Diverse earliest Triassic ostracod fauna of the non‐microbialite‐bearing shallow marine carbonates of the Yangou section, South China XINCHENG QIU, LI TIAN , KUI WU, MICHAEL J. BENTON, DONGYING SUN, HAO YANG AND JINNAN TONG Qiu X., Tian L., Wu K., Benton M.J., Sun D., Yang H. & Tong J. 2019: Diverse earliest Triassic ostracod fauna of the non‐microbialite‐bearing shallow marine carbonates of the Yangou section, South China. Lethaia, Vol. 52, pp. 583–596. Since diverse ostracod faunas in the immediate aftermath of the latest Permian mass extinction are mainly found within Permian–Triassic boundary microbialites (PTBMs), the idea of an ostracod ‘microbial‐related refuge’ has been proposed. Here, we report a diversified earliest Triassic ostracod fauna from the Yangou section in South China, where no PTBMs were deposited, providing evidence inconsistent with this ‘microbial‐ related refuge’ hypothesis. In addition, a significant ostracod extinction is recorded, corresponding with the earliest Triassic mass extinction (ETME). This ETME of ostra- cods is associated with size increases and a length/height ratio (L/H) decrease, indicat- ing varied evolutionary patterns of shape and size of ostracods through the Permian– Triassic (P‐Tr) extinction events. Although the nature of these biotic changes is some- what unclear, the temporally varied ‘refuge zone’ scenario provides us with a window to reconstruct the environmental dynamics of ecosystem changes during the P‐Tr transition. □ Body size, Hollinella, marine ecosystem, mass extinction, recovery, refuge zone. Xincheng Qiu [[email protected]], Li Tian ✉ [[email protected]], Kui Wu [kuiwu@ cug.edu.cn], Hao Yang [[email protected]], and Jinnan Tong ✉ [jntong@ cug.edu.cn], State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan 430074, China; Li Tian [[email protected]], and Michael J. Benton [[email protected]], School of Earth Sciences, University of Bristol, Bristol BS8 1 RJ, UK; Dongying Sun [[email protected]], College of Tourism and Territorial Resources, Jiujiang University, Jiujiang 332005, China; manuscript received on 13/09/2018; manuscript accepted on 8/03/2019. As one of the major components of the marine ben- 1978; Kozur 1991; Crasquin et al. 2007a,b; Yuan thic ecosystem, ostracods were heavily affected by the 2008; Crasquin & Forel 2014; Chu et al. 2015; Forel Permian–Triassic (P‐Tr) crisis, in which over 90% of et al. 2015a). Dozens of publications have reported marine invertebrate species and 70% of terrestrial P‐Tr ostracod faunas in various facies, including ner- animals became extinct (e.g. Erwin 1994; Benton & itic clastic (Hao 1994), shallow carbonate (e.g. Cras- Twitchett 2003; Algeo et al. 2011; Shen et al. 2011; quin et al. 2005, 2006, 2007b, 2018; Forel et al. 2009, Payne & Clapham 2012; Benton & Newell 2014). The 2013a,b, 2015a,b), and slope and deeper facies (Yuan extinction rates of the latest Permian ostracod faunas et al. 2007; Crasquin et al. 2010). exceeded 70% and may be as high as 100% in some The exceptionally high abundance and diversity of localities (Crasquin & Forel 2014) during the major ostracods in Permian–Triassic boundary micro- phase of the P‐Tr crisis, corresponding to the latest bialites (PTBM) were thought to be evidence for a Permian mass extinction (LPME) of Song et al. ‘microbial‐related refuge’ scenario in the immediate (2013). aftermath of the LPME (Forel et al. 2009, 2013a,b; Many researchers have attempted to explore sur- Forel 2012). However, here we present a newly dis- vival and recovery patterns by studying well‐pre- covered diversified ostracod fauna in the earliest Tri- served fossil faunas of the extinction aftermath in the assic non‐microbialite, shallow marine carbonates of Early Triassic (e.g. Benton et al. 2004, 2013; Twitch- the Yangou section, as a means of testing the ‘micro- ett et al. 2004; Brayard et al. 2010, 2017; Song et al. bial‐related refuge’ scenario. Meanwhile, temporal 2011; Hautmann et al. 2015; Godbold et al. 2017; Ji diversity and abundance changes as well as size vari- et al. 2017; Zhang et al. 2017). Ostracods have also ations are analysed to explore the impact on shallow attracted great research interest because of their wide marine ostracods of the second phase of the P‐Tr cri- habitat ranges and ecological significance (e.g. Wang sis, the earliest Triassic mass extinction (ETME; DOI 10.1111/let.12332 © 2019 Lethaia Foundation. Published by John Wiley & Sons Ltd 584 Qiu et al. LETHAIA 52 (2019) identified by Song et al. (2013)). Furthermore, our outcrop exposure for cement quarrying was biostrati- quantitative data also show some inconsistent results graphically constrained at decimetre‐level resolution with the newly established ontogeny model of Holli- (Fig. 3) using conodonts by Sun et al. (2012). Since nella lungcamensis (Crasquin et al. 2018), providing then, foraminiferal extinction (Tian et al. 2014b), additional evidence for systematic re‐evaluation of carbon isotope variation (Song et al. 2012), facies Hollinella. and microfacies diagnoses (Tian et al. 2014a, 2015), elemental compositions (Li et al. 2017) and the car- bonate diagenesis history (Li & Jones 2017) have Geological setting and methods been explored in detail in the Yangou study section. Field sampling and bulk sample dissolution were The Yangou section was located on the northern conducted for study of the ostracods and conodonts, margin of the Yangtze Platform during the P‐Tr as introduced in detail by Sun et al. (2012). The transition (Fig. 1), in the eastern‐most Palaeotethys acetic acid we used was at 10% concentration, and realm (Fig. 1). The studied P‐Tr successions are the use of dilute acid has ensured that dozens of composed of the Upper Permian Changxing Forma- exceptionally well‐preserved specimens, like Holli- tion and the Lower Triassic Daye Formation, with nella (Fig. 4A–H), were retained. All specimens are continuous deposition of carbonates (Fig. 2). The deposited in the collections of the State Key Labora- Changxing Formation is dominated by medium‐ to tory of Biogeology and Environmental Geology, thick‐bedded limestones, overlain by the Daye For- China University of Geosciences (Wuhan). The col- mation and its thin‐ to medium‐bedded limestones lections have been numbered in this way: LY(section and muddy limestones. Two oolitic limestone beds name)**(bed number)+**(sampling number)i** were found at the base of the Daye Formation, (ostracod specimen number). In all specimens, we implying shallow seawater level (Tian et al. 2014a,b). measured the length (L) and height (H) for quantita- These lithologies and microfacies analyses suggest a tive size analyses. Size variations are shown in plotted platform‐margin facies (Fig. 3; Tian et al. 2014a). boxes including length, height, Log(length × height). The study interval can be well constrained by The specimens belonging to Hollinella were plotted other well‐studied P‐Tr sections in South China, both in a height/length diagram. lithologically and biostratigraphically (Song et al. 2012; Sun et al. 2012; Tian et al. 2014a, 2015), pro- viding a solid stratigraphical basis for further Results and discussion palaeontological study and correlations. Ostracods and other fossils of the P‐Tr successions in the study In total, 238 well‐preserved ostracod specimens were areas were initially reported by Zhu (1999), but the found from nine sampling horizons (Fig. 3). These results were unreliable in the absence of fossil pho- ostracods are all from the wackestones of the Daye tographs and detailed descriptions. Renewed Formation (Fig. 2B, C; diagnosed as restricted inner‐ A B Fig. 1. Palaeogeographical maps of the global plates (A) and South China Block (B) during the Permian–Triassic transition. Map B shows the study section location (star) on the Yangtze Platform: YG, Yangou. The GSSP Meishan section is also marked by MS. The South China Block has rotated nearly 90° clockwise since the Permian–Triassic. Map A is adapted from Ron Blakey (<http://jan.ucc.nau.edu/rcb7/>), while map B is modified after Yin et al. (2014). [Colour figure can be viewed at wileyonlinelibrary.com] LETHAIA 52 (2019) Earliest Triassic ostracod fauna of Yangou 585 A B C Fig. 2. Photography of the macro- and micro- sedimentary compositions. A, field photograph of the study section and ostracod‐enriched intervals. B, C, photographs of thin sections. Os, ostracods. [Colour figure can be viewed at wileyonlinelibrary.com] shelf shallow marine facies by Tian et al. 2014a) that H. panxiensis (Wang 1978), according to the descrip- lack microbial‐related structures in the matrix (Fig. 2 tions of these two species by Wang (1978). In the B, C). In total, 34 species belonging to eight genera of review of the history of H. tingi by Crasquin et al. ostracods have been identified and measured (Figs 4, (2018), it is very difficult to trace the real H. tingi 5; Table 1). from the indefinite descriptions by Patte (1935) and Hou (1954). Crasquin et al. (2018) interpreted all New materials for Hollinella classification specimens assigned to H. tingi to be H. panxiensis, since Wang (1978) established H. panxiensis with The ostracod Hollinella is often present in many P‐ clear descriptions and he indicated differences from Tr boundary sections globally (Crasquin et al. 2018). H. tingi. However, a long dorsal spine and an obtuse However, especially for Hollinella tingi, a proposed spine reaching over the dorsal margin were noted for biostratigraphical index fossil for the P‐Tr boundary H. panxiensis and H. tingi, respectively, by Wang (Kozur 1985), the systematic classification is con- (1978). In the absence of such spines in our Holli- fused (Crasquin et al. 2018). Morphological varia- nella sp. 1 (Fig. 4A–D), it cannot be either H. panx- tions of Hollinella specimens are presented here as a iensis or H. tingi. means of testing aspects of their systematic classifica- Hollinella sp.