06

Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39 – 55 www.elsevier.com/locate/palaeo

Conodont biostratigraphy across the Permian–Triassic boundary at Chaotian, in Northern Sichuan, China Zhansheng Ji a,⁎, Jianxin Yao a , Yukio Isozaki b , Tetsuo Matsuda c , Guichun Wu a a b

Institute of Geology, Chinese Academy of Geological Sciences, Beijing, China Department of Earth Science and Astronomy, The University of Tokyo, Japan c Kyoei Consultant Corporation, Toyama, Japan Accepted 30 November 2006

Abstract Detailed conodont biostratigraphy of the Permian–Triassic boundary (PTB) interval was analyzed at the fossiliferous Chaotian section in northern Sichuan, China. The PTB interval at Chaotian is composed of the uppermost Permian mudstone/limestone of the Dalong Formation and the lowermost Triassic marl/limestone of the Feixianguan Formation both representing deep basinal facies. The PTB interval is enriched in ammonoids and conodonts, and is divided into 6 conodont zones including 1) Clarkina changxingensis changxingensis–C. subcarinata Assemblage Zone, 2) C. postwangi–C. sp. B Assemblage Zone, 3) C. taylorae–C. zhejiangensis–C. changxingensis yini Assemblage Zone, 4) barren intra-zone, 5) Hindeodus parvus Zone, and 6) Isarcicella Zone in ascending order. The first four zones correspond to the Changhsingian, whereas the latter two to the Lower Induan (Griesbachian). The first occurrence (FO) of H. parvus in the lowermost Feixianguan Formation is regarded as the conodont biostratigraphic boundary. From the uppermost Dalong Formation, various taxa of the genus Clarkina occurred abundantly, and suddenly disappeared at the top of the Dalong Formation, therefore, the top of the Dalong Formation is interpreted as a horizon of the eventostratigraphic boundary. On the basis of conodont zones, the Chaotian section is correlated with the Shangsi and Meishan (GSSP of the PTB) sections in South China, and the Zal section in Iran. © 2007 Elsevier B.V. All rights reserved. Keywords: Permian–Triassic boundary; Conodont; Clarkina; Hindeodus parvus; South China

1. Introduction The Permian–Triassic boundary (PTB) mass extinction was the largest in magnitude in the Phanerozoic (Sepkoski, 1984; Erwin, 1993; Hallam and Wignall, 1997). Conodont animals also suffered from the PTB mass extinction, but they survived into the Triassic (Orchard and Krystyn, 1998). Conodonts are one of the most important fossil taxa for dating and high-resolution ⁎ Corresponding author. E-mail address: [email protected] (Z. Ji). 0031-0182/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2006.11.033

correlation of the PTB interval (Sweet, 1970a,b; Matsuda, 1981; Clark et al., 1986; Yin et al., 1986; Wang, 1994, 1995; Yin et al., 1996; Mei et al., 1998; Kozur, 1998; Jin et al., 2000; Kozur, 2004; Yin et al., 2001a,b). The conodont biostratigraphy around the PTB has been studied in detail in several well-known sections in the world, e.g., Transcaucasia and Iran, Southern Alps, Salt Range in Pakistan, Guryul Ravine in Kashmir, Tibet, and South China, as compiled in Erwin (1993) and Hallam and Wignall (1997). Fundamental contributions for establishing the conodont biostratigraphic framework of the PTB was made by some pioneers,

40

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

such as Sweet (1970a,b), Kozur et al. (1975), Kozur and Pjatakova (1976) and Matsuda (1981), and refinement at a higher resolution followed afterwards (e.g., Mei et al., 1998; Orchard and Krystyn, 1998; Nicoll et al., 2002; Kozur, 2004). Among the previously described areas, South China appears to be an ideal area for the PTB research because of the rich conodont data, and more than 50 PTB sections described to date (e.g., Yang et al., 1987). In South China, the conodont zones have been established in shallow marine limestone-dominant facies referred to the Upper Permian Changxing Formation (Fm). The major studied sections include the well known Meishan section in Zhejiang province (Zhao et al., 1981; Wang and Wang, 1981; Wang and Chen, 1983; Yang et al., 1987; Zhang, 1987; Wang, 1994, 1995; Zhang et al., 1995, 1996; Mei et al., 1998; Nicoll et al., 2002; Wu et al., 2003); The section D in Meishan was selected as the Global Stratotype Section and Point (GSSP) of the PTB (Yin et al., 2001a,b). On the other hand, South China has other types of PTB sections of a much deeper basinal facies with limestone, siliceous limestone and shale referred to the Upper Permian Dalong Fm (Yang et al., 1987). Such depositional sites, in general, are less affected by subtle sea-level changes and thus have better chance to preserve a continuous succession, however, they have been less studied in detail compared with the shallow marine facies. In this regard, the Shangsi section in Sichuan is the only one that has been studied in detail for the PTB conodont biostratigraphy (Zhang et al., 1984;

Yang et al., 1987; Li et al., 1989; Ding, 1992; Nicoll et al., 2002). However, the originally described section along a local road was completely covered by farmer's construction in the 1990s. As the continuity of the strata near the PTB in the Meishan section was questioned recently by several workers (Metcalfe et al., 2001; Nicoll et al., 2002; Kozur, 2004), more detailed conodont biostratigraphy in the basinal facies is definitely needed in the PTB study in South China. The Chaotian section in northern Sichuan is a good research target of such a deep basinal facies in this respect as pointed out by Isozaki et al. (2007-this issue), and in our research since 1998, we obtained a good conodont succession associated with ammonoids across the PTB. In this paper, we report the conodont biostratigraphy of the PTB section in the Chaotian section, and make correlation with other representative PTB sections in South China and Iran. 2. Geologic setting During the Late Permian, South China was isolated from other continents the eastern opening of Tethys around the equator (Yang et al., 1987), accommodating an extraordinarily thick pile of Paleozoic sedimentary rocks. Sedimentological analyses clarified that northern Sichuan was in a basin facies in the Changhsingian, surrounded by much shallower carbonate platforms (Zhu et al., 1999). In the Early Triassic, the sea level dropped to change the deep-water basin to shallow-water near-shore environments (Li et al., 1989). Nonetheless, this area has kept a

Fig. 1. Index map of the study section at Chaotian in northern Sichuan, China.

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

well-preserved continuous marine succession spanning from the Middle Permian to the Lower Triassic. The studied section at Chaotian is located in northern Sichuan, ca. 30 km to the north of the Guangyuan city (Fig. 1). This area corresponds to the northwestern margin of the Yangtze platform, immediately to the south of the Qinling collisional suture. The Shangsi section (Li et al., 1989), that was once proposed as a GSSP candidate, is located about 60 km to the west of the Chaotian section and belongs to the same geotectonic domain on the Yangtze Craton. Due to the deposition in deep-water facies, the Chaotian section yielded various ammonoid and conodont faunas, allowing preliminary paleontological studies by several workers (Zhao et al., 1978, 1981; Yang et al., 1987; Li et al., 1989; Lai et al., 1996). Lately, Isozaki et al. (2004) re-studied the Chaotian section, focusing on the PTB mass extinction event, and added more data of ammonoids and conodonts. The Chaotian section consists of the Middle to Late Permian and the overlying Lower Triassic rocks. These rocks are subdivided into 4 formations; i.e., the Maokou Fm (Guadalupian), Wujiaping Fm (Wuchiapingian), Dalong Fm (Wuchiapingian to Changhsingian), and Feixianguan Fm (Lower Induan), in ascending order (Isozaki et al., 2004, 2007-this issue). In this study, we focused on the PTB interval that is composed of the uppermost Changhsingian (the upper part of the Dalong Fm) and the lowermost Induan (the lower Feixianguan Fm). On the basis of a detailed observation of the continuous outcrop on the eastern bank of the Jialingjiang river at Chaotian (Fig. 1), the PTB interval at Chaotian is lithologically divided into 7 units, Unit A to Unit G in ascending order (Fig. 2). Unit A (N 0.9 m) and Unit B (2.8 m) are composed of black mudstone, Unit C (1.4 m) of bedded black limestone, Unit D (2.3 m) of bedded light gray limestone, Unit E (1.4 m) of bedded light gray marl, Unit F (1.7 m), Unit G (N 1.4 m) of bedded gray limestone, respectively. For more details of Units A–G, refer to Isozaki et al. (2007this issue). 3. Conodont biostratigraphy Fifty-five limestone samples were collected from the 12 m-thick study section across the PTB at Chaotian. Average weight of each sample was about 2.5 kg. Each sample was crushed and dissolved in 10% acetic acid. After sieving between 20 and 200 mesh, conodonts were picked under a binocular microscope. Conodonts are abundant in Unit C and D, but quite rare to absent in the rest of the study interval. The sample horizons, occurrences of conodonts, and stra-

41

tigraphic range of each taxon are shown in Fig. 2. The main Clarkina species are illustrated in Pl. I and II. Six conodont zones are discriminated from Unit B to Unit G. They are, in ascending order, 1) Clarkina changxingensis changxingensis–C. subcarinata Assemblage Zone, 2) C. postwangi–C. sp. B Assemblage Zone, 3) C. taylorae–C. zhejiangensis–C. changxingensis yini Assemblage Zone, 4) barren intra-zone, 5) Hindeodus parvus Zone, and 6) Isarcicella Zone. The former three correspond to the uppermost part of the Dalong Fm, and the latter three to the basal Feixianguan Fm (Fig. 2). In Unit C and D, conodonts of the genus Clarkina were predominant without the genus Hindeodus, while in Unit F and G those of Hindeodus and Isarcicella characterized the basal Triassic without the genus Clarkina. Besides these platform elements, ramiforms also occur, but they are not dealt with here. 1) C. changxingensis changxingensis–C. subcarinata Assemblage Zone (below C7). This zone is defined by the occurrences of C. changxingensis changxingensis and C. subcarinata. Its base is not confirmed because of the conodont-poor nature of the underlying rocks, while its top is clearly defined by the first occurrence (FO) of C. postwangi at the C8 horizon. 2) C. postwangi–C. sp. B Assemblage Zone (from C8 to D6). This zone is defined by the occurrence of C. postwangi below C. changxingensis yini and C. taylorae. Accordingly, its base is at the FO of C. postwangi, and its top at FO of C. taylorae and C. changxingensis yini. 3) C. taylorae–C. zhejiangensis–C. changxingensis yini Assemblage Zone (from D7 to D24). This zone is defined by the range of C. taylorae and C. changxingensis yini. In this zone, C. changxingensis changxingensis, C. meishanensis zhangi, C. meishanensis meishanensis, and C. deflecta occur abundantly. In addition, C. zhejiangensis, C. orchardi occur. 4) Barren intra-zone (from E1 to E18). This zone corresponds to the less calcareous and fossil-poor marl of Unit E. Conodonts are rare to almost absent in this interval, except for some Clarkina specimens found in the E2 horizon. 5) Hindeodus parvus Zone (F1 to H55). This zone is defined by the range of H. parvus. Its base is defined by the FO of H. parvus and the top is by the FO of I. staeschei. In this zone, conodonts are rare, and the index taxon is accompanied only by some ramiforms. 6) Isarcicella Zone (above H55). The base of this Zone is defined the FO of I. staeschei. I. isarcica and H. parvus co-occurred in the bed H59. The top of this zone is not defined. This zone probably includes

42

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

Fig. 2. Stratigraphic distribution of conodonts in the Permian–Triassic boundary interval at Chaotian. See Isozaki et al. (2007-this issue) for lithostartigraphy and ammonoid zones.

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

I. staeschei Zone and I. isarcica Zone, however, it is not divided in detail for the limited data. According to the ammonoids from the Chaotian section (Isozaki et al., 2004, 2007-this issue), Units A–C belong to the Pseudostephanites–Tapashanites Zone, while Unit D to the Pseudotirolites–Pleuronodoceras Zone of the Changhsingian. The basal part of Unit E belongs to the Hypophiceras Zone of the uppermost Changhsingian. The rest part of Unit E is barren of both conodonts and ammonoids. The age of Units F to G is constrained by conodonts, instead of ammonoids, to the Induan. The PTB is temporarily put at the base of the Unit F on the basis of the first occurrence of H. parvus, although the PTB may lie at some horizon in the Unit E above E2 because there is no age-determining conodont obtained from Unit E. If the FO of H. parvus is regarded as its FAD (first appearance datum, i.e., the exclusive biostratigraphic criterion for the PTB), the PTB at Chaotian is set at the base of Unit F. The conodont biostratigraphy across the PTB at Chaotian displayed a clear pattern of replacement of the Permian to the Triassic fauna. The Permian gondolellids fauna suddenly disappeared, together with other fossils, at the top of D24 or the base of Unit E, then none or rare conodont occurred in Unit E, then Hindeodus fauna occurred from the Unit F. Thus we can distinguish two biostratigraphically important horizons, i.e., the event horizon at the boundary between Unit D and Unit E, and the appearance horizon of the new Triassic fauna at the boundary between Unit E and Unit F. 4. Remarks on the Changhsingian Clarkina at Chaotian In the Late Permian conodont biostratigraphy, gondolellids are the most important taxon among all; however, the classification of gondolellids is still in discussions. Kozur (1989) proposed the genus Clarkina for the Late Permian gondolellids, whereas Orchard and Krystyn (1998) and Orchard and Rieber (1999) preferred to stay on usage of the genus Neogondolella for Late Permian through Middle Triassic gondolellids. Mei et al. (2004) also agreed to use the genus Clarkina for the Late Permian gondolellids. The classification of gondolellids is beyond the topic of the gondolellids near the PTB at Chaotian, thus Clarkina (= Neogondolella) is adopted herein to report the conodont biostratigraphy results. The species concepts of Clarkina are another controversial topic. Mei et al. (1998) classified the shapes of the posterior termination of Clarkina into 4 types; i.e., round, square, narrow, and transitional

43

(between square and round). These types correspond to the form species concepts of changxingensis (round), deflecta (square), wangi (narrow), and subcarinata (transitional). However, Wardlaw and Mei (2000) and Wardlaw et al. (2003) abandoned these form species concepts lately, and suggested that the outline shape is less important. By introducing the concepts of the evolutionary progression of denticulation in the Pa elements of Clarkina, they re-classified the previously described Clarkina species into C. wangi, C. subcarinata, and C. changxingensis. Mei et al. (2004) also emphasized that the configuration of the denticles of Clarkina is the most stable characteristics and introduced a sample-population approach in defining the base of the Changhsingian Stage that emphasized the carinal denticle configuration, but recognized that other characteristics are also useful. However, the traditional concepts of morphologic form species concepts are still widely accepted. In addition to the denticulation, shapes of the platform and other morphologic characteristics have been all taken into the consideration to define a species (Orchard and Krystyn, 1998; Wang, 2001; Kozur, 2004). For the Upper Changhsingian in Iran, Kozur (2004) recently proposed much finer-scale zones for the Upper Changhsingian in Iran based on the traditional morphologic form concepts. Although both approaches aim for the more precise dating and correlation of the PTB interval, they are unfortunately not compatible with each other completely, and in fact, both approaches are not yet obtained full acceptances (Wang, 2001; Chen et al., 2005). Under the circumstances, we should test the validity and compatibility of the two new approaches in various sections in the world. The identifications of Clarkina herein mainly rely on the traditionally morphologic form species concepts, while the configuration of denticles is also checked. On the basis of our material from the PTB interval at Chaotian, here we mention five critical remarks on the Changhsingian Clarkina relevant to zoning. 4.1. Brim-bearing Clarkina around the PTB The upper Changhsingian at Chaotian is entirely characterized by the predominant occurrence of C. subcarinata, C. changxingensis, and C. deflecta, however, these three species are all long-ranged thus not appropriate for fine-scale zoning. C. taylorae with a clear brim along the posterior end of platform, has a much shorter range, thus this taxon was selected as a nominal species for the uppermost Permian zone at Chaotian. Clarkina with a clear brim along the posterior platform margin occur particularly in the uppermost

44

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

Changhsingian and the lowermost Induan in the world. C. taylorae is a representative of the brim-developed Clarkina and was originally described in the lowermost Triassic H. parvus Zone in the Selong section, Tibet by Orchard et al. (1994). This species occurs from the uppermost Dorashamian to the I. isarcica Zone of the Lower Triassic in Iran (Kozur, 2004), also occurs in the lowermost Triassic in Spiti (Orchard and Krystyn, 1998), and in Lower Triassic Jinjiling Fm at Shaanxi province, South China (Neogondolella changxingensis, pl. 49, Figs. 10–15 and N. zhenanensis, pl. 48, Figs. 13– 16, 20–23, by Li et al., 1989, are re-identified here as C. taylorae). C. taylorae occurs in the uppermost Permian and lowermost Triassic widely in peri-Gondwana, South China, and Arctic Canada (Henderson and Baud, 1997; Orchard and Krystyn, 1998; Kozur, 2004). In addition, similar brim-bearing Clarkina occurs in the latest Changhsingian and earliest Triassic in some sections, such as in Tibet (C. sheni by Mei and Henderson, 2001) and from the latest Permian in Hunan, South China (C. xiangxiensis by Tian, 1993a,b). Therefore, it is noteworthy that the brim-bearing Clarkina occurs from the PTB interval regardless of the Permian conodont provincialism (Mei and Henderson, 2001). In the conodont zonation and correlation for the Changhsingian, especially for the late Changhingian, the cusp-terminated Clarkina lineage was emphasized in the Meishan section (Wardlaw and Mei, 2000;

Wardlaw et al., 2003), whereas the brim-bearing types of Clarkina, such as C. iranica and C. abadehensis, were emphasized in Iran (Kozur, 2004). The mode of occurrence of the brim-bearing Clarkina, especially those in South China described herein, appears helpful on this issue. The development of a brim is an unnegligible character in the gondolellids species concepts in the Triassic and also in the Permian. It seems likely that the brim becomes more and more evident from the latest Permian to the earliest Triassic. According to Tian (1993b), the brim of Neogondolella (=Clarkina) xiangxiensis is more clearly developed than that of N. postwangi in the latest Permian. At Chaotian, the brim of C. taylorae is more developed than C. postwangi in the latest Permian, too. Judging from its abundance (Fig. 2), it seems likely that C. taylorae can serve as a key taxon in the uppermost Permian conodont zonation, at least for deep-water basinal facies in South China. 4.2. C. zhejiangensis in the latest Permian C. zhejiangensis occurs in the Pseudotirolites– Pleuronodoceras Zone at Chaotian, although it was previously reported from a much higher horizon, the Hypophiceras Zone at Meishan (Mei, 1996; Wang, 1994, 1995; Zhang et al., 1995; Mei et al., 1998). Orchard and Krystyn (1998) reported C. zhejiangensis in the lowermost Triassic Otoceras latilobatum Zone

Plate 1. SEM photos of conodonts from the PTB interval of the Chaotian section. Scale bar below each specimen is 50 μm in Plates 1 and 2. Each specimen below is labeled with specific name, sample number, and SEM photo number. All the conodonts illustrated in Plates 1 and 2 are housed in the Institute of Geology, Chinese Academy of Geological Sciences. (see on page 45) 1–3, ?6. 4. 5. 7. 8. 9. 10–13. 14–15. 16–18. 19. 20–22. 23–27.

Hindeodus parvus, 1, H55, 04539; 2, H55, 05927; 3, H59, 05827; 6, H59, 05825; H. praeparvus, H59, 05828; H. parvus anterodentatus, H59, 05824; Isarcicella staeschei, H55, 05926; I. isarcica, H59, 05826; Clarkina deflecta, D24, 05921; C. tulongensis, 10, C1, 04690; 11, C1, 04671; 12. D6, 05841; 13, C1, 04545; C. changxingensis changxingensis, 14. D19, 05896; 15, D11, 05968; C. postwangi. 16, D10, 05865; 17, D10, 05867; 18, D9, 05853; C. sp. B, D15, 05890; C. sp. A, 20, D9, 05951; 21, D9, 05950; 22, D9, 05854; C. meishanensis zhangi, 23, C1, 04697; 24, D10, 04550; 25, D15, 05888; 26, D15, 05889; 27, D24, 04634.

Plate 2. SEM photos of conodonts from the PTB interval at Chaotian. (see on page 46) 1,2. 3–5. 6–11. 15–18. 19–21. 22–23. 24.

Transitional specimens between C. meishanensis zhangi and C. meishanensis meishanensis, 1, D6, 05942; 2, C1, 04702; 12–14, C. meishanensis meishanensis, 3, C1, 04706; 4, D22, 05914; 5, D11, 05873; 12, C1, 04698; 13. C1, 04686; 14, C1, 04691; C. changxingensis yini, 6. D24, 04632; 7, D22, 04646; 8, D24, 04559; 9, D24, 04648; 10, D24, 04654; 11, D15, 05891; ?26. C. subcarinata, 15, D14, 05882; 16, D14, 04658; 17, D14, 05876; 18, D14, 05880; 26, D24, 04644; C. zhejiangensis, 19, D9, 05952; 20, D10, 05868; 21, D10, 04669; 25 C. orchardi, 22, D10, 04552; 23, D10, 05870; 25, D10, 04551; C. taylorae, D10, 04553.

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

Plate 1 (caption on page 44 ).

45

46

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

Plate 2 (caption on page 44 ).

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

and O. woodwardi Zone in Spiti. Recently, Xia et al. (2004) reported C. zhejiangensis from ca. 20 cm below the base of the H. parvus Zone at the Huangshi section in south China, too. Thus the range of C. zhejiangensis is not restricted in the Hypophiceras Zone and above but extend further downward than previously thought. C. changxingensis has been generally regarded as the last taxon of the Clarkina lineage in the Changhsingian (e.g., Wardlaw and Mei, 2000; Wardlaw et al., 2003). C. changxingensis has more reduced carina in height than C. subcarinata and completely discrete and depressed denticles in the posterior end. On the contrary, C. zhejiangensis has carinal denticles that become smaller, lower and more isolated both to the anterior and posterior end, and its blade also becomes lower than C. changxingensis. Such a denticulation of C. zhejiangensis appears more advanced than that of C. changxingensis during the latest Permian to earliest Triassic. 4.3. Problems with C. meishanensis C. meishanensis meishanensis was described only from Beds 25 and 26 at Meishan, i.e., the Hypophiceras Zone (Mei et al., 1998); however, at Chaotian, C. meishanensis meishanensis occurs in an interval from D11 to D22 that corresponds to the Pseudotirolites–Pleuronodoceras Zone (Fig. 2). An exceptional solitary occurrence from C1 suggests that its range may extend down further. C. meishanensis zhangi at Chaotian also has a much longer range than at Meishan (Fig. 2). As this species is generally small in size, it is difficult to distinguish it from juvenile forms of other Clarkina species (See in detail the remarks of this species). Under the circumstances, we avoid using both C. meishanensis meishanensis and C. meishanensis zhangi as key species in zoning at Chaotian. 4.4. Revised definition of C. changxingensis yini The taxonomy of C. changxingensis yini is also in confusion. This subspecies was originally described by Mei et al. (1998), but other workers tend to put all of such morphotypes into C. changxingensis (e.g., Wardlaw and Mei, 2000) as Wang (2001) pointed out. Later, Kozur (2004) re-evaluated the morphological difference between C. changxingensis changxingensis and C. changxingensis yini and proposed that the latter may represent a noteworthy independent species. However, he also admitted that C. yini is generally difficult to be distinguished from C. changxingensis solely by the previously proposed characteristics, such as erect cusp and the flat posterior platform, therefore, that C. yini is not likely suitable for a nominal species of a conodont zone.

47

Although C. yini was still used as an important species (e.g., Kozur, 2004), the more restricted definition of C. yini is definitely needed. Based on the specimens in the upper part of Unit D at Chaotian and their comparison with the original definition and illustrated figures (Mei et al., 1998), here we use C. changxingensis yini as a subspecies of C. changxingensis. On the other hand, C. changxingensis yini has another significance concerning the evolution of gondolellids across the PTB. First, all the gondolellids from the uppermost Permian to the lowermost Triassic are relatively small in size even in adult forms. For example, the latest Permian Clarkina specimens from Chaotian are mostly small in size. In addition, the smallsized gondolellids are dominant in the lowermost Triassic according to the reported data by others (Tian, 1993b; Orchard et al., 1994; Orchard and Krystyn, 1998). Second, some lowermost Triassic gondolellids in Spiti and in South China, such as Neogondolella nevadaensis, N. planata, N. carinata etc. (Tian, 1993b; Orchard and Krystyn, 1998) have wider and shorter platform than the relatively slender Clarkina in the Late Permian. In this regard, C. changxingensis yini with a short, wide platform may represent the latest Permian missing link in the gondolellid lineage between the Permian Clarkina and Triassic Neogondolella. 4.5. Diverse cusp morphology of Clarkina near the PTB Concerning the gondolellids with a single and terminated cusp, Wardlaw and Mei (2000) and Wardlaw et al. (2003) pointed out that the cusp became more isolated during the Changhsingian. However, the denticle configuration on the posterior end of platform in the gondolellids was considerably diverse in the latest Permian and the earliest Triassic. For example, C. taylorae has a brim behind the cusp that makes the cusp not terminated; C. zhejiangensis has some additional denticles behind the cusp; C. sp. A has two denticles on the position of cusp; C. tulongensis has one or two lateral posterior carina instead of single cusp. The gondolellid specimens with the two-denticle-cusp are tentatively classified in more than two different species in this article, but it should be emphasized that all of them occurred in the uppermost Permian and the lowermost Triassic (see in the remarks of C. sp. A). 5. Correlation On the basis of the new conodont data, here we attempt to correlate the Chaotian section with two representative sections in China, the Shangsi section in

48

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

Sichuan and the GSSP of the PTB at Meishan in Zhejiang, and with another important section at Zal in Iran (Fig. 3). 5.1. With the Shangsi section The Shangsi section has almost the same lithostratigraphy as well as ammonoid stratigraphy (Li et al., 1989) as the Chaotian section. The original described section by Li et al. (1989) was covered now. However, a substitute section is observed along the neighboring stream, and its detailed conodtont biostratigraphy of the PTB interval will be reported in future. We herein compare the conodont biostratigraphy at Chaotian with that of the old section reported by Li et al. (1989) and with the data on the new section by Nicoll et al. (2002). The H. parvus Zone and Isarcicella Zone are easily correlated between the Chaotian and Shangsi sections. The sudden disappearance of Clarkina occurs at the topmost of the Dalong Fm in both sections, in Bed D24 at Chaotian as well as in Bed 27 at Shangsi (Li et al., 1989; Lai et al., 1996; Nicoll et al., 2002). The unique marl unit of the lowermost Feixianguan Fm yields Hypophiceras in both sections (Li et al., 1989; Isozaki et al., 2004). The marl unit at Chaotian (Unit E) can be correlated to the

Fig. 3. Correlation of the PTB interval between the Chaotian section in Sichuan and the Meishan section (the GSSP of the PTB) in Zhejiang. Note that the interval between the main extinction horizon of Permian taxa and the FO (or FAD) of the first Triassic index fossil (H. parvus) has considerably different thickness between the two sections.

lowermost 4.5 m of the Feixianguan Fm at Shangsi lithostratigraphically. In this interval, no good data were obtained at Chaotian, however, Hindeodus priscus, H. latidentatus and H. eurypyge were reported by Nicoll et al. (2002) at Shangsi. Yang et al. (1987) and Li et al. (1989) reported various gondolellids from the topmost Dalong Fm in Shangsi in the name of C. subcarinata, C. changxingensis, C. deflecta etc., however, these species names should be amended in future in the light of the latest concepts of conodont taxonomy. The detailed correlation with the Chaotian section will be reported later with re-description of the gondolellids. 5.2. With the GSSP at Meishan On the basis of the full dataset on lithostratigraphy and biostratigraphy (e.g., Zhao et al., 1981; Wang and Wang, 1981; Wang and Chen, 1983; Yang et al., 1987; Zhang, 1987; Wang, 1994, 1995; Zhang et al., 1995, 1996; Mei et al., 1998), the Meishan section was selected as the GSSP of the PTB (Yin et al., 2001a,b). The first appearance datum (FAD) of H. parvus was applied as the ultimate criterion for the biostratigraphically-defined PTB (Yin et al., 2001a). For the uppermost Permian correlation, the ammonoid succession indicates that the C. taylorae–C. zhejiangensis–C. changxingensis yini Assemblage Zone (about 1.8m thick) at Chaotian is probably correlated to the C. changxingensis yini–C. meishanensis zhangi Assemblage Zone (about 1.5 m thick) at Meishan (Mei et al., 1998). Likewise, the C. postwangi–C. sp. B Assemblage Zone at Chaotian is correlated roughly to the upper part of the C. changxingensis changxingensis–C. parasubcarinata–C. postwangi–C. deflecta Assemblage Zone (Mei et al., 1998) at Meishan. The C. changxingensis changxingensis–C. subcarinata Assemblage Zone is also correlated to the lower part of the C. changxingensis changxingensis–C. parasubcarinata–C. postwangi–C. deflecta Assemblage Zone at Meishan. Thus the Clarkina fauna of the upper part of the Dalong Fm at Chaotian could be correlated to that in the upper part of the Changxing Fm at Meishan. Concerning the evolutionary change in denticulation of the Changhsingian Clarkina, however, the occurrence of C. zhejiangensis in the uppermost Dalong Fm indicates that a relatively later evolutionary stage of Clarkina likely has existed after the C. changxingensis stage, and that the Chaotian section preserves a more complete lineage of Clarkina than the Meishan section. The Clarkina fauna in the uppermost Dalong Fm at Chaotian, dominated by C. taylorae, C. orchardi, C. tulongensis, C. zhejiangensis and C. changxingensis yini is more closely related with

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

the lowermost Triassic fauna rather than that of the uppermost part of the Changxing Fm at Meishan. The thickness of the interval between the main extinction horizon and the FO of H. parvus is nearly ten times thicker at Chaotian than at Meishan, as previously emphasized by Isozaki et al. (2004) (Pl. 1). The main extinction horizon of the Permian taxa is at the top of Bed D24 of the topmost Dalong Fm at Chaotian, and at the top of Bed 24 of the topmost Changxing Fm at Meishan (Clark et al., 1986; Yang et al., 1987; Wang, 1994; Jin et al., 2000) respectively. This horizon can be doubtlessly correlated between the two sections as an event-stratigraphic boundary. In contrast, the FO of H. parvus is about 1.4 m higher than the main extinction horizon at Chaotian, whereas it is merely 16 cm above the extinction horizon at Meishan. The Shangsi section discussed above also has a greater thickness about 4.5 m for this interval (Nicoll et al., 2002). This apparent contrast in thickness between the Chaotian–Shangsi sections and the Meishan section can be explained in two ways; 1) the boundary marl at Chaotian and Shangsi may contain rare H. parvus which has not been found yet (i.e., FO of H. parvus does not coincide with FAD because of the scarcity of conodonts), or 2) the Meishan section is lacking a certain interval of the topmost Permian or at least highly condensed. The above-mentioned brim-bearing Clarkina of the Chaotian section supports the latter option. Although the H. parvus Zone and Isarcicella Zone at Chaotian are easily correlated with those in Meishan, there is a remarkable difference in thickness again between the two sections. The H. parvus Zone at Chaotian is ca. 3.7 m thick, while that in Meishan is no more than 8 cm. This also suggests the possible incompleteness or highly condensed nature of the Meishan section. See Isozaki et al. (2007-this issue) for more detailed discussion. 5.3. With the Zal section in Iran By clarifying detailed Clarkina succession in the Zal section in NW Iran with descriptions of some new species and subspecies of Clarkina, Kozur (2004) and Korte et al. (2004) emphasized that the Zal section has a complete sequence of strata immediately across the PTB. In the Dorashamian rocks in Zal, six conodont zones were discriminated; i.e., 1. C. bachmanni Zone, 2. C. changxingensis–C. deflecta Zone, C. nodosa Zone, 3. C. yini–C. zhangi Zone, 4. C. iranica Zone, 5. C. hauschkei Zone, and 6. C. meishanensis–H. praeparvus Zone, in ascending order. Later Chen et al. (2005) criticized that the new taxa used in Kozur's zonal scheme may have some problem with endemism.

49

Although the strata were studied almost bed by bed across the PTB at Chaotian, we could not obtain typical index taxa, e.g. C. hauschkei, C. iranica. The correlation between the Chaotian and the Zal section is not straightforward, however, some species may allow us to make correlation as follows. (1) C. tulongensis common in the upper Dalong Fm at Chaotian may represents a typical cool-water species that migrated from the peri-Gondwana domain into the low-latitude provinces like South China during the latest Permian in the interval of the C. hauschei Zone (Kozur, 2004). (2) Kozur (2004) speculated that C. taylorae may occur in the C. meishanensis–H. praeparvus Zone from the uppermost Dorashamian to I. isarcica Zone in periGondwana, South China, Iran, and Arctic Canada. As C. taylorae occurs commonly in the uppermost Dalong Fm at Chaotian together with Changhingian ammonoids, this part may possibly be correlated with the uppermost Dorashamian (Changhsingian) in Zal. 6. Conclusion The PTB interval at Chaotian are divided into 6 conodont zones: Clarkina changxingensis changxingensis–C. subcarinata Assemblage Zone, C. postwangi–C. sp. B Assemblage Zone, C. taylorae–C. zhejiangensis– C. changxingensis yini Assemblage Zone, barren intrazone, Hindeodus parvus Zone, and Isarcicella Zone in ascending order. The main extinction horizon of the Changhsingian conodonts and ammonoids was identified immediately below the Hypophiceras Zone. The FO of H. parvus is tentatively regarded here as the conodont-based biostratigraphic PTB at Chaotian. The first 4 zones correspond to the Changhsingian, whereas the last two to the Induan. On the basis of these conodont zones, the Chaotian section is correlated with the Shangsi section in Sichuan, the Meishan section (GSSP of the PTB) in Zhejiang, and the Zal section in Iran. Acknowledgements We appreciate Harutaka Sakai, Noriei Shimizu, Hiroshi Nishi, Hodaka Kawahata, Tomomi Kubo, and Masao Takano for their help in fieldwork. The first author thanks Heinz Kozur and Shugang Tian for their presentations of their articles and hints on the conodont identifications. We also thank Michael Orchard, Charles Henderson and Daniel Lehrmann for their critical comments and constructive suggestions on the manuscript. This paper is dedicated to Tetsuo Matsuda, one of our members who deceased in 2002. This study was supported by National Natural

50

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

Science Foundation of China (project no. 40502004 and 49972014) and by Japan Society for the Promotion of Science (project no. 12573011 and 16204040). Appendix A. Taxonomy of the main Clarkina species

type was not emphasized afterward. C. deflect is identified herein by: 1) continuous carina comprising densely-spaced denticles in almost same size, 2) carina gradually deflecting from the middle of the platform to end, 3) the asymmetric bilateral platforms.

A.1. Clarkina changxingensis (Wang and Wang, 1981)

A.4. Clarkina meishanensis (Zhang et al., 1995)

(figs. 3–14, 15) 1981 Neogondolella subcarinata changxingensis Wang and Wang, p.117, pl. 1, figs 13–16 1987 Gondolella changxingensis — Zhang, p.199, pl. 1, figs. 16, 18 1993b Neogondolella changxingensis —Tian, p. 190, pl. 4, figs. 12–19 1994 Neogondolella changxingensis — Orchard at al., p. 833, pl. 1, figs. 7, 8, 10, 13–16,18,19 1998 Clarkina changxingensis changxingensis — Mei et al., p. 217, pl. I, B–D, F–H, K; pl. II, G 2004 Clarkina changxingensis (Wang and Wang) — Kozur, p.43, pl. 3, figs. 6, 8, pl. 4, figs. 15?, 17?; pl. 15, fig. 15 Remarks: After the original definition (Wang and Wang, 1981), its diagnosis was revised (Mei et al., 1998; Wardlaw and Mei, 2000; Wardlaw et al., 2003). Wang (2001) pointed out that several form species are included in the C. changxingensis of Wardlaw and Mei (2000). Kozur (2004) agreed with the original diagnosis, and also agree with the identification of C. changxingensis by Mei et al. (1998) and disagreed the emendation of C. changxingensis by Wardlaw et al. (2003). The identification of the species mainly refers to Wang and Wang (1981), Mei et al. (1998) and Kozur (2004).

(fig. 3–3, 4, 5, 12, 13, 14) 1995 Clarkina meishanensis sp. nov., Zhang, Lai, Ding, Wu and Liu, p.674, pl. 2. II. 5, figs. 4–6. 1996 Clarkina meishanensis — Zhang et al., p. 62. pl. II. 7, figs. 5, 12, 13. 1998 Clarkina meishanensis meishanensis — Mei et al., p. 223, pl. IV, figs. C–G, H? Remarks: Clarkina meishanensis resembles C. changxingensis in the narrow, tapered anterior platform. However, in C. meishanensis, the main part of the lateral margins of the platform is subparallel, the most width is near the posterior end, the adcarinal grooves are deep, and the cusp is both prominent and distinct. (Zhang et al., 1995, 1996). It is distinguished from C. meishanensis zhangi by that the cusp is larger relative to the posterior denticles, much set off from the carina and less reclined (Mei et al., 1998). The C. meishanensis in Fig. 5 of Yin et al. (2001b) has a high carina with densely-spaced denticles, and the cusp is almost connected with the carina. This specimen is identified as C. sp. (pl. II. 7, fig. 4) by Zhang et al. (1996). Its denticulation is different from the illustrated C. meishanensis (pl. II. 7, fig. 5, 12, 13) by having a high carina with narrow-spaced denticles and the cusp that is almost connected with the carina. The free blade of C. meishanensis is high with 3 to 4 denticles fixed, except the tip of the denticles (Zhang et al., 1995, 1996), and judging from the illustrated figures, the free blade is short or moderate in length, and the posterior denticles on the carina are smaller, lower and more sparse in the holotype.

A.2. Clarkina changxingensis yini (Mei et al., 1998) (fig. 4–6, 7, 8, 9, 10, 11) 1998 Clarkina changxingensis yini subspecies nov. Mei, Zhang and Wardlaw, p. 217, pl. IV, L–N Remarks: The reasons why this species should be more restricted have been introduced in the text. The small size and large width to length ratio of the platform are emphasized in the diagnoses. A.3. Clarkina deflecta (Wang and Wang, 1981) (fig. 3–9) 1981 Neogondolella deflecta sp. nov., Wang and Wang, p. 117, pl. 2, figs. 9,10 Remarks: In the original definition, the specimen with a straight carina deflects abruptly at the end of the platform was also included in this species; however, this

A.5. Clarkina meishanensis zhangi (Mei et al., 1998) (fig. 3–23, 24, 25, 26, 27) 1998 Clarkina meishanensis zhangi Mei subsp. nov., Mei et al., p.218, pl. I, figs. Aa, Ab, Plate IV, Ka, Kb 2004 Clarkina zhangi Mei, 1998— Kozur, P. 49, pl. 5, figs. 1, 2, 4, 8 Remarks: The holotype (Pl. IV, Ka, Kb) is a broken specimen without anterior part, and the other specimen (pl. I, figs. Aa, Ab) is regarded as a young specimen by the establisher. Its diagnosis could be summarized as: 1) narrow platform with acute posterior margin, 2) denticles decreasing gradually in size and height toward

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

the posterior, 3) carina is connected with cusp, 4) prominent, reclined and terminated cusp. In addition, the small size of the individual and the developed free blade are also considered during our identification. C. meishanensis zhangi is regarded as an independent species and not as a subspecies of C. meishanensis (Kozur, 2004), however, the specimens of the two subspecies displayed some transitional character morphologically in our conodont collection. The specimen (fig. 4–1) has acute posterior platform margin and inclined cusp, which is the character of C. meishanensis zhangi, at the same time, its carina is low with quite separated denticles and the cusp much set off from the carina, which is the character of C. meishanensis meishanensis. The specimen (fig. 4–2) has an erect or a little inclined cusp, which is the character of C. meishanensis meishanensis, however, the carina is almost connected with the cusp, which is the character of C. meishanensis zhangi. It is difficult to classify them into one species on basis of the present definition by the establishers. C. meishanensis zhangi is thought to be a very characteristic form of the C. yini–C. zhangi Zone which allows a direct correlation of the South Chinese intraplatform basin with the open sea Tethyan faunas during this upper Dorashamian time interval (Kozur, 2004). The C. meishanensis zhangi Zone is only less than 20 cm in the strata in the Huangshi section (Xia et al., 2004). C. meishanensis zhangi maybe ranges from Mc-40 to Mc-43, about 2m thick, at the topmost of Changxing Fm in the Meishan section (Mei et al., 1998). C. meishanensis zhangi occurred in the Upper Dorashamian C. yini– C. zhangi Zone of Iran and Transcaucasia (Kozur, 2004). In the Chaotian section, C. meishanensis zhangi ranges from the bed C1 to the bed D24. A.6. Clarkina postwangi (Tian, 1993a,b) (fig. 3–16, 17, 18) 1993a Neogondolella postwangi sp. nov., Tian, p.147, pl. II, figs. 2,3 1993b Neogondolella postwangi sp. nov.—Tian, p.186, pl. 5, figs. 1–5, 11a–12b Remarks: Mei et al. (1998) pointed out that Tian (1993a,b) had not assigned a holotype and did not give a detailed description of the denticulation and therefore emended the diagnosis. Kozur (2004) thought that C. postwangi (Mei et al., 1998) is different from the original species and assigned the specimen (pl. 5, fig. 11, Tian, 1993) as the holotype. The first author consulted Dr. Tian on the identification of C. postwangi. In his opinion, the cusp is erect and separated, a little

51

brim of the posterior platform and the shuttle outline of the platform are the main characters of C. postwangi. We followed his suggestions during the identification. A.7. Clarkina orchardi (Mei, 1996) (fig. 4–22, 23, 25) 1996 Clarkina orchardi sp. nov., Mei, p.144, pl. 18.1, figs. 6–12 1998 Neogondolella orchardi— Orchard et al., p. 360, pl. 1, figs. 1–4, 9 Remarks: This species is diagnosed by a narrow and elongate platform, evenly curved platform margins, a narrowly rounded to pointed posterior end, a carina with largely discrete denticles that slightly rise anteriorly to a low blade, and an erect cusp located at the posterior end of the platform (Mei, 1996). This species is included in the Narrow-morphotype and thought to be the immediate progenitor of C. taylorae by the establisher. In our collections, the species is characteristic by that 1) the low carinal denticles are separated more than C. changxingensis and less than C. taylorae. 2) the posterior platform is narrower than both C. changxingensis and C. taylorae. 3) the cusp is obviously separated from the carina and is not as distinguished as C. changxingensis and C. taylorae. The specimen with brim is also included in this species for the narrow posterior platform and less separated carinal denticles. C. praetaylorae was established and assumed to be the forerunner of C. taylorae too (Kozur, 2004). C. praetaylorae and C. orchardi are both charactered by low carina. In C. praetaylorae, the posterior carinal denticles are separated and the anterior carinal denticles are higher and more fused, and the small and erect cusp lies at the flat posterior margin or a little in front of it, the platform becomes rapidly, rarely gradually narrower in the anterior third of the unit. Our specimens may be the transitional morphotype between these two species. A.8. Clarkina subcarinata (Sweet, 1973) (fig. 4–15, 16, 17, 18, ?26) 1981 Neogondolella subcarinata subcarinata Sweet— Wang and Wang, p. 117, pl. I, figs. 8–11, pl. II, figs. 5–6 1994 Neogondolella subcarinata— Orchard et al., p. 835, pl. 1, figs. 3–5, 9 1998 Clarkina subcarinata— Mei et al., p. 225, pl. I, E, pl. II, I, K Remarks: Based on the description by Wang and Wang (1981), C. subcarina subcarinata has a wide and short platform. According to Mei et al. (1998), C.

52

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

subcarinata has two morphotypes: one is represented by the figure (pl. I, fig. E), which is characteristic by the slowly narrowing bilateral margins of the anterior platform and undeveloped free blade, widest at the posterior platform; the other is represented by the figures (pl. II, figs. Ia, Ib, Ka, Kb), which are characteristic by the rapidly narrowing bilateral margins of the anterior platform and the well developed free blade, widest in the middle of the platform. Wardlaw and Mei (2000) and Wardlaw et al. (2003) re-defined C. subcarinata based on the evolutionary progression of denticulation. Mei et al. (2004) explained in more detail on C. subcarinata with the same opinion. However, Kozur (2004) pointed out that the C. subcarinata should be the slender forms with continuous narrowing and lowering of the platform toward the anterior end, and its platform does not have brim and the cusp overreaches the platform end or not overreach sometimes. It is quite difficult for us to make a selection among the different opinions. The relatively wide and short platform is emphasized. A.9. Clarkina taylorae (Orchard et al., 1994) (fig. 4–24) 1994 Neogondolella taylorae Orchard n. sp., Orchard et al., p.833, pl. 2, figs. 9, 10, ?11, ?12, 15, 16 1996 Clarkina taylorae — Mei, p. 145; pl. 18. 2, fig. 15 1998 Clarkina taylorae — Orchard and Krystyn, p. 360, pl. 2, figs. 1–9, 13–22 2004 Clarkina taylorae — Kozur, p. 48, pl. 2, fig. 14 Remarks: This species was established at the Selong section, Tibet in the lower Triassic Otoceras latilobatum bed by Orchard et al. (1994). The species is redefined in more detail by Orchard et al. (1998). C. taylorae is distinguished by: 1)an elongate-oval shaped platform, mostly rounded and evenly tapered posterior outline, 2) mostly discrete Carinal nodes rising to a low anterior blade, 3) a prominent, rounded, upright cusp terminating near the posterior end of the platform. 4) a developed brim behind the cusp at the posterior platform. Kozur (2004) reported that it is common in the latest Dorashamian in Iran. This species is different from the C. zhejiangensis by the relatively big cusp, and more rounded posterior end of the platform and the developed brim. A.10. Clarkina tulongensis (Tian, 1982) (fig. 3–10, 11, ?12, 13) 1982 Neogondolella tulongensis sp. nov., Tian, p. 159, pl. 1, figs. 15, 17

1994 Clarkina tulongensis — Orchard et al., p. 835, pl. 2, figs. 1–4 1996 Clarkina tulongensis— Mei, p.145, pl. 18, figs. 1–5 1998 Clarkina tulongensis— Orchard at al. p. 363– 364, pl. 3, figs. 11–17; pl. 3, figs. 1, 3, 4 2004 Clarkina tulongensis— Kozur, p.49, pl. 3, figs. 16–18, 21 Remarks: After the original definition, Orchard et al. (1998) and Kozur (2004) redefined the species more precisely. We distinguished the species from C. deflecta by: 1) The bilateral platforms are more symmetric in C. tulongensis than in C. deflecta; the subrectangular outline of C. tulongensis is against the wedge-like outline in C. deflecta (Kozur, 2004). 2) The carina is more straight in C. tulongensis than in C. deflecta; the carina deflects abruptly at the far end in C. tulongensis while the carina deflects gradually from the middle of the platform; the carina could be separated into the middle carina and the lateral posterior carina in C. tulongensis while the carina is composed of the continuous and compressed denticles in C. deflecta. Kozur (2004) also pointed out that the carina of the Permian forms is variable. In some forms, the size of the last denticle on the middle carina could be larger than the foregoing denticles. The specimen (fig. 3–13) displays this character. C. tulongensis seems to be included in the original definition of the C. deflecta (pl. II, figs. 12) by Wang and Wang in 1981. However, this morphotype is not emphasized lately. Kozur (2004) thought this species is common in the upper Dorlashmian C. hauschkei Zone of Iran and Transcaucasia and rare in the C. yini–C. zhangi Zone and in the C. meishanensis–H. praeparvus Zone of South China and Salt Range. It is common in the Gangetian (Lower Triassic) of Perigondwana. The lateral posterior carina is separated from the middle carina in C. tulongensis. This phenomenon maybe represent one of the evolutional trend of “cusp” of the asymmetric Clarkina, that is, the “cusp” evolved more separated from the carina and becomes to be more complicated as the “cusp” in the symmetric Clarkina such as C. sp. A. A.11. Clarkina zhejiangensis (Mei, 1996) (fig. 4–19, 20, 21) 1996 Clarkina zhejiangensis sp. nov, Mei, p.145, pl. V, figs. A, G, N 1998 Clarkina zhejiangensis— Orchard et al., p.364, pl. 1, figs. 16, 22–25

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

? 2004 Clarkina zhejiangensis— Xia et al., p.37, pl. 1, fig. 1 Remarks: Mei (1996) re-assigned C. changxingensis (Wang, 1995) to be C. zhejiangensis. According to Mei (1996), it can be distinguished from the C. changxingensis by: 1) largely discrete denticles, 2) a small and erect cusp, which is terminally located or sometimes followed by a small accessory denticle, 3) rather low carina. This species is different from C. taylorae mainly by its relatively small cusp. It is distinguished from the C. sp. A by that the latter has a more prominent “twodenticle cusp” and a higher carina. A.12. Clarkina sp. A (fig. 3–20, 21, 22) Description: Clarkina with a symmetric and elongate oval platform, whose posterior end is bluntly rounded. The free blade is obvious in the most specimens. The denticles on carina decrease in size and height toward the posterior. The posterior denticles are separated. Two small but obvious denticles occupy the position of cusp at the terminal or near the terminal of the platform. An obvious gap exits between the carina and the twodenticle cusp. Remarks: The Clarkina with two-denticle cusp was not emphasized by previous authors, however, it is easy to be separated from the single-denticle cusp type of Clarkina in the latest Permian. The similar Clarkina is reported elsewhere in different species in the uppermost Permian and the lowermost Triassic. In the uppermost Changxing Fm at Meishan, the specimen of C. jesmondi (pl. II. 7, fig. 2) illustrated by Zhang et al. (1996) show the similar character of the two-denticle cusp. This specimen is not like the typical C. jesmondi (pl. 3, figs. 7, 9, 10, 13–15), which has a big cusp and an obvious brim (Beyers and Orchard, 1991). The specimen (pl. 1, fig. 22) from the sample C-158625 in the topmost Changxing Limestone at Meishan has two-denticle cusp and an obvious gap between the cusp and carina (Orchard et al., 1994); however, it is interpreted as late growth stages of Neogondolella changxingensis Morphotype 2. Mei (1996) mentioned the similar character of two-denticle cusp by the description that “ a small and erect cusp sometimes followed by a small accessory denticle” in C. zhejiangensis (pl. 18. 2, fig. 17) in the uppermost Permian and the transitional bed of the Permian–Triassic at Meishan. Some specimen in C. praetaylorae (pl. 3, fig. 14) also displayed the two-denticle cusp and the obvious gap between the cusp and the carina in the uppermost

53

Permian in Iran (Kozur, 2004). Orchard et al. (1998) also mentioned the similar character of two-denticle cusp in C. zhejiangensis (pl. 1, figs. 24, 25) in the lowermost Triassic in Spiti. A.13. Clarkina sp. B (fig. 3–19) Description: The specimen is characterized by: 1) the acute posterior platform, which make the outline of the platform alike C. wangi, 2) the more reduced posterior carina than C. wangi in height, 3) no obvious cusp and related brim behind it. Remarks: The C. sp. B seems not to be included in C. postwangi Tian, meanwhile, we are also hesitant to identify it as C. postwangi (Mei et al., 1998) for the argument of Kozur (2004). However, C. sp. B displays closer relationship with C. wangi morphologically, it seems to be the further evolutionary denticulation stage of C. wangi by the reduced height of the posterior carina with the unchanged shuttle outline of the platform. References Beyers, J.M., Orchard, M.J., 1991. Upper Permian and Triassic conodont faunas from the type area of the Cache Creek complex, south-central British Columbia, Canada. In: Orchard, M.J., McCracken, A.D. (Eds.), Ordovician to Triassic Conodont Paleontology of the Canadian Cordillera. Geological Survey of Canada Bulletin, vol. 417, pp. 269–298. Chen, J., Henderson, C.M., Shen, S.Z., 2005. Discussion on Late Permian–Early Triassic conodonts: morphological variation and evolutionary succession. Permophiles 45, 22–25. Clark, D.L., Wang, C.Y., Orth, J., Gilmore, J.S., 1986. Conodont survival and low Iridium abundances across the Permian–Triassic boundary in South China. Science 233, 984–986. Ding, M.H., 1992. Conodont sequences in the Upper Permian and Lower Triassic of South China and the nature of conodont faunal changes at the systemic boundary. In: Sweet, W.C., Yang, Z.Y., Dickins, J.M., Yin, H.F. (Eds.), Permian–Triassic Events in the Eastern Tethys-Stratigraphy, Classification and Relations with the Western Tethys. Cambridge University Press, Cambridge, pp. 109–119. Erwin, D.H., 1993. The great Paleozoic crisis. Columbia University Press, New York. 327 pp. Hallam, A., Wignall, P.B., 1997. Mass Extinction and Their Aftermath. Oxford University Press, Oxford. Henderson, C.M., Baud, A., 1997. Correlation of the Permian–Triassic boundary in Arctic Canada and comparison for the Permian and Triassic stratigraphy. Courier Borsch. Institut Senckenberg, vol. 117, pp. 385–408. Isozaki, Y., Yao, J.X., Matsuda, T., Sakai, H., Ji, Z.S., Shimizu, N., Kobayashi, N., Kawahata, H., Nishi, H., Takano, M., Kubo, T., 2004. Stratigraphy of the Middle–Upper Permian and Lowermost Triassic at Chaotian, Sichuan, China. Proceedings of the Japan Academy 80 (1), 10–16 (Series B). Isozaki, Y., Yao, J.X., Shimizu, N., Ji, Z.S., Matsuda, T., 2007. EndPermian extinction and volcanism-induced environmental stress:

54

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55

Permian-Triassic boundary interval of a lower-slope facies at Chaotian, South China. Paleogeography, Paleoclimatology, Paleoecology 252, 39–55 (this issue). doi:10.1016/j.palaeo.2006.11.051. Jin, Y.G., Mei, S.L, Wang, W., Shang, Q.H., Cao, C.Q., Erwin, D.H., 2000. Pattern of marine mass extinction near the Permian–Triassic boundary in South China. Science 289, 432–436. Korte, C., Kozur, H.W., Partoazar, H., 2004. Negative carbon istope excursion at the Permian/Triassic boundary section at Zal, NWIran. Hallesches Jahrbuch fur Geowissenschaften. Reihe B Beiheft vol. 18, 69–71. Kozur, H., 1989. The taxonomy of the gondolellid conodonts in the Permian and Triassic. Courier Forschungsinstut Senckenberg 117, 409–469. Kozur, H., 1998. Some aspects of the Permian–Triassic boundary (PTB) and of the possible causes for the biotic crisis around this boundary. Palaeogeography, Palaeoclimatology, Palaeoecology 143, 227–272. Kozur, H., 2004. Pelagic uppermost Permian and the Permian–Triassic boundary conodonts of Iran. Part 1: Taxonomy. Hallesches Jahrbuch fur Geowissenschaften. Reihe B Beiheft 18, 39–68. Kozur, H., Pjatakova, M., 1976. Die conodontenart Anchignathodus parvus n. sp., eine wichtige Leitform der basalen Trias. Koninkl. Nederl. Akademie van Westenschappen 79 (2), 123–127 (Series B). Kozur, H., Mostler, J., Rahimi-Yazd, A., 1975. Beitrage zur Mikrofauna permotriadischer schichtfolgen, teil 2, Neue conodonten aus dem Oberperm und der basalen Trias von nord-und zentral Iran. Geologische-Palaeontologische Mitteilungen Innsbruck 5 (3), 1–23. Lai, X.L., Yang, F.Q., Hallam, A., Wignall, P.B., 1996. The Shangsi section, candidate of the Global Stratotype Section and point of the Permian–Triassic boundary. In: Yin, H.F. (Ed.), The Palaeozoic– Mesozoic Boundary Candidates of Global Stratotype Section and Point of the Permian–Triassic Boundary. China University of Geosciences Press, Wuhan, pp. 113–124. Li, Z.S., Zhan, L.P., Dai, J.Y., Jin, R.G., Zhu, X.F., Zhang, J.H., Huang, H.Q., Xu, D.Y., Yan, Z., Li, H.M., 1989. Study on the Permian–Triassic biostratigraphy and event stratigraphy of Northern Sichuan and Southern Shaanxi. Geol. Publ. House, Beijing, pp. 1–435 (In Chinese, with English abstract). Matsuda, T., 1981. Early Triassic conodonts from Kashmir, India, Pt.1, Hindeodus and Isarcicella. Journal of Geosciences, Osaka City University 24 (3), 75–108. Mei, S.L., 1996. Restudy of conodonts from the Permian–Triassic boundary beds at Selong and Meishan and the natural Permian– Triassic boundary. In: Wang, H.Z., Wang, X.L. (Eds.), Centennial Memorial Volume of Prof. Sun Yunzhu: Palaeontology and Stratigraphy. China University of Geosciences Press, Beijing, pp. 141–148. Mei, S.L., Henderson, C.M., 2001. Permian conodont provincialism, evolution and their controlling factors. Acta Palaeontologica Sinica 40 (4), 471–485. Mei, S.L., Zhang, K.X., Wardlaw, B.R., 1998. A refined succession of Changhsingian and Griesbachian neogondolellid conodonts from the Meishan section, candidate of the global stratotype section and point of the Permian–Triassic boundary. Palaeogeography, Palaeoclimatology, Palaeoecology 143, 213–226. Mei, S.L., Henderson, C.M., Cao, C.Q., 2004. Conodont samplpopulation approach to defining the base of the Changhsingian Stage, Loping Series, Upper Permian. In: Beaundoin, A.B., Head, M.J. (Eds.), The Palynology and Micropalaeontology of Boundaries. Geological Society, London, Special Publications, vol. 2, 30, pp. 105–121.

Metcalfe, I., Nicoll, R.S., Mundil, R., Foster, C., Glen, J., Lyons, J., Wang, X.F., Wang, C.Y., Renne, P.R., Black, L., Qu, X., Mao, X.D., 2001. The Permian–Triassic boundary and mass extinction in China. Episodes 24 (4), 239–244. Nicoll, R.S., Metcalfe, I., Wang, C.Y., 2002. New species of the conodont Genus Hindeodus and the conodont biostratigraphy of the Permian–Triassic boundary interval. Journal of Asian Earth Sciences 20, 609–631. Orchard, M.J., Krystyn, L., 1998. Conodonts of the Lowermost Triassic of Spiti, and new zonation based on Neogondolella successions. Rivista Italiana di Paleontologia e Stratigrafia 104 (3), 341–368. Orchard, M.J., Rieber, H., 1999. Multielement Neogondolella (Conodonta, Upper Permian–Middle Triassic). Bollettino della Societa Paleontologica Italiana 37 (2–3), 475–478. Orchard, M.J., Nassichuk, W.W., Rui, L., 1994. Conodonts from the Lower Griesbachian Otoceras latilobatum Bed of Selong, Tibet and the position of the Permian–Triassic boundary. Canadian Society of Petroleum Geologists, Mem. 17, 823–843. Sepkoski Jr., J.J., 1984. A kinetic model of Phanerozoic taxonomic diversity. III, Post-Paleozoic families and mass extinctions. Paleobiology 10, 246–267. Sweet, W.C., 1970a. Permian and Triassic conodonts fauna from a section at Guryul Ravine, Vihi district, Kashmir. University of Kansas Palaeontological Institute 49, 1–10. Sweet, W.C., 1970b. Uppermost Permian and Lower Triassic conodonts of the Salt Range and Trans Indus Ranges, West Pakistan. University of Kansas, Department of Geology, Special publication 4, 207–275. Sweet, W.C., 1973. Conodontophorida Eichenberg, 1930. In: Teichert, C., Kummel, B., Sweet, W.C. (Eds.), Permian-Triassic strata, Kuhe-Ali Bashi, northwestern Iran. Bulletin Museum of comparative Zoology, Harvard University, vol. 145, pp. 423–472. Tian, C., 1982. Triassic conodonts in the Tulong section from Nyalam County, Xizang (Tibet), China. Contribution to the Geology of the Qinghai-xizang (Tibet) Plateau 7, 153–165. Geological Publishing House, Beijing. Tian, S.G., 1993a. The Pemo-Triassic boundary and conodont zones in Northwestern Hunan Province. Bulletin of the Chinese Academy of Geological Sciences 26, 133–150 (In Chinese, with English abstract). Tian, S.G., 1993b. Evolutions of conodont genera Neogondolella, Hindeodus and Isarcicella in northwastern Hunan, China. Stratigraphy and Paleontology of China 2, 173–190. Wang, C.Y., 1994. A conodont-based high-resolution eventostratigraphy and biostratigraphy for the Permian–Triassic boundaries in South China. Palaeoworld 4, 234–248. Wang, C.Y., 1995. Conodonts of Permian–Triassic boundary beds and biostratigraphic boundary. Acta Palaeontologica Sinica 34 (2), 129–151. Wang, C.Y., 2001. Too premature to define the base of the Changhsingian Stage. In: Yan, J.X., Peng, Y.Q. (Eds.), The international Sympsium on the globle stratotype of the Permian– Triassic boundary and the Paleozoic–Mesozoic Events, Changxing, China, 10–13, pp. 106–111. August. Wang, A.D., Chen, X.H., 1983. Changxingian conodonts from northern Zhejiang Province. Journal of Zhejiang University 4, 47–59 (In Chinese, with English abstract). Wang, C.Y., Wang, Z.H., 1981. Permian conodont from the Longtan Fm and Changsing Fm of Changxing, Zhejiang and their stratigraphical and paleoecological significance. Seclected papers on the 1st Convention of Micropaleontological Society of China. Science Press, Beijing, pp. 114–120 (In Chinese, with English abstract).

Z. Ji et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 39–55 Wardlaw, B.R., Mei, S.L., 2000. Conodont definition for the basal boundary of the Changhsing Stage. In: Jin, Y.G. (Ed.), Conodont definition on the basal boundary of Lopingian stages: A report from the International Working Group on the Lopingian Series, Permorphiles, vol. 36, pp. 37–40. Wardlaw, B.R., Henderson, C.M., Mei, S.L., 2003. Progress on the conodont taxonomy and zonation for the lower part of the Changhsingian. Permophiles 42, 33–35. Wu, G.C., Yao, J.X., Ji, Z.S., 2003. Conodont Fauna of Late Upper Permian in Xinfeng Area, Jiangxi Province. Acta Scientiarum Naturalium Universitatis Pekinensis 39 (2), 211–218 (In Chinese, with English abstract). Xia, W.C., Zhang, N., Wang, G.Q., Youshitaka, K., 2004. Pelagic radiolarian and conodont biozonation in the Permo-Triassic boundary interval and correlation to the Meishan GSSP. Micropaleontology 50 (1), 27–44. Yang, Z.Y., Yin, H.F., Wu, S.B., Yang, F.Q., Ding, M.H., Xu, G.R., 1987. Permian–Triassic boundary stratigraphy and Fauna of South China. Geological Memoirs, Series, vol. 2,6. Geol. Publ. House, Beijing, pp. 1–380 (In Chinese, with English abstract). Yin, H.F., Zhang, K.X., Tong, J.N., Yang, Z.Y., Wu, S.B., 2001a. The Global Stratotype Section and Point (GSSP) of the Permian– Triassic Boundary. Episodes 24, 102–114. Yin, H.F., Zhang, K.X., Tong, J.N., Yang, Z.Y., Wu, S.B., 2001b. The Global Stratotype Section and Point (GSSP) of the Permian–Triassic Boundary. In: Yan, J.X., Peng, Y.Q. (Eds.), The international Sympsium on the globle stratotype of the Permian– Triassic boundary and the Paleozoic–Mesozoic Events, Changxing, China, vol. 10–13, pp. 1–19. August. Zhang, K.X., 1987. The Permo-Triassic conodont fauna in Changxing area, Zhejiang Province and its stratigraphic significance. Earth

55

Science—Journal of China University of Geosciences 12 (2), 193–200 (In Chinese, with English abstract). Zhang, J.H., Dai, J.Y., Tian, S.G., 1984. Biostratigraphy of Late Permian and Early Triassic conodonts in Shangsi, Guangyuan County, Sichuan, China. Scientific papers on Geology for the 27th International Geological Congress, vol. 1. Geol. Publ. House, Beijing, pp. 163–178 (In Chinese, with English abstract). Zhang, K.X., Lai, X.L., Ding, M.H., Liu, J.H., 1995. Conodont sequence and its global correlation of Permian–Triassic boundary in Meishan section, Changxing, Zhejiang Province. Earth Science — Journal of China University of Geosciences 20 (6), 669–676 (In Chinese, with English abstract). Zhang, K.X., Ding, M.H., Lai, X.L., Liu, J.H., 1996. Conodont sequences of the Permian–Triassic Boundary Strata at Meishan Section, South China. In: Yin, H.F. (Ed.), The Palaeozoic– Mesozoic Boundary Candidates of Global Stratotype Section and Point of the Permian–Triassic Boundary. China University of Geosciences Press, Wuhan, pp. 57–64. Zhao, J.K., Sheng, J.Z., Yao, Z.Q., Liang, X.L., Chen, C.Z., Rui, L., Liao, Z.T., 1981. The Changhsingian and Permian–Triassic boundary of South China. Bulletin of Nanjing Institute of Geology and Paleontology, Academia Sinica 2, 1–85 (In Chinese, with English abstract). Zhao, J.K., Liang, X.L., Zheng, Z.G., 1978. Late Permian cephalopods of South China. Palaeontologica Sinica 12, 1–194 (In Chinese, with English abstract). Zhu, T.X., Huang, Z.Y., Hui, L., 1999. The Geology of Late Permian Period Biohermal Facies in Upper Yangtze Tableland. Geol. Publ. House, Beijing, pp. 1–110 (In Chinese, with English abstract).