Upper Ordovician (Hirnantian) to Lower Silurian (Telychian, Llandovery) graptolite biostratigraphy of the Tielugou section, Shennongjia anticline, Hubei Province, China

The Tielugou section, Shennongjia Anticline, Hubei Province (China) includes a relatively complete succession of Hirnantian (latest Ordovician) to basal Telychian (Llandovery, early Silurian) graptolite faunas. The section shows the first record of a fauna of the late Aeronian Stimulograptus halli Biozone from South China, even though the index species was not reported. The Stimulograptus sedgwickii Biozone may not be represented, indicating a possible gap at the base of the Stimulograptus halli Biozone. The interval yields a number of taxa that are elsewhere reported to originate only in the Stimulograptus halli Biozone. The youngest graptolitic levels are included in the Spirograptus guerichi Biozone based on specimens of Parapetalolithus dignus and Parapetalolithus palmeus not known from earlier intervals. Spirograptus guerichi is not represented in the section. The Tielugou section provides the first detailed information on the faunas and thickness of the encountered biostratigraphic units for the Shennongija region.


Introduction
Graptolitic successions of Late Ordovician to Early Silurian age are well known from the Yangtze platform of China and a number of faunal descriptions of the Late Ordovician to Early Silurian graptolites exist (Ni 1978;Chen and Lin 1978;Fang et al. 1990;Ge 1990;Li 1995Li , 1999. Interpretations by Chen et al. (2015Chen et al. ( , 2017 and Nie et al. (2017) indicated that the Ordovician/Silurian boundary interval is strongly condensed and the thickness of some biostratigraphic intervals shows values of sometimes less than 10 cm per graptolite biozone (Chen et al. 2006). A number of successions on the eastern side of the Huangling massif ( Fig. 1a) provide detailed information pertaining to this interval. Also, the GSSP section for the international Hirnantian Stage of the Ordovician System, the Wangjiawan section, is defined in this area. According to Maletz et al. (2019), the Rhuddanian interval is condensed and incomplete in the YD-1 drill core east of the Huangling massif of Hubei Province, China, whereas an extremely thick mid-Aeronian Lituigraptus convolutus Biozone was documented. Younger graptolites have not been registered in the drill core, unfortunately. Thus, the precise biostratigraphic correlation of some intervals is uncertain. Loydell (2012) indicated that the Stimulograptus sedgwickii and Stimulograptus halli biozones of the late Aeronian have not been documented unequivocally from South China. The Spirograptus guerichi Biozone is present as the oldest Telychian time interval, whereas the Cyrtograptus sakmaricus Biozone of late Telychian is the youngest interval recognized. Most of this information, however, originated from the northern Sichuan Basin successions (Chen 1984;Ge 1990;Fu et al. 2000;Chen et al. 2003;Wang et al. 2014) and the successions in the Huangling and Shennongjia regions are less well known.
Handling Editor: Mike Reich.

3
Little information is available about the Lower Palaeozoic succession from the rim of the Shennongjia anticline (Fig. 1a). Bai et al. (2011) discussed the Huangling-Shennongjia area in the western Hubei province as part of the Yangtze continental nucleus with early Precambrian crystalline basement and a Proterozoic to Phanerozoic sedimentary cover. Ye et al. (2019) considered that the tectonic relationships between the Shennongjia terrane and the Huangling massif are unclear. Zhang and Dong (2016), however, interpreted the Shennongjia-Huangling massif as a single, rigid structure in the southern Dabashan unit. Fan et al. (2011) discussed the Qingquan section, ca. 10 km east of Shennongjia and recognized the Parakidograptus accuminatus Biozone at the base of the Lungmaxi Shale. The Lungmaxi Shale is overlain by yellowish or gray-green shales and mudstones with a Spirograptus guerichi Biozone fauna. Information on the thickness of the lithostratigraphic and biostratigraphic units was not provided and the faunas have not been described. Thus, the here reported succession of the Tielugou section provides for the first time a detailed overview of the graptolite faunal succession of the lower Silurian of the region. It shows that the succession starts at least in the Upper Ordovician Metabolograptus persculptus Biozone and ranges upwards into the Spirograptus guerichi Biozone (basal Telychian). The index species Spirograptus guerichi was not recognized, but appears to be present in the nearby Qingquan section ).

The Tielugou section
The Tielugou section (probably the Shennongjia section of Chen et al. 2014) is a roadside section on the Provincial Road S307 to the west of Shennongjia  in the Shennongjia Forestry district (listed in the UNESCO's World Network of Biosphere Reserves). The section is about 15 km to the west of the city center of Shennongjia, north of the conspicuous U-turn in the road (Fig. 1c). The section measures ca. 57 m from the Metabolograptus persculptus Biozone to the top of the accessible outcrop. The higher part of the succession consists of silty, light colored siltstones and mudstones and does not bear any graptolite faunas. More than 70 samples with graptolites have been collected from the section (Figs. 2, 3), some of which include a considerable number of identifiable specimens.  The graptolite biostratigraphy of the Tielugou section. Open circles indicate uncertain identifications due to poor preservation

Lithology
The succession was documented recognizing 114 lithostratigraphic units that were carefully collected for fossils and described lithologically in the field. A shortened report is given here for the lithologies. The strata are in part strongly weathered, but parts closer the road are less weathered due to the fairly recent construction. . Black thin-bedded carbonaceous, siliceous mudstone inter-bedded with thin-bedded carbonaceous mudstone (Fig. 2b).
CGS27-64 (13-32 m). Dark gray-black siltstone interbedded with carbonaceous mudstone and thin-bedded siltstone. Pyrite bands or nodules are present (Fig. 2a). . Black carbonaceous mudstone inter-bedded with thin-bedded siltstone or fine-grained sandstone lenses. A medium-thin-bedded (30 cm in total thickness) fine-grained sandstone at the base with plenty of scattered pyrite crystals. . Gray-dark gray silty mudstone or carbonaceous mudstone inter-bedded with thin-bedded siltstone (Fig. 2c). . Gray thin-bedded mudstone with a few very thin-bedded siltstones or small siltstone lenses (0.5-5 mm) in the lower part; dark gray to gray, thinbedded muddy siltstones inter-bedded with a few black, thinbedded carbonaceous mudstones at the middle; gray-dark gray thin-bedded mudstones in the upper part.

Graptolite faunas and preservation
The graptolite samples were taken from 20 to 50 cm thick intervals. A fairly precise graptolite biostratigraphy (Fig. 3) can be established through the samples. The biostratigraphic coverage of the samples shows that the graptolite record is very good in the lower 38 m of the section, whereas it is poor to extremely poor in the upper 20 m. Up to sample CGS76, nearly every collected sample bears graptolite specimens in various amounts, but above this level, only a single sample with graptolites has been secured (sample CGS88).
The invariably flattened graptolites are largely preserved in dark gray to black, often silty shales and siltstones and are generally easily spotted on the rock surface ( Fig. 4) due to the contrast of the silvery shining fusellum to the dark background ( Fig. 4g-i). In weathered material, the bedding surfaces are lighter colored and the graptolite fusellum appears darker (Fig. 4c). In these specimens, the fusellum is often broken into small pieces and only small amounts of the fusellum are preserved. The preservation ranges from extremely poor in some of the siltstones and silty shales to fair in finer grained sedimentary rocks. The specimens are randomly oriented on the sediment surface and distinct current orientations are rare. Problems in the species identification are based either on the poor preservation of the material or in the amount of specimens on some slabs, but also on the fragmentation of many specimens (cf. Fig. 4a, b: Lituigraptus convolutus fragments). A few layers are so crowded with graptolites that it is not possible to identify individual specimens. The graptolite specimens range in size from single siculae to complete and undistorted biserial and uniserial graptolite specimens more than 10 cm in length, but fragmented material appears to be most common.
The preservation of the graptolites indicates some tectonic distortion and thermal heating of the sedimentary rocks (cf. Hartkopf-Fröder et al. 2015;Maletz and Steiner 2015;Maletz 2020) through the generally silvery color of the graptolite fusellum in less weathered material. Thus, the material ( Fig. 4g-i) can be compared with the preservation of more strongly tectonized material from Thuringia, Germany, in which apart from the coalification, stronger tectonic distortion modifies the material (cf. Fig. 4d, e). Clearly, the Thuringian specimens of Pseudorthograptus radiculatus (Fig. 4d, e) are strongly elongated, while the specimens from the Tielugou section (Fig. 4g) show more densely spaced thecae. The color indicates graptolite reflectance reaching values of 3.0% or anchizone metamorphism (Hartkopf-Fröder et al. 2015;Maletz 2020). The fusellum is visible as a thin film of organic material, often surrounded by whitish pressure shadow minerals. In some cases, the tectonic strain is visible through parallel fractures in the fusellum of the specimens (e.g., Fig. 4h: Parapetalolithus dignus), but is difficult to observe in other specimens (Fig. 4i). This may in case provide considerable distortion of specimens and make taxonomic identification difficult. Slabs are also weathered on the surface to a variable degree, depending to the length of surface exposure of the material in the field. The strata are not horizontal, but are inclined towards the NE. Along the outcrop the inclination of the beds does not change considerably. However, no measurements are taken and further information on the tectonic situation is not available.

Repositories
All illustrated graptolites from the Tielugou section are preserved in the type collection at Nanjing Institute of Geology and Palaeontology, Academia Sinica (NIGPAS), Nanjing, China (NIGP). Additional illustrated material belongs to the type collection at Technische Universität Bergakademie Freiberg (BAF) and the Royal Belgian Institute of Natural Sciences, Brussels, Belgium (IRSNB: Manck collection).
The Metabolograptus extraordinarius Biozone of South China (cf. Chen et al. 2006) with its mixture of DDO and M graptolite faunas (see Melchin and Mitchell 1991;now Diplograptina and Neograptina: Štorch et al. 2011) has not been discovered in the Tielugou section. It appears that the characteristic Kuanyinchiao Limestone with the Hirnantian fauna separating the Upper Ordovician Wufeng Formation and the Llandovery Longmaxi (Lungmachi) Formation (see Wang 1978Wang , 1987Chen et al. 2005Chen et al. , 2006 in the region is not exposed in this section. Thus, it is unclear, whether the base of the Metabolograptus persculptus Biozone is represented in the local succession as older Ordovician graptolites have not been collected from the region (Fan et al. 2011: fig. 4). Unfortunately, the succession cannot be followed further downwards and additional sections in the region are not available. The oldest graptolites are from sample CGS1, yielding a small number of poorly preserved taxa that belong to the Metabolograptus persculptus Biozone, but most specimens are indeterminable at the species level.

Ordovician: Hirnantian
Metabolograptus persculptus Biozone (0-1.25 m). The samples CGS1 -CGS3 are here included in the Metabolograptus persculptus Biozone (Fig. 5). The faunas are extremely poor in preservation and the specimens therefore are hard to identify. It is unclear, whether the base of the Metabolograptus persculptus Biozone is represented in the section, as older faunas have not been discovered. Possible specimens of Metabolograptus persculptus (Fig. 5p) have been recognized, but good proximal ends and specimens showing thecal details are not present. The identification of this species is surprisingly uncertain as can be seen by the comparison of material from various descriptions. Štorch and Loydell (1996) discussed this species in some detail including relief material. Loxton (2017) found the species only in the lower part of his Metabolograptus persculptus Biozone and illustrated a single specimen that shows much more densely spaced proximal thecae and is not comparable with the Štorch and Loydell (1996) material. Thus, his Metabolograptus persculptus Biozone does not contain any true specimens of the index species. The material of Rickards and Riva (1981) also has to be questioned. It shows extreme tectonic deformation and is specifically indeterminable.
Normalograptus mirnyensis (Fig. 5o) is common in the interval as is Korenograptus elegantulus   (Fig. 5a, c). A single well-preserved specimen identified as Normalograptus sp. (Fig. 5b), might be referrable to Avitograptus acanthocystus (Fang et al., 1990) as described by Muir et al. (2020). A number of robust specimens are here identified as Neodiplograptus sp. (Fig. 5r), but as they do not provide any further information than the general shape of the colony, they cannot be identified to species level.
A number of slender normalograptids were identified as Avitograptus avitus (Fig. 5g, k) following the description in Melchin et al. (2011). Chen et al. (2006) identified comparable material from the Wangjianwan section as Normalograptus rhizinus. Chen et al. (2005) indicate a very short range of Normalograptus rhizinus in the top of the Metabolograptus persculptus Biozone at Wangjiawan. Better material of these taxa is necessary to support or reject the synonymy of both species and the identity of the genus Avitograptus and its various species.
Metabolograptus persculptus Biozone faunas have been described in some detail by Štorch et al. (2011) from north-central Nevada, where they show a considerable mixing of members of the Diplograptina and Neograptina. Chen et al. (2006: fig. 6) indicated the presence of Anticostia uniformis and Paraorthograptus brevispinus in the Metabolograptus persculptus Biozone, but these faunal elements were not illustrated. Previously, Chen et al. (2000) indicated the presence of the Diplograptina only reaching into the Metabolograptus extraordinarius Biozone. As no specimens of the Diplograptina were recorded in the Metabolograptus persculptus Biozone at Tielugou, a precise correlation of the interval may remain uncertain. However, diplograptine faunal elements were only discovered in some regions in the Metabolograptus persculptus Biozone, but not in others (e.g., Štorch et al. 2011). Silurian: Llandovery (Rhuddanian) Loydell (2012) differentiated four graptolite biozones in the Rhuddanian of South China, based on data in Chen (1984), Chen et al. (2003) and Fu et al. (2000). The Akidograptus ascensus and Parakidograptus acuminatus biozones are here combined, as they cannot be differentiated in the Tielugou section. Both species appear only in the higher part of the interval. The graptolite faunas around the base of the Rhuddanian have been thoroughly investigated in the last decades and a revision of the original GSSP definition was proposed differentiating an Akidograptus ascensus Biozone overlain by a Parakidograptus acuminatus Biozone in the Dobb's Linn section (Rong et al. 2008). The GSSP is defined with the FAD of Akidograptus ascensus in the stratotype section. While the Cystograptus vesiculosus Biozone is easily recognizable, it is preferred to identify the overlying interval as the Coronograptus leei Biozone and not the Coronograptus cyphus Biozone, as the latter species is not present.
The base of the Silurian is difficult to determine due to the lack of important index taxa. It is here identified at the FAD of Normalograptus anjiensis ( Fig. 5f), but Akidograptus ascensus and Parakidograptus acuminatus first appear much higher in the succession. The local FAD of Parakidograptus acuminatus is in CGS 11, while Akidograptus ascensus appears slightly lower, in CGS 09. These first appearances of the index species may be based on the poor preservation and difficulty to identify most biserial graptolites of the interval. It may also be based on insufficient material available for the investigation and the possibility remains that there is a gap at the base of the biozone. Loxton (2017) noted that Normalograptus anjiensis first appears in the Akidograptus ascensus Biozone in China and in Yukon, Canada. The species was initially described from the Akidograptus ascensus/Glyptograptus bifurcus Biozone of South China, but later found also in the Kurama range of Usbekistan (Koren' and Melchin 2000). It is similar to Normalograptus trifilis (Manck, 1923) as described by Štorch and Serpagli (1993) and Štorch and Feist (2008) and was identified as Normalograptus lubricus in the past (Loxton 2017). The Akidograptus ascensus/Parakidograptus acuminatus Biozone interval contains a number of biserial taxa, but includes also the earliest monograptids, here referred to Atavograptus sp. (Fig. 6h). The material is extremely poor and thecal details are not available. The index taxa of the interval, Akidograptus ascensus (Fig. 6i, l) and Parakidograptus acuminatus (Fig. 6j) are uncommon in the interval, but a number of easily recognizable species also appear. Poor specimens of Normalograptus trifilis (Fig. 6k) are present in sample CGS9 (4.25 m). Paramplexograptus kiliani (Fig. 6b, d) is a frequent member of the fauna, ranging through the higher part of the interval. Loxton (2017) reported the species only from the Akidograptus ascensus Biozone in Canada. Songxigraptus elongatus (Fig. 6f) is restricted to a single level, in which it is a common member of the fauna. The known material of Songxigraptus elongatus is very poor and details of the development are unknown. Maletz (2019) synonymized the genus with Talacastograptus Cuerda et al., 1988, but the latter may be considerably younger as the associated fauna indicates (cf. Melchin 2007). Loxton (2017) referred the species to the Akidograptus ascensus Biozone and did not report it from higher intervals. Neodiplograptus lanceolatus (Štorch and Serpagli, 1993) (Fig. 5d, i, l, q) is common in the Akidograptus ascensus/ Parakidograptus acuminatus Biozone. The species may be identical to Neodiplograptus shangchongensis (Li, 1984).
Slightly higher is the first appearance of the spinose Hirsutograptus sinitzini (Fig. 6g), associated with the robust Cystograptus ancestralis. Several species of Normalograptus are present in the interval, but often show only the scalariform view, making a specific identification difficult. Poor specimens of Neodiplograptus sp. and Rickardsograptus lautus ( Fig. 6c) are also present in the upper part of the interval. A number of specimens have been identified as Korenograptus lungmaensis (Sun, 1933) (Fig. 6a), but may be more closely related to Korenograptus laciniosus, based on their characteristic thecal style.
The base of the Akidograptus ascensus Biozone at the base of the Silurian System has been described in great detail from South China. The faunas are especially well documented from the GSSP of the Hirnantian at Wangjiawan, where the Akidograptus ascensus Biozone is just about 10 cm thick, while the following Parakidograptus acuminatus Biozone measures more than 1 m in thickness (e.g., Chen et al. 2005Chen et al. , 2006. Based on Wang et al. (1983) and Wang (1987), the Parakidograptus acuminatus Zone is 40 cm thick in the Wangjiawan section. Akidograptus ascensus only appears in the lower 17 cm of the zone, and Parakidograptus acuminatus in the upper 23 cm. In the Tielugou section, the combined Akidograptus ascensus and Parakidograptus acuminatus biozone interval measures about 5 m in thickness.
The Akidograptus ascensus/Parakidograptus acuminatus Biozone interval is well known from numerous sections around the world and will not be discussed here in detail. Štorch (1996) discussed the distribution in peri-Gondwanan Europe and Masiak et al. (2003) added details of the interval from the Holy Cross Mountains of Poland. Štorch and Feist (2008)  Cystograptus vesiculosus Biozone (7.3-9.8 m). The first appearance of Cystograptus vesiculosus (Fig. 7j) defines the base of the interval. The top can be seen in sample CGS20. The Cystograptus vesiculosus Biozone is a widely distributed and easily recognizable interval due to the characteristic index species with its robust tubarium and the extensive nematularium often found in this species (Fig. 7j). The interval is about 2.5 m thick in the section and includes silty black shales with randomly distributed, often large graptolite specimens. Numerous slender monograptid fragments are typical of the interval in the Tielugou section, but proximal ends have not been collected.
Cystograptus vesiculosus appears to have a fairly long biostratigraphic range, originating at the base of the Cystograptus vesiculosus Biozone and ranging into the Monograptus revolutus Biozone (Zalasiewicz et al. 2009). In the Tielugou section, slender specimens from the higher part of the Coronograptus leei Biozone have been identified as Cystograptus penna (Fig. 7c). The differentiation of both species is difficult, as it is mainly based on the dorso-ventral width of the colonies and the everted thecal apertures in Cystograptus penna (Hopkinson, 1869), which may be the result of the preservation as internal casts in relief (see Jones and Rickards 1967). Hutt (1974b: 46) considered both taxa to be extreme variants of a single species.
Bulmanograptus erectus only appears in the uppermost sample of the interval, associated with Paraclimacograptus innotatus (Fig. 7h) and early specimens of Dimorphograptoides physophora (Fig. 7g). While Paraclimacograptus innotatus has been discovered only at a single level, the other two taxa range into the overlying Coronograptus leei Biozone.
A number of robust specimens of Rickardsograptus lautus (Fig. 7i) have been identified in the interval. The species has been described from the Cystograptus vesiculosus Biozone of the Montagne Noire, but Štorch and Feist (2008: 948) indicated that the species is also present in the ascensus-acuminatus Biozone, as it is in the Tielugou section.
The Cystograptus vesiculosus Biozone can be correlated with the Atavograptus atavus and Huttagraptus acinaces biozones of Britain (Zalasiewicz et al. 2009;Loydell 2012). Štorch and Feist (2008) described the fauna of the Cystograptus vesiculosus Biozone from the Montagne Noire in some detail and provided information on the biostratigraphic ranges of a number of faunal elements. Štorch et al. (2018) discussed the upper part of the Cystograptus vesiculosus Biozone from the Czech Republic, underlain by an unconformity in the proposed GSSP section at Hlásná Třebaň. Even though enough faunas of the Cystograptus vesiculosus Biozone interval have been described worldwide (Schauer 1971;Chen and Lin 1978;Ni 1978;Wang 1987;Loydell 2007;Loydell et al. 2017), precise biostratigraphic ranges are rarely given. Bjerreskov (1997, 1999) provided information on the monograptid radiation in the Cystograptus vesiculosus Biozone using Bornholm, Denmark and the southern Urals, Russia as example. Melchin (1989) and Lukasik and Melchin (1997) provided additional information. They recognized both the Atavograptus atavus and Lagarograptus acinaces biozones. Thus, the Lagarograptus acinaces Biozone of Lukasik and Melchin (1997) is the equivalent of the upper part of the Cystograptus vesiculosus Biozone. Legrand (2003) referred the interval of the North African part of Gondwana to the Neodiplograptus africanus Biozone, but Neodiplograptus africanus appears much earlier in Jordan (Loydell 2007), indicating that there is a considerable biogeographic differentiation. Biozone (10.4-21.4 m). The base of the Coronograptus leei Biozone is found in sample CGS21 with the first occurrence of Coronograptus leei. As Demirastrites triangulatus appears in sample CGS44 (21.6 m), the top of the Coronograptus leei Biozone can be seen in CGS43 (21.4 m). The interval is identified as the Coronograptus leei Biozone following Hsü (1934) and Ni (1978). Coronograptus leei is characterized by its thecal style and differs considerably from the better-known Coronograptus cyphus. Coronograptus leei (Hsü, 1934) is widely distributed in China (see Hsü 1934;Ni 1978;Fu 1986;Fu and Song 1986;Mu et al. 2002), but has not been identified in other regions. Wang (1987) identified the interval as the Huttagraptus acinaces Biozone in the Yangtze region. Chen and Lin (1978) and Chen (1984)  Coronograptus leei (Fig. 8l) is especially common in the lower part of the interval, but a number of further species of Coronograptus occur in the zone. Coronograptus hipposideros (Fig. 8n), Coronograptus cirrus (Fig. 8m) and Coronograptus gregarius have been identified. Species of Pernerograptus and Pribylograptus are present, but the material is largely not identifiable to species level. A few poor specimens are referred to Huttagraptus acinaces. Apart from a number of taxa surviving from the Cystograptus vesiculosus Biozone (Fig. 3), a number of additional taxa appear. Rhaphidograptus minutus (Fig. 8k) is present at the base of the zone, as is Sudburigraptus sp., Agetograptus hubeiensis (Fig. 8f) and Bulmanograptus compactus (Fig. 8d) appear slightly higher in the succession. Rickardsograptus lautus (Fig. 8b) can be found through the whole interval. The robust Pseudorthograptus obuti (Fig. 8o) is found in a number of fairly well-preserved specimens. The specimens show the distinct ancora hub of the retiolitids, but the typical more complex ancora development described by Štorch (2015) is not present in any of the specimens. A single long specimen also bears a long and slender nematularium and a distally narrowing tubarium. Štorch (2015: 862) indicated a possible origin of the petalolithids from Pseudorthograptus obuti, as the very similar Petalolithus minor (Fig. 9e) from the Demirastrites triangulatus Biozone appears to be one of the earliest species of this genus and is quite similar in Fig. 7 Graptolites characteristic of the Cystograptus vesiculosus Biozone. a Bulmanograptus swanstoni (Lapworth, 1876), NIGP 173522, 7.3 m. b ?Bulmanograptus erectus (Elles and Wood, 1908), NIGP 173523, 9.8 m. c Cystograptus penna (Hopkinson, 1869) (Nicholson, 1868), NIGP 173528, 9.8 m. h Paraclimacograptus innotatus (Nicholson, 1869), NIGP 173529, 9.8 m. i Rickardsograptus lautus (Štorch and Feist, 2008), NIGP 173530, 10.4 m. j Cystograptus vesiculosus (Nicholson, 1868), NIGP 173531, 9.8 m. k Coronograptus cf. cyphus (Lapworth, 1876), NIGP 173532, 8.8 m. l Huttagraptus solidus (Štorch and Feist, 2008), NIGP 173533, 8.8 m. m Atavograptus atavus (Jones, 1909), NIGP 173534, 7.8 m. n Huttagraptus solidus (Štorch and Feist, 2008), NIGP 173535, 9.8 m. Scale bar is 1 mm ◂ shape. Petalolithus ovatoelongatus (Kurck, 1882) has a less protracted proximal end and more horizontal thecal apertures proximally, but was not found during this investigation. Dimorphograptoides physophora (Fig. 8e) extends from the Cystograptus vesiculosus Biozone into the lower part of the Coronograptus leei Biozone. It can be recognized through the short uniserial proximal end including a single theca. Elles and Wood (1908); Schauer (1971) and Koren' and Rickards (1996) described material of this species with a distinct ancora development from the Cystograptus vesiculosus to the Coronograptus gregarius Biozone of western Europe and the southern Urals. Štorch and Feist (2008) described the lowermost Silurian graptolite faunas from the Montagne Noire, France and referred Dimorphograptus swanstoni and Dimorphograptoides physophora to the higher part of the Cystograptus vesiculosus Biozone.

Coronograptus leei
The genus Coronograptus ranges from the Coronograptus cyphus Biozone to the Lituigraptus convolutus Biozone and a number of species have been differentiated (see Lukasik and Melchin 1997;Sennikov 1998;Štorch 2015). Lukasik and Melchin (1997) showed Coronograptus cyphus to be restricted to the zone bearing its name, but did not report any specimens from their samples. Štorch (1988 , Table 1) listed it from the Coronograptus cyphus Biozone. Lenz (1982) did not find Coronograptus cyphus in the Northern Canadian Cordillera. Thus, it seems that Coronograptus cyphus is rare in North America and may not be very useful as an index species in most regions. However, Melchin (1989) illustrated two fragments under this name from the Cape Phillips Formation of the Canadian Arctic Islands. Zalasiewicz and Tunnicliff (1994: 697) revised the Coronograptus cyphus Biozone of Britain and indicated that the index species ranges downwards into the Huttagraptus acinaces Biozone. The authors regarded the incoming of monograptids of the revolutus/austerus group as indicative of the base of the Coronograptus cyphus Biozone (see also Zalasiewicz et al. 2009).

Silurian: Llandovery (Aeronian)
The base of the Aeronian is defined at the base of the Demirastrites triangulatus Biozone (e.g., Štorch 1994;Melchin et al. 2012;Štorch et al. 2018), while the top is defined by the base of the Telychian at the FAD of Spirograptus guerichi. However, the Rhuddanian/Aeronian boundary is under discussion due to problems with the original GSSP section (Cocks et al. 1984;Melchin et al. 2012;Štorch et al. 2018). Melchin et al. (2018) provided a report to assess the Rheidol Gorge section as a replacement for this GSSP section, as it includes a good succession of graptolite faunas from the middle Pernerograptus revolutus or Coronograptus cyphus Biozone to the middle Demirastrites triangulatus Biozone. Štorch et al. (2018) proposed a section in the Czech Republic as a replacement of the original GSSP section. Štorch and Melchin (2019) supported the data with a documentation of the precise succession of the Demirastrites species in this section and provided an understanding of the evolution of the group.
A number of local biostratigraphic schemes have been established to subdivide the Aeronian time interval into graptolite biozones (Loydell 2012). While the Avalonia/ Baltica succession is subdivided into six biozones, the South China succession appears to be fairly incomplete. Loydell (2012: fig. 4) identified the lower part as an extensive Coronograptus gregarius Biozone interval, followed by the Lituigraptus convolutus Biozone, but he indicated that the upper part of the Aeronian has not been verified by described faunas.
The specimens here assigned to Demirastrites triangulatus (Fig. 9k) most closely match the material identified as Demirastrites triangulatus early form by Štorch and Melchin (2019). They are associated with common Rastrites guizhouensis (Fig. 9h). Other monograptids in the interval include Rastrites longispinus with its very slender metathecae (Fig. 9g), as well as fragments of Pernerograptus and slender specimens, probably belonging to Pristiograptus. Demirastrites campograptoides (Fig. 9j) has recently been described from the Demirastrites triangulatus Biozone of the Czech Republic (Štorch and Melchin 2019) and nothing is known on its further biostratigraphic and palaeobiogeographic distribution. Thus, the record in the Tielugou section adds important new information.
Specimens of Coronograptus gregarius (Fig. 9i) are not uncommon through the interval. A number of biserial species are present, including Petalolithus minor (Fig. 9e), Pseudorthograptus insectiformis (Fig. 9c), Rivagraptus sp. (Fig. 9a) and, in the higher part of the interval, Agetograptus longicaudatus (Fig. 9d) and Agetograptus primus (Fig. 10d). Pseudorthograptus radiculatus (Manck, 1918) (= Pseudorthograptus finneyi Štorch and Kraft, 2009) is common in the lower part of the interval (Fig. 4g). The earliest specimen of this species is present in the underlying Coronograptus leei Biozone at 23.1 m (CGS27) (Fig. 8i). Manck (1918) described the species from a number of well-preserved, flattened specimens found in the Demirastrites triangulatus Biozone of Thuringia, but the name has not been used since its original description. Štorch and Melchin (2019) recently discussed the biostratigraphic importance of Demirastrites triangulatus as index of the zone of its name. The authors documented in great detail the faunas from a single section in the Czech Republic and suggested evolutionary relationships or lineages of the various species recognized. It is uncertain, however, how precisely the succession can be matched in other areas. Little is known on the intraspecific variation of the species and their biostratigraphical changes, even though early and late forms of certain species have been illustrated from the Hlásná Třebaň section in the Prague Synform of the Czech Republic.
The Demirastrites triangulatus Biozone has been previously discussed for South China (Wang 1978(Wang , 1987Chen and Lin 1978;Li 1995), but illustrated specimens of Demirastrites triangulatus appear to be misidentified according to Štorch and Melchin (2019), who considered the presence of this taxon in China as unproven. The here illustrated specimens can be compared with the early form of Demirastrites triangulatus. As the species of the genera Demirastrites, Rastrites and Petalolithus appear all at the same level, there might be the chance of a condensation or a gap at the base of the Demirastrites triangulatus Biozone. In other regions like the Czech Republic and in Wales, species of Rastrites and Petalolithus appear after the FAD of Demirastrites triangulatus (Zalasiewicz et al. 2009;Štorch et al. 2018). Thus, the presence of Demirastrites triangulatus in the Tielugou section may be regarded as a late appearance.
A similar situation appears to be present in Arctic Canada. Loydell (2012) used the Campograptus curtus Biozone for the interval of the Coronograptus gregarius Biozone of South China, based on Melchin (1989). Melchin (1989) and Melchin et al. (2017) subdivided the Campograptus curtus Zone into lower and upper subzones, the upper one being the Rastrites orbitus Subzone. Recently, Melchin (in Strauss et al. 2020) recognized the Rastrites orbitus Zone as a distinct zone between the Demirastrites triangulatus and Lituigraptus convolutus biozones in Yukon, Canada.
Lituigraptus convolutus Biozone (23.7-29.8 m). The base of the Lituigraptus convolutus Biozone is defined in sample CGS48. The eponymous species is not uncommon in the interval, but complete specimens have not been recognized (Fig. 4a, b). The interval is about 7 m thick and bears a quite diverse fauna that has been documented recently by Maletz et al. (2019) from the YD-1 drill core on the eastern limb of the Huangling anticline, where the thickness of the Lituigraptus convolutus Biozone is considerably higher, measuring about 385 m.
A number of new faunal elements appear at the base of the Lituigraptus convolutus Biozone, but also quite a number of long-ranging taxa (e.g., Normalograptus scalaris, Metaclimacograptus hughesi, Pseudoretiolites sp., Korenograptus nikolayevi (Fig. 12d) extend their ranges into this interval.
Petalolithus intermedius (Fig. 11c) and Petalolithus ulstae (Fig. 11i) appear at the base of the zone as do Glyptograptus serratus (Fig. 11l), Monograptus mirificus and Pribylograptus argenteus. Petalolithus ulstae can be differentiated from other species of Petalolithus through its long sicula. It has been described from the Pribylograptus leptotheca Biozone of the Kaliningrad region, Russia (Suyarkova, 2017). Paramonoclimacis sidjachenkoi (Fig. 11n) and Metaclimacograptus sculptus (Fig. 11k) appear at the same level. Thus, a subdivision of the Lituigraptus convolutus Biozone based on these taxa (cf. Maletz et al. 2019) cannot be provided and the presence of a gap at the base of the interval cannot be excluded. Cephalograptus cometa (Fig. 11d) appears in the middle part of the zone. Petalolithus dubovikovi (Fig. 11a) and Petalolithus clandestinus (Fig. 11g) are also present in the interval. Petalolithus clandestinus was first described by Štorch (2001) from the lowermost Stimulograptus sedgwickii Biozone of the Czech Republic. It can be recognized by a robust tubarium with a very small ancora and everted, ventrally facing apertures. The similar Parapetalolithus globosus (Fig. 13b) can be distinguished by its simple virgella. Petalolithus clandestinus is not uncommon in the higher part of the Lituigraptus convolutus biozone in the Tielugou section.
A gap in the Llandovery graptolite succession is known in southern Scandinavia (Bjerreskov 1975;Loydell et al. 2017;Walasek et al. 2018) and possibly in China (see Loydell 2012), but more complete successions across the Stimulograptus sedgwickii and Stimulograptus halli biozones can be found in the Southern Uplands of Scotland and in further regions worldwide as discussed by Loydell et al. (2015).
Stimulograptus halli Biozone (30.3-36.7 m). Loydell (2012: fig. 4) questioned the presence of the late Aeronian Stimulograptus sedgwickii and Stimulograptus halli biozones in South China, but did not provide details for this interpretation. Wang (1987) reported a Stimulograptus sedgwickii Biozone that was 138.9 m thick in the Wangjiawan section. The fauna was not described and the single fragment illustrated as Stimulograptus sedgwickii is unidentifiable. Stimulograptus sedgwickii (Fig. 12s-u) is common at Tielugou, often preserved as long straight fragments. The interval, however, might be referred to the Stimulograptus halli Biozone based on the record of Pristiograptus xiushanensis (see Loydell et al. 2015: 780) and especially of the genus Parapetalolithus first appearing in the Stimulograptus halli Biozone (Lenz et al. 2018).
Stimulograptus sedgwickii is commonly associated with Pristiograptus regularis (Fig. 12q-r) and a number of biserial graptolites in the interval, but Stimulograptus halli has not bee recognized so far. Several species of Parapetalolithus can be found in the interval, of which Parapetalolithus mui (Fig. 13a, h) and Parapetalolithus fusiformis (Figs. 12b, 13g) may be the most common species. A few specimens of Parapetalolithus sp. A (Fig. 12f) and Parapetalolithus sp. B (Fig. 12g) were found in the middle of the interval. The specimens strongly remind of the genus Cephalograptus, but their proximal ends are less strongly elongated. They do not possess an ancora and a closer relaitonship to the genus Cephalograptus may not be reasonable to suggest. Similarities can be seen to Parapetalolithus aknisos Loydell et al., 2015 with its elongated, slender and low inclined thecae, but the specimens show a distinct distal widening of the tubarium and certainly not represent this species.
Funding Open Access funding enabled and organized by Projekt DEAL.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.