mush with mollusks. I. Experiment results, showing the way 2a; cf. Fig. 7). best preservation in the lower intertidal environments; the leathery microbial mats do not preserve tracks at all, or else the trackmakers break through the mats, leaving un- shaped forms. tions about the track types, or the trackmakers and their palaeoecology (e.g. [12, 13, 16, 19, 21–45]). The landmark analysis technique (see [46]) used for dis- makers was first discussed early in the 19th century tinguishing ‘ichnospecies’ appears to work for selected (e.g. [4, 48]). Euparkeria capensis [49], the name given to ‘best preserved’ material, at least on ichnogenus levels, an African archosauriform skeleton, was the first species but only because this analysis does not include footprints to be identified as a probable trackmaker for the Chi- from individuals of different ages or variations in footprint rotherium tracks (e.g. [17]); this species’ trackmaker at- preservation. The method’s usefulness in deriving phy- tribution is revised herein using data from newly pre- logenetic or evolutionary trends, as suggested by Klein pared skeletons. Remains of this archosauriform have also and Haubold [46], is therefore highly questionable in some been reported from the Middle Triassic of Russia, suggest- cases, and it has not been used further in this study, ex- ing a Pangaean distribution covering Laurasia and Gond- cept for the separation of ichnogenera. wana [50–52]; however, it currently remains unknown if Several potential trackmakers have been previously pro- the small quantity of archosauriform remains recovered posed for chirotherid tracks. Nöggerath [47] criticized the from the Upper Buntsandstein Roet Formation of Wald- first, incorrect theory that the ‘human hand-like’ prints shut, Germany [53] also belong to this genus. from Hildburghausen were the footprints of apes (called At least two more archosauriform reptiles have recently ‘Quadrumanen’). The possibility of reptiles as the track- become known from articulated skeletal material in cen-

499 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

tral Europe. The older of these two is from the Anisian– Ladinian boundary (Illyrian–Fasanian boundary, middle Middle Triassic) of Monte San Giorgio, Switzerland, where the carcass of a complete suchian archosauriform, Ticinosuchus ferox [54], which had drifted along the coast and come to rest in lagoonal black shales, was discovered. This archosauriform was proposed to have produced chirotherid trackways [55] based on its comparable foot anatomy, and has been identified as a likely trackmaker for Isochirotherium footprints [15]. Its identity as a trackmaker of Chirotherium tracks is revised herein. Several skulls, and even articulated postcranial skeletal remains including the pedal bones, of Batrachotomus kupferzellensis [56] have been recovered in recent years from coastal, freshwater influenced, marine lagoonal deposits in southwestern Germany (mainly at Kupferzell, Vellberg, and Baden-Wuerttemberg) [56–58]. These were found in the slightly younger layers of the Erfurt Forma- tion (Longobardian, Lower Keuper, Lower Ladinian, uppermost Middle Triassic). In this paper, the trackways and pedal anatomy of Batrachotomus are also compared, as they are very similar to other chirotherid trackways. Other models suggesting that aetosaurs produced ‘Brachychi- rotherium’ trackways [59] do not fit the gait, trackway width, or shape of the Lower Keuper Brachychirotherium trackways. However, the figured tracks of B. thuringiacum used by Lucas and Heckwert [59] (Fig. 7E) are Upper Car- nian in age, and these already have a different anatomy with digits I and IV longer, which does match well with aetosaurs as the likely trackmakers. Therefore, those tracks should in future be attributed to another ichnogenus. Although chirotherid footprints are known from all over Figure 6. The uppermost Lower Triassic (Aegean–Bithynian bound- Europe in the Early and Late Triassic, there appeared ary, Lower Anisian) archosauriform footprint types and main features from the Germanic Basin of central Eu- to be an absence of reptile discoveries from the middle rope (A–B Positives; C–E negatives). A. Manus/pes set part of the Middle Triassic [16, 45, 60]. This absence of Isochirotherium herculis [21] from the lowermost Mid- dle Muschelkalk, preserved on marine biolaminites (Karl- can be demonstrated as simply a gap in the fossil record, stadt Formation, Pelsonian) from Bernburg, D (coll. MB). which has now been filled by important new discoveries B. Manus/pes set of Chirotherium barthii Kaup, 1835 from at Bernburg, and other German sites. This gap previously the lowermost Middle Muschelkalk, preserved on marine biolaminites (Karlstadt Formation, Pelsonian) from led to a number of incorrect conclusions about the global Bernburg, D (coll. MB). C. Manus/pes set imprint of archosaur footprint record (cf. Klein and Haubold [46], Lu- Isochirotherium herculis [21] from the uppermost Upper Bunter terrestrial layers (Roet Formation, Aegean) of Hild- cas [60]), especially since such conclusions also ignored burghausen, Germany (coll. NMS). D. Manus/pes set important material such as Synaptichnium manus/pes sets imprints of Chirotherium barthii Kaup, 1835, from the uppermost Upper Bunter terrestrial layers (Roet Forma- (= Brachychirotherium in Demathieu and Oosterink [61]) tion, Aegean) of Hildburghausen, Germany (coll. MB). E. from Winterswijk in the Netherlands (Fig. 6E), and ex- Manus/pes set imprints of Synaptichnium hildburghausen- cluded Middle Triassic German chirotherid material on sis [90], from the uppermost Upper Bunter, preserved on marine biolaminites (Winterswijk Formation, Aegean) from the basis of incorrect, insufficiently detailed, or lacking Winterswijk, NL (coll. MF). chronostratigraphic dating. New evidence is presented herein concerning archosaur footprint types and their preservation, as well as their body fossils and palaeobiogeographical record. These conclusions are based on important new chirotherid trackway discoveries (the longest ever mapped in Europe), in-

500 Cajus G. Diedrich

may have been responsible for a food-chain reaction in the Germanic Basin [62].

2. Material and methods

Comparative studies were carried out on chirotherid tracks from various intertidal-flat track localities of the Middle Triassic Germanic Basin (Fig. 1), including the quarry localities at Winterswijk (Netherlands; Museum Fredricks = MF collection), Grossenlueder 1 and Germete (Hesse and Westphalia, Germany; Naturkundemuseum Kas- sel = NMK collection), Karsdorf (Saxony-Anhalt, Ger- many; Martin-Luther-University Institute for Geosciences = MLU.IfG collection), and the most recently discovered and material-rich site at Bernburg (Saxony-Anhalt, Ger- many; Landesmuseum Sachsen-Anhalt = LSDA collection). A critical revision of (mostly incomplete) single, non-associated Middle Triassic chirotherid tracks from intertidal biolaminites at the above-mentioned sites is also discussed in relation to the well-preserved material seen in several long trackways at Bernburg. Some 75 other German tracksites have also been included (Fig. 1), even though most of them have not yet provided chirotherid trackways, only single tracks; in most cases, this may simply be due to the limited sizes of the outcrops. Since the year 2000, there have been small surface exca- Figure 7. Map of the 2008 excavation, showing the Middle Triassic vations of several square metres extent at the Winterswijk chirotherid trackways in four track beds of the Karlstadt site in the Netherlands [63], at Borgholzhausen in Ger- Formation (lower Middle Muschelkalk, uppermost Pelso- nian) at Bernburg (modified after Diedrich [117]) (coll. many [64], and at Grossenlueder 1, also in Germany [65], LSDA). although at the latter site only a single archosaur footprint has been collected. Of all these localities, only the Bern- burg tracksite has yielded a large number of long, well cluding some perfectly preserved new material with partial preserved, chirotherid trackways, together with a large digital-pad and skin-structure details, collected in 2008 number of other tracks. and 2010 at Bernburg in Saxony-Anhalt, central Germany The largest systematic Middle Triassic track excavation in ([15]; Fig. 1). From these new Middle Triassic chirotherid Europe was conducted at Bernburg in June 2008 cover- trackways, in which the footprints show all kinds of vari- ing a surface area of approximately 2000 square metres ability and degrees of preservation, it will be demon- (Fig. 7), revealing one of the most important Middle Tri- strated that all of the known Middle Triassic archosaur assic tracksites in the world. This tracksite is also rich in footprint ichnospecies should be reduced to just the four horseshoe-crab trackways, which have been discussed in valid and distinct, Middle Triassic chirotherid track ichno- a separate publication [62]. The 20–40 m long chirotherid genera of decreasing size: Isochirotherium, Brachychi- trackways were first coloured in detail in the field and rotherium, Chirotherium and Synaptichnium. These be- then photographed. The photos were then equalized and ginnings of a ichnogenus-based revision also clarifies the compiled in AUTOCAD to create track maps from which all palaeobiogeography, distribution, and few biodiversity of the trackways were redrawn, although only the chirotherid the Middle Triassic archosaurs, from which a new interpre- reptile tracks are figured and discussed here. The exca- tation of global archosaur palaeobiology can be presented. vation of this site has demonstrated that only large areas Included in this interpretation are the newest discover- of outcrop or extensive surface excavations are likely to ies of horseshoe-crab Koupichnium tracks from Bernburg, yield the relatively rare, larger tracks. Two or three Chi- which are abundant, have long and different behavioural rotherium trackways (numbers 2, 9 and ?11), and three trackways, and their possible trackmakers, Tachypleus, Isochirotherium trackways (numbers 1, 12 and 13) were

501 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

In one case, after documentation a cast was made of one Isochirotherium trackway over a length of about 10 m. A large quantity of selected original slabs and cast material of 2008 is stored in the Amt fuer Denkmalpflege, Saxony- Anhalt, in Halle/Saale, together with a detailed report on the excavation, including all maps and data on the techniques employed [15]. The track area excavated in 2008 was completely destroyed in 2009, after documentation. From the second excavation in 2010, three of the trackways were sampled nearly completely (Fig. 8), with the smallest rescued with at least three following manus/pes sets. This material is deposited in the Museum Bernburg (= MB). To understand the Muschelkalk, and the general archosaur trackway preservation of coastal intertidal carbonate environments, experiments were made on the lower intertidal to lower sabkha of the Al Dabb’ iya peninsula, west of Abu Dhabi (UAE), in spring of 2010 (Fig. 5). In the experiment, a human walked over different sediment types, which were flooded at least twice during high tides. The footprints are ‘similar in shape and trackways’ to that of archosaurifom reptiles, and some of the reptiles were also partly similar in weight (Euparkeria and small Tici- nosuchus individuals were around 100 kg). Those track and trackway patterns were documented photographically to demonstrate differences in the shape and depth of impression into the substrate, and to explain the method of their nondestruction in intertidal zones. Figure 8. Map of the 2010 excavation, showing the Middle Triassic chirotherid trackways in track-bed 7 of the Karlstadt For- Finally, skeletons of possible archosaur trackmakers were mation (lower Middle Muschelkalk, uppermost Pelsonian) studied at the South African Museum of Natural History at Bernburg (coll. MB). (SAMNH) and the Paläontologische Institut der Univer- sutät und Museum Zürich (PIUMZ). Also examined was the skeletal material and reconstruction of the loricate documented on this surface in 2008, but isolated indi- archosauriform Batrachotomus in the Staatliche Samm- vidual tracks also were found in five separate excavated lungen Naturkunde in Stuttgart, with the skeletal pedal layers (Figs 2 and 7) of the 15 cm thick late Pelsonian anatomy of particular interest for comparison to the digital intertidal deposits, which are themselves a small part of formula of the chirotherid tracks. the complete biolaminite series of about 2.5 m thickness. Finally, there were four more trackways found in only one track layer, during a 10–30 m, smaller, second excavation 3. Geology in 2010 (Fig. 8), of which only the chirotherid material is discussed here. Those Chirotherium trackways, two of 3.1. Stratigraphy and sedimentology adult individuals and two of smaller sized animals, were mapped as between 7–15 m in length, although these can The Middle Triassic at Bernburg consists of a 110 m thick only be figured and discussed preliminarily here. At an- carbonate series, mainly limestones of the Jena Forma- other area of the quarry, a single Isochirotherium tackway tion [66], but also including carbonates and dolomites from was mapped at a length of about 8 m. the underlying Dorndorf Member and overlying Karlstadt In both excavation campaigns at Bernburg it was impossi- Formation (Fig. 2). ble to excavate all of the thinly bedded biolaminite layers. At Bernburg, the Aegean to Pelsonian carbonates of Hence, the excavation has only provided a preliminary in- the Jena Formation are approximately 90 m thick, and sight into this very large track locality, covering just a consist of shallow-marine deposits, mainly ‘Wellenkalke’ small part of the Karlstad Formation megatracksite that and shell- or ooid-rich limestones [66, 67]. In Bern- is being continually destroyed by quarrying activities [15]. burg, 19 sediment layers in the Jena Formation are de-

502 Cajus G. Diedrich

bris flows, mass flows and intraclast conglomerates, or ing different epicentres. The two slickenside trends seen slickensided beds with sinusoidal, deformed, veins. These in the seismic layers of the Karlstadt Formation at Bern- features are typical of seismically influenced sediments burg also correspond to these same orientations. Previous in shallow subtidal zones [68–72]. The positions of seis- studies [74, 79, 80] have postulated the two main epicen- mic epicentres during the deposition of the Jena Forma- tres as being in the Alps and in the Silesian Gate. tion have been previously inferred using slickenside ori- Originally grown, millimetre-thick, bedded biolaminites, entations in the shallow-submarine carbonates, which, are comparable to modern examples with known early diage- mainly south-southwest–north-northeast in the northern netic cementation (cf. [81]), have typical polygonal mud- Germanic Basin, and closer to east–west in the southern cracks, such as those found in the most important track- part of the basin [74]. bed, number 7 (Fig. 2C). In Bernburg, biolaminites vary in The Karlstadt Formation carbonates, assigned to the Ju- the carbonate particle sizes up to sand size, with carbon- dicarites zoldianus cephalopod biozone at the top of ate sand material consisting of thin, intercalated, ooid lay- the Pelsonian (middle Anisian), can be easily correlated ers, demonstrating a close relationship between the tidal throughout the entire Germanic Basin and up to the north- flats and the extensive, shallow-marine sand-bar deposits ern Tethys [15]. The chrono-, bio- and lithostratigraphic that formed in the central basin of the Lower Muschel- data and track-bed stratigraphy were additionally dated kalk in the Berlin area [82]. This carbonate sand was through the correlation of ceratite zones (cf. [15, 75]) transported during times of flood and was deposited on and by using absolute radiometric data from volcanic the tidal flats; however, such carbonate sand layers are ash layers in the northern Tethys Ocean, south of the absent further to the west, as this area was too far away Alps [76, 77]. The Karlstadt Formation possibly rep- from the shallow submarine sand bars at the time of de- resents deposition within a Milankovitch cycle of about position [15, 75]. 400 000 years (cf. carbonate cycles/sequences in [78]). Therefore, the Bernburg track-beds can be dated to an 3.2. Differences between track-beds 7 and 8 absolute age of about 243.5 – 243.9 Ma. (Figs 2 and 3). The Karlstadt Formation carbonate sediments are 5.15 m The geology of the most important of the 16 recognized of lagoonal platy dolomitic limestones interbedded with a track layers (although there are certainly more layers to 2.5 m thick, mud-cracked biolaminites and a thin, micritic be discovered), track-beds 7 and 8, are considered here, as to arenitic, series of carbonate layers (Fig. 2). both are different in their genesis (Figs 2D–E), and there- At the base of the Karlstadt Formation, three major bone- fore result in different track preservation and assemblages. bed layers occur within massive dolomitic and lagoon These tectonically and tide-influenced ‘mini-sequences’ platy limestones. Marine sauropterygian reptile (no ter- (1–5 cm thick; Fig. 2C) are repeated several times within restrial reptiles) and fish remains are found mainly con- the Karlstadt Formation biolaminite series, also explain- centrated in shallow subtidal channels with the monofau- ing the track abundancy. Tracks were generally preserved nistic, small, hypersaline-adapted bivalve, Neoschizodus in the lower intertidal regions. orbicularis. The track-bed 7 biolaminite (Figs 2D and 4) contains At its top, the biolaminite sequence is built of interbedded many and most of the reptile trackways, such as nearly autochthonous biolaminites and rapidly sedimented aren- all of the perfectly preserved Chirotherium trackways pre- ite layers. Seismically influenced arenite (tempestite or sented here. The trackways recognized on this biolami- tsunamite) layers have been recognized to be highly con- nite surface include several Rhynchosauroides, a large centrated (Fig. 2A). The 1–4 cm thick arenites have char- and a small Isochirotherium, and Chirotherium trackways, acteristic slickenside vein structures covering their sur- as well as a large number of subaquatic trackways of faces, such as those seen in track-bed 8 layer (Fig. 2D) large horseshoe-crabs [15, 62]. This track-rich surface in the form of parallel vein-lineations about 20 cm apart also has several long drift marks of uncertain origin, in- that cover the trackways on track-bed 8 (chirotherid track dicating palaeocurrents flowing in a north-northwesterly layer) and other layers (Fig. 2D). Sixteen arenite lay- direction. These are at right angles to the slickensides, ers have similar slickensided surfaces: these structures again suggesting formation due to submarine earthquakes start at the surface but do not extend far into the under- or storm events, and their resulting waves perpendicu- lying biolaminites or tempestites. Between track-bed 8 lar to the coastline. As the tide ebbed, the biolami- and track-bed 7, the trend of these structures changes nite layer of track-bed 7 would have dried out and ter- rapidly from north-northeast–south-southwest to nearly restrial archosauromorph reptiles such as Macrocnemus east–west (Fig. 2A). Structures vary between these two (=Rhynchosauroides trackmaker) or archosauriforms like main directions several times in the upper section, reflect- Arizonasaurus or Ticinosuchus (=Isochirotherium / Chi-

503 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

rotherium trackmakers, respectively) would have ventured whole of the Jena Formation at Bernburg, although this out onto the broad tidal area [15]. Track-bed 7 can there- changed towards the end of this formation’s deposition. fore be seen as having preserved a full ebb-tide cycle The earliest Middle Triassic chirotherid tracks (extremely (possibly even several) as a result of having been rapidly rare records of Synaptichnium and Isochirotherium) occur covered by the flood that deposited track-bed 8. Both in the Winterswijk Formation [84] Figs 3 and 6E, which layers were then subjected to an earthquake shock that consists of intertidal biolaminites, sabkha dolomites, and resulted in the formation of vein structures that even pen- platy to massive lagoonal dolomites. The shallow, fully etrated the mud-cracked surface of track horizon 8 when marine ‘Wellenkalke facies’, typical of the eastern Os- this was already semi-consolidated by early diagenetic nabrücker and Jena Formations, is absent in Winterswijk, cementation. indicating a condensed section in the far western part of The arenite of track horizon 8 (Figs 2E and 4) is only 2 cm the basin forming thick intertidal deposits rich in tracks thick, and thus quickly covered the underlying track-bed 7, and bone-beds [15, 75]. These layers can be correlated which forms the top of a 2 cm thick biolaminite package. using stratigraphic and sedimentary sequences, includ- Only the tracks of large chirotherids (Isochirotherium and ing tsunamite or tempestite layers, which are composed Chirotherium) are found on track-bed 8, all faintly pre- of bone-bed layers on the coastlines (Winterswijk), and served but also in extramorphological preservation in soft bioclastics with 5–25 cm thick bone-bearing limestone substrate. After the documentation of the 2008 excava- beds (e.g. Upper Basal Conglomerate Bed, Upper Oolithic tion surface, a crossing small and giant chirotherid track- Bed; Fig. 3) in the Osnabrück and Jena Formations [69]. way was observed where both individuals overstepped the The two trackway records in the Aegean-aged intertidal slickenside structures. biolaminites of Winterswijk [84], one of Isochirotherium (described as ‘Coelurosaurichnus’ in Demathieu and Os- 3.3. Intertidal carbonate flat megatracksites terink [61]) and one identified here as Synaptichnium (described previously as ‘Brachychirotherium’; Demathieu and their age-dating and Osterink [61]), can be correlated with the first extensive megatracksite (track-beds I and II; Aegean–Bithynian The chirotherid tracks of the Middle Triassic Germanic boundary). This megatracksite resulted from a marine Basin have all been found in marine intertidal-flat to lower ingression that formed the first extensive intertidal and sabkha carbonate deposits, mainly biolaminites, from the sabkha environments around the margins of the Germanic coastal zones [15]. They formed extensive megatracksites Basin [15]. From similar layers at the Grossenlueder 1 site (defined as isochronous, wide, extended, track-rich layers (Fig. 1), a single chirotherid track was found [65], which or surfaces) at the Upper Bunter – Lower Muschelkalk can be now identified as an Isochirotherium manus/pes (track-beds I–II), Lower–Middle Muschelkalk (track-beds set. XVII–XVIII), and Middle–Upper Muschelkalk (track-beds XXI–XXII) boundaries (numbering from [15] Fig. 3). Glob- Although there are many track layers (track-beds III– ally, the extension and type of intertidal environments XVI) in the Osnabrück and Jena Formations (Bithynian– seen in the Germanic Basin are quite unique for the Tri- Pelsonian; Fig. 3), no chirotherid tracks have yet been assic record. found from this interval in Germany. Their eventual discovery is to be expected; however, outcrops of this age are In the eastern part of the Germanic Basin, a larger number generally too small to get larger tracks and trackways, es- of track-beds are present in the Aegean to Illyrian (about pecially when surfaces are not well exposed. 22 track-beds; [15, 71] Fig. 3) as a result of a carbonate ramp that shallowed towards the west. Towards the west, The next (and best) Middle Triassic record of chirotherid the higher number of track-beds present in the Aegean tracks is from the base of the basal Middle Muschelkalk to Illyrian (about 31) is also a result of of this carbon- Karlstadt Formation (uppermost Pelsonian) at Bernburg ate ramp in the intercontinental Germanic Basin [15]. On (15; Fig. 3), in which the chirotherid trackways described the shallowest part of the ramp, near the western Ger- here were discovered within another megatracksite that manic Basin margin [83], the typical subtidal carbonates extends across the Germanic Basin (track-beds XVIII– — the ‘Wellenkalke’ of the Jena Formation — developed XIX). The tracks were found in the Judicarites zoldianus into the Osnabrück Formation with a far higher propor- cephalopod biozone (middle Anisian), which can easily be tion of intertidal biolaminite and sabkha sediments, which correlated throughout the entire Germanic Basin and all itself finally disappears in the far west, such as at Winter- the way to the northern Tethys [15, 72, 73, 75]. swijk in the Netherlands [15, 84] Fig. 3. In contrast, the The final record of a chirotherid track (with a well- Wellenkalk dominates the continuous, shallow-marine and preserved manus, but poorly preserved pes) occurs full-marine sediments that were deposited throughout the within the uppermost Middle Muschelkalk (track-bed XXI,

504 Cajus G. Diedrich

Diemel Formation; middle Illyrian; Fig.3) of Germete, been much farther south than it is today. Low-energy where another extensive megatracksite was developed in tides in the shallow, intracratonic, Germanic Basin would the central Germanic Basin [15]. have had almost no impact on tracks, as was demonstrated During the Upper Muschelkalk, the morphology of the both in experiments with modern lizard trackways on the Germanic Basin changed completely, with the develop- Arabian Gulf coast of the UAE (cf. [86]), and in the exper- ment of steeper ramps and consequently an absence of iments conducted in coastal intertidal regions of the UAE shallow, intertidal biolaminite-dominated zones [15]. This in 2010. change in environment explains the absence of any footprint records in the Upper Illyrian, the Fassanian, and even the Lower Longobardian in this basin, with the ex- Recent experiments with present-day Iguana lizards, both ception of tracks found along the Bavarian–Hessian De- in natural carbonate intertidal environments [86] and in pression and on the connected Burgundian Gate coast- laboratory situations on clay and sandy substrates [87, lines. 88], have demonstrated the enormous variability in trackway preservation, with trackway variations for the same trackmaker influenced mainly by the locomotion type and 4. Palaeontology speed, the sediment type, the sediment’s moisture content, and the degree of cementation. 4.1. Track preservation within intertidal zones and actualistic experiments In 2010, new experiments, again conducted in the lower Variations in track preservation can be seen in the ar- intertidal to lower sabkha environments of the Al Dabb’iya chosaur tracks from Bernburg, particularly from the bio- Peninsula west of Abu Dhabi (UAE), were made along laminites of track-bed 7 and arenite layers in track-bed 8 several kilometres of the extended intertidal zone (Fig. 5) (Fig.4). During the tidal ebb and drying out of the car- with the hope of clarifying the existing confusion in bonates, chirotherid trackmakers walked over the tidal-flat chirotherid track ichnotaxonomy (see partial compilation surfaces. Large archosauriforms weighing around 100 kg in [16]). The author himself walked over different sedi- (e.g. Euparkeria) and up to 300 kg (e.g. Ticinosuchus) ment types, covering the lower, middle, and upper inter- did not sink more than a 2 cm into the biolaminite or tidal and lower sabkha environments (Figs 5A–F), all of arenite sediment layers, although the fine details of these which were smoothly flooded during the next few days chirotherid tracks can often be seen in the biolaminites during the high storm-tides (Fig. 5G) that happen regu- (Fig.4). Therefore, the stabilization of these biolaminite larly at that time of year. The footprints are preserved carbonates must have been an early sedimentary process, differently in all four environments, including quite differ- and they must have become well-stabilized very rapidly, ent surface structures. In the lowermost lower intertidal in contrast to the rapidly deposited, 2–4 cm thick, muddy zone on carbonate mush, the footprints were only about arenites in which archosaurs left only faint impressions 3 cm deep with toe details; after flooding, mangrove tree (Fig.4). In many cases, the regular tidal flooding of these leaves were washed into the track depressions (Fig. 5E). carbonate intertidal-flats did little damage to the tracks, This is similar to the situation at Bernburg, where large and some prints even show the detail of skin impressions. Triassic archosaur footprints in track-bed 8 acted as traps Several chirotherid tracks on biolaminites reveal details for fish and reptile bones in the lower to middle inter- of the digital pads, and also claw marks or scratches, fea- tidal zone. In the modern Abu Dhabi environments, the tures previously unknown from chirotherid fossils. dark-brown middle intertidal zone is polygonally cracked The intertidal sediments are best for detailed track preser- and covered by leathery biomats (Fig. 5H), although the vation due to the rapid covering of the carbonate surfaces, underlying sediments are still a carbonate mush (Fig. 5H) rapid cementation due to arid conditions with high tem- into which the author sank about 10–15 cm (Fig. 5F). To- peratures of up to 50◦C, and the catalytic cementation wards the margin of the uppermost upper intertidal zone, processes seen in cyanobacterial/algal mats (e.g. [71, 85]. the polygons become smaller and more wrinkled; here the Preservation would have also been assisted by generally tracks were only about 5–10 cm deep (Fig. 5C) due to the smooth tidal flooding in the shallow intertidal zone, as sedimentary change to the lower sabkha. In this latter seen in the modern southern Arabian Gulf, where high zone where gypsum crystals are the dominant sediment tides are only expected to reach up to 50 cm; this is mineral, faint tracks are imprinted about 5 cm deep, and particularly supported by the Middle Triassic palaeogeo- a thin but hard halite crust also protects the tracks from graphic position of the Germanic Basin, which would have destruction (Fig. 5B).

505 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

4.2. Systematics (there are both large and small Chirotherium tracks; Fig. 13). Therefore, there are no clear arguments that The confusing and inconsistent chirotherid ichnotaxonomy Isochirotherium represents larger Chirotherium tracks remains a problem, with even recent revisions not revising formed by the same track producer. The characteristics of track species and genera conscientiously (cf. [43]). This track size, pedal rotation axis, polygon shape, differences confusion is mainly a result of the lack of international in the digit lengths, and the close or wide position of comparisons between tracks of different preservations, in the digits (partly dependant on the substrate and the different facies types, and with different types and rates animal’s speed) are figured herein, all of which allow of locomotion, also taking into account the variations in only a few of the Aegean to Illyrian ichnogenera to be different authors’ outline drawings and interpretations (cf. distinguished (Fig. 6). compiled redrawings in [16]). Therefore, in the future, a revision of this track ichnotaxonomy will be required for Chirotherium Kaup, 1835 all Triassic chirotherid track ‘ichnospecies’. Chirotherium barthii Kaup, 1835 This paper aims to start the critical revision and reduc- Figs 9–12 and 15D–J; Table 1 tion to a few ichnospecies. This revision should be made using detailed intertidal (in some cases also clay preser- Material: The 2008 excavation delivered a single track- vations) track preservations, which allow the redefinition way from a large individual (Fig.7, Trackways 2a of, at least, the ichnogenera Synaptichnium, Chirotherium, and 2b; Figs 9 and 13G) in track-bed 7, consisting of and Isochirotherium. The tracks preserved in the inter- 17 manus/pes sets (Table 1; coll. LSDA) printed to a tidal environments have more preserved detail, and in- medium depth, with some sliding and interruption or de- clude tracks found in the terrestrial northern American struction of details. Another trackway, also from track- redbeds of the Moenkopi Formation/Group [43], and the bed 7, is from a smaller individual (Figs 7 and 13H, Track- Upper Buntsandstein Roet Formation of Europe (e.g. 45, way 3), and consists of shallow imprints of 64 manus/pes 89). With the exception of Synaptichnium, which is named sets (Table 1 only partly saved, coll. LSDA). A third track- ‘Protochirotherium’ when it has pedal-pad skin details way, from track-bed V, might also belong to this track type, preserved (see 90; Fig. 6E), other preservation types have with deep imprints consisting of 48 manus/pes sets and received a confusing ichnotaxonomy based on a variety several other prints occurring within a trampled area (Ta- of different features (e.g. 90; 91, fig. 2; 43). For exam- ble 1; Figs 16F–G and 13D; mostly saved, coll. LSDA). In ple, in many illustrations and tracks, the digit V varies 2010, four more trackways were discovered on track-bed 7 greatly and digit IV is sometimes longer or even equal in a new excavation area northwest of the first excavation in length to digit III, and as such, these features can- surface (Figs 8, 10–12, and 13E–F,I–J; Table 1). These not be used for ichnotaxonomic clarification. Also, the tracks all fall within the range of sizes and proportions digit lengths are demonstrated here to vary between Chi- seen for medium to large sized individuals. rotherium and Isochirotherium mainly due to rotation dur- Description: Manus and pes are plantigrad, with a ratio of ing locomotion, and the curving of the legs (which may also 1:2.5 in manus/pes proportions (Fig. 6). The footprint sizes result in more curved digit imprints), causing their feet to depend on the age of the individual, ranging in Bernburg imprint at different angles on the substrate. Furthermore, material from the smallest manus/pes lengths of 6/14 cm, the tracks described in those papers seem to reflect vari- to the largest lengths of 8/19 cm (Table 1). Typically, the ations in preservation type and locomotion. As a result, claw marks are clearly visible for digits I to IV in the pes, Klein and Lucas [43] agreed to synomymize Synaptich- and only for digits II and III in the manus. Whereas in nium (less deeply printed tracks without preserved der- the pes all digits are mostly printed, in the manus prints, mal skin structures) with ‘Protochirotherium’ sensu Fichter the outer digits, especially digit I, are generally missing and Kunz [90] (tracks with digital pad and skin preserva- in tracks recorded on biolaminites. tion). The three mapped trackways from the 2008 excavations After a study of the holotypes, including material from all have between 17 (trackway 1, track-bed 7) and 64 over the world, only two ichnogenera — Chirotherium and (trackway 2, track-bed 7) manus/pes sets (Table 1). In Isochirotherium — and their type ichnospecies are con- trackway 3, several tracks have fish remains (mainly sidered valid, with the range of these taxa extending from cranial elements) washed into the depressions. The the Olenekian to basal Anisian. The Chirotherium and manus/pes distances range between 2–3.5 cm, but this, Isochirotherium tracks do not reflect ‘ontogenetic stages’ and the 44–49 cm width of the pes–pes distances, of the same trackmaker where pes bones have positive are dependent on the individual’s speed. The pes– and negative allometric growth, and in Chirotherium at pes distance is also dependant on the trackmaker’s least different aged animals seem to have been present size, as are the 16–18 cm wide trackway width, the

506 Cajus G. Diedrich

Figure 9. Chirotherium barthii Kaup, 1835 trackway 2 (2008 excavation; cf. Fig. 7), preserved on polygonally mud-cracked biolaminite (trackbed 7), showing the tracks of a large individual trackmaker. The trackway consists of two parts, A and B, and is interrupted over a length of about 8 m (coll. LSDA); all tracks are positive imprints. The animal walked at the end 0◦ towards north. A. Trackway map; B. trackway part, in dorsal view; C-N. selected single manus/pes sets of the trackway, demonstrating the strong variability in shape and manus presence/absence, resulting from slipping and partial bipedal walking.

507 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

Figure 10. Chirotherium barthii Kaup, 1835 trackway II (2010 excavation; cf. Fig. 8), preserved on polygonally mud-cracked biolaminite (trackbed 7), showing the tracks of a large individual trackmaker. The trackway consists of two parts, A and B, and is interrupted over a length of about 2 m (coll. MB); all tracks are positive imprints. The animal walked towards north. A. Trackway map; B. trackway photographed diagonally walking towards viewer; C-D. trackway parts, in dorsal view; E–L. selected single manus/pes sets of the trackway.

508 Cajus G. Diedrich

Figure 11. Chirotherium barthii Kaup, 1835 trackway I (2010 excavation; cf. Fig. 8), preserved on polygonally mud-cracked biolaminite (trackbed 7), showing a large individual trackmaker. The trackway consists of two parts, A and B, with 23 manus/pes sets interrupted over a length of about 8 m (coll. MB). The animal walked 130◦ towards the southeast. A. trackway map; B. trackway photographed diagonally walking away from viewer; C–D. trackway parts, in dorsal view; E–L. selected single manus/pes sets of the trackway.

509 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

Figure 12. Chirotherium barthii Kaup, 1835 trackway III (2010 excavation; cf. Fig. 8), preserved on polygonally mud-cracked biolaminite, indicating a medium-sized individual trackmaker. The trackway consists of 22 tracks (including one feeding place on a horseshoe crab, and 6–7 destroyed tracks) and 18 manus/pes sets being interrupted over a length of about 1 m (coll. MB). The animal walked towards the north-northeast. A. trackway map; B. trackway part, in dorsal view; C. redrawn ‘feeding’ trace (archosauriform feeding on a horseshoe crab) redrawn; D–G. selected single manus/pes sets of the trackway.

510 Cajus G. Diedrich

Figure 13. Isochirotherium herculis (A–C) and Chirotherium barthii (D–J) archosauriform trackways of different sizes and preservation types from the Middle Triassic (upper Pelsonian) tracksite at Bernburg (Germany). Footprint shape types reflect differences in substrates (arenite/biolaminite), ecological behaviours  such as trampling, feeding, and slipping on microbial intertidal mats  rates of motion, and individual ages (sizes); finally, scratch marks from claws are important in understanding limb locomotion.

159–167◦ pace angulation, and 86–99 cm stride (Table 1). Discussion: Most of the imprints are illustrated herein be- The four trackways discovered in 2010 fall into the same cause they contain preservational details rare in Middle range of measurements (Table 1). Triassic material, both in Europe and globally [43]. The imprints show a high variability in shape (Fig. 4), and, in several cases, show evidence of pes slippage on the When completely imprinted, the typical features of the biolaminites (Fig. 9), suggesting that these surfaces were ichnogenus Chirotherium (Fig. 6) are the smaller foot ro- covered with slippery biomat on which the Chirotherium tation, and the increasing lengths of digits I to III, the II trackmaker was insecure in its walking or running. being the longest, with digit IV longer than digit I (polyg- The tracks from Bernburg were compared with original onal analyses). The digital pad of V can be sometimes material on the track slab from Hildburghausen (Fig. 6D), subdivided in two parts (Fig. 6B). The digital pads in- including material which had not been illustrated or listed dicate that the trackmaker has a phalange formula, from by Haubold [13], comprising one slab in the Bernburg digit I to V, of 3, 3, 4, 5, 4.

511 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

Table 1. Comparison of features of the Middle Triassic (Pelsonian) Isochirotherium and Chirotherium trackways from Bernburg (Germany).

Manus/pes Number of Manus Pes Manus/pes Pes/pes Trackway Pace Individual set manus/pes length length distance distance width (cm) angulation size and sets (cm) (cm) (cm) (cm) speed Isochirotherium 34 11.5 34 1 84 37.5 172.5◦ Very large, (trackway 12) fast moving: Fig. 13A Isochirotherium 47 10 22 0.5 64 30 160◦ Large, (trackway 1) slow moving: Fig. 13B Isochirotherium 12 8 24 3 65 32 162.5◦ Large, trackway slow moving: Fig. 13C Manus/pes Number of Manus Pes Manus/pes Pes/pes Trackway Pace Individual set manus/pes length length distance distance width (cm) angulation size and sets (cm) (cm) (cm) (cm) speed Chirotherium 48 8 18 3.5 46 18 167◦ Large, (trackway 11) slow moving: Fig. 13D Chirotherium 27 7.5 17 2 48 24.5 165◦ Large, (trackway II) slow moving: Fig. 13E Chirotherium 12 8 18.5 2 52 17 169◦ Large, (trackway I) fast moving: Fig. 13F Chirotherium 17 7 17 2 49 16 160◦ Large, (trackway 2) fast moving: Fig. 13G Chirotherium 64 - 14 - 44 16 159◦ Small, (trackway 3) slow moving: Fig. 13H Chirotherium 15 8 18.5 3 50 23 163◦ Small, (trackway III) fast moving: Fig. 13I Chirotherium 29 - 13.5 - 45 18 158◦ Small, (trackway IV) slow moving: Fig. 13J

Museum and another small slab in the Stadtmuseum at or an argument for a different ichnospecies. The figures Saalfeld. Both slabs, as well as the main material from in of C. sickleri in Klein and Lucas [43] resemble a set the Museum für Naturkunde der Humboldt-Universität, of different Synaptichnium and Chirotherium tracks in Berlin (coll. MB no. IVC 30, 31, 32–34, 35; Fig. 6D), ‘badly preserved’ material only; therefore, C. sickleri is indicate that the Middle Triassic Bernburg material not considered a separate ichnotaxon. In contrast, Middle can be easily attributed to Chirotherium barthii [3] Triassic tracks of C. barthii Kaup, 1835, illustrated (Fig. 6), with no recognizable differences other than from the Burgundian Gateway, on the eastern border of in the state of preservation, which would logically be the Massif Central in France [92], should be attributed different between fluvial terrestrial clays and marine instead to Isochirotherium. This track assemblage from tidal-flat biolaminites. It has also been noted that the France appears to be from the upper Middle Muschelkalk C. sickleri Kaup, 1835 tracks from Hildburghausen [19] (Longobardian), which would make these well-preserved appear simply to represent a smaller individual, in which C. barthii tracks the youngest known Chirotherium tracks the perceived differences in foot rotation [13] may reflect in Europe (Fig. 18). Chirotherium has also been recorded a smaller body-size or individual anatomical differences, in the Spanish (Barcelona, Cuenca, Guadalajara, Mal- but is not considered indicative of a different trackmaker lorca, Teruel, and Zaragoza) Anisian (‘Buntsandstein and

512 Cajus G. Diedrich

Figure 15. A. Digital 3D animation reconstruction of the suchian archosauriform Ticinosuchus, seen in dorsal, lateral, and Figure 14. A. Digital 3D animation reconstruction of the poposauroid ventral views. B. Trackways of Chirotherium (trackway archosauriform Arizonasaurus, seen in dorsal, lat- 2, cf. Figs 7 and 9) from Bernburg, used for anatomical eral, and ventral views. B. Trackways of Isochi- locomotion reconstruction (Trackway ©PaleoLogic; Ani- rotherium (trackway 1; cf. Fig. 14) from Bernburg, mation ©Haas). used for anatomical locomotion reconstruction (Track- way ©PaleoLogic; Animation ©Haas).

it was possible only to save two footprint sets from this Muchelkalk’; e.g. [45]) with C. barthii, which underlines strongly weathered layer (Fig. 6A). the time-range presented herein (Fig. 18). Finally, C. Description: Manus and pes are plantigrad, with a ratio lulli (Baird, 1954) from the upper Lower to lower Middle of 1:3 in the manus/pes proportions (Fig. 6; Table 1). The Triassic Moenkopi Formation of northern America also footprint sizes can be given only for the Bernburg track- appears to be a differently preserved track of the C. way, which seems to be from an adult animal ranging from barthii type. Therefore, there are several chirotherid of 8–10/22–24 cm in manus/pes lengths (Table 1). Claw trackways that should be attributed to the C. barthii marks are also clearly visible in digits I–IV in the pes, ichnospecies, though others should instead be attributed and in digits II and III in the manus. Whereas in the pes to Isochirotherium. In many cases, incomplete preser- digits I–IV are well printed and digit V is variable in its vation or poorly imprinted forms do not allow a clear shape and size, the manus is generally printed less com- ichnospecies identification. pletely and deeply, with digit I and V missing quite often from the tracks recorded on biolaminites. The preservation Isochirotherium Haubold, 1971 of coarse, granular skin impressions in seven manus/pes Isochirotherium herculis (Egerton, 1838) sets of the long trackway (trackway 1, Figs 7 and 13B) Figs 13–14 and 15A–C; Table 1 is unique in the global track record; the best material is Material: One trackway from the 2008 excavation, pre- figured here in detail (Fig. 17). served on biolaminites of track-bed 1 (Trackway 1, Fig. 7), The two mapped trackways from the 2008 and 2010 exca- consists of 47 manus/pes sets mapped over a 40 m length vations have between 12 (trackway 2, track-bed 1) and 47 (Table 1. This trackway shows considerable variation in (trackway 1, track-bed 7) manus/pes sets (Table 1). The its preservation, but still reveals many details such as manus/pes distances range between 0.5–2 cm; the width dermal structures and scratch marks from claws (Figs 17 of the pes–pes distances is about 64 cm; the trackway and 13B). A second trackway from the biolaminites of width is between 30 and 33 cm; the pace angulation is track-bed 4 preserves fewer details; three manus/pes sets 160◦; and the stride of the very narrow gauged trackways have been mapped on the track-bed surface, and a further is 122–126 cm, with large foot rotations such as for Chi- six have been preserved. In the 2010 excavation, another rotherium. trackway from track-bed 4 with 12 manus/pes sets was Where completely printed, the typical features of the documented and measured (Fig. 13C; Table 1), although ichnogenus Isochirotherium (Fig. 6) includes a larger max-

513 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

sandstones of Storeton and Tarporley in Cheshire, Eng- land [21, 30, 31]. These Isochirotherium forms have also been found in the Upper Bunter (Roet Formation) of Hild- burghausen (Fig. 6C) and other German localities (Fig. 1), and have been described from different sized individuals with different types of preservation. There are similar, large isochirotherid-like footprints elsewhere in the Up- per Bunter [16, 42], but the small number of trackways and their poorly preserved shapes, especially in terrestrial beds, do not allow clear attribution to the Isochirotherium ichnogenus, or even to the same ichnospecies or trackmaker. I. archaeum from France [93], together with I. je- nense, I. hessbergense, and I. soergeli from Germany [19], and I. marshalli or I. coltoni from the Moenkopi Forma- tion/Group of Northern America [43], all appear to represent trackways of younger individuals or different types of track preservation. C. moquinense, I. rex, I. marshalli, and others (e.g. [22, 43], as well as the small I. coltoni forms, seem to represent the largest and smallest individual tracks, respectively, left by one and the same trackmaker species, and all have been found together with the smaller questionable ichnospecies in the Middle Trias- sic terrestrial sandstones in France [28]. However, these tracks all fall within the range of shapes described herein for the material from Bernburg. A similar situation exists with the late Middle Triassic Isochirotherium ichnospecies Figure 16. Intersecting Isochirotherium herculis and Chirotherium I. felenci, I. coureli, I. combelei, I. circademathieui (cf. [24], barthii trackways (cf. trackway 12; Fig. 7) preserved and I. inferni [35], all of which fall within the range of on slickensided arenite (trackbed 8) at Bernburg (Ger- many), also with marine shark and marine sauroptery- preservation and individual sizes of Middle Triassic isochi- gian reptile remains flooded into the large track depres- rotherid tracks. I coureli and other Isochirotherium tracks sions. All positive imprints (surfaces mapped are completely removed, coll. LSDA). A. Isochirotherium track- have also been recorded from the Anisian (‘Buntsand- way map; B–C. Isochirotherium trackway, photographed stein’) of Spain (Barcelona and Guadalajara; e.g. [45]). diagonally; D–E. selected single manus/pes sets of the Isochirotherium trackway; F–G. Chirotherium trackway All have been established on the basis of a small number photographed diagonally. of tracks, mainly using single manus/pes sets, often cho- sen selectively, and often with incomplete and not clearly printed tracks. However, the Isochirotherium footprints described from the Middle Triassic of Italy also demon- imum pes length up to 35 cm, with smaller manus prints strate a full range of different sizes [34, 35]. There are (ratio 1/3) in comparison to Chirotherium (which has pes also many other Chirotherium and Isochirotherium foot- imprints not longer than 19 cm). Other features different prints that appear to have been attributed to the oldest from Chirotherium include a stronger foot rotation, and the nomenclature of I. herculis [21], although most of these development of similarly lengthed digits I and IV (polygo- records are too incomplete, too unclear in their stratig- nal analyses). Isochirotherium resembles Chirotherium in raphy, or too incompletely preserved to warrant further the structure of the granular dermal pads (Figs 17I–J). The discussion. However, the ichnospecies definition of I. her- digital pad V of Isochirotherium is very variable, and when culis has been extended here, especially taking into ac- complete this pad lies further behind digits I to IV and is count data on skin impression structures, individual size more bean-shaped, resulting in a higher foot-rotation an- variations, and other variability demonstrated within long gle (Fig. 6B). The digital pads (Figs 17D–I) suggest a trackways from Bernburg. trackmaker with a phalange formula, from digit I to V, of 2, 3, 4, 5, 4. Discussion: I. herculis [21], to which the Bernburg material is attributed, is a well-known ichnospecies, first described from the Lower to Middle Anisian (Middle Triassic)

514 Cajus G. Diedrich

Figure 17. Isochirotherium herculis Egerton, 1838 trackway (cf. trackway 1; Fig. 7), preserved on polygonally mud-cracked biolaminites (trackbed 4) at Bernburg (Germany). All positive imprints (surfaces mapped are partly removed as originals or are partly casted, coll. LSDA). A. Trackway map; B. trackway photographed diagonally, animal walking towards viewer; C. trackway part, in dorsal view; D–J. selected single manus/pes sets of the trackway, demonstrating strong variability in shape and the presence/absence of dermal structures or claw marks

515 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

5. Discussion intertidal-flat biolaminites in Europe. The facies at Win- terswijk is proximal intertidal to lagoonal, with little ma- 5.1. Archosaur track record during the Trias- rine influence. Tracks of this type have also been reported in other European localities, with well-dated ma- sic terial only known from the Lower Triassic, such as from Spain (e.g. [45] or England (e.g. [31]). Discussions on the stratigraphic occurrence and ‘evolutionary trend with toe reduction’ of archosaurs, presented The first occurrences of Chirotherium trackways are from by Klein and Haubold [46], appears to be incorrect in a the Upper Bunter (Aegean; Roet Formation in Europe, and number of features, with gaps left in the Middle Triassic Moenkopi Formation/Group in North America); the new, footprint record through the omission of many important well-dated finds from Bernburg extend this ichnogenus’ discoveries and incompletely imprinted footprints, and the range into the Middle Muschelkalk (Illyrian; Karlstadt exclusion of variations that were simply due to differences Formation). The youngest records are probably from the in preservation or facies. These variations, in particular, uppermost Upper Muschelkalk and Lower Keuper of the are key to understanding the facies and environmental de- Middle Triassic (upper Fassanian/lower Longobardian) in pendency of chirotherid tracks in the Germanic Basin, at the eastern Massif Central, although, as mentioned pre- least for the late Early to late Middle Triassic (Figs 18 viously, the dating of these discoveries is based solely on and 19). their ‘track assemblages’ [28] and not yet on any bios- Due to an improved understanding of the palaeogeo- tratigraphic or chronostratigraphic evidence. However, graphic development of the Germanic Basin, including the the presence of both dinosauriform and archosaur tracks formation of important and extensive coastal intertidal and could indicate an Upper Muschelkalk or Lower Keuper sabhka zones [15, 62], a clear relationship can now be seen (Longobardian–Carnian) age, in which case this ‘assem- between the distribution of these particular facies types blage’ would appear to be much younger than Bernburg and that of Synaptichnium, Chirotherium, Isochirotherium, and the other German sites. Therefore, current knowl- and Brachychirotherium trackways. In these coastal envi- edge of this ichnospecies indicates a stratigraphic range of ronments, which existed for about 10 m.y., these tracks and 247.2 – 238.0 Ma (chronostratigraphy after [77]; Fig. 19). their most probable trackmakers can be correlated hypo- Isochirotherium tracks range in age from the Upper Bunter thetically to the known skeletal record, and particularly (upper Olenekian to Aegean; Roet Formation in Europe, pedal anatomy (Fig. 18). The majority of these trackways and Moenkopi Formation/Group in North America: [43, 46] have been found in Europe, partly due to the high levels through to the lower Ladinian [15]. The largest known of research and collecting activities in this continent, but Middle Triassic chirotherid track form, ‘Prosauropodichnus also as a result of the quite unique intracratonic setting of bernburgensis’ from Bernburg (15; Figs 16A–E), can be the Germanic Basin, which in the Middle Triassic formed seen to be equivalent to large ‘I. rex/I. herculis’ tracks also a bridge of a lowland and intertidal-flat environments be- found in North American sandstones of a similar age [22]. tween the two supercontinents of Laurasia and Gondwana Another giant trackway seen on the slickensided arenite (Fig. 19). of Bernburg’s track-bed 8, consisting of 34 manus/pes sets The stratigraphic range of Synaptichnium tracks was cor- (Trackway 12, Fig. 7) with pes lengths of 35 cm, falls into rectly presented by Klein and Haubold [46] by taking into the known range of the aforementioned chirotherid Isochi- account tracks from the Lower Triassic, even though these rotherium track types, but was also first described as ‘P. tracks from the Upper Bunter (Roet Formation, Aegean) bernburgensis’[15]. This trackform is thought to have of Hildburghausen had previously been incorrectly deter- been produced by the largest poposauroid archosauri- mined as ‘Brachychirotherium’ by Haubold [19, 26]. Some form Arizonasaurus [75] Fig. 14. At Grossenlueder, in of these tracks are in fact attributable to Synaptichnium the Hess Province of Germany, a single Isochirotherium or its younger synonym ‘Protochirotherium’ sensu the def- track (identified merely as ‘chirotherid track’; [65]) was inition presented by Fichter and Kunz [90], whereas oth- discovered in biolaminites of a similar age (again from ers just represent different track preservations. With their the Aegean Roet Formation). Another rare chirotherid large manus prints and manus/pes ratio of 1:1.75, the track found in sabkha biolaminites at Karsdorf, central tracks assigned to B. parvaparvum (=B. preparvum) from Germany, from the Aegean–Bithynian boundary [94], can a single trackway found in the latest Aegean at Winter- also be attributed to Isochirotherium. The tracks at Bern- swijk in the Netherlands (Fig. 6E) are reidentified here burg fall within the accepted age-range for this ichno- as Synaptichnium tracks (Fig. 6); although these have genus (Fig. 18), having absolute ages between 247.5 and been clearly dated as latest Aegean [84], these are the 243.8 Ma (cf. chronostratigraphy: [76]). Other large Isochi- only presently known records of Synaptichnium tracks on rotherium tracks have been reported from the ‘Middle Tri-

516 Cajus G. Diedrich

Figure 18. A. Euparkeria possibly produces Synaptichnium tracks (skeleton in the Natural History Museum South Africa). B. In the terrest red beds Chirotherium footprints dominate, possibly left mainly by the Suchia Ticinosuruchus (skeleton in the Palaeontological Museum of the University Zurich). The changing palaeogeography and habitats of Archosauriforms in the Middle Triassic Germanic Basin, between Aegean and Longobardian times (“Lower to Upper Muschelkalk”), within a time frame of about 7.8 million years (246.8 to 239.0 Ma). With the development of the marine Germanic Basin during the Lower to Middle Muschelkalk, and seismic peaks around the Pelsonian/Illyrian boundary, extensive intertidal zone habitats were formed: abundant Isochirotherium tracks were left on these coastal zones, possibly by the Poposauroidea Arizonasaurus. C. With the regression and fresh water influenced lagoons at the end of the Upper Muschelkalk and into the Lower Keuper, the Loricata Batrachotomus may have been the Brachychirotherium trackmaker (see Archosauriform pedal anatomy in E). 517 Middle Triassic chirotherid trackways on earthquake influenced intertidal limulid reproduction flats of the European Germanic Basin coasts

Figure 19. A. The uppermost Middle Triassic Archosauriform predator Batrachotomus kupferzellensis from southwest Germany (coll. SMNS). B. Stratigraphic occurrence of the chirotherid track ichnogenera, in central Europe and their possible Archosauriform trackmakers (chronostratigraphy after Kozur and Bachmann, [77]). C. Holotype skeleton of Euparkeria capensis from the Lower Triassic of South Africa (coll. SAMNH; photo R. Smith). D. Holotype skeleton of Ticinosuchus ferox from the Middle Triassic of Switzerland (coll. PIUMZ; photo H. Furrer). E. Known pedal skeletons and track types discussed herein in comparisons.

518 Cajus G. Diedrich

crease in the size of Brachychirotherium tracks throughout the Triassic [25], such as into the Longobardian (basal Upper Triassic) material of B. gallicum (uppermost Upper Muschelkalk to Lower Keuper; Ladinian) from Spain [45]. This assertion will need to be validated using a greater amount of material from pre-Longobardian times, where they are well recorded in the southern Germanic Basin, Burgundian Gate, and Spain (Jaén; e.g. [45]), from both large and juvenile trackways [28]. The total time-range for Brachychirotherium tracks in Europe appears to be younger than that for the other chirotherid track types (Longobardian–Carnian), and they appear to be the only chirotherid tracks to have persisted into the Carnian [32]. As has been recently suggested, upper Lower to Mid- dle Triassic ‘chirotherid’ tracks assigned to the ichnogenus Synaptichnium may not have been produced by archosauriform reptiles nor by Tanystrophaeus (cf. [96]); the latter reptile is well known from bone records and coinci- dent tracks in coastal sediments across much of the Ger- manic Basin and northern Tethys (e.g. [75, 96]). Synap- tichnium best ﬁts the pes skeletal anatomy (manus not known; Fig. 18E) of Euparkeria, and this reptile is herein suggested as the trackmaker. The smaller chirotherid trackways and tracks ﬁt well to these small-sized (less then 2 m long) archosaurs, and there is a clear stratigraphic overlap in the track and skeletal record.

5.2. Palaeoenvironment and archosaur bio- geography

During the Early Triassic, non-dinosaurian footprint assemblages — dominated by archosaur tracks — were typically found in arid, intramontane areas (‘chirotherid track assemblage’; [16, 97] Fig. 18). Isochirotherium, Chi- rotherium, Synaptichnium, small Rotodactylus and Rhyn- chosauroides tracks, and tracks of Capitosaurioides, are Figure 20. Reconstruction of an archosauriform carnivore feed- the most common non-dinosaur reptile footprints (e.g. [16, ing on a horseshoe crab (possibly Tachypleus; cf. Diedrich [118]) in the Middle Triassic intertidal-mud flats 60]). The most probable trackmaker for Synaptichnium that formed limulid reproduction sites (Map ©PaleoLogic tracks, Euparkeria capensis ([49]; Figs 18C), has been from Diedrich [118]; Animation ©Haas). abundantly documented from both footprint and body- fossil records around the world [50]. Chirotherid tracks of Isochirotherium, Chirotherium, and Synaptichnium are assic’ of Italy [95], but their exact stratigraphic position is abundant in Europe and have been described from Ger- unclear. The youngest forms probably occur in the east- many e.g. [1, 18, 19], France e.g. [36, 92, 98, 99], England ern Massif Central (Burgundian Gate), but these discover- and Scotland e.g. [12, 30, 42, 100], and Poland [33, 38, ies have not yet been dated chronostratigraphically, only 101]. They are also known from North America [22, 102], by their association as an ‘early dinosaur and archosaur The ‘oldest tridactyl tracks from Poland’ [103] are noth- mixed track assemblage’ ([28]; Figs 18 and 19). ing else than badly and incompletely printed preserved Brachychirotherium tracks are not convincingly figured or chirotherid tracks in redbed sand facies. described before the middle Middle Triassic (Longobar- The Middle Triassic was a period of marine ingression dian). As there is only a small quantity of material avail- and major environmental change in Europe, and this was able, there is currently no evidence for the suggested in- the time when the first probable dinosaur tracks appeared

519 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

in what is called the ‘slightly mixed, archosaur-dominated to the extreme hypersaline and gypsum-enriched environ- track assemblages within the Longobardian in Europe’ [15, ments (except mangroves in modern intertidals). 60, 75, 104]. As mentioned previously, archosauriform reptiles were carnivorous [51, 52] and may have relied on a variety of smaller reptile species as their food source, which they The 22 trackbeds from the middle Middle Triassic inter- would have hunted in terrestrial [16], sabkha, and intertidal biolaminites of northern Germany and the Nether- tidal environments [15]. With the regression and low- lands, including the new material from Bernburg, are the stand at the beginning of the Late Pelsonian, the tidal best dated Middle Triassic tracksites in the world ([15]; flats and sabkha zones likely became increasingly ex- Fig.3). These track-beds range in age from the early to tended around the hypersaline basin centres (Fig. 21B), in the middle Middle Triassic (Aegean–Illyrian), and include which massive layers of salt and gypsum evaporated [66]; two different chirotherid ichnogenera — Isochirotherium Fig. 21C. At this time, the initial Alpine tectonics started and Chirotherium — although large footprints are gener- to create shallow submarine and coastal mountain chains, ally rare as a result of the areally restricted outcrops. The and intramontane basins parallel to the northern Tethys development of new, extensive intertidal-flat habitats and coastline (Fig. 19). A seasonal ‘migration bridge’ was sabkhas (Figs 21, 18, and 19B) during the Middle Trias- thereby opened between the Palaeozoic massifs, allowing sic (Aegian to Illyrian) must have affected both the reptile more east–west migration over the ancient, drying, inter- populations and the food chain. It appears the archosaurs continental sea for all kinds of reptiles [71, 72]. During did not retreat into the fully terrestrial ecosystems of this extreme sealevel low-stand, seasonal reptile migra- the Pangaean supercontinent, but adapted to the extreme tions may also have occurred between the northern Tethys coastal environment with their vegetation-free habitats. and the Germanic Basin, through travel around the Euro- Because large regions were being flooded by the Ger- pean coastlines (Figs 21 and 19). With this palaeogeogra- manic Basin, the terrestrial habitats shrank continuously phy migration, connections were in place for both circum- (Fig. 21), although the originally carnivorous archosauri- central European and Pangaean migrations by the end forms were able to hunt other reptiles within the inter- of the Pelsonian, by which time the evolution of the di- tidal zones, which were likely inhabited by large numbers nosaurs must have already begun. The chirotherid reptile of prey in the form of small reptiles such as Macrocne- tracks recorded from the middle Middle Triassic of Italy mus and possibly Heschleria [15, 71, 86]. Athough the have been ascribed to a Pelsonian age [35], but actually archosauriform reptiles’ terrestrial fluvial habitats disap- appear to be younger, and may be Illyrian or Fassanian peared over large areas of central Europe (Fig. 21), the in age, which would fit better with the palaeogeography Bernburg site shows that at least three different species of Europe during those times. The most likely trackmaker or ‘groups’ of archosaurs must have survived, and may for the Chirotherium track records in Germany, France, have even migrated over much of Pangaea during this and the Italian Alps is Ticinosuchus ferox, which is known period (Fig. 19). Evidence in support of this theory is from a skeleton found in Alpine fore-arc lagoonal sedi- provided by the global distribution of similar large tracks ments from the Illyrian–Fasanian boundary at Monte San attributable to Chirotherium, and Isochirotherium. These Giorgio (Fig. 19), and which has a pedal skeletal anatomy tracks may only represent 0.3% of all tracks preserved in that best fits those tracks (Fig. 18E). As indicated for other the tidal flats [15], but they are much more abundant, and archosaurs and their possible footprints, the track record even dominant, in terrestrial redbed-facies footprint as- overlaps the body-fossil record, which provides at least semblages, such as those in central Germany [19], North one optimal method for linking tracks to their possible America [22], and England [31]. The only intertidal-flat trackmaker, although projection of the skeletal anatomy site with relatively abundant chirotherid tracks is the re- into the best preserved tracks has also been attempted cently discovered Bernburg site, but even here they rep- using the pedal anatomy of known skeletons (Fig. 18E). resent not more than 5% of all vertebrate tracks present. Nevertheless, this an unusually high proportion for inter- The last chirotherid tracks of Pangaea occur in the up- tidal tracksites in the Germanic Basin [96]. However, it permost Middle to lowermost Upper Triassic layers in Eu- should also be taken into account that it has not been pos- rope. These are predominantly Brachychirotherium tracks sible to explore other sites over such a large outcrop area. (cf. [32]), found especially on the steeper, sandy beaches of Bernburg was far from any island or mainland (e.g. the the southwestern Germanic Basin’s Burgundian Gate, and Rhenish or Bohemian Massifs; Fig. 21B), and plant-eating in the northwestern Tethys of the southern Alps. In North reptiles could not have survived in these plant-free zones, America, similar tracks have been found in terrestrial en- which must have been similar to modern, arid, carbonate vironments e.g. [24, 28, 35, 39, 105, 106], and these tracks tidal-flats and sabkhas where no higher plants survive due are also from Brazil [107]. It appears from the pedal skele-

520 Cajus G. Diedrich

Figure 21. A. Middle Triassic (Aegean to Longobardian) sites with Protochirotherium, Isochirotherium, Chirotherium, Brachychirotherium tracks, and Archosauriform skeleton sites for Arizonasaurus (North-America), Euparkeria (South Africa), Ticinosuchus (Switzerland) and Barachotomus (Germany). Most of the skeletons and tracks are from the European Germanic Basin which was an intra cratonic basin forming an exchange bridge between Laurasia and Gondwana (compiled from the following: North and South America [22, 43, 49, 102]; Germany – [72, 73]; France – [36]; Switzerland – [65]; Italy – [25, 35]; Russia – [50]; China – [119]). B. The Germanic Basin with its unique carbonate intertidal ﬂats and food chain reaction as result of Tachypleus horse shoe crab seasonal reproduction migrations which eggs were consumed by Macrocnemus which was killed or only scavenged by the predators (B and C modiﬁed from Diedrich [62]).

521 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

ton anatomy of Batrachotomus kupferzellensis, which has Whether these seismic movements resulted in low-wave digit I shorter than digit IV ([58]; Figs 21C and 18E) — a tsunamis as a result of the shallow basin morphology is feature seen in Chirotherium but absent in Isochirotherium presently unclear and difficult to prove from the sediments, and Synaptichnium (Fig. 18E) — that this European ar- but floods due either to storms or tsunamis may also have chosauriform, found in coastal swamp and lagoonal beach destroyed, or at least impacted, reptile populations over environments [56–58], left the abundant footprints along extensive areas through rapid flooding of their permanent the coastlines surrounding the Bavarian–Hessian depres- habitats. Within seconds, these flat coastal areas would sion, and also in the eastern Massif Central and northern have been been flooded over distances of tens of kilo- Italian regions (Fig. 21C). In Italy, all three chirotherid metres, washing up the carcasses of marine reptiles, fish, track types have been reported from sandy near-shore sharks, and limulid crustaceans that would have become a environments (e.g. [25, 35]). A similar, sandy, coastal fa- food source for the carnivorous archosauriforms (Fig. 20). cies and footprint assemblage from the uppermost Middle This must be considered a plausible ecological interpreta- Triassic has been studied in southeastern France, where tion for such an extensive coastal environment of extremely typical chirotherid trackways have been found in asso- low relief, and would also explain the discovery of articu- ciation with more highly developed early coelurosaurid lated marine sauropterygian skeletons in these sediments. and theropod footprints, as well as possible prosauropod Such articulated remains were found at different sites, dinosaur footprints [28, 106], apparently from the latest even in biolaminites, together with footprints [64, 94, 114]. Middle Triassic (Fassanian–Longobardian). Terrestrial animals and marine sauropterygians could have been caught up in waves with their carcasses carried for many kilometres onto the tidal flats or sabkhas; in such 5.3. Coastal hazards, global archosaur ex- hazardous situations, the marine reptiles that survived the tinction, and the emergence of the dinosaurs flooding would have had little chance of returning to the sea. Many kilometres of low-relief intertidal zones around The Chirotherium and Isochirotherium footprints and their the Germanic Basin would have been flooded in just a few probable trackmakers were not well defined previously seconds, as has been observed in similar intertidal zones (e.g. [19, 28, 108]), but as presented here, appear to and sabkhas of the modern Arabian Gulf coast [115]. The have disappeared around the boundary between the Mid- high density of seismically shocked sediment layers at dle and Upper Triassic (Fig. 18), when the archosaur Bernburg reflects a period of much higher tectonic activ- group was overwhelmed by the radiation of the dinosaurs ity in central Europe around the transition from the Lower (e.g. [51, 52, 103, 109]). Despite this, chirotherid track- to Middle Muschelkalk, as has previously been discussed makers certainly survived into the Upper Triassic in other by Szulc [80]. This appears to have been a result of north- forms (e.g. [110]), such as Batrachotomus kupferzellen- ern Tethyan tectonic development and the collision of the sis [111] or other species found, for example, in North African and European plates that marked the beginning of America [112]. The late Middle and early Late Triassic Pangaean supercontinental break-up; this is also the rea- was a time of gradual change in reptile groups world- son why the Germanic Basin became so flat, with salinas wide [103], with archaic forms, such as the few diversified in its central regions and extensive carbonate tidal-flats archosauriform reptiles described herein becoming almost and sabkhas in the coastal zones [66, 76, 80]. Catastrophic extinct across Pangaea [51] due to competition with di- earthquakes in the Middle Triassic must have had a ma- nosaurs at the end of the Triassic (Carnian); only one jor effect on organisms living along the coast of the Ger- branch of the archisauriforms still survives as the modern manic Basin, and the shock-waves would have resulted in crocodiles [103, 110, 113]. a stressful situation for their populations (Figs 16 and 19). To date, there has been no model to explain these evo- During the tidal ebb and the drying out of the carbonates, lutionary developments. One possible influence on this large- and medium-sized archosaur chirotherid trackmak- gradual evolutionary change was the impact of the Pan- ers walked over the tidal-flat surfaces, treading on the gaea break-up, which started gradually during the Mid- slickensides and vein-following mud-cracks, a chronology dle Triassic, leaving the Germanic Basin between two that can be clearly observed in track horizon 8 (Fig. 16). new supercontinents and resulting in drastic changes to This proves that these terrestrial reptiles were mobile im- both coastal and intramontane habitats ([15]; Figs 21 mediately after the seismic events, while the sediments and 18). Intramontane playa basins were destroyed by were still drying out. It is interesting to note that the marine ingressions (Fig. 21), and additional hazards such large, heavy chirotherid did not sink deeper than 2 cm as earthquake-induced tsunamis may also have periodi- into the sediment layer, which means that the underly- cally impacted the extensive intertidal habitats found in ing biolaminates must have already been well stabilized, regions such as the Germanic Basin (Figs 21 and 19).

522 Cajus G. Diedrich

in contrast to the newly deposited arenite. This early also present in Synaptichnium (=Protochirotherium) and consolidation of the biolaminites also explains why the Chirotherium tracks, permit the identification of Isochi- vein structures that resulted from the seismic shock-waves rotherium as an archosauriform footprint. In other lepi- reached only 3 cm more from the arenite track horizon 6 dosauriform archosaur tracks, such as Rhynchosauroides, into underlying biolaminites. Such vein structures have those pad structures are polygonal not round. Compar- been found in the modern sediments of active fault zones, isons of pedal skeletons with more general reconstruc- proving formation in unconsolidated sediments [116]. The tions of the valid chirotherid footprint types allow them to underlying biolaminites are therefore shown to have been be attributed to their most probable archosauriform track- rapidly consolidated, as has been described for modern makers. examples [81], and were cemented sediments forming a The skeleton of the upper Lower Triassic small archosaur, stable substrate to walk on, despite being formed within Euparkeria capensis, comparing using skeletons from the middle to upper intertidal or lower sabkha zones. South Africa, indicates that it was the most probable trackmaker of the small-chirotherid Synaptichnium tracks, 5.4. Archosauriform predators at the end of which globally are predominantly found in upper Lower to the food chain basal Middle Triassic terrestrial redbeds, and more rarely seen in intertidal-facies types. The middle Middle Tri- An explanation for the presence of archosauiform carni- assic skeleton of Ticinosuchus ferox, found in the near- vores on these intertidal flats most likely lies in the food coastal lagoons of Switzerland (central Europe), was much chain. There must have been millions of horseshoe-crab larger and appears to be the most probable producer of the eggs laid in the intertidal, coastal reproduction zones medium-sized chirotherid tracks Chirotherium, which are around the Germanic Basin, as indicated by the large most commonly found in coastal intertidal facies, but also numbers of limulid trackways recorded at Bernburg [62]. occur in terrestrial playa facies, also worldwide through- Being the bottom of the food chain, these eggs appear out Pangaea. The Lower Middle Triassic (Lower Anisian) to have been consumed mainly by two small reptiles, giant poposaurid Arizonasaurus, found in Northern Amer- Macrocnemus and Hescheleria, for which these environ- ica and most probably in England and Germany (large ments must have provided a primary abode, judging by poposaurid remains, at least, are known) was the most the large numbers of footprints that they left behind [71]. probable trackmaker for Isochirotherium, the largest chi- These common small reptiles seem to have been the main rotherid tracks known from many terrestrial and coastal prey that attracted carnivorous archosauriforms onto the sites in America and Europe. Finally, the upper Mid- intertidal flats. Additionally, the crabs themselves must dle Triassic and lower Upper Triassic archosaur Batra- have been fed upon by large archosauriform reptiles, as chotomus kupferzellensis, found in Germany (central Eu- indicated by a trackway and ‘feeding’ place seen at Bern- rope), was the most probable maker of the Early Keu- burg (Fig. 20). per (Longobardian to Lower Carnian) Brachychirotherium tracks (medium-sized chirotherid tracks), found mainly in coastal swamps and alluvial-fan sands of the Late Middle 6. Conclusions to Late Triassic, also in America and Europe.This common occurrence again indicates the continuous exchange of archosauriforms across Pangaea until the Late Trias- Several extremely detailed trackways of Chirotherium sic. Aetosaurs also seem to have produced tracks that are barthii and Isochirotherium herculis are described from a much more similar in track shape to Brachychirotherium, new middle Middle Triassic (Karlstadt Formation; upper although those tracks will require a different ichnogenus Pelsonian) site at Bernburg, in central Germany. These name due to their different rotation trackways and longer examples serve to establish a range of variability for these digits I and IV. two track types, caused primarily by: sediment differences between biolaminites and arenites, changes in locomotion In the Middle Triassic, dramatic changes took place in the type, and variations in the size of the individual. The Germanic Basin and, within a time frame of about 10 m.y., Bernburg material consists of long mapped trackways with this basin went from intracratonic to a shallow-marine, 15–64 manus/pes sets, with only the millimetre-thick bi- tectonically controlled basin with extensive intertidal car- olaminites of track-bed 7 showing detailed preservation bonate zones. Horseshoe crabs produced millions of eggs with skin impressions, whereas the 2 cm thick, slick- in Middle Triassic coastal reproduction areas within these ensided, seismic shocked arenites of track-bed 8 form intertidal zones, establishing a completely new food chain coarse, poorly shaped imprints. New details, such as cir- in which archosauriform predators such as Ticinosuchus, cular and granular dermal patterns of the feet, features Euparkeria, and Arizonasaurus appear to have hunted the

523 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

smaller permanent inhabitants, such as Macrocnemus and Acknowledgements Hescheleria, which fed on the horseshoe-crab eggs. A Chirotherium trackmaker is even demonstrated to have fed I would like to thank Prof. Dr H. Meller from the Lan- upon the crabs themselves. These plant-free coastal habi- desamt fuer Denkmalpflege, Saxony-Anhalt, and Prof. Dr tats were occasionally in danger of being flooded, over J.-H. Olbertz from the Ministry of Culture of Saxony- tens of kilometres and within a just few seconds, by sub- Anhalt, who supported and financed the excavations. marine, seismically induced shallow tsunamis, or else by Parts of the research were funded by Solvay Chemicals subtropical storm-waves. These events may have killed GmbH, and strongly supported by J. Lischka. I am grate- larger marine and terrestrial reptile populations over much ful to Dr S. Friederich for coordinating the excavation pro- of the extensive tidal coastline of the Germanic Basin gram. Trackway experiments and tidal-flat expeditions in (central Europe). These habitats changed again in the the UAE were possible due to the sponsorship for an ARTE Illyrian, when a new marine ingression and deepening of documentary film project ‘Saurier Code’, made by the film the basin resulted in the disappearance of most of these producer company Hofrichter and Jacobs. Dr D. Henning intertidal zones, probably only leaving those in the north- kindly provided access to the slab in the Stadtmuseum, ern Tethys. Saalfeld, and Dr R. Wiermann provided access to the slab in the Schlossmuseum, Bernburg. I would like to thank With the subsequent regression in the Fassanian (Mid- Prof. Dr G.H. Bachmann and Dr N. Hauschke for scientific dle/Early Late Triassic), the Germanic Basin became and geological discussions. Dr H. Hagdorn kindly allowed a shallow lagoon with freshwater influences, and a the study of archosaur material in the Muschelkalkmu- lagoonal remnant in the Bavarian–Hessian depression seum at Ingelfingen. The University of Zurich kindly per- around which the archosaur Batrachotomus must have mitted photography of the archosaur skeleton from Monte hunted small reptiles. From the footprint record, mainly San Giorgio. The South African Museum of Nature pro- in Italy (northern Tethys), eastern France (Burgundian vided photographs of the unpublished specimen of Eupark- Gate), and southern Germany (Germanic Basin), these last eria. I am also grateful to Dr R. Schoch for permission to archosaurs were in competition with small theropod and study and illustrate the archosaur skeleton exhibited in prosauropod dinosaurs, and possibly even with pterosaurs, the Staatliche Sammlungen Naturkunde in Stuttgart. G. at least since the Fassanian time. ‘Rinaldino’ Teichmann illustrated the trackmakers and the tsunami scenario. E. Manning carried out proofreading of The emergence and diversification of the dinosaurs and the original manuscript. Finally, I thank I. Díáz-Martinez disappearance of archaic reptiles in the late Middle Tri- for a critical review, and a second unknown reviewer. assic may have resulted from these marked, drastic, but natural environmental changes, which were due to plate tectonics, marine ingressions and regressions, habitat changes, and resulting changes in the food chain. It ap- References pears to have been the slow, gradual adaptation of the better constructed dinosaurs that led to their survival and [1] Sickler F.K.J., Sendenschreiben an Dr. Blumen- the eventual demise of the archosaurs during the Late Tri- bach über die höchst merkwürdigen Reliefs der assic. Fährten urweltlicher Tiere in den Hessberger Sand- steinbrüchen bei Hildburghausen. Neues Jahrbuch Whether the conditions in the Germanic Basin provided für Mineralogie Geologie und Paläontologie, 1835, the main driving force behind the emergence of the di- 230–232 nosaurs, or whether it was the result of general changes to [2] Bernhardi R., Thierfaerten bei Hildburghausen. coastal zones around Pangaea remains a matter for spec- Neues Jahrbuch für Mineralogie Geognosie und Ge- ulation. Nevertheless, the intracratonic Germanic Basin ologie, 1834, 642–643 undoubtedly formed an exchange bridge between Laura- [3] Kaup J.J., Über Tierfaehrten bei Hildburghausen. sia and Gondwana, over which predatory archosauriform Neues Jahrbuch für Minerlaogie Geologie und reptiles may have successfully migrated seasonally over Palaeontologie, 1835, 327–328 millions of years, until the full break-up of Pangaea. The [4] Humboldt von, A., Not sur les empreintes de pieds development of the Germanic Basin, its changing reptile d‘une quadrupede dans le grès bigarre de Hild- populations, and its abundant Triassic skeletal and foot- burghausen. Annales des Sciences Naturelle de print records are certainly essential elements to be con- Paris B, 1835, 4, 134–138 sidered if we are to understand fully the emergence and [5] Link O., Sur les traces des pieds d‘animeaux incon- global domination of the dinosaurs. nues a Hildburghausen. Comptes Rendues Academie

524 Cajus G. Diedrich

de Science Paris, 1835, 1, 1–258 [21] Egerton S., On two casts in sandstone of the im- [6] Nöggerath J., Das urweltliche Handthier von pressions of the hind foot of a gigantic Chirotherium Hildburghausen. Rheinländische Provinzial Blätter from the New Red Sandstone of Chechire. Proceed- Neue Folge, 1835, 4, 212–213 ings of the Geological Society of London, 1838, 59, [7] Voigt F.S., Tierfährten von Hildburghausen. Neues 14–15 Jahrbuch für Mineralogie Geologie und Paläontolo- [22] Peabody F.E., Reptile and amphibian trackways gie, 1835, volume, 322–326 from the Lower Triassic Moenkopi Formation of Ari- [8] Wigmann A., Tierfährten im Hildburghauser Sand- zona and Utah. University of California Publications stein. Archiv der Naturgeschichte, 1835, 1, 395–397 in Geological Sciences, 1948, 27, 295–468 [9] Virlet T., Note sur les empreintes de oieds dàn- [23] Gand G., Sur le matériel ichnolique récolté dans imeaux dans le grés bigarré des environs de Hild- le Muschelkalk de Culles-les-Roches. Bulletin de burghausen, es Saxe. Bulletin de la Societé Geolo- la Societé d’ Histoire Naturelle Creusot, 1977, 35, gie France, 1836, 7, 220–224 21–44 [10] Engelhardt B., Über die Formation, in welcher die [24] Gand G., Description de deux nouvelles traces Tatzenabdrücke vorweltlicher Tiere in der Nähe von d’Isochirotherium observées dans le grès du Trias Hildburghausen auftreten. Neues Jahrbuch für Ge- moyen de Bourgogne. Bulletin de la Societé d’ His- ologie und Paläontologie, 1837, 379–384 toire Naturelle Creusot, 1979, 37, 13–25 [11] Rühle von Lilienstein, H.,. Fährten und Spuren [25] Brandner R., Tertapodenfährten aus der unteren im Chirotheriumsandstein von Südthüringen. Mitteltrias der Südalpen. Veröffentlichungen Uni- Fortschritte in Geologie, 1839, 40, 239–387 versität Innsbruck, 1973, 86, 57–71 [12] Swinton W.E., The history of Chirotherium. Geol. J., [26] Haubold H., Chirotherium-Fährten aus dem 2007, 2, 443–473 Buntsandstein im "Mauritianum" in Altenburg. Ab- [13] Haubold H., Die Saurierfährten Chirotherium barthii handlungen und Berichte für Natrukunde Museum Kaup, 1835 — das Typusmaterial aus dem Altenburg, 1969, 6, 21–36 Buntsandstein bei Hildburghausen/Thüringen und [27] Demathieu G., Archosaurier-Fährtenfaunen der das “Chirotherium-Monument“. Veröffentlichungen Trias: Die Bedeutung ihrer Ähnlichkeiten und ihrer des Naturhistorischen Museum Schleusingen, 2006, Verschiedenheiten; ihre mögliche Verwendung im 21, 3–32 Rahmen der Stratigraphie der Trias. Int. J. Earth Sci., [14] Koch K., Schmid E., Die Fährtenabdrücke im Bunten 1982, 71, 741–746 Sandsteine bei Jena. Hochhausen, Jena, 1841 [28] Demathieu G., Trace fossil assemblages in Middle [15] Diedrich C., Die Saurierspuren-Grabung im basalen Triassic marginal marine deposits, eastern border Mittleren Muschelkalk (Anis, Mitteltrias) von Bern- of the Massif Central, France. Societé Economique burg (Sachsen-Anhalt). Archäologie in Sachsen- Paläontologie et Mineralogie Speciale Publication, Anhalt Soderband, 2012, 15, 1–63 1985, 35, 53–66 [16] Haubold H., Saurierfährten. Neue Brehm Bücherei, [29] Thulborn R.A., Dinosaur tracks. Chapman and Hall, Wittenberg Lutherstadt, 1984 London - New York - Tokyo - Melbourne - Madras, [17] Soergel W., Die Fährten der Chirotheria. Gustav 1990. Fischer, Jena, 1925 [30] Tresise G., ‘Chirotherium herculis’ — problems [18] Demathieu G., Leitz F., Wirbeltierfährten aus dem posed by the first finds. Ann. Sci., 1991, 48, 565– Röt von Kronach (Trias, NO-Bayern). Mitteilungen 57 der Bayerischen Staatssammlung für Paläontologie [31] Tresise G., Sarjeant W.A.S., The tracks of Triassic und Historische Geologie, 1982, 22, 63–89 vertebrates. Fossil evidence from North-West Eng- [19] Haubold H., Die Tetrapodenfährten des Buntsand- land. The Stationary Office, London, 1997 steins in der Deutschen Demokratischen Republik [32] Karl C., Haubold H., Saurierfährten im Keuper und in Westdeutschland und ihre Äquivalente in der (Obere Trias) Frankens, die Typen von Brachy- gesamten Trias. Paläontologische Abhandlungen A chirotherium. Berichte der Naturwissenschaflichen Paläozoologie, 1971, 4, 395–548 Gesellschaft Bayreuth, 2000, 24, 91–120 [20] Haderer F.O., Demathieu G., Böttcher R., [33] Ptaszynski T., Lower Triassic vertebrate footprints Wirbeltier-Fährten aus dem Rötquarzit (Oberer from Wióry, Holy Cross Mountains, Poland. Acta Buntsandstein, Mittlere Trias) von Hardheim bei Palaeontol. Pol., 2000, 45,151–194 Wertheim/Main (Süddeutschland). Stuttgarter [34] Avanzini M., Lockley M., Middle Triassic archosaur Beiträge zur Naturkunde, 1995, 230,1–31 population structure: interpretation based on

525 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

Isochirotherium delicatum fossil footprints (Southern senschaften, 2003, 25, 21–36 Alps, Italy). Palaeogeogr. Palaeoecol., 2002, 185, [47] Nöggerath J., Angebliche Fährten urweltlicher Af- 391–402 fen (Quadrumanen) in Sandsteinen. Rheinländische [35] Avanzini M., Leonardi N., Isochirotherium inferni Provinzial Blätter Neue Folge, 1835, 4, 143–148 ichnosp. nov. in the Illyrian (Late Anisian, Middle [48] Owen R., Description of the skeleton of the genus Triassic) of Adige Valley (Bolzano, Italy). Bolletino Labyrinthodon; with remarks of the probable iden- Societa Paleontologica Italiana, 2002, 41, 41–50 tity of the Chirotherium with that genus of extinct [36] Demathieu G.R., Demathieu P., Chirotheria and batrachians. Proceedings of the Geologisal Society other Ichnotaxa of the European Triassic. Ichnos, of London, 1841, 3, 389–397 2004, 11, 79–88 [49] Ewer R.F., The anatomy of the thecodont reptile [37] King M., Sarjeant W., Thompson D., Tresise G.A., Euparkeria capensis Broom. Philosophical Transac- Revised systematic ichnotaxonomy and review of the tions of the Royal Society London B, 1965, 248, vertebrate footprint ichnofamily Chirotheriidae from 379–435 the British Triassic. Ichnos, 2005, 12, 241–299 [50] Gower D.J., Sennikov A.G., Early Archosaurs from [38] Valdiserr D., Avanzini M.A., Tetrapod ichnoassocia- Russia. In: Benton M.J., Kurochkin E.N., Shishkin tion from the Middle Triassic (Anisian, Pelsonian) of M.A., Unwin D.M. (Eds.), The Age of Dinosaurs in Northern Italy. Ichnos, 2007, 14, 105–116 Russia and Mongolia: Cambridge University Press, [39] Kuleta M., Niedźwiedzki G., Krebs B., Ptaszyński T., Cambridge, 2000, 140–159 Nowe stanowisko z tropami kręgowców z górnego [51] Nesbitt S.J., The early evolution of archosaurs: re- pstrego piaskowca Gór Swiętokrzyskich.` Przegląd lationships and the origin of major clades. B. Am. Geologiczny, 2005, 53, 151–155 Mus. Nat. Hist., 2011, 352, 1–292 [40] Hunt A.P., Lucas S.G., The Triassic tetrapod record: [52] Irmis R.B., Evaluating hypotheses for the early di- ichnofaunas, ichnofacies and biochronology. New versification of dinosaurs. Earth Env. Sci. T. of R. Mexico Museum of Natural History Science Bul- So., 2011, 101, 397–426 letin, 2007, 41, 78–87 [53] Ebel K., Falkenstein F., Haderer F.O., Wild [41] Niedźwiedzki G., Ptaszyński T., Large Chirotheri- R., Ctenosauriscus koeneni (v. Huene) und der idae tracks in the Early Triassic of Wiory, Holy Rauisuchier von Waldshut — Biomechanische Deu- Cross Mountains, Poland. Acta Geol. Pol., 2007, 57, tung der Wirbelsäule und Beziehungen zu Chi- 325–342 rotherium sickleri Kaup. Stuttgarter Beiträge zur [42] Clark N.D.L., Corrance H., New discoveries of Isochi- Naturkunde B, 1998, 261, 1–18 rotherium herculis and a reassessment of chi- [54] Krebs B., Ticinosuchus ferox n. g. n. sp. Schweiz- rotheriid footprints from the Triassic of the Isle of erische Paläontologische Abhandlungen, 1965, 81, Arran, Scotland. Scot. J. Geol., 2009, 45, 69–82 1–141 [43] Klein H., Lucas S.G., Review of the tetrapod ich- [55] Krebs B., Zur Deutung der Chirotherium-Fährten. nofauna of the Moenkopi Formation/Group (Early– Natur und Museum, 96(10):389–396. Middle Triassic) of the American Southwest. New Avanzini, M. 2000. Synaptichnium tracks with skin Mexico Museum of Natural History and Science impressions from the Anisian (Middle Triassic) of the Bulletin, 2010, 50, 1–67 Southern Alps (Val di Non, Italy). Ichnos, 1966, 7, [44] Gand G., De La Horra R., Galán-Abellán B., López- 243–251 Gómez J., Barrenechea J.F., Arche A., Benito M.I., [56] Gower D., The cranial and mandibular osteology of New ichnites from the Middle Triassic of the Iberian a new rauisuchian archosaur from the Middle Tri- Ranges (Spain): palaeoenvironmental and palaeo- assic of southern Germany. Stuttgarter Beiträge zur geographical implications. Historical Biology, 2010, Naturkunde B, 1999, 280, 1–49 22, 40–56 [57] Hagdorn H, Muschelkalkmuseum Ingelfingen. Edi- [45] Díaz-Martínez I., Pérez-García A., Estudio de las ic- tion Lattner, Ingelfingen, 2004 nitas de vertebrado triásicas de Espana. In: Pérez- [58] Gower D.J., Schoch R.R., Postcranial Anatomy of the García A., Gascó F., Gasulla J.M., Escaso F. (Eds.), Rauisuchian Archosaur Batrachotomus kupferzel- Viajando a Mundos Pretéritos. Ayuntamiento de lensis. J. Vertebr. Paleontol., 2009, 29, 103–122 Morella, 2011, 111–122 [59] Lucas S.G., Heckert,A.B., Late Triassic aetosaurs as [46] Klein H., Haubold H., Differenzierung von aus- the trackmaker of the tetrapod footprint ichnotaxon gewählten Chirotherien der Trias mittels Land- Brachychirotherium. Ichnos, 2011, 18, 197–208 markanalyse. Hallesches Jahrbuch für Geowis- [60] Lucas S.G., Tetrapod footprint biostratigraphy and

526 Cajus G. Diedrich

biochronology. Ichnos, 2007, 14, 5–38 2008, 259, 410–423 [61] Demathieu G.R., Oosterink H.W., New discoveries of [73] Diedrich C., Palaeogeographic evolution of the ichnofossils from the Middle Triassic of Winterswijk marine Middle Triassic marine Germanic Basin (the Netherlands). Geologie en Mijnbouw, 1988, 67, changements — with emphasis on the carbonate 3–17 tidal flat and shallow marine habitats of reptiles in [62] Diedrich C., The palaeogeographic reconstructions Central Pangaea. Global Planet. Change, 2009, 6, of the Middle Triassic tectonical controlled carbon- 27–55 atic Germanic Basin of Central Europe — a northern [74] Föhlisch K., Überlieferungen seismischer Aktivität Tethys connected cratonic marine Basin – coastal im Unteren Muschelkalk. Beiträge zur Geologie basin margin mappings by the use of reptile foot- Thüringens Neue Folge, 2007, 14, 55–83 print rich intertidal and sabkha environments. 5th [75] Diedrich C., The Middle Triassic marine reptile bio- International Conference on the Geology of Tethys, diversity in the Germanic Basin in the centre of the Quena, Egypt, Abstract Book, 2010, 3-5 Pangaean world. Central European Journal of Geo- [63] Diedrich C., Oosterink H., Bergings- en docu- sciences, 2012, 4, 9–46 mentatietechniek van Rhynchosauroides peabodyi [76] Bachmann G.H., Kozur H.W., The Germanic Tri- (Faber) — Sauriersporen op de grens Boven- assic: correlations with the international chronos- Buntsandsteen/Onder Muschelkalk van Winterswijk. tratigraphic scale, numerical ages and Mi- Grondboor and Hamer, 2000, 54, 125–130 lankovitch cyclicity. Hallesches Jahrbuch für Ge- [64] Diedrich C., Die Wirbeltierfährtenfundstelle owissenschaften B, 2004, 26, 17–62 Borgholzhausen (Teutoburger Wald, NW- [77] Kozur H.W., Bachmann G.H., Updated correlation of Deutschland) aus der Oolith-Zone (Bithyn, Unterer the Germanic Triassic with the Tethyan scale and Muschelkalk). Paläontologische Zeitschrift, 2002, assigned numeric ages. Berichte der Geologischen 76, 35–56 Bundesanstalt Wien, 2008, 76, 53–58 [65] Diedrich C., Fichter J., Eine erste Saurierfährten- [78] Aigner T., Bachmann G.H. Sequence stratigraphic Grabung im Unteren Muschelkalk (Anis, Mitteltrias) framework of the German Triassic. Sediment. Geol., von Größenlüder, Nordhessen (NW-Deutschland). 1992, 80, 115–135 Philippia, 2003, 11, 109–132 [79] Kurze M., Zum Problem der Entstehung von Wellen- [66] Bachmann G.H., Geologie von Sachsen-Anhalt. strifen und Querplattung im Muschelkalk. Zeitschrift Schweizerbart, Stuttgart, 2008 der deutschen geologischen Wissenschaften, 1981, [67] Schwarz U., Sedimentary structures and facies anal- 9, 489–499 ysis of shallow marine carbonates (Lower Muschel- [80] Szulc J., Anisian–Carnian evolution of the Germanic kalk, Middle Triassic, SW-Germany). Contributions Basin and its eustatic, tectonic and climate controls. to Sedimentology, 1975, 3, 1–100 Zentralblatt für Geologie und Paläontologie, 1998, [68] Kurze M., Zum Problem der Entstehung von Wellen- 7–8, 813–852 strifen und Querplattung im Muschelkalk. Zeitschrift [81] Riding R., Microbial carbonates: the geological der deutschen geologischen Wissenschaften, 1981, record of calcified bacterial–algal mats and biofilms. 9, 489–499 Sedimentology, 2000, 4, 179–214 [69] Knaust D., Signatures of tectonically controlled sed- [82] Lukas V., Die Terebratel-Bänke (Unterer Muschel- imentation in Lower Muschelkalk carbonates (Mid- kalk, Trias) in Hessen — ein Abbild kurzzeit- dle Triassic) of the Germanic Basin. Zentralblatt für iger Faziesänderungen im westlichen Germanischen Geologie und Paläontologie I, 1998, 9–10, 893–924 Becken. Geologisches Jahrbuch Hessen, 1991, 119, [70] Föhlisch K., Voigt T., Synsedimentary deformation in 119–175 the Lower Muschelkalk of the Germanic Basin. Spe- [83] Knaust D., Trace fossils and ichnofabrics on the cial Publications of the International Association of Lower Muschelkalk carbonate ramp (Triassic) of Sedimentologists, 2001, 31, 279–297 Germany: tool for high resolution sequence stratig- [71] Bachmann G.H., Aref M.A.M., A seismite in Trias- raphy. Geologische Rundschau, 1998, 87, 21–31 sic gypsum deposits (Grabfeld Formation, Ladinian), [84] Diedrich C., Vertebrate track bed stratigraphy of the Southwest Germany. Sediment. Geol., 2005, 180, Upper Bunter and basal Lower Muschelkalk (Mid- 75–89 dle Triassic) of Winterswijk (East Netherlands). Ge- [72] Diedrich C., Millions of reptile tracks — Early ologie en Mjinbouw//Netherlands Journal of Geo- to Middle Triassic carbonate tidal flat migration sciences, 2001, 80, 31–39 bridges of Central Europe. Palaegeogr. Palaeoecol., [85] Chafetz H.S., Buczynski C, Bacterially inducted

527 Middle Triassic chirotherid trackways on earthquake inﬂuenced intertidal limulid reproduction ﬂats of the European Germanic Basin coasts

lithification of microbial mats. Palaios, 1992, 7, 277– [97] Mader D., Charakteristische Fossilien des mitteleu- 293 ropäischen Buntsandsteins. Naturwissenschaften, [86] Diedrich C., Actuopalaeontological trackway exper- 1984, 71, 69–78 iments with Iguana on intertidal flat carbonates of [98] Maubeuge P., Nouvelle découverte d’ empraintes de the Arabian Gulf — a comparison to fossil Rhyn- reptiles dans les grès du Trias inférieur de l’ Est chosauroides tracks of Triassic carbonate tidal flat de la France. Bulletin Academie Societé Lorraine, megatracksites in the European Germanic Basin. 1972, 11, 203–237 Senckenbergiana Maritima, 2005, 35, 203–220 [99] Durand M., A propos de quelques traces de pas [87] Fichter J., Aktuopaläontologische Untersuchungen et figures sédimentaires dans le Buntsandstein an den Fährten einheimischer Urodelen und Lac- supérieur du Sud-Ouest des Vosgeses. Bulletin d’ ertilier. Teil I: Die Morphologie der Fährten in Ab- Academie Societé Lorrain, 1975, 14, 23–25 hängigkeit von der Sedimentbeschaffenheit. Mainzer [100] Clark N.D.L., Aspen P., Corrance H., Chirotherium naturwissenschaftliches Archiv, 1982, 20, 91–129 barthii Kaup, 1835 from the Triassic of the Isle of [88] Brand L.R., Variations in salamander trackways re- Arran, Scotland. Scot. J. Geol., 2002, 38, 83–92 sulting from substrate differences. J. Palaeont., 1996, [101] Fuglewicz R., Lower Triassic footprints from the Swi- 70, 1004–1010 etokrzyskie (Holy Cross) Mountains, Poland. Acta [89] Haubold H., Die Saurierfährten Chirotherium Palaeontol. Pol., 1990, 35, 109–164 barthii Kaup, 1835 – das Typusmaterial aus dem [102] Peabody F.E., Occurrence of Chirotherium in South Buntsandstein bei Hildburghausen/Thüringen und America. Bulletin of the Geological Society of Amer- das ”Chirotherium-Monument”. Veröffentlichungen ica, 1955, 66, 239–240 des Naturhistorischen Museum Schleusingen, 2006, [103] Brusatte S.L., Benton M.J., Lloyd G.T., Ruta M., 21, 3–32 Wang. S.C., Macroevolutionary patterns in the evo- [90] Fichter J., Kunz R., New genus and species of lutionary radiation of archosaurs (Tetrapoda: Diap- chirotheroid tracks in the Lower Triassic (Middle sida). Earth Env. Sci. T. R. Soc., 2011, 101, 367–382 Bunter, Detfurth-Folge) of Central Germany. Ichnos, [104] Klein H., Haubold H., Archosaur footprints — po- 2004, 11, 183–193 tential for biochronology of Triassic continental se- [91] Haubold H., Klein H., The dinosauroid track- quences. New Mexican Museum of Natural History ways Parachirotherium – Atreipus – Grallator from Sciences Bulletin, 2007, 41, 120–130 the lower Middle Keuper (Upper Triassic: La- [105] Bock W., Triassic reptilian tracks and trends of lo- dinian, Carnian, ? Norian) of Frankonia (Germany). comotive evolution. J. Peleontol., 1952, 26, 395–433 Hallesches Jahrbuch für Geowissenschaften B, 2000, [106] Courel L., Demathieu G., Gand G., Contribution ap- 22, 59–85 portée par l’ ichthyologie a quelques disciplines de [92] Demathieu G., Prémiers traces de reptiles ar- la géologie — example des traces de Vertébrés du chosauriens dans le Trias autochthone des Argilles Trias moyen de la bordue orientale du Massif Cen- Rouges (Col des Corbeaux, Vieil Em osson, Suisse). tral. Université Dijon Mémoires, 1982, 7, 313–326 Comtes Rendues d’ Academie de Science Paris, [107] Leonardi G., Isochirotherium sp.: Pista de um gi- 1977, 285, 649–652 gentesco Tecodonte na Formacao Antenor Navarro, [93] Demathieu G., Haubold H., Reptilfährten aus dem Soisa, Brazil. Revista Brasilia Geologica, 1980, 10, Mittleren Buntsandstein von Hessen. Hallesches 186–190 Jahrbuch für Geowissenschaften, 1982, 7, 97–110 [108] Carrano M.T., Wilson J.A., Taxon distributions and [94] Diedrich C., Trostheide F., Auf den Spuren the tetrapod track record. Paleobiology, 2001, 27, der terresten Muschelkalksaurier und aquatischen 563–581 Sauropterygier vom obersten Röt bis zum Mittleren [109] Gauthier J.A., Saurichian monophyly and the origin Muschelkalk (Unter-/Mitteltrias) von Sachsen- of birds. Memoires of the Californian Acadamy of Anhalt. Abhandlungen und Berichte für Naturkunde Science, 1986, 8, 1–55 Magdeburg, 2007, 30, 5–56 [110] Benton M.J., Clark J.M., Archosaur phylogeny and [95] Avanzini M., Cavin L., A new Isochirotherium track- the relationships of the crocodylia. Systematic As- way from the Triassic of Vieux Emosson, SW sociation Special Volume, 1988, 35, 295–338 Switzerland: stratigraphic implications. Swiss J. [111] Gower D., The cranial and mandibular osteology of Geosci., 2009, 102, 351–362 a new rauisuchian archosaur from the Middle Tri- [96] Wild R., The fossil deposits of Kupferzell, southwest assic of southern Germany. Stuttgarter Beiträge zur Germany. Mesozoic Vertebrate Life, 1980, 1, 15–18 Naturkunde B 1999, 280, 1–49

528 Cajus G. Diedrich

[112] Nesbitt S.J., The anatomy of Effigia okeeffeae (Ar- marine Middle Triassic marine Germanic Basin chosauria, Suchia), theropod-like convergence, and changements With emphasis on the carbonate tidal the distribution of related taxa. B. Am. Mus. Nat. flat and shallow marine habitats of reptiles in Cen- Hist., 2007, 302, 1–84 tral Pangaea. Global and Planetary Change, 2009, [113] Sereno P.C., Basal archosaurs: phylogenetic rela- 65, 27–55 tionships and functional implications. Journal of Ver- [118] Diedrich C., Middle Triassic limulid crustacean re- tebrate Palaeontology Memoirs, 1995, 2, 1–53 production intertidal flats of Europe and contri- [114] Oosterink H.W., Berkelder W., Jong C. de Lankamp J., bution to the variability of Kouphichnium Nopcsa, Winkelhorst H., Sauriens uit de Onder-Muschelkalk 1923 tracks — early Mesozoic limulid mating mi- van Winterswijk. Staringia 2003, 11, 1–145 grations into the Germanic Basin and reptile food [115] Warren J.K., Evaporite sedimentology. Prentice Hall chain reactions. Biological Journal of the Linnaean Inc., New Jersey, 1989 Society London, 2011, 103, 76–105 [116] Brothers R.J., Kemp A.E., Maltman A.J., Mechanical [119] Lu H., Zhang Y., Xiao J, Chirotherium: fossil foot- development of vein structures due to the passage of prints of primitive reptiles in the Middle Triassic earthquake waves through poorly-consolidated sed- Guanling Formation, Zhenfeng, Guizhou Province, iments. Tectonophysics, 1996, 266, 227–244 China. Acta Geol. Sin., 2004, 78, 468–474 [117] Diedrich C., Palaeogeographic evolution of the

529