On the geological age of mammalian fossils from Shanmacheng, Gansu Province

ZHANG Zhao-Qun WANG Jian

(Key Laboratory of Vertebrate Evolution and Human Origins of Chinese Academy of Sciences,Institute of Vertebrate Paleontology and Paleoanthropology,Chinese Academy of SciencesBeijing 100044 zhangzhaoqun@ivpp.ac.cn)

On the geological age of mammalian fossils from Shanmacheng, Gansu Province

ZHANG Zhao-Qun WANG Jian

(Key Laboratory of Vertebrate Evolution and Human Origins of Chinese Academy of Sciences,Institute of Vertebrate Paleontology and Paleoanthropology,Chinese Academy of SciencesBeijing 100044 zhangzhaoqun@ivpp.ac.cn)

The mammalian fossils (MimolagusandAnagalopsis) from Shanmacheng in the Jiuxi Basin described by Bohlin in 1951 have long been enigmatic in systematics and controversial in the geological age. New survey in this area suggests the fossils were discovered from the Shanmacheng Member of the Huoshaogou Formation. Comparisons of the Shanmacheng fossils with new specimens ofMimolagusfrom the Irdin Mahan Formation, andAnagalefrom the Late Eocene Ulan Gochu Formation, Nei Mongol, suggest an Eocene age for the Shanmacheng fossils. Reinterpretation of the fossil horizon and the paleomagnetic data of Dai et al. (2005) indicates the polarity zones from Huoshaogou section can be correlated to GPTS Chrons 13-18. Therefore, the Huoshaogou Formation may cover late Middle Eocene to Late Eocene and Bohlin’s fossil horizon can roughly be correlated to Chron18n (～39-40 Ma) of late Middle Eocene. The late Middle Eocene age ofMimolagus rodensnarrows the time gap with its close relativeGomphosthat have been recorded from Early to Middle Eocene. The archaic group Anagalidae may have become extinct before Oligocene.

Shanmacheng, Jiayuguan, Gansu; Eocene;Mimolagus,Anagalopsis; geological age

1 Introduction

Bohlin (1951) described three mammalian fossil taxa, e.g.Mimolagusrodens,Anagalopsis kansuensisand an ungulate, from Shih-her-ma-ch’eng (Shanmacheng), Hui-hui-p’u area, western Kansu (Gansu), China collected during the Sino-Swedish expedition. However, due to the uniqueness of the taxa, the geological age of this locality was undetermined, and a tentative correlation to Early Tertiary was made. Later references of its geological age are controversial, with correlation either to Early Tertiary (Hu, 1962), Early Oligocene (Zhou et al., 1977), earlier than Oligocene (Meng et al., 2009), or to Late Eocene or earlier (Wang, 1997). Stratigraphic position of the fossil horizon is also controversial. It has been associated either to Baiyanghe Formation (Zhai and Cai, 1984) or Huoshaogou Formation (Liang et al., 1992). The Jiuxi Basinwhere the fossils were found is an important area for studying the uplift processes of Qinghai-Tibet Plateau and also for oil exploration (Dai et al., 2005; Zhu et al., 2006). Therefore, the geochronology of this fossil locality is important for both geology and evolution of mammals. To relocate the fossil locality, we made a short survey in the early April, 2015. In this paper, we will provide our relocation of the fossil horizon, and discuss the biochronology based on updated understanding of related taxa, biostratigraphy and magnetostratigraphic data.

2 Geographic and geological relocation of the fossil locality

Bohlin (1951) stated that the fossils were collected at Shih-her-ma-ch’eng, on the right bank of the Po-yang-ho (Baiyanghe), some hundred meters north of the main road. We extracted part of Bohlin’s illustration from 1940 (Bohlin, 1940: pl. I-1) and compared it with the Google satellite image (Fig. 1). The gulley Bohlin figured is now locally called Shanmacheng gulley after the name of Shanmacheng castle while the Hui-hui-p’u village is named as Xinminpu. Terrestrial sediments are well exposed along the right bank of the gulley, dipping towards south as Bohlin observed. Cretaceous sediments are well exposed in the lower reaches of the gulley. The overlying Paleogene sediments of Huoshaogou, Baiyanghe, and Shulehe formations are well developed. After about 80 years, the main road Bohlin mentioned is not visible, but may be near the castle relics. Tracing from the castle relics northwards, we found “the brick red sandstone or fi ne conglomerate laid down in very thick bed” on the right bank of the gulley. Referring to the photo (Bohlin, 1951: pl. VII-7), we relocated the fossil horizon and the possible locality (N39°55′31″, E97°46′44″) (Fig. 1). Stratigraphically, the brick red sandstone bed is in the lower part of Huoshaogou Formation, about 40 m above its lower boundary with Cretaceous sediments.

3 Systematics of Mimolagus and Anagalopsis

Mimolagushas long been an enigmatic taxon. Bohlin (1951) described in detail the morphology of the partial skull, named it asMimolagusrodens, suggesting both lagomorph and rodent characters, and listed it as a duplicidentate. However, he referred the postcranial bones from the same locality as rodent inc. sed. Bleefeld and McKenna (1985) revised these materials and consideredMimolagusas a genus of Lagomorpha. When studying the fossils from Qianshan, Anhui Province, Li (1977) named a new family Mimotonidae, includingMimotonaandMimolagus, showing their similarity and close relationship.Mimolaguswas also considered as a monotypic genus in the family Mimolagidae (Szalay, 1985; Erbajeva, 1986). Asher et al. (2005) placedMimolagusat the basal position of Lagomorpha in Glires, which was followed and reinforced by Wible (2007) and Fostowicz-Frelik et al. (2015).

Recently Fostowicz-Frelik et al. (2015) reported a new species,Mimolagusaurorae, from the lower beds of Irdin Mahan Formation, Erlian Basin, Nei Mongol. With only a few teeth and some postcranials, the new species show strong similarity with the type speciesM.rodensin many characters. The dI2 is narrow and lacks the anterior groove which is developedin Lagomorpha, upper cheek teeth are unilateral hypsodonty. The ankle bones are close in both morphology and size. The calcaneus size ofM.rodensfalls in the variation ofM.aurorae, though the astragalus is slightly smaller. Therefore, the two species are very close if not conspeci fi c.

Fig. 1 Geographic map of the fossil locality near Shanmacheng, west of Jiayuguan City, Gansu Province The Google satellite image shows the location of the fossil locality with the partial fi gure from Bohlin (1940) for comparison. The fi eld photo at the right bank of Shanmacheng gulley indicates exposure of“the brick red sandstone or fi ne conglomerate laid down in very thick bed”. The red star marks the fossil level of Bohlin (1951)

Anagalopsiswas also an enigmatic taxon for Bohlin. He cautiously listed it under Mammalia insertae sedis. In the description and comparison, Bohlin stated the similarities withAnagaleof Anagalidae, but refrained from referring it to Anagalidae. With more materials found, Zhou et al. (1977) comparedAnagalopsiswithLinnanialofoensisfrom the Paleocene of Nanxiong Basin and grouped them in Anagalidae, and further statedAnagalopsishas more derived characters such as larger size, more molarized P4/p4, and reduced M3 thanLinnaniaandAnagale. Hu (1993) confirmed the systematic position ofAnagalopsisusing cladistic analysis, which is nested in the clade of Anagalidae.

4 Geological age of Anagale

The type specimen (well preserved skull, jaws and postcranials) was discovered from the lower part of Ulan Gochu Formation, at Twin Oboes, Nei Mongol (Inner Mongolia) by Chinese assistant “Buckshot” during Central Asian expedition in 1928 (Simpson, 1931). Together withAnagale, some remains ofEmbolotheriumof Brontotheriidae were also found. Ulangochuian age was named by Romer (1966) after the fauna from Ulan Gochu Formation, as representing the Early Oligocene Asian Land Mammal Age. The Ulangochuian was correlated with Chadronian age of North America and Sannoisian age of Europe. Hence, Ulangochuian has been widely cited as Early Oligocene till 1997 Wang Banyue introduced the updated dating of the E/O boundary and the realigned age of Chadronian of North American Land Mammal Age, and revised the Ulangochuian as representing Late Eocene.

5 Magnetostratigraphic study in Jiuxi Basin

Huang et al. (1993) did magnetostratigraphical work in Jiudong and Jiuxi basins including a section in Huoshaogou (Fig. 1), where the Huoshaogou Formation was originally named. Referring the mammal fossils and pollen data from Huoshaogou Formation, they correlated the magnetozones 10-12 to Early Oligocene. The most recent work in the Huoshaogou section was carried out by Dai et al. (2005). Totally 11 normal and 11 reversal polarity zones were recognized in the Huoshaogou Formation. They cited the Shanmacheng fossils in the basal Baiyanghe Formation, as indication of no later than Early Oligocene. They further cited typical Oligocene mammalian data from a locality of Baiyanghe Formation a few hundred kilometers away. Hence, they correlated the polarity zones in the Huoshaogou Formation to Chrons 13-18n respectively.

6 Reinterpretation of the geological age of Mimolagus and Anagalopsis

By comparison with the EoceneAnagaleandMimolagusfrom Nei Mongol,MimolagusrodensandAnagalopsiskansuensisfrom Shanmacheng may also be Eocene in age. Our survey shows the locality should be in the lower part of Huoshaogou Formation, not in the lower part of Baiyanghe Formation. The brick red sandstone and conglomerate level in the Shanmacheng section can be correlated to the Shanmacheng Member of Huoshaogou Formation in the Huoshaogou gulley. With the updated understanding of the biostratigraphy, the polarity zones of Dai et al. (2005) showing dominantly normal polarity, can be correlated to GPTS Chrons 13-18. Therefore, Bohlin’s fossil horizon can roughly be correlated to Chron 18n (～39-40 Ma), late Middle Eocene in age (Fig. 2).

Fig. 2 Reinterpretation of the paleomagnetic data of Dai et al. (2005) with the fossil horizon marked

7 Conclusive remarks

With the updated data and new interpretation of the paleomagnetic data, the Huoshaogou Formation in the Jiuxi Basin may be referred to late Middle Eocene-Late Eocene in age. The late Middle Eocene age ofMimolagusrodensnarrows the time gap with its close relativeGomphosthat have been recorded from Early to Middle Eocene (Meng et al., 2004, 2009). The archaic group Anagalidae may have become extinct before Oligocene.

Acknowledgements Sincere thanks are due to Dr. Wang Weitao, Zhang Zhuqi, Zhang Ning from Institute of Geology, China Earthquake Administration for their collaboration in the field work. Discussions with Prof. Li Chuankui from IVPP and Dr. Fostowicz-Frelik from Institute of Paleobiology, Polish Academy of Sciences greatly improve the understanding of fossils concerned in the text. We also thank reviewers and Dr. Anu Kaakinen for constructive comments.

Asher R J, Meng J, Wible J et al., 2005. Stem Lagomorpha and the antiquity of Glires. Science, 307: 1091-1094

Bleefeld A R, Mckenna M C, 1985. Skeletal integrity ofMimolagusrodens(Lagomorpha, Mammalia). Am Mus Novit,2806: 1-5

Bohlin B, 1940. Notes on the hydrography of western Kansu. Sino-Swed Exped Publ 10, III. Geol, 3: 1-54

Bohlin B, 1951. Some mammalian remains from Shih-her-ma-ch’eng, Hui-hui-p’u, western Kansu. Sino-Swed Exped Publ 35, VI. Vert Palaeont, 5: 1-47

Dai S, Fang X M, Song C H et al., 2005. Early tectonic uplift of the northern Tibetan Plateau. Chinese Sci Bull, 50(15): 1642-1652

Erbajeva M, 1986. Lagomorphs of Cainozoic (Systematic review). Proc Zool Inst, USSR Acad Sci, 156: 157-165

Fostowicz-Frelik L, Li C K, Mao F Y et al., 2015. A large mimotonid from the Middle Eocene of China sheds light on the evolution of lagomorphs and their kin. Sci Rep, 5: 9394: 1-9

Gradstein F M, Ogg J G, Schmitz M D et al., 2012. The Geologic Time Scale. Amsterdam: Elsevier. 1-1176

Hu C K, 1962. Cenozoic mammalian fossil localities in Kansu and Ningshia. Vert PalAsiat, 6(2): 162-172

Hu Y M, 1993. Two new genera of Anagalidae (Anagalida, Mammalia) from the Paleocene of Qianshan, Anhui and the phylogeny of anagalids. Vert PalAsiat, 31(3): 153-182

Huang H F, Peng Z L, Lu W et al., 1993. Paleomagnetic division and comparison of the Tertiary system in Jiuxi and Jiudong basins. Acta Geol Gansu, 2(1): 6-16

Li C K, 1977. Paleocene eurymyloids (Anagalida, Mammalia) of Qianshan, Anhui. Vert PalAsiat, 15: 103-120

Liang S J, Wang F T, Hu T et al., 1992. A new idea on the division of Tertiary formation in Jiuquan Basin. Acta Petrol Sin, 13(2): 102-108

Meng J, Bowen G J, Ye J et al., 2004.Gomphoselkema(Glires, Mammalia) from the Erlian Basin: evidence for the Early Tertiary Bumbanian Land Mammal Age in Nei-Mongol, China. Am Mus Novit, 3425: 1-24

Meng J, Kraatz B, Wang Y Q et al., 2009. A new species ofGomphos(Glires, Mammalia) from the Eocene of the Erlian Basin, Nei Mongol, China. Am Mus Novit, 3670: 1-11

Romer A S, 1966. Vertebrate Paleontology. Chicago: University of Chicago Press. 1-468

Simpson G G, 1931. A new insectivore from the Oligocene, Ulan Gochu horizon, of Mongolia. Am Mus Novit, 505: 1-22

Szalay F S, 1985. Rodent and lagomorph morphotype adaptations, origins and relationships: some postcranial attributes analyzed. In: Luckett W P, Hartenberger J L eds. Evolutionary Relationships among Rodents - a Multidisciplinary Analysis. New York: Plenum Press. 83-132

Wang B Y, 1997. Problems and recent advances in the division of the continental Oligocene. J Stratigr, 21(2): 81-90

Wible J, 2007. On the cranial osteology of the Lagomorpha. Bull Carnegie Mus Nat Hist, 39: 213-234

Zhai Y P, Cai T L, 1984. The Tertiary system of Gansu Province. Gansu Geol, 2: 1-40

Zhou M Z, Zhang Y P, Wang B Y et al., 1977. Mammalian fauna from the Paleocene of Nanxiong Basin, Guangdong. Palaeont Sin, New Ser C, 20: 1-100

Zhu L D, Wang C S, Zheng H B et al. 2006. Tectonic and sedimentary evolution of basins in the northeast of Qinghai-Tibet Plateau and their implication for the northward growth of the Plateau. Palaeogeogr Palaeoclimat Palaeoecol, 241: 49-60

甘肃骟马城哺乳动物化石地质年代的新认识

张兆群 王 健

(中国科学院脊椎动物演化与人类起源重点实验室，中国科学院古脊椎动物与古人类研究所 北京 100044)

步林(Bohlin, 1951)描述的酒西盆地骟马城(又译为十二马厂)化石(Mimolagus和Anagalopsis)在分类位置与地质年代上长期存在争议。步林认为仅依据这些化石无法确定其年代，只是推测可能为第三纪的早期。后人根据Anagalopsis与内蒙古发现的Anagale对比，认为该化石点大致时代为早渐新世。化石出自哪个层位也有不同的认识。有人认为化石发现在白杨河组下部，也有人认为化石层位属于火烧沟组。鉴于其重要意义，我们于2015年4月赴酒西盆地进行了调查。根据步林的地理位置描述与照片(Bohlin, 1940, 1951), 我们初步确定化石地点应位于骟马城沟内，火烧沟组下部骟马城段。对比内蒙古二连盆地伊尔丁曼哈组新发现的Mimolagus以及晚始新世乌兰戈楚组的Anagale, 认为步林描述的化石时代也应为始新世。根据对化石产出层位的重新认识，并对比戴霜等(2005)在火烧沟剖面的磁性地层学资料，推测火烧沟组的时代为中始新世晚期至晚始新世，骟马城化石层位的地质年代大致为中始新世晚期。Mimolagus的生存年代与其近亲Gomphos的差距大大减小，古老的Anagalidae也应该在渐新世之前就已全部灭绝。

甘肃嘉峪关骟马城，始新世,Mimolagus,Anagalopsis, 地质年代

Q915.873, P534.613

A

1000-3118(2016)04-0351-07

2015-06-12

Zhang Z Q, Wang J, 2016. On the geological age of mammalian fossils from Shanmacheng, Gansu Province. Vertebrata PalAsiatica, 54(4): 351-357

国家自然科学基金(批准号：41472003)和国家重点基础研究发展计划项目(编号：2012CB821904)资助。