1 FIRST XIPHOSURAN TRACEWAY IN THE MIDDLE MUSCHELKALK 2 FACIES (MIDDLE TRIASSIC) OF THE CATALAN BASIN (NE IBERIAN 3 PENINSULA) 4 5 PRIMER RASTRO DE XIFOSURO EN LAS FACIES MUSCHELKALK MEDIO 6 (TRIÁSICO MEDIO) DE LA CUENCA CATALANA (NE DE LA PENÍNSULA 7 IBÉRICA) 8 9 Chabier De Jaime-Soguero1*, Eudald Mujal1,2 & Josep Fortuny1 10 11 1 Institut Català de Paleontologia Miquel Crusafont, ICTA-ICP Building, c/de les 12 columnes, s/n, E-08193 Cerdanyola del Vallès, Catalonia, Spain; 13 chabierdejaime@gmail.com; josep.fortuny@icp.cat 14 2 Staatliches Museum für Naturkunde Stuttgart, Rosenstein 1, D-70191 Stuttgart, 15 Germany; eudald.mujalgrane@smns-bw.de 16 * Corresponding author 17 18 Abstract 19 In the last decade, the first ichnoassamblages from the middle Muschelkalk facies (upper 20 Anisian–middle Ladinian) of the Catalan Basin (NE Iberian Peninsula) have been 21 discovered. Herein, the first xiphosuran trace fossils are described from the locality of 22 Penya Rubí, a newly discovered ichnosite from the Catalan Basin. The finding opens a 23 window into peri-Tethys ecosystems with coastal influence. The traceway is referred to 24 the ichnogenus Kouphichnium: a locomotion trace attributed to xiphosurans. The 25 traceway preserves telson grooves and different imprint morphologies from the various 1 26 appendages. The traceway pattern and arrangement of the different traces suggest a 27 crawling locomotion style. The sedimentology suggests a coastal zone with areas 28 influenced by tides (intertidal flat). The morphological variations of the ichnites are 29 correlated to substrate rheology, the locomotion of the tracemaker and environmental 30 conditions. 31 32 Keywords 33 Invertebrate ichnology, Kouphichnium, telson, Middle Triassic, coastal 34 palaeoenvironment 35 36 Resumen 37 En la últma década, se han descubierto las primeras icnoasociaciones de las facies 38 Muschelkalk medio (Anisiense superior–Ladiniense medio) de la Cuenca Catalana (NE 39 de la Península Ibérica). Aquí se describe el primer rastro fósil de xifosuro de la localidad 40 de Penya Rubí, un icnoyacimiento descubierto recientemente en la Cuenca Catalana. El 41 hallazgo abre una ventana dentro de los ecosistemas del peri-Tethys con influencia 42 costera. El rastro ha sido identificado como Kouphichnium: una traza de locomoción 43 asociada a xifosuros fósiles. El rastro preserva surcos del telson e impresiones de los 44 distintos apéndices, que resultan en morfologías diferentes. El patrón del rastro y la 45 disposición de las distintas trazas sugiere un estilo de locomoción de reptación 46 (“crawling”). La sedimentología sugiere una zona costera, con áreas de influencia mareal 47 (intermareal). Las variaciones morfológicas de las icnitas están correlacionadas con la 48 reología del sustrato, la locomoción del productor de las huellas y las condiciones 49 ambientales originales. 50 2 51 Palabras clave 52 Icnología de invertebrados, Kouphichnium, telson, Triásico Medio, paleoambiente 53 costero 54 55 1. INTRODUCTION 56 57 The order Xiphosura Laterille, 1802, generally known as horseshoe crabs, consists 58 of aquatic chelicerate arthropods with a fossil record extending from the Ordovician to 59 the present (Rudkin et al., 2008; Lamsdell, 2016). Although xiphosurans have inhabited 60 in many different environments through most of the Phanerozoic, and the group has 61 survived several mass extinction events, currently only four marine species of the family 62 Limulidae survive (Gaillard, 2011). Because, the morphology of the group remained 63 largely unchanged through much of its evolution, the extant species are considered as 64 “living fossils” (Gaillard, 2011; Kin & Błażejowski, 2014). The xiphosurans show a 65 maximum diversity during the Carboniferous and Triassic periods (Moore et al., 2007). 66 Xiphosuran trace fossils provide additional information about the palaeobiology of these 67 animals and their habitats (Gaillard, 2011; Fernández & Pazos, 2013; Mujal et al., 2018a; 68 Bicknell et al., 2019; Shibata & Varricchio, 2020). 69 The Triassic fossil record of Xiphosura is particularly interesting as it helps to 70 elucidate how this group survived after the end-Permian mass extinction (Błażejowski et 71 al., 2017; Mujal et al., 2018a). Even more important, in general terms, after the end- 72 Permian mass extinction, the diversity of life shows a general diminishment, with a very 73 poor terrestrial record, particularly in central Pangea (Benton & Newell, 2014). 74 Biodiversity recovery in marine ecosystems was faster than in terrestrial ecosystems due 3 75 to a high increase of marine productivity after the mass extinction event (Shen et al., 76 2015). 77 Considering on these concepts, the discovery of a new fossil traceway referred to 78 Xiphosura from the north-eastern Iberian Peninsula (Western Tethys) is of special 79 interest, because it was recovered from the middle Muschelkalk facies of the Catalan 80 Basin. Triassic facies are diachronic at European level due to the different influence of 81 sea level (i.e., different transgressions and regressions) among the Western Tethys basins 82 (Virgili et al., 1977; Escudero-Mozo et al., 2015; Franz et al., 2015). In the particular 83 case of the Catalan Basin, a regression episode occurred during the late Anisian–middle 84 Ladinian affected coastal environments, creating numerous coastal river floodplains (Solé 85 de Porta et al., 1987; Calvet & Marzo, 1994; Ortí et al., 2018). This regression is 86 represented in the sedimentary record as red beds and evaporite deposits between the 87 marine Muschelkalk facies (Calvet & Marzo, 1994; Ortí et al., 2018). 88 In the NE Iberian Peninsula few arthropod remains have been recovered (both 89 direct and indirect evidences). Three taxa from the Muschelkalk facies (Ladinian, upper 90 Middle Triassic) have been described in the Catalan Basin to date (Vía-Boada & Villata, 91 1966; Romero & Vía-Boada, 1977; Vía, 1987). Also, several xiphosuran traceways have 92 been described from one locality of the terrestrial lower Buntsandstein Pyrenean Basin 93 dated from the Lower Triassic (upper Olenekian) (Mujal et al., 2018a). 94 The present study analyses the first record of a xiphosuran traceway from the 95 middle Muschelkalk of the Iberian Peninsula. From this unit of the Catalan Basin different 96 ichnoassociations of tetrapod trace fossils have been reported so far (Mujal et al., 2015, 97 2018b). Thus, the newly discovered traceway adds relevant data on the palaeodiversity 98 of these ecosystems. The main objective of this study is the ichnotaxonomy of the 99 traceway recovered from the ichnosite of Penya Rubí (Vallirana, Barcelona province). 4 100 Further, this traceway provides information about the palaeoenvironment and 101 palaeogeography of its tracemaker. 102 103 2. GEOLOGICAL SETTING 104 105 The outcrop of Penya Rubí is located in the Catalan Basin, currently forming the 106 Catalan Coastal Ranges (Fig. 1). The Catalan Basin is situated at the northeast of the 107 Iberian Peninsula and began to develop during the middle to late Permian due to the 108 extension of the Iberian plate (Galán-Abellán et al., 2013). The Triassic sedimentary 109 succession of the Catalan Basin is composed of the so-called Germanic facies, recognised 110 at European scale, which comprise the following lithostratigraphic units: Buntsandstein, 111 lower (M1), middle (M2) and upper (M3) Muschelkalk, Keuper and Imón Formation 112 (Calvet & Marzo, 1994). The Triassic succession presents a thickness of 500–800 meters 113 (Calvet et al., 1990). 114 During the Middle Triassic the sedimentary environmental conditions changed 115 drastically in different areas of the Iberian massif, but especially affecting the Catalan 116 Basin. Such changes are reflected in the three different Muschelkalk facies successions 117 within the Catalan Coastal Ranges: They represent marine environments (M1 and M3) 118 separated by the coastal and alluvial environments (M2) (Calvet & Marzo, 1994; Ortí et 119 al., 2018). Following these events, during the Early Triassic, remains of braided and 120 meandering river settings under arid and semiarid climatic conditions are represented by 121 the Buntsandstein facies (Galán-Abellán et al., 2013; Mujal et al., 2017). During the early 122 Anisian (Middle Triassic) there were abundant fluvial systems that gradually change into 123 coastal and marine settings (lower Muschelkalk facies – M1), but later emerging again 124 during the late Anisian–middle Ladinian as a result of a short regression episode, as 5 125 represented by the middle Muschelkalk (M2) facies; afterwards, a major transgression 126 adjusted coastal and marine settings (upper Muschelkalk facies – M3) (Calvet & Marzo, 127 1994; Ortí et al., 2018). This is the reason why the continental environments gain presence 128 in the Catalan Basin (but not in other Iberian basins), and the upper Anisian and lower 129 Ladinian are characterised by the predominance of distal alluvial, coastal and shallow 130 marine successions (Escudero-Mozo et al., 2015; Ortí et al., 2018). Herein, the studied 131 outcrop corresponds to the middle Muschekalk facies, dated as late Anisian–middle 132 Ladinian based on palynological and biostratigraphic analysis (Solé de Porta et al., 1987; 133 Márquez-Aliaga et al., 2000; Mujal et al., 2015, 2018b). 134 135 3. MATERIAL AND METHODS 136 137 An in-situ slab with a xiphosuran traceway was discovered in the Penya Rubí 138 ichnosite (Vallirana, Catalan Basin). The slab was recovered from a reddish sandstone 139 stratum (50–60 cm thick). The slab dimensions are: 50.8 cm long, 29.45 cm wide and 140 4.13 cm thick. The ichnological study was carried out in the lab. The slab is stored at the 141 Institut Català de Paleontologia Miquel Crusafont (Sabadell, Catalonia, Spain) and 142 labelled as IPS110272. 143 A stratigraphic log was performed using a Jacob staff and meter stick. The slab 144 was located in its exact stratigraphic level. After measuring, the data were translated into 145 a stratigraphic sequence using the software Inkscape (v. 0.92.3). 146 Once in the lab, the traces were measured by caliper and with digital measuring 147 tools from the image manipulating software GIMP (v. 2.10.0). Also, light was shone from 148 different directions to appreciate all the details of the impressions. The traceway was 149 digitised photogrammetrically (see Supplementary Information). This method consists of 6 150 creating a three dimensional model using photographs. The photographs were obtained 151 with a digital reflex camera Panasonic Lumix DMC-FZ200 following the procedures of 152 Falkingham (2012) and Mallison & Wings (2014). The three dimensional models was 153 processed with three programs, following the procedures of Mujal et al. (2020, see 154 references therein for a review): Agisoft Photoscan (standard version 1.1.4), to generate 155 a dense point cloud, MeshLab (v. 2016.12) to edit the mesh (cleaning, scale and 156 orientation) and ParaView (v. 5.5.0) to generate a coloured depth map and contours. The 157 final three dimensional meshes were used to complement the morphological study of each 158 trace and their deepness. These meshes, refined and improved following some procedures 159 (Marcé-Nogué et al., 2015), can be also used in biomechanical studies to evaluate 160 tracemaker locomotion. 161 162 4. OUTCROP OF PENYA RUBÍ (VALLIRANA, CATALAN BASIN) 163 164 The Triassic outcrops in the studied area include Buntsandstein, Muschelkalk 165 (M1, M2, M3) and Keuper facies, and the recovered traceway is located in the M2 unit 166 (Figs. 2, 3). The middle Muschelkalk facies are composed of clays and sandstones 167 interbedded with evaporite rocks (Ramón & Calvet, 1987; Ortí et al., 2018). The 168 sandstones are fine-grained, with a clayey matrix and limestone cement. Sandstone grains 169 are, in decreasing order, composed of quartz, feldspar and mica. The mudstones are 170 mainly fine detrital sediments and appear alternated with sandstone strata (Calvet & 171 Marzo, 1994; Ortí et al., 2018). 172 In the Penya Rubí ichnosite a repichnial traceway of a xiphosuran has been 173 recognised (Figs. 3, 4). These imprints are preserved in a reddish sandstone stratum (50– 174 60 cm), near to the upper limit between sandstones and a stratified mudstone interval 7 175 (Figs. 2, 3). The trace-bearing sandstone presents parallel stratification in its lower part 176 and wave ripples in the topmost part, including the surface where the traceway is 177 impressed. In the overlying strata, in a greyish interval of dolostones and mudstones, 178 tetrapod footprints appear. They are currently under study, preliminarily referred to 179 Rhynchosauroides, Procolophonichnium and Rotodactylus. This interval presents a 180 millimetric stratification with alternating thick (greyish) and thin (reddish) laminae 181 building up centimetric sequences. The thick levels are composed of grains coarser than 182 those of the thin levels, and from the bottom to the top, the grain size decreases (fining- 183 upwards sequences). 184 The overlying interval displays similar stratification than the previous one but it 185 presents reddish colouration; on top of it the first medium-grained sandstone horizons are 186 present. These sandstone beds contain Chirotherium footprints, currently under study, 187 and desiccation cracks at their base. In many of the strata ripples are present. In the finely 188 laminated sequences climbing ripples appear, whereas the coarser and thicker sandstones 189 are characterised by unidirectional ripples. 190 As a whole, the succession can be interpreted as a coastal area, with bidirectional 191 flows and a certain cyclic sedimentation, as indicated by successive layers of fine grained- 192 sandstones with ripples indicating opposite palaeoflows. Most probably, the succession 193 was deposited in an intertidal area with relatively low energy, as already stated by Ortí et 194 al. (2018) (see also references therein). 195 196 5. SYSTEMATIC PALAEOICHNOLOGY 197 198 5.1. Ichnogenus Kouphichnium 199 8 200 Referred material. The ichnites belong to a relatively regular repetition of 201 symmetrical imprints (IPS110272). This set of imprints is considered to be a traceway 202 preserved in concave epirelief (Fig. 4a-e) (see three-dimensional model in Supplementary 203 Information). 204 Description. One straight and symmetric traceway, reaching up to 240 mm long 205 and 26.8 mm wide. The traceway is composed of five nearly straight parallel rows of 206 ichnites (Fig. 4a-4b), which show opposite symmetry. The central row consists of six 207 rectilinear grooves (mean length: 20.6 mm). These linear traces are the deepest imprints 208 and present a profound region in the middle of the groove. On both sides, with a 209 symmetric display, numerous smaller prints close to the straight grooves are arranged in 210 two rows in each side. These rows tend to interchange up to four types of imprints (Fig. 211 4c-4d). The first type shows a bifid shape (mean length: 3.3 mm), composed of a central 212 point from which two curved scratches emerge. The second type is consist of linear 213 scratches, much shorter than the central grooves (mean length: 4.2 mm). The shape of 214 these imprints ranges from straight to curved lines. Despite their linear shape, the extreme 215 anterior of the imprint displays a deep and punctuated impression. The third type of 216 imprints shows pointed ichnites (mean length: 2.0 mm). Finally, at the edges of the 217 traceway arrow-shaped ichnites make up the fourth type (mean length: 5.4 mm). This last 218 imprint type shows the direction of the traceway, starting in the middle of the slab and 219 finishing at one of the ends, i.e., the traceway advances towards the bifid and arrowed end 220 of the traces. 221 The position of each type of imprint is not completely regular along the rows, but 222 it seems to follow a trend. In both lateral rows, the pointed and bifid imprints tend to be 223 in the inner part of the row, while the lineal and arrow-shaped frequently appear in the 224 outer side. The arrow shaped and linear traces tend to appear in the posterior part of the 9 225 telson trace while the bifid and pincer-like imprints are located in the medium part, and 226 the pointed imprints tend to be in the anterior part, near to the telson trace; the inner traces 227 are relatively aligned, outlining a diagonal approaching the central linear groove from the 228 posterior to the anterior. Also, in the upper part of the anterior side is also frequent to find 229 a bifid imprint. The long linear traces are always the central row (Fig. 4a-4e). 230 Discussion. The morphological characters of the imprints reveal that the traceway 231 corresponds to locomotion traces (repichnia) of an arthropod. The presence of a “tail 232 trace”, corresponding to the central linear grooves, and the disposition and morphology 233 of the imprints in each side allow to assign the traceway to Kouphichnium. This 234 ichnogenus is interpreted as a repichnion, likely produced by xiphosurans (Caster, 1938; 235 Goldring & Seilacher, 1971; Seilacher, 2007; Fernández & Pazos, 2013; Rindsberg & 236 Martin, 2015; Mujal et al., 2018a; Shu et al., 2018; Shi et al., 2019). The body of the 237 xiphosurans is composed of three main parts: the prosoma (anterior part), the abdomen 238 (central part) and a relatively long telson (posterior part) (Gaillard, 2011) (Fig. 4e). In the 239 abdomen, xiphosurans present six pairs of appendages that contribute to various 240 functions. The first pair of appendages are the chelicere, which end with pincers, being 241 useful for grasping food. The second to fifth pair of appendages are chelate walking legs, 242 all of which display bifid ends (Shi et al., 2019; Shibata & Varricchio, 2020). The 243 principal function of these appendages is for locomotion. Finally, the sixth pair is known 244 as the pusher legs. They are the strongest appendages and push the animal forward over 245 (commonly soft) substrates (Caster, 1938). This strong heteropody is the cause of the 246 large variability of the resulting traceway (Caster, 1938; Goldring & Seilacher, 1971; 247 Rindsberg & Martin, 2015). The anterior five pairs consist in small pincer-like chelipeds, 248 leaving characteristic V-shaped impressions seen in xiphosuran traceways (Shi et al., 10 249 2019). On the other hand, the “pusher” legs produce many variable imprints with digitate 250 and flabellar forms (Gaillard, 2011). 251 As indicated in the description, the position of each type of imprint tends to be 252 regular. The imprints show a descending line from the posterior part of the central row to 253 the anterior part, being the linear and arrow shaped imprints the furthest ones from the 254 telson at the anterior part, whereas the bifid, pincer-like and pointed imprints are the 255 closest at the posterior part. This trend has been represented in an ideal reconstruction of 256 the traceway (Fig. 4e). 257 Comparing the anatomy of living limulids and the different types of imprints, we 258 can establish a correlation between trace and tracemaker. The central row of straight 259 grooves would be produced by the telson. According to Caster (1938), nowadays 260 xiphosurans would use the telson to raise the posterior part of the prosoma when the rest 261 of the appendix is buried. This behaviour would describe the telson as a key appendix in 262 the locomotion of the xiphosurans. Similarly, Gaillard (2011) suggested such traceways 263 with discontinuous telson impressions as crawling behaviour. Also, according to Caster 264 (1938) and Goldring and Seilacher (1971), due to the shallow drag of the telson, its 265 impression is only preserved in the superficial laminae of the substrate, hence these 266 imprints usually do not appear as “undertraces” (see also Seilacher, 2007). Otherwise, the 267 two sets of imprints from each side (Fig. 4a, 4b) of the traceway most likely correspond 268 to the walking appendages. The multiple variation of these traces, herein grouped in four 269 types, would represent the locomotion stages of the appendages (Shi et al., 2019). 270 271 6. DISCUSSION 272 273 6.1. Palaeoecology and palaeoenvironment 274 11 275 The trace fossils were found in a laminar interval composed by reddish sandstones 276 and mudstones (Fig. 3a-b). The presence of sandstone layers with ripples showing 277 opposite palaeoflows, the finely laminated mudstones alternated with sandstones, as well 278 as the presence of climbing ripples, give support to the interpretation of the Penya Rubí 279 outcrop as an intertidal setting, as suggested in other middle Muschelkalk successions 280 (Ortí et al., 2018) and other areas elsewhere (Davis & Dalrymple, 2012). As explained in 281 the previous section, due to the shallow drag of the telson in xiphosurans, these imprints 282 most commonly do not appear in “undertraces”, thus the traceway is most probably 283 preserved in the originally trampled surface. 284 In addition, the ethology and ecology of extant limulids provides an excellent 285 model for determining the behavioural characteristics and palaeoecology of the ancient 286 xiphosurans (Martin & Rindsberg, 2007; Seilacher, 2007). Limulids can live in a wide 287 range of environments from freshwater to marginal marine settings (Lamsdell, 2016; Shi 288 et al., 2019). But during their life, limulids inhabit different environmental niches at 289 different ontogenetic stages. Mature individuals mostly live in distal areas of the 290 continental shelf and return to the shore and tidal areas to mate and to lay their eggs in 291 the breeding season (Caster, 1938, 1944; Shuster, 1982; Shi et al., 2019). Juveniles are 292 benthic and live in coastal areas for two years before moving into deeper waters as they 293 grow older (Shi et al., 2019). Thus, if it is assumed that the herein studied xiphosuran 294 traceway is the result of similar behaviour, the tracemaker could be a juvenile that lived 295 in the intertidal area or mature specimen that returned to the coast for the breeding season. 296 According to Martin & Rindsberg (2007) on the studies of recent juvenile limulid 297 traceways, likely producing traces assignable to Nereites, we favour the latter 298 interpretation. 299 12 300 6.2. Palaeogeographic implications 301 302 In recent years, the fossil record from the Muschelkalk facies of the Catalan Basin 303 and other Iberian basins has been enhanced, particularly due to the discovery of tetrapod 304 footprints (Gand et al., 2010; Fortuny et al., 2011; Mujal et al., 2015, 2018b; Berrocal- 305 Casero et al., 2018). The Penya Rubí ichnosite, where the studied traceway was 306 recovered, is not an exception and several tetrapod footprints have been also recovered 307 here. Otherwise, xiphosuran remains have been historically restricted to the upper 308 Muschelkalk (M3) of Alcover (Catalan Basin, Tarragona province) outcrops (Vía-Boada 309 et al., 1977 and references therein). More recently, xiphosuran trace fossils were also 310 discovered in the lower part of the Buntsandstein facies, in non-marine fluvial deposits 311 from the Pyrenean Basin, also stratigraphically and geographically close to tetrapod 312 footprints (Mujal et al., 2017, 2018a). 313 In the Catalan Basin remains of three limulid taxa have been recovered from the 314 upper Muschelkalk: Tachypleus gadeai (Vía-Boada & Villalta, 1966) (first described as 315 Heterolimulus gadeai), Tarracolimulus rieki Romero & Vía-Boada, 1977 and 316 Mesolimulus crespelli Vía, 1987. These species have been described from the quarry of 317 Alcover-Montral (Tarragona, Spain), in Ladinian dolomites. The outcrops of Alcover- 318 Montral (Vía, 1987) make up the first area where Triassic Iberian xiphosuran remains 319 were discovered. Nevertheless, these specimens are slightly younger than the herein 320 reported ichnites. 321 Outside the Iberian Peninsula, other xiphosuran remains have been recovered in 322 Middle Triassic deposits from peri-Tethys basins. At the German locality of Bernburg, in 323 the Central European Basin, limulid traceways related to shallow marine and intertidal 324 successions have been documented (Diedrich, 2011). A new species of limulid from the 13 325 Anisian–lower Ladinian of Tunisia has also been described (Błażejowski et al., 2017). 326 This species, Limulitella tejraensis, probably lived in freshwater environments, and 327 would be phylogenetically related with another lower Middle Triassic (Anisian) limulid 328 from France, Limulitella bronni (see Błażejowski et al., 2017). 329 Beyond Europe and North Africa, some Triassic xiphosuran remains have been 330 discovered. Recently in China, ichnites from the Daye Formation of Guiyang (South 331 China) have been described. Shi et al. (2019) described two different types of traceways 332 referred as Kouphichnium from the Lower Triassic. The palaeoenvironmental conclusions 333 suggest that the tracemakers would inhabit the seafloor at 50-60 m deep, in a complex 334 ichnoassociation with other invertebrates. In North China, Shu et al. (2018) described 44 335 well-preserved traceways attributed to Kouphichnium from the upper Permian–Lower 336 Triassic successions. Also, in upper Permian–Lower Triassic terrestrial successions 337 (freshwater settings) from the Central Transantarctic Mountains (Antarctica) several 338 horseshoe crab traceways have been attributed to the ichnospecies Koupichnium 339 natalensis by Hasiotis (2012). Finally, in Madagascar were recovered two specimens of 340 a Lower Triassic limulid from the Ankitokazo Basin, tentatively determined as the genus 341 Limulitella by Hauschke et al. (2004), and in New South Wales (Australia), the limuloid 342 Dubbolimulus peetae, which would have inhabited freshwater settings during the Middle 343 Triassic, was described (Pickett, 1984). 344 All these fossil remains suggest that, after the Permian mass extinction, 345 xiphosurans recovered and radiated during the Middle Triassic, with species occupying 346 both marine environments and, of special note, continental freshwater environments (cf. 347 Błażejowski et al., 2017; Mujal et al., 2018a). Most of the known Triassic fossil remains 348 of xiphosurans from the present Northern Hemisphere have been described from Europe, 349 with few additional records from North Africa and China. 14 350 351 6.3. Size of the tracemaker 352 353 The size of the tracemaker can be inferred from the external width of the traceway, 354 following the formula I = W ×1.5 (Malz, 1964; Shi et al., 2019), where I is the width of 355 the prosoma of the tracemaker and W is the external width of the traceway. In the herein 356 reported specimen, the traceway width is 26.8 mm, so the estimated prosoma width is 357 40.2 mm. Also, using the formula L = I ×2.12 (see Shi et al., 2019), the length of the 358 tracemaker (L) can be calculated, being 85.2 mm for the tracemaker of the studied 359 specimen. The mean L value obtained in Kouphichnium specimens from the Lower 360 Triassic in Daye Formation (China) is 192.7 mm (Shi et al., 2019). Mujal et al. (2018a) 361 described Kouphichnium specimens from the upper Lower Triassic of the Catalan 362 Pyrenees of up to 70 mm wide. In comparison, the size of the tracemaker from Penya 363 Rubí would be among the smallest found in the Triassic ichnofossil record (Table 1). It 364 suggests that the traceway might correspond to a juvenile specimen or a small xiphosuran 365 taxon. However, according to Martin & Rindsberg (2007), juvenile specimens of 366 xiphosurans might produce traces attributable to Nereites, thus the option of a small 367 xiphosuran taxon as the producer of the studied traceway is preferred. 368 The Iberian species Mesolimulus crespelli (width measures: 35 mm for the 369 holotype, and 31 and 53 mm for the other specimens) presents the most similar size to the 370 studied traceway (Table 2). Also, Tarracolimulus reiki shows a similar size, although it 371 is slightly smaller. Otherwise, T. gadeai shows the most different size, being larger than 372 140 mm wider than the Penya Rubí specimen. Therefore, M. crespelli shows the most 373 similar size to the inferred size of the tracemaker of Penya Rubí Kouphichnium traceway, 374 but we cannot ignore that T. reiki had an almost identical size. 15 375 376 7. CONCLUSIONS 377 378 The arthropod traceway recovered from the Penya Rubí ichnosite of the Catalan 379 Basin has been identified as Kouphichnium isp., a locomotion trace fossil (repichnia) 380 attributed to xiphosurans. This is the first invertebrate traceway described from the middle 381 Muschelkalk (Middle Triassic) of the Catalan Basin. The traceway is composed of three 382 parallel rows of imprints, the central one represented by straight grooves (telson 383 impressions) and the lateral rows (two in each side of the telson impression) being 384 composed of straight and pincer-shaped (bifid) imprints (internal row of each side) and 385 pointed and arrow-shaped imprints (external row of each side). The lateral rows evidently 386 correspond to the impressions of the different appendages. According to the general 387 traceway pattern and the arrangement of the different appendages and telson traces, a 388 crawling behaviour was the most probable locomotion style of the tracemaker. 389 The animal that produced this trace fossil was probably roaming an intertidal area 390 with shallow, quiet water. Considering the size of the imprints, the tracemaker would be 391 a small xiphosuran of 40.2 mm width, which could possibly be an adult of a small-sized 392 taxon in a breeding season. Comparing the size of the traceway, the Triassic species 393 Mesolimulus crespelli and Tarracolimulus rieki present the most similar size to the 394 theoretical size of the tracemaker of Penya Rubí ichnites, but it is not possible to 395 confidently correlate the traceway with a specific taxon. 396 397 8. ACKNOWLEDGMENTS 398 16 399 First of all, thanks to the Spanish Society of Palaeontology (SEP), for granting financial 400 aid to carry out this investigation. We are deeply indebted to Dr. Oriol Oms (Universitat 401 Autònoma de Barcelona), discoverer of the slab, and also to Dr. Arnau Bolet (University 402 of Bristol and ICP), Manel Méndez (ICP) and Marc Riccetto for helping us during the 403 fieldtrips. Also to Dr. Chabier de Jaime for logistics related to the project. J.F. is supported 404 by the Spanish Agencia Estatal de Investigación and the European Regional Development 405 Fund of the European Union (AEI/FEDER EU, project CGL2017-82654-P). J.F. is 406 member of the consolidated research group 2017 SGR 86 GRC of the Generalitat de 407 Catalunya. We acknowledge support from the CERCA programme (ICP) from the 408 Generalitat de Catalunya, and the project “Evolució dels ecosistemes durant la transició 409 Paleozoic–Mesozoic a Catalunya” (ref. CLT009/18/00066), financially supported by the 410 Departament de Cultura (Generalitat de Catalunya). We are grateful to Dr. Andrew K. 411 Rindsberg and Dr. Alberto Pérez-López, who reviewed the manuscript and provided 412 valuable and constructive comments that substantially improved a previous version of 413 this work. 414 415 9. BIBLIOGRAPHY 416 417 Benton, M.J. & Newell, A.J. 2014. Impacts of global warming on Permo-Triassic 418 terrestrial ecosystems. Gondwana Research, 25, 1308-1337. DOI: 419 10.1016/j.gr.2012.12.010 420 Berrocal-Casero, M., Arribas, M. & Moratalla, J.J. 2018. Didactic and divulgative 421 resources of the Middle Triassic vertebrate Tracksite of Los Arroturos (Province of 422 Guadalajara, Spain). Geoheritage, 10(3), 375-384. DOI: 10.1007/s12371-017- 423 0244-1 17 424 Bicknell, R.D.C., Brougham, T., Charbonnier, S., Sautereau, F., Hitij, T. & Campione, 425 N. 2019. On the appendicular anatomy of the xiphosurid Tachypleus syriacus and 426 the evolution of fossil horseshoe crab appendages. The Science of Nature, 106, 38. 427 DOI: 10.1007/s00114-019-1629-6 428 Błażejowski, B., Niedźwiedzki G., Boukhalf, K. & Soussi, M. 2017. Limulitella 429 tejraensis, a new species of limulid (Chelicerata, Xiphosura) from the Middle 430 Triassic of southern Tunisia (Saharan Platform). Journal of Paleontology 91(5), 431 960-967. DOI: 10.1017/jpa.2017.29 432 Calvet, F., Tucker, M.E. & Henton, J.M. 1990. Middle Triassic carbonate ramp systems 433 in the Catalan Basin, northeast Spain: facies, systems tracts, sequences and controls. 434 Special Publications International Assotiation of Sedimentology, 9, 79-108. 435 Calvet, F. & Marzo, M. 1994. El Triásico de las Cordilleras Costero Catalanas: 436 Estratigrafía, Sedimentología y Análisis Secuencial. Cuaderno de Excursión. III 437 Coloquio de Estratigrafía y Paleoestratigrafía del Pérmico y Triásico de España. 438 Field Guide, 1-53. 439 Caster, K.E. 1938. A restudy of the Tracks of Paramphibius. Journal of Paleontology, 440 12, 3-60. 441 Caster, K.E. 1944. Limuloid trails from the upper Triassic (Chinle) of the Petrified Forest 442 National Monument, Arizona. American Journal of Science, 242, 74-84. 443 Davis, R.A. Jr. & Dalrymple, R.W. (eds.) 2012. Principles of tidal sedimentology. 444 Springer, London. 445 Diedrich, C. G. 2011. Middle Triassic horseshoe crab reproduction areas on intertidal 446 flats of Europe with evidence of predation by archosaurs. Biological Journal of the 447 Linnean Society, 103, 76-105. 18 448 Escudero-Mozo, M.J., Márquez-Aliaga, A., Goy, A., Martín-Chivelet, A., López-Gómez, 449 J., Márquez, L., Arche, A., Plasencia, P., Pla, C., Marzo, M. & Sánchez-Fernández, 450 D. 2015. Middle Triassic carbonate platforms in eastern Iberia: Evolution of their 451 fauna and palaeogeographic significance in the Western Tethys. Palaeogeography, 452 Palaeoclimatology, Palaeoecology, 417, 236-260. DOI: 453 10.1016/j.palaeo.2014.10.041 454 Falkingham, P.L. 2012. Acquisition of high resolution 3D models using free, open- 455 source, photogrammetric software. Paleontologia Electronica, 15, 1-15. 456 DOI:10.26879/264 457 Fernández, D.E. & Pazos, P.J. 2013. Xiphosurid trackways in a Lower Cretaceous tidal 458 flat in Patagonia: Palaeoecological implications and the involvement of microbial 459 mats in tracefossil preservation. Palaeogeography, Palaeoclimatology, 460 Palaeoecology, 375, 16-29. DOI: 10.1016/j.palaeo.2013.02.008 461 Fortuny, J., Bolet, A., Sellés, A. G., Cartanyà, J. & Galobart, À. 2011. New insights on 462 the Permian and Triassic vertebrates from the Iberian Peninsula with emphasis on 463 the Pyrenean and Catalonian basins. Journal of Iberian Geology, 37 (1), 65-86. 464 DOI: 10.5209/rev_JIGE.2011.v37.n1.5 465 Franz, M., Kaiser, S.I., Fischer, J., Heunisch, C., Kustatscher, E., Luppold, F.W., Berner, 466 U. & Röhling, H.-G. 2015. Eustatic and climatic control on the Upper Muschelkalk 467 Sea (late Anisian/Ladinian) in the Central European Basin. Global and Planetary 468 Change, 135, 1-27. DOI: 10.1016/j.gloplacha.2015.09.014 469 Gand, G., De La Horra, R., Galán-Abellán, B., López-Gómez, J., Barrenechea, J.F., 470 Arche, A. & Benito, I. 2010. New ichnites from the Middle Triassic of the Iberian 471 Ranges (Spain) paleoenvironmental and paleogeographical implications. Historical 472 Biology, 22, 40-56. DOI: 10.1080/08912961003644096 19 473 Gaillard, C. 2011. A giant limulid trackway (Kouphichnium lithographicum) from the 474 lithographic limestones of Cerin (late Kimmeridgian, France): Ethological and 475 environmental implications. Swiss Journal of Geosciences, 104(Suppl. 1), 57-72. 476 DOI: 10.1007/s00015-010-0032-2 477 Galán-Abellán, B., López-Gómez, J., Barrenecha, J.F., Marzo, M., De la Horra, R. & 478 Arche, A. 2013. The beginning of the Buntsandstein cycle (Early-Middle Triassic) 479 in the Catalan Ranges, NE Spain: Sedimentary and palaegeographic implications. 480 Sedimentary Geology, 296, 86-102. DOI: 10.1016/j.sedgeo.2013.08.006 481 Goldring, R. & Seilacher, A. 1971. Limulid undertracks and their sedimentological 482 implications. Neues Jahrbuch für Geologie und Paläontologie Abhandlungen, 483 137(3), 422-442. 484 Hasiotis, S. T. 2012. Horseshoe crabs lived in Permo-Triassic Antarctic freshwater rivers 485 and lakes: Trace fossil evidence from the Buckley and Fremouw Formations 486 Beardmore Glacier Area, Central Transantarctic Mountains (CTAM), Antarctica. 487 Geological Society of America, Abstracts with Programs, 44(7), 500. 488 Hauschke, N., Wilde, V. & Brauckmann, C. 2004. Triassic limulids from Madagascar 489 missing links in the distribution of Mesozoic Limulacea. Neues Jahrbuch für 490 Geologie und Paläontologie, 87-94. 491 Kin, A. & Błażejowski, B. 2014. The horseshoe crab of the genus Limulus: Living fossil 492 or stabilomorph? PLoS ONE, 9, e108036. DOI: 10.1371/journal.pone.0108036 493 Lamsdell, J.C. 2016. Horseshoe crab phylogeny and independent colonizations of fresh 494 water: Ecological invasion as a driver for morphological innovation. 495 Palaeontology, 59(2), 181-194. DOI: 10.1111/pala.12220 496 Mallison, H. & Wings, O. 2014. Photogrammetry in paleontology, a practical guide. 497 Journal of Paleontological Techniques, 12, 1-31. 20 498 Malz, H. 1964. Kouphichnium walchi, die Geschichte einer Fährte und ihres Tieres: Natur 499 und Museum, 94, 81–97. 500 Marcé-Nogué, J., Fortuny, J., Gil, L. & Sánchez, M. 2015. Improving mesh generation in 501 finite element analysis for functional morphology approaches. Spanish Journal of 502 Palaeontology, 30(1), 117-132. 503 Márquez-Aliaga, A., Valenzuela-Rios, J.I., Calvet, F. & Budurov, K. 2000. Middle 504 Triassic conodonts rom northeastern Spain; biostratigraphic implications. Terra 505 Nova, 12, 77-83. 506 Martin, A.J. & Rindsberg, A.K. 2007. Arthropod tracemakers of Nereites? - 507 Neoichnological observations of juvenile limulids and their paleoichnological 508 applications. In Trace fossils: Concepts, problems, prospects (Ed. W. Miller), 478- 509 491. 510 511 Moore, R.A., McKenzie, S.C. & Lieberman, B.S. 2007. A Carboniferous synxiphosurine 512 (Xiphosura) from the Bear Gulch Limestone, Montana, USA. Palaeontology, 50(4), 513 1013-1019. DOI: 10.1111/j.1475-4983.2007.00685.x 514 Mujal, E., Fortuny, J., Rodríguez-Salgado, P., Diviu, M., Oms, O. & Galobart, À. 2015. 515 First footprints occurrence from the Muschelkalk detritical unit of the Catalan 516 Basin: 3D analyses and palaeoichnological implications. Spanish Journal of 517 Palaentology, 30, 97-107. 518 Mujal, E., Fortuny, J., Bolet, A., Oms, O. & López, J.Á. 2017. An archosauromorph 519 dominated ichnoasssemblage in fluvial settings from the late Early Triassic of the 520 Catalan Pyrenees. PLoS ONE, 12(4), e0174693. DOI: 521 10.1371/journal.pone.0174693 21 522 Mujal, E., Belaústegui, Z., Fortuny, J., Bolet, A., Oms, O. & López, J.Á. 2018a. 523 Ichnological evidence of a horseshoe crab hot-spot in the Early Triassic 524 Buntsandstein continental deposits from the Catalan Pyrenees (NE Iberian 525 Peninsula). Journal of Iberian Geology, 44, 139-153. DOI: 10.1007/s41513-017- 526 0026-2 527 Mujal, E., Iglesias, G., Oms, O., Fortuny, J., Bolet, A. & Méndez, J.M. 2018b. 528 Rhynchosauroides footprint variability in a Muschelkalk detrital interval late 529 Anisian–middle Ladinian) from the Catalan Basin (NE Iberian Peninsula). Ichnos, 530 25(2-3), 150-161. DOI: 10.1080/10420940.2017.1337571 531 Mujal, E., Marchetti, L., Schoch R.R. & Fortuny, J. 2020. Upper Paleozoic to Lower 532 Mesozoic tetrapod ichnology revisited: Photogrammetry and relative depth pattern 533 inferences on functional prevalence of autopodia. Frontiers in Earth Science, 8, 534 248. DOI: 10.3389/feart.2020.00248 535 Ortí, F., Salvany, J.M., Rosell, L., Castelltort, X., Inglès, M. & Playà, E. 2018. Middle 536 Triassic evaporite sedimentation in the Catalan Basin: implications for the 537 paleogeographic evolution in the NE Iberian Platform. Sedimentary Geology, 374, 538 158-178. DOI: 10.1016/j.sedgeo.2018.07.005 539 Pickett, J.W. 1984. A new freshwater limuloid from the Middle Triassic of New South 540 Wales. Palaeontology, 27(3), 609-621. 541 Ramón, X. & Calvet, F. 1987. Estratigrafía y sedimentología del Muschelkalk inferior 542 del dominio Monserrat-Llobregat (Catalanides). Estudios Geológicos, 43, 471-487. 543 Rindsberg, A.K. & Martin, A.J. 2015. Caster’s Plasters: Neoichnological Experiments by 544 Kenneth Caster on Limulids in 1937. In ICHNOLOGY: Papers from ICHNIA III: 545 Geological Association of Canada, Miscellaneous Publications (Eds. McIlroy, D.), 546 197-210. 22 547 Romero, A. & Vía-Boada, L. 1977. Tarracolimulus rieki, nov. gen., nov. sp., nuevo 548 limúlido del Triásico de Montral-Alcover. Cuadernos de Geología Ibérica, 4, 239- 549 246. 550 Rudkin, D.M., Young, G.A. & Nowlan, G.S. 2008. The oldest horseshoe crab: A new 551 xiphosurid from Late Ordovician Konservat-Lagerstätten deposits, Manitoba. 552 Canada. Palaeontology, 51(1), 1-9. DOI: 10.1111/j.1475-4983.2007.00746.x 553 Seilacher, A. 2007. Trace fossil analysis. Berlin, Springer. 554 Shen, J., Schoepfer, S. D., Feng, Q., Zhou, L., Yu, J., Song, H., Wei, H. & Algeo, T.J. 555 2015. Marine productivity changes during the end-Permian crisis and Early Triassic 556 recovery. Earth-Science Reviews, 149, 136-162. DOI: 557 10.1016/j.earscirev.2014.11.002 558 Shi, G., Woods, A. D., Yu, M.-Y., Li, X.-W., Wei, H.-Y. & Qiao, D. 2019. Lower Triassic 559 limulid trackways (Kouphichnium) from the southwestern margin of the Yangtze 560 carbonate platform: paleoenvironmental and paleoecological implications. Palaios, 561 34(4), 229-243. DOI:10.2110/palo.2018.081 562 Shibata, M. & Varricchio, D.J. 2020. Horseshoe crab trace fossils from the Upper 563 Cretaceous Two Medicine Formation of Montana, USA, and a brief review of the 564 xiphosurid ichnological record. Journal of Paleontology, 1-19. DOI: 565 10.1017/jpa.2020.16 566 Shu, W., Tong, J., Tian, L., Benton, M.J., Chu, D., Yu, J. & Guo, W. 2018. Limuloid 567 trackways from Permian-Triassic continental successions of North China. 568 Palaeogeography, Palaeoclimatology, Palaeoecology, 508, 71-90. DOI: 569 10.1016/j.palaeo.2018.07.022 570 Shuster, C.N. Jr. 1982. A pictorial review of the natural history and ecology of the 571 horseshoe crab Limulus Polyphemus, with reference to other Limulidae. In: 23 572 Physiology and biology of horseshoe crabs. Studies on normal and environmentally 573 stressed animals (Eds. Bonaventura, J., Bonaventura, C. & Tesh, S.). New York: 574 Alan R. Liss, Inc, 1-55. 575 Sole de Porta, N., Calvet, F. & Torrento, L. 1987. Análisis palinológico del Triásico de 576 los Catalanides (NE España). Cuadernos de Geología Ibérica, 11, 237-254. 577 Vía, L. 1987. Artrópodos fósiles triásicos de Alcover-Montral. II. Limúlidos. Cuadernos 578 de Geología Ibérica, 11, 281-282. 579 Vía-Boada, L. & Villata, J.F. 1966. Heterolimulus gadeai nov. gen., nov. sp., 580 representante de una nueva familia de LIMULACEA, en el Triásico español. Acta 581 geológica hispánica, 1, 9-11. 582 Vía-Boada, L., Villalta, J.F. & Cerdá, M.E. 1977. Paleontología y paleoecología de los 583 yacimientos fosilíferos del Muschelkalk Superior entre Alcover y Mont-Ral 584 (Montañas de Prades, Provincia de Tarragona). Cuadernos de Geología Ibérica, 4, 585 247-256. 586 Virgili C., Sopeña, A., Ramos, A. & Hernando, S. 1977. Problemas de la 587 cronoestratigrafía del Trías en España. Cuadernos de Geología Ibérica, 4, 57-87. 588 589 10. TABLES 590 591 Table 1. W value (mm) I value (mm) L value (mm) Outcrop Traceway width Prosoma width Tracemaker length Penya Rubí 26.8 40.2 85.2 (this work) Catalan Pyrenees 70 (maximum) 105 222.6 24 (Mujal et al., 2018a) Daye Formation 46.9–79.2 mm 89 (average) 192.7 (average) (Shi et al., 2019) 592 593 594 Table 2. Taxon of fossil limulid Width measure data (mm) Mean width measure (mm) Tachypleus gadeai 181, 168 174.5 Tarracolimulus rieki 36.9 36.9 Mesolimulus crespelli 35 (holotype), 31, 53 39.7 Kouphichnium isp. 40.2 40.2 (IPS 110272) 595 25 596 11. TABLE AND FIGURE CAPTIONS 597 598 Table 1. Size comparison. Obtaining the width of the prosoma (I value knowing the width 599 of the traceway, W, using the formula I = W × 1.5) and the length of the tracemakers (L 600 value, using the formula L = I × 2.12) in the xiphosuran traceways from the Lower and 601 Middle Triassic. 602 603 Table 2. Width of the Middle Triassic Iberian limulid fossils (all from the upper 604 Muschelkalk of the Catalan Basin) and inferred width of the potential tracemaker for the 605 Penya Rubí traceway. Tachypleus gadeai (measures from Vía-Boada & Villata, 1966). 606 Tarracolimulus rieki (measures from Romero & Vía-Boada, 1987). Mesolimulus 607 crespelli (measures from Vía, 1987). 608 609 Figure 1. Geographic location and simplified geological maps of studied areas. a. 610 Location of the Catalan Basin in the Iberian Peninsula. b. Geological map of the Triassic 611 around Penya Rubí (Vallirana) in the Catalan Basin. Maps modified from the Institut 612 Cartogràfic i Geològic de Catalunya (ICGC). 613 614 Figure 2. Stratigraphic log of the middle Muschelkalk facies from Penya Rubí (Vallirana, 615 Catalan Basin) where tetrapod footprints and invertebrate traces appear (uppermost part 616 of the outcropping section). Md: Mudstone; Ss: Sandstone; vf:very fine; f:fine; 617 m:medium; c:coarse; vc:very coarse; Cgl: Conglomerate. 618 619 Figure 3. a. Overview of the middle Muschelkalk sedimentary succession at Penya Rubí 620 (Vallirana) outcrop. b. Detail of the stratigraphic interval where the sandstone stratum 621 preserving the traceway was found. The arrow points to the exact level. 26 622 623 Figure 4. Kouphichnium isp. (IPS110272) from the middle Muschelkalk facies of Penya 624 Rubí, advancing from left to right in the photos. a. Photo of the whole traceway. b. 625 Interpretative outline of the traceway. c. Detail of the traceway showing the linear ichnites 626 produced by appendages II-V. The arrow points to a curvilinear ichnite. d. Detail of the 627 traceway showing the arrow-shaped imprints produced by appendages VI (upper arrow), 628 and pointed, lineal and bifid- imprints produced by appendages II-V (lower arrow). e. 629 Ideal traceway using Mesolimulus crespelli (redrawn from Vía, 1987; scaled to fit the 630 traceway width) as potential tracemaker. 631 27 632 12. SUPPLEMENTARY INFORMATION 633 A three-dimensional photogrammetric model of the slab (IPS110272) that contains 634 Kouphichnium isp. is available from MorphoBank 635 http://morphobank.org/permalink/?P3770. 636 28