Comparison with other morphologically similar Arthropoda

Based on the morphological characters observed in Duraznovis gallegoi n. gen. n. sp. (Figs. 3, 4), affinities to either Xiphosura Latreille, Reference Latreille1802 (Chelicerata) (= Merostomata Dana, Reference Dana1852) and Notostraca Sars, Reference Sars1867 (Crustacea, Branchiopoda) can be inferred.

Duraznovis gallegoi n. gen. n. sp. (Figs. 3, 4) could be placed close to xiphosurans (Fig. 5.3) by the presence of a body differentiated into tagmatas, terminating in a long, styliform tail-like structure (Fig. 5.1, 5.2, Ta); large shield (Fig. 5.1, 5.2, S); and six thoracic segments (Fig. 5.1, 5.2, se) that probably bore six pairs of ‘appendages,’ ventrally attached to the shield (Dunlop and Selden, Reference Dunlop, Selden, Fortey and Thomas1997) (Fig. 5.3). The anterior region (Fig. 5.1, 5.2, An) and posterior region (Fig. 5.1, 5.2, Po) in Duraznovis gallegoi n. gen. n. sp. could be considered analogous to the ‘prosomal’ dorsal shield (cephalothorax) and abdomen (opisthosoma/thoracetron) in xiphosurans, respectively (Fig. 5.3). The shield (Fig. 5.1, 5.2, S) in both xiphosurans and Duraznovis gallegoi n. gen. n. sp. has a semicircular anterior margin and lateral margins oblique to the median axis, covering the anterior region (Fig. 5.1, 5.2, An), but in Duraznovis gallegoi n. gen. n. sp., the shield is guitar-shaped, whereas in xiphosurans it has the shape of a horseshoe (Fig. 5.3). The holotype specimen (IANIGLA-PB 203, Figs. 3, 5.1) has an irregularity on the left margin of the shield (Fig. 5.1, c), like the cephalothoracic injuries found in extant xiphosurans (see Bicknell et al., Reference Bicknell, Pates and Botton2018, fig. 2.5). Such injuries could be self-inflicted (from mating or predation), teratological (developmental and genetic), or pathological (disease or parasites). In both of our specimens, the impression of two undetermined structures are visible on the anterior section (Fig. 5.1, 5.2, a), similar to xiphosuran compound eyes (Fig. 5.3, ce), but these are too bulbous, not associated with an ophthalmic ridge, and differ in their position, and therefore are here left as undetermined.

Possible anterior muscle structures (Fig. 5.1, d) are observed in the holotype specimen, resembling those described in Yunnanolimulus luopingensis Zhang et al., Reference Zhang, Hu, Zhou, Lv and Bai2009 (see Hu et al., Reference Hu, Zhang, Feldmann, Benton, Schweitzer, Huang, Wen, Zhou, Xie, Lü and Hong2017, fig. 4c, e), a xiphosuran from the Middle Triassic of China. A central area in Duraznovis gallegoi n. gen. n. sp. (Fig. 5.1, 5.2, b) resembles that of xiphosurans in the presence of six segments (Fig. 5.1, 5.2, se) that are probably related to the walking appendages (Fig. 5.3, app). The posterior region (Fig. 5.1, 5.2, Po) in Duraznovis gallegoi n. gen. n. sp. can be compared with the opisthosoma of xiphosurans (Fig. 5.3), however in Duraznovis gallegoi n. gen. n. sp., this region is more elongated than in xiphosurans, which could be a consequence of postmortem deformation. In both the holotype and paratype specimens, there is a broad area similar to the opisthosomal margin of xiphosurans (Fig. 5.1–5.3, opm). The abdomen of Duraznovis gallegoi n. gen. n. sp. has a broad area (Fig. 5.1, 5.2, e) that could be interpreted as an equivalent to the genital operculum of xiphosurans (Fig. 5.3, gen. op.). Following, an area transversally crossed by numerous lineations could be indicating the position of the gills (Fig. 5.1, 5.2, f, 5.3, gill.op.). The distal placement of the gills in our specimens is in accordance with Lamsdell (Reference Lamsdell2013), who stated that in aquatic chelicerates, the opisthosomal appendages are modified into flattened opercula that bear the respiratory organs (‘book gills’). The posteriormost part of the abdomen in Duraznovis gallegoi n. gen. n. sp. ends in a heart-shaped plate with sharp to rounded lateral extensions (Fig. 5.1, 5.2, g), somewhat comparable to those observed in xiphosurans (Fig. 5.3, po. part). Lastly, the fossil specimens have long, styliform tail-like structure (Fig. 5.1, 5.2, Ta) analogous to the telson of xiphosurans (Fig. 5.3), although in Duraznovis gallegoi n. gen. n. sp., the tail is divided distally in holotype IANIGLA-PB 203, and more anteriorly in paratype IANIGLA-PB 168. The asymmetrical split occurring at different lengths of this structure in the two fossil specimens suggests that it could be taphonomic breakage rather than a taxonomic feature. Similarly, divided telsons in extant xiphosurans have been reported as teratological abnormalities (see Bicknell et al., Reference Bicknell, Pates and Botton2018, fig. 1), so this possibility should not be ruled out.

The main difference observed between Duraznovis gallegoi n. gen. n. sp. and other xiphosurans is their body size (7–10 mm), being smaller than all other known Triassic specimens, although not markedly differing from ‘Paleolimulus’ fuchsbergensis Hauschke and Wilde, Reference Hauschke and Wilde1987 from the Upper Triassic of northern Germany (12 mm long). The remaining species are considerably larger (21–146.5 mm), i.e., Dubbolimulus peetae Pickett, Reference Pickett1984 from Middle Triassic of Australia (21 mm); and Austrolimulus fletcheri Riek, Reference Riek1955, Middle Triassic of Australia (146.5 mm). The variation of size observed in fossil xiphosurids could be explained by the great morphological disparity that they reached during Triassic times (Bicknell et al., Reference Bicknell, Žalohar, Miklavc, Celarc, Križnar and Hitij2019). Other typical morphological structures of xiphosurans that are not visible in our specimens are the opisthosomal spines or the genal spines (Fig. 5.3) (Dunlop and Selden, Reference Dunlop, Selden, Fortey and Thomas1997; Lamsdell, Reference Lamsdell2013). Nevertheless, the absence of ornamentation in our specimens could be due to a taphonomic issue, because spines and other structures are most likely to be lost during the process of fossilization (Bicknell et al., Reference Bicknell, Pates and Botton2018).

At a different level, Duraznovis gallegoi n. gen. n. sp. presents a ‘notostracan-like’ (Crustacea, Branchiopoda) overall appearance (Figs. 3–5), e.g., the presence of a large shield (Fig. 5.1, 5.2, S) that is strongly drawn out laterally and posteriorly; an elongate trunk that carries the appendages (Fig. 5.1, 5.2, 5.4, app); and a long, tail-like structure analogous to the caudal furca of some members of the extant family Triopsidae Keilhack, Reference Keilhack and Brauer1909 (Fig. 5.1, 5.2, Ta; 5.4). At closer look, however, notostracans have an elongated shield with a rounded-acute proximal margin (Fig. 5.4), differing from the guitar-shaped shield of Duraznovis gallegoi n. gen. n. sp. Another difference between these specimens and notostracans refers to total size; extant adult notostracans range from 1–10 cm in length (Rogers, Reference Rogers and Likens2009). The ventral organization also shows variations; for example, the number of appendages on the thorax of both fossil and extant notostracans ranges from 10–80 (Rogers, Reference Rogers and Likens2009; Lagebro et al., Reference Lagebro, Gueriau, Hegna, Rabet, Butler and Budd2015), markedly contrasting from the potential presence of only six pairs in Duraznovis gallegoi n. gen. n. sp. Additionally, the abdomen in notostracans is composed of several segments, surpassing the posterior limit of the protective shield and leaving the more distal segments exposed (Fig. 5.4, h). In comparison, the abdomen of Duraznovis gallegoi n. gen. n. sp. is shorter and almost completely covered by the shield. However, Hegna (Reference Hegna2012) suggested that some abdominal segments in notostracans can be lost during the fossilization process, or potentially, shortened by desiccation. Because both of our fossil specimens are complete and articulated, we consider this as highly unlikely. On the other hand, Wagner et al. (Reference Wagner, Haug, Sell and Haug2018) mentioned that a short trunk is probably related with early juvenile notostracans stages. Unfortunately, the developmental stages of our specimens are hard to tell from the fossils, because no gonads are recognized, and more specimens are needed to show size variations that could clarify this point. Lastly, the long, styliform tail-like structure visible in Duraznovis gallegoi n. gen. n. sp. (Fig. 5.1–5.2, Ta) is centrally inserted to the heart-shaped plate (Fig. 5.1, 5.2, f), whereas in notostracans, the ‘tail’ is composed of two caudal furcas laterally attached to a telson (Fig. 5.4).

In summary, these fossils show a combination of features found in two disparate arthropod groups, i.e., Xiphosura and Notostraca, making correct taxonomic allocation difficult. Therefore, until additional, better-preserved specimens are recovered, which can shed light about development stages, reproductive structures, or even clarify the still-undefined structures mentioned herein, we prefer to maintain their systematic classification as indeterminate.

Taphonomic and paleoenvironmental implications

Duraznovis gallegoi n. gen. n. sp. comes from Facies Association A2 described by Lara et al. (Reference Lara, Cariglino and Zavattieri2017), which refers to ponds or swamps developed within the shallow embayments during deltaic flooding pulses in the lower delta plain. These ponds never dried out completely, as suggested by the absence of desiccation cracks, but their shallow, usually submerged substrates were probably episodically exposed, as the presence of incipient paleosols indicate. Moreover, numerous Euestheria forbesi (Crustacea: Spinicaudata) specimens (Figs. 3, 4, spin.) recorded in this facies present wrinkles in their valves, producing desiccation patterns similar to those described by Astrop (Reference Astrop2014) for the upper Permian spinicaudatan Megasitum harmonicum Novojilov, Reference Novojilov1970 from Kazakhstan.

The material studied herein was found associated with an abundant well-preserved flora, a rich fauna of spinicaudatan crustaceans (Figs. 3, 4, spin.), numerous diverse insects, and a few fish remains (Lara et al., Reference Lara, Cariglino and Zavattieri2017). The low abundance of these enigmatic arthropods at the Quebrada del Durazno locality might be related to the presence of fishes in the water pools, which probably acted as top predators in these habitats. Other explanations could be that the high abundance of spinicaudatans inhibited the diversification of the group by competition for the same resources; or that during extreme flooding events, these specimens were transported accidentally from farther areas. Finally, we do not rule out a potential collection or taphonomic bias.

The faunistic associations of spinicaudatans with notostracans, as well as spinicaudatans with xiphosurans, have been previously recorded from other Triassic localities worldwide. The former association was reported for ephemeral pool communities from the Middle Triassic of the Grès à Voltzia Formation in France (Gall and Grauvogel-Stamm, Reference Gall and Grauvogel-Stamm2005), whereas taphocenosis with both xiphosurans and spinicaudatans have been described from freshwater to brackish-water environments from the Middle Triassic of southern Tunisia (Błażejowski et al., Reference Błażejowski, Niedźwiedzki, Boukhalfa and Soussi2017).

In addition to the similar morphological characteristics mentioned above, Duraznovis gallegoi n. gen. n. sp. was found in a paleoenvironment not very different from those registered for notostracans and xiphosurans. With a fossil record starting in the Devonian, notostracans have developed a wide range of forms that inhabit almost exclusively temporary freshwater bodies on all continents (including Antarctica) (Gueriau et al., Reference Gueriau, Rabet, Gaël, Lagebro, Vannier, Derek, Briggs, Charbonnier, Olive and Béthoux2016), being occasionally found in marine environments (Adamowicz and Purvis, Reference Adamowicz and Purvis2005; Olesen, Reference Olesen2007; Hegna, Reference Hegna2012; Rogers and Padhye, Reference Rogers and Padhye2015). On the other hand, the record of xiphosurans extends to the Lower Ordovician (Van Roy et al., Reference Van Roy, Orr, Botting, Muir, Vinther, Lefebvre, el Hariri and Derek2010). This group of arthropods is nowadays found living in marginal environments, and the same is observed for their fossil counterparts, recovered from marginal marine to brackish paleoenvironments (Fisher, Reference Fisher and Nitecki1979; Babcock et al., Reference Babcock, Merriam and West2000), and more rarely from freshwater settings (Pickett, Reference Pickett1984; Błażejowski et al., Reference Błażejowski, Niedźwiedzki, Boukhalfa and Soussi2017). Lamsdell (Reference Lamsdell2016) suggested that xiphosurans have independently colonized the nonmarine realm many times during their evolutionary history, like Psammolimulus gottingensis Lange, Reference Lange1923, from the Triassic of Germany, which inhabited freshwater floodplain habitats along braided river systems (Lerner et al., Reference Lerner, Lucas and Lockley2017). Therefore, the freshwater paleoenvironment at the Quebrada del Durazno locality would not be a barrier to finding xiphosurans. If in the near future new fossils validate Duraznovis gallegoi n. gen. n. sp. as belonging to Xiphosurida, this site will add new evidence of freshwater xiphosurans as a result of the post-Permian extinction environmental pressure (Foster and Twitchett, Reference Foster and Twitchett2014; Lamsdell, Reference Lamsdell2016; Błażejowski et al., Reference Błażejowski, Niedźwiedzki, Boukhalfa and Soussi2017).

The complete, articulated state and the quality preservation of Duraznovis gallegoi n. gen. n. sp. suggest that the specimens were quickly covered by sediments, protecting them from postmortem processes and/or biological attacks. An analogous example was given by Lagebro et al. (Reference Lagebro, Gueriau, Hegna, Rabet, Butler and Budd2015), who mentioned that the lack of damage in fossil notostracans implies in situ deposition where little or no scavenging or turbulence occurred. Additionally, Hegna (Reference Hegna2012) evaluated how notostracans tolerated disarticulation and fragmentation through decay experiments under laboratory conditions (euthanized by asphyxiation, covered with freshwater in a tray at room temperature; some specimens were dried to assess the effects of desiccation) showing that after 16 days from their death, no fully articulated animals remained. For the xiphosurans, Babcock et al. (Reference Babcock, Merriam and West2000) mentioned that only under exceptional circumstances (i.e., Lagerstätten deposits), the cuticular exoskeleton and disarticulation of body parts will remain relatively low. Likewise, the remarkable preservation as a 3D mold of the holotype of Duraznovis gallegoi n. gen. n. sp. suggests a different taphonomic history from spinicaudatans recorded at the same site (Figs. 3, 4, spin.). The latter taxa are represented by isolated and poorly preserved prints of valves (dorsal shields), which probably corresponded to postmortem disarticulated bodies, deposited in a water depression. The small size observed in Duraznovis gallegoi n. gen. n. sp. points to a juvenile instar stage rather than a mature adult, although this cannot be fully ascertained until we have more specimens that help us to understand their complete life cycle. For instance, Błażejowski (Reference Błażejowski, Carmichael, Botton, Shin and Cheung2015) mentioned that juvenile xiphosurids spend their first years in shallow waters, rapidly increasing in size before moving to deeper waters. Unfortunately, ontogenetic changes in the fossil record are enormously difficult to observe from a limited number of specimens (Allen and Feldmann, Reference Allen and Feldmann2005).