Fossil Mammals of Asia: Neogene Biostratigraphy and Chronology

Introduction

Toward a Continental Asian Biostratigraphic and Geochronologic Framework

XIAOMING WANG, LAWRENCE J. FLYNN, AND MIKAEL FORTELIUS

Strategically located between North America, Europe, and Africa, Asia is at the crossroads of intercontinental migrations of terrestrial mammals. Asia thus plays a crucial role in our understanding of mammalian evolution, zoogeography, and related questions about first appearances of immigrant mammals in surrounding continents and their roles as major markers of biochronology. As the largest continent, Asia is the locus of origination for many groups of mammals and/or a site of significant subsequent evolution. The temporal and spatial distributions of these mammals in Asia thus provide a vital link to related clades in surrounding continents (figure I.1; see figure I.3). Such a strategic role is particularly apparent during the Neogene (~23–2.6 Ma) when Asia was intermittently connected to Africa and North America, and widely connected to Europe. Asia also occupies the greatest range of climates and habitats, from tropics to arctic and from rainforests to desert zones, often boasting the most fossiliferous regions with fantastic exposures and producing some of the richest fossil mammal localities in the world. It is therefore no exaggeration that Asia is central to a global understanding of mammalian history.

Such importance and opportunity notwithstanding, with the exception of a few instances (such as northern Pakistan), Asian mammalian biostratigraphy lags behind that science in Europe and North America for historical reasons, and many unresolved issues become bottlenecks for a detailed understanding of mammalian evolution elsewhere. Despite a relatively late start, a tremendous surge has been seen in recent decades in indigenous research, with international collaborations. Asian mammalian biostratigraphy is at a stage where local or regional frameworks are beginning to take shape, but there is no attempt at linking these regional syntheses to derive a continent-wide perspective. Asian vertebrate paleontologists are largely operating within the borders of their own countries, with infrequent communication across political boundaries. This is in contrast to situations in North America and Europe, where fluid exchange of information and ideas results in continuous refinement of continent-wide chronological schemes that are widely accepted among practitioners (e.g., Woodburne 1987; Steininger et al. 1996; Steininger 1999; Woodburne 2004).

During the last 30 years, an indigenous continental mammalian chronological system has been emerging, mostly based on existing, relatively well-studied faunas in China (Chiu et al. 1979; Li et al. 1984; Qiu 1989; Qiu and Qiu 1990, 1995; Tong et al. 1995; Qiu et al. 1999; Deng 2006). These compilations, however, suffer from some shortcomings. Foremost is constant looking to Europe as a reference for relative correlations. To a certain extent, this is inevitable as Asia and Europe constitute essentially a single continent during much of the Neogene and at any given time, the two continents share many faunal characteristics. However, this tendency of looking to the West for guidance also breeds a reluctance to build indigenous systems. As a result, discussions about chronology tend to make references to the European Neogene Mammal units (MN system), as if the latter’s stamp of approval would somehow make a more reliable age determination. This is unfortunate because many Asian faunas are derived from basins with long and continuous sections, which, with careful magnetic calibrations, can offer chronological control superior to the long-distance correlations that the MN system ever can achieve.

Figure I.1 Main Neogene vertebrate fossil-producing regions or localities in Asia discussed in this volume. A traditional definition of the continent of Asia is adopted here (areas without shade), even though such a definition is somewhat arbitrary and often does not represent natural boundaries of faunal provinces. The Aral Sea is currently much smaller than it is shown in this map.

Afghanistan: (12) Khurdkabul Basin; (13) Kabul Basin. China: (51) Botamoyin (XJ99005) section, Junggar Basin, Xinjiang Autonomous Region; (52) Chibaerwoyi section, Junggar Basin, Xinjiang Autonomous Region; (53) Dingshanyanchi section, Junggar Basin, Xinjiang Autonomous Region; (54) Xishuigou section, Tabenbuluk (Danghe) Basin, Gansu Province; (55) Olongbuluk section, Qaidam Basin, Qinghai Province; (56) Tuosu Nor section, Qaidam Basin, Qinghai Province; (57) Shengou section, Qaidam Basin, Qinghai Province; (58) Huaitoutala section, Qaidam Basin, Qinghai Province; (59) Bulong Basin, Tibetan Autonomous Region; (60) Kunlun Pass Basin, Qinghai Province; (61) Gyirong Basin, Tibetan Autonomous Region; (62) Zanda Basin, Tibetan Autonomous Region; (63) Guide Basin, Qinghai Province; (64) Xining Basin, Qinghai Province; (65) Linxia Basin, Gansu Province; (66) Zhangjiaping-Duitinggou section, Lanzhou Basin, Gansu Province; (67) Quantougou section, Lanzhou Basin, Gansu Province; (68) Tongxin Basin, Ningxia Autonomous Region; (69) Leijiahe, Lingtai area, Gansu Province; (70) Renjiagou, Lingtai area, Gansu Province; (71) Lantian area, Shaanxi Province; (72) Baode Basin, Shanxi Province; (73) Fugu area, Shaanxi Province; (74) Yushe Basin, Shanxi Province; (75) Jingle Basin, Shanxi Province; (76) Nihewan Basin, Hebei Province; (77) Damiao area, Inner Mongolia Autonomous Region; (78) Aoerban area, Inner Mongolia Autonomous Region; (79) Gashunyinadege area, Inner Mongolia Autonomous Region; (80) Tunggur Tableland, Inner Mongolia Autonomous Region; (81) Baogeda Ula area, Inner Mongolia Autonomous Region; (82) Jurh area, Inner Mongolia Autonomous Region; (83) Huade area, Inner Mongolia Autonomous Region; (84) Gaotege area, Inner Mongolia Autonomous Region; (85) Shanwang area, Shandong Province; (86) Sihong area, Jiangsu Province; (87) Nanjing area, Jiangsu Province; (88) Huainan area, Anhui Province; (89) Xiaolongtan Basin, Yunnan Province; (90) Lufeng Basin, Yunnan Province; (91) Yuanmou Basin, Yunnan Province; (92) Zhaotong (Chaotung) Basin, Yunnan Province. Georgia: (5) Bazaleti; (6) Eldari. India: (17) Ramnagar; (18) Nurpur; (19) Haritalyangar; (20) Chandigarh and Haripur Khol areas; (21) Kalagarh. Iran: (4) Maragheh. Japan: (93) Kani Basin, Gifu Prefecture; (94) Mizunami Basin, Gifu Prefecture; (95) Sasebo area, Nagasaki Prefecture; (96) Sendai area, Miyagi Prefecture; (97) Tochio area, Niigata Prefecture; (98) Aikawa area, Kanagawa Prefecture; (99) Iga-Omi Basin, Mie Prefecture; (100) Awaji Island, Hyogo Prefecture. Kazakhstan: (35) Aktau Mountain area; (36) Kalmakpay; (37) Pavlodar; (39) North Aral region; (40) northern Ustyurt region. Kyrgyzstan: (33) Ortok; (34) Djilgyndykoo and Akterek. Mongolia: (41) Altan-Teli and Hyargas Nor; (42) Valley of Lakes; (43) Kholobolchi Nor and Hung Kureh; (44) Shamar. Myanmar: (23) Chaungtha; (24) Yenangyaung; (25) Magway. Nepal: (22) Dang Valley. Pakistan: (14) Bugti; (15) Zinda Pir; (16) Potwar Plateau. Russia: (38) Novaya Stanitsa and Isakovka; (45) Tuva; (46) Tagay and Sarayskoe; (47) Aya Cave; (48) Tologoi 1; (49) Udunga; (50) Beregovaya. Saudi Arabia: (8) Al Jadidah; (9) Jabal Midra ash-Shamali; (10) Ad Dabtiyah; (11) As Sarrar. Tajikistan: (30) Daraispon; (31) Magian and Pedjikent. Thailand: (26) Li Mae Long Basin; (27) Mae Moh Basin; (28) Chiang Muan Basin; (29) Mun River Sand Pits. Turkey: (1) Pas¸alar; (2) Sinap; (3) central and western Anatolia. United Arab Emirates: (7) Al Gharbia. Uzbekistan: (32) Kairakkum.

This book is thus a coming-of-age attempt to synthesize the state of the art. By compiling mammal faunas from all major fossil-producing countries and regions in Asia, we hope to demonstrate that an Asian system can stand on its own, or at the very least be a starting point for further refinements that can ultimately build a major continental system in its own right. This book is the result of a collaborative effort by leading mammalian paleontologists of the world, who gathered in Beijing in 2009 and 2010 for two international conferences for the purpose of formulating an initial framework of Asian continental biostratigraphy (see following section). The complex nature of such a task, which often has to contend with incomplete information, makes it necessarily an interim solution intended to encourage additional research and further debate. A timely publication of this volume, however incomplete it may be in particular areas, stands to gain the most by laying down the principles and practices of mammalian biostratigraphy and geochronology from all regions and countries. Toward this goal, we are confident that a well-established mammalian biostratigraphic framework in Asia will contribute to a global picture of mammalian evolution in a refined chronological context.

BACKGROUND FOR BEIJING WORKSHOPS AND GENESIS OF THIS VOLUME

The idea of an Asian Neogene biostratigraphic meeting in Beijing with Asia-wide participation came up in late June 2007, while the senior author (X. W.) was in Beijing. The main impetus was the recognition that there is, thus far, no Asia-wide forum to discuss the feasibility of an Asian land mammal age system. As an emerging leader, China seems a natural place to take the initiative, as the country embarks on an unprecedented economic development with attendant re naissance in basic research. China also happens to straddle the mid-latitude desert zones that are often the best hunting grounds for vertebrate fossils in the world. Its long history of dragon bone hunting, going back hundreds of years in traditional medicine, gives it a head start in vertebrate paleontology.

Given these favorable conditions, a meeting proposal, with endorsements from Zhan-xiang Qiu, Zhu-ding Qiu, and Tao Deng, was submitted to the Institute of Vertebrate Paleontology and Paleoanthropology (IVPP) in early July 2007. A symposium volume was also included in the proposal. However, or gan i za tion al efforts did not begin in earnest until April 2008, when IVPP noted that such a meeting would be opportune as a celebration of its 80th anniversary. At this point, co-editors of this volume (LJF and MF) agreed to be involved in the meeting organization and editing of the symposium volume. The main challenge was to raise substantial funds to pay for participants who were otherwise unable to attend. Toward that end, we secured funding from the National Science Foundation (NSF, U.S.), National Natural Science Foundation (NSFC, China), and Society of Vertebrate Paleontology, as well as institutional support from the IVPP. In particular, we adopted the Critical Transitions workshop (a NSF–NSFC cofunded workshop series on the critical transitions in the history of life) as a unifying theme for international collaborations.

The Neogene terrestrial mammalian biostratigraphy and chronology in Asia—a workshop and symposium toward the establishment of a continent-wide stratigraphic and chronologic framework was convened at the IVPP over 3 days, June 8–10, 2009, followed by a 4-day post-conference field trip to the Linxia Basin in Gansu Province. More than 70 scholars and graduate students participated in the workshop, with representation from 19 countries, including Austria, China, Finland, France, Germany, Great Britain, Greece, India, Iran, Japan, Mongolia, Pakistan, Russia, Spain, Sweden, Thailand, Turkey, United Arab Emirates, and the United States.

It became apparent during the workshop that existing Chinese mammalian biostratigraphic divisions possess the best potential as the core of an Asian framework, as summarized by Woodburne: The background of China’s long and fundamental role in developing a chronologic system was clearly recognized in this regard, and the array of approaches to developing chronological systems portrayed at this conference provided the Chinese organizers with considerable examples to draw upon in furthering their goals (unpublished report to the Society of Vertebrate Paleontology by M. O. Woodburne). As one of the chief architects of the Chinese system developed during the past 20 years, Zhan-xiang Qiu was tasked to form a working group for creating such a framework (Qiu et al., chapter 1, this volume). However, it was clear from the beginning, as well as in reviews of various draft s of manuscripts circulated during the workshop, that serious disagreements exist regarding conceptual issues as well as practical problems. Another forum would thus be necessary to give a full airing of the controversies. Toward that end, a second workshop was organized, again funded by the NSF-NSFC critical transitions theme. This second workshop was held at IVPP, March 8–9, 2010, and attended by a small group of key participants from the United States, Finland, and China.

This book, following a similar volume on North American mammals (Woodburne 2004), is the culmination of these efforts. It attempts to bring together state-of-the-art Asian biostratigraphy and geochronology with the widest representation possible.

SUMMARY OF WORKSHOP DISCUSSIONS AND RESOLUTIONS

One of the distinguishing features of the workshops is the open discussion about concepts and practices, as well as the diversity of opinions. Much reflection is given to practices elsewhere in the world. In particular, the European Neogene Mammal units (MN system) and North American Land Mammal Ages (NALMA) are closely scrutinized for strengths and weaknesses in the hope of building a better system. Many of the comments during the workshop are indicative of current sentiments regarding historic developments, and they are briefly summarized here as extractions from meeting minutes with original commentators cited in parentheses when appropriate.

There is general recognition that the European MN system, although very practical and widely used, has some serious limitations, mostly out of necessity rather than by design. By its own nature and often for lack of long stratigraphic sections with unambiguous superpositional relationships, the MN system is a formulation of biozonation that cannot distinguish diachrony even in cases of precise correlation, and the system would not be able to distinguish time differences in correlative faunas (L. Werdelin). Furthermore, correlation errors can be as much as two MN units above and below (M. Fortelius). Whenever possible, therefore, an Asian system should avoid the deficiencies in the MN zonation, which is undergoing revision to improve the basis of those units. For example, current work by Spanish colleagues is recalibrating MN units to base them on a true biostratigraphic framework (J. Agustí), but where this is done, the Iberian equivalents are often younger than those in mainland Europe.

Given the shortcomings of the MN system, the widely used chronostratigraphic stage (golden spikes and associated concepts) in the marine realm seems an attractive approach (M. Böhme). Furthermore, most of the marine Neogene stages have been ratified, and the All-China Stratigraphic Commission has been in full agreement with this approach and has attempted to set Chinese continental Neogene research in motion toward that goal (Z.-x. Qiu). However, there is strong opposition against a chronostratigraphic system by several participants, particularly those who champion an independent system as exemplified by the NALMA. The main problem with golden spikes is that, once nailed, they are no longer flexible, and an Asian system should be based on true biostratigraphy in multiple long sections that can be further refined and revised as new advancements come along (M. Woodburne; see further discussion in Some Conceptual Issues).

Given the often messy developments of continental land mammal systems, some openly wonder if we should not simply do away with a land mammal age system and use numeric ages instead (F. Bibi). In fact, a biozonation has never been given a high priority in the Siwalik sequence (J. Barry), and people working in South Asia are generally content in talking about absolute ages rather than land mammal ages (L. Flynn). However, most seem to recognize that land mammal ages will always have a place in the formulation of a chronological system because the biological component can never be subjugated under isotopic dating or paleomagnetic dating (M. Woodburne).

Another issue of major concern is the spatial distribution of mammal fossils. Geography is of paramount importance for a super-continent as Eurasia that spans great longitudes and latitudes and crosses many climatic zones. In South Asia and Southeast Asia, roughly the modern Oriental Zoogeographic Province, mammals share much greater similarities during much of the Neogene, whereas the low latitude faunas in southern China and southeastern Asia are generally unlike those from mid-latitudes in north China and the rest of central Asia (R. Hanta; L. Flynn). Nonetheless, mid-latitude faunas can oft en be recognized along great longitudinal spans, such as the Pikermian chronofauna originally recognized from late Miocene localities in Greece, which have comparable equivalents in north China (M. Fortelius).

While conventional biostratigraphic approaches are perhaps best employed to generate regional stratifications, the rising field of computational ordering (seriation) of localities based on taxonomic presence/absence information (e.g., Alroy 1992, 1994, 1998, 2000; Fortelius et al. 2006; Puolamäki et al. 2006) may well offer more general systems. Ultimately based on the essentially irreversible evolution of lineages and communities (climate driven or intrinsic), computational approaches are especially attractive as a potential route toward a future, continent-wide mammal chronology. Indeed, a preliminary study by Alroy et al. (1998) already suggested that the biochronological signal was stronger for Western Eurasia as a whole than for its western and eastern parts separated at 20 degrees eastern longitude. Such a result reflects well-known phenomena of the fossil record: regional persistence of core lineages and the connectedness of coeval communities through long-range dispersal of species. For the operational detection of both of these, two requirements are critical: standardized taxonomy and presence of exceptionally well-sampled Rosetta localities (Alroy 1992). Therefore, a key priority for computational as well as for conventional approaches to biostratigraphy is coming to grips with problems of synonymy and regional taxonomic dialects.

For the time being, it is clear that an Asian land mammal system faces some challenges common to all continents (fossil mammals are rare; sampling errors are high; diachrony is common) as well as unique challenges in Asia (uneven studies in different countries; lack of marine interface; shortage of datable volcanic rocks interbedded in sediments; high degree of zoogeographic differentiation; some degree of endemism). Recognizing these challenges, the workshop participants adopted the following resolutions by unanimous consent:

(1) an Asian chronologic system, independent from the European MN units, is needed; (2) such a system should be mainly based on biological events, associated with paleomagnetic and isotopic dates where available; (3) the existing Chinese system, imperfect as it is, can serve as a starting point that can evolve through time; (4) the benefit of such a system is a common framework in which hypotheses of biological events across the continent can be rigorously tested; (5) a committee headed by Zhan-xiang Qiu, Tao Deng, Zhuding Qiu, Chuan-kui Li, Zhao-qun Zhang, Ban-yue Wang, and Xiaoming Wang (additional expertise will be recruited as need arises) will work toward the above goal; and (6) additional subcommittees of relevant specialists to clean up taxonomies should be established.

SOME CONCEPTUAL ISSUES

Mammalian biostratigraphy has been and still is the primary means for Cenozoic terrestrial geochronology. Continental mammalian biostratigraphic frameworks are integral to related disciplines such as mammalian evolution, zoogeography, paleoecology, and paleoenvironment. Various chronologic frameworks have been established in all continents except Antarctica, but their qualities (precision and internal consistencies) vary greatly, with European and North American systems being the most mature and those of other continents far less so. In developing an Asian land mammal system, much of the focus, both in workshop discussion and in subsequent manuscript development, has thus centered on the best practices in Europe and North America.

Although the Chinese land mammal age system has implicitly or explicitly adopted certain aspects of the European or North American practices, past iterations have mostly been concerned with articulations of the empirical evidence instead of an examination of the methodologies (e.g., Qiu 1989; Qiu and Qiu 1990; Qiu et al. 1999). An introspective assessment of current practices in the world thus represents a welcomed first step to construct a thoughtful system that is both methodologically defensible and practically useful.

From the beginning of the first workshop, it became clear that a European-style MN unit system has serious shortcomings because of its general lack of biostratigraphic underpinnings. The MN system, while widely practiced, offers little guidance as a model for Asia. Asia, like North America, possesses all the potential for developing a framework based on biostratigraphy in long stratigraphic sections. Nonetheless, the MN system is by far the most influential in Asian biochronologic developments due to the wide connections between the two continents and the large number of shared taxa in various ages. So pervasive are the MN units that it is not uncommon for Asian faunas to be directly compared to European ages/MN units or simply to be labeled with MN designation.

Unity with International Code vs. Regional In de pen dence

One of the most controversial subjects during the two Beijing workshops is the desire to follow the International Stratigraphic Guide (ISG; Hedberg 1976; Salvador 1994). Intense debates center on the suitability of a chronostratigraphic system in continental settings with golden spikes (Global Stratotype Section and Point, or GSSP) nailed in a physical lithostratigraphic section. The debate is set against a background of recent trends in the Chinese stratigraphic community to adopt the ISG protocol, buoyed by the establishment of several Chinese GSSPs for the Mesozoic and Paleozoic eras (e.g., Yin et al. 2001; Chen et al. 2006). The All-China Stratigraphic Commission (2001) went as far as selecting many existing Chinese land mammal units as stages and briefly characterized each (within Neogene the following were included: Xiejian, Shanwangian, Tunggurian, Baodean, Gaozhuangian, and Mazegouan). To push these efforts further, the commission distributed grants to the IVPP to flesh out Cenozoic stages in China, which resulted in some preliminary boundary selections, mostly coinciding with those endorsed by the ISG (e.g., Deng et al. 2003; Deng et al. 2004; Deng et al. 2006; Meng et al. 2006; Deng et al. 2007).

Whereas the GSSP standard promoted by the ISG is largely accepted in the marine stratigraphic community, it is far from certain how a continental system should proceed given its inherent problems in depositional gaps, rareness of fossils, patchiness in distribution, and insularity of paleoenvironments. While there is general agreement that such factors call for regionally limited chronological systems, commonly at the continental scale or smaller, opinions are deeply divided regarding how to construct such a system and whether such a system should be consistent with the ISG recommendations. A prominent example is the North American Land Mammal Age system, which enjoys wide acceptance among North American vertebrate paleontologists but is at variance from the recommendations of the ISG. Fundamental to the premise of the NALMA is the recognition that there is no inherent reason why events in land mammal evolution should coincide with those of marine organisms from half a world away. In fact, part of the initial impetus by the Wood Committee to establish the North American provincial ages is an attempt to avoid the dangerous ambiguity, cumbersome circumlocution, or both when trying to correlate to the European standard time scale (Wood et al. 1941:2).

Following the recommendations by the All-China Stratigraphic Commission (2001), Qiu et al. (chapter 1, this volume) propose a Chinese Regional Land Mammal Stage/Age system that they envision will ultimately transition to one fully consistent with the ISG standards. Chronostratigraphic boundaries of such a system are based on multiple criteria of lithostratigraphy, magneto-stratigraphic reversals, and mammalian first appearances and faunal characterizations. In doing so, Qiu at al. point out that the NALMA also uses lithologic criteria, at least in the case of the lower boundary of the Arikareean. They further argue that land mammal ages cannot be equated to biochrons. In fact, in their opinion, biochrons have no place in a regional chronostratigraphic system. As a step further in making all land mammal stage/age systems conformable to the international standard, Qiu et al. propose that for those mammal ages whose lower boundaries are near the standard international boundaries of a higher rank, such as the Oligo-Miocene and Mio-Pliocene boundaries, the mammal age boundaries should coincide with the epoch boundaries.

Bringing their vast experience in the North American land mammal age system to bear, Woodburne, Tedford, and Lindsay (chapter 2, this volume) proposed a framework of an endemic North China mammalian biochronologic system as an evolving standard of temporal intervals that accounts for all of Neogene time without gaps or overlaps. They suggest that such a system represents informal biochronologic units, and until this system has been widely tested, formalized international chronostratigraphic standards should not be applied. Woodburne et al.’s premise is that a land mammal age system should always give fossil mammals prominent consideration. Methodologically, they strongly advocate for a single taxon definition of mammal age boundaries in order to minimize potential gaps and overlaps.

In a compromise approach, Meng et al. (chapter 3, this volume) used the Xiejian as an example to illustrate their single-criterion, single-taxon definition, but largely within the chronostratigraphic framework recommended by the ISG. As such, Meng et al.’s scheme allows future adjustments of boundary definition but it must be tied to a specific stratotype section. Their Xiejian example also explores the case where a stage/age in question roughly coincides with a major international boundary of higher rank (in this case, Oligo–Miocene boundary). They treat these two boundaries as strictly separate entities and place the Xiejian lower boundary 0.5 myr above the international Oligo–Miocene boundary.

This controversy pits chronostratigraphic boundary definition as a convention serving to standardize nomenclature against a more dynamic land mammal age scheme (as practiced by North American paleontologists), emphasizing empirical evidence and flexibility of shift ing boundaries. To a certain extent, the former seems to signal a desire to move toward an internationally accepted, marine invertebrate norm, whereas the latter represents a more self-confident approach to a regional, continental mammal-based system divorced from ISG standards.

CURRENT STATE OF ASIAN BIOCHRONOLOGY

Primarily as a working hypothesis and overview aid, we compile a generalized chart to summarize the state of continental Neogene mammalian biostratigraphy and chronology, usually based on the most recent published updates in the respective regions, including those in this volume (figure I.2). Not intended as an original synthesis, these diagrammatic summaries provide a measure of consistency in presentation of existing stratigraphic frameworks and thus serve as a quick index for existing works on fossil-producing basins. Efforts are made to preserve a sense of lithostratigraphic (formations, suites, etc.) and biostratigraphic relationships (fossil localities, faunas, faunistic complexes, etc.). Although we cite the original sources for individual columns, any errors or misinterpretations are entirely our own. Nor is this an exhaustive account of all Asian sites, although most of the well-known sites are included. Such an exercise invariably fails to capture the complexities and nuances of the regions being depicted, and readers are urged to examine the original sources (and citations within) for each locality or basin. In many basins, controversies exist for faunal interpretations, in some cases, with discrepancies of millions of years. In presenting individual stratigraphic columns, we did not attempt to analyze each regional scheme, although we did occasionally reinterpret magnetic correlations. The intention of this exercise is to put together, for the first time, all major fossil-mammal-producing regions in a series of charts, to draw attention to the different conceptual frameworks and different constructions of faunal relationships. We hope this will serve as a starting point to integrate various stratigraphic schemes.

Figure I.2a Asian terrestrial Neogene vertebrate-producing strata, mammalian faunas, and faunistic complexes (abbreviated as F and F C) or fossil-producing horizons (placed within a box), and their chronologic relationships. Solid lines above and below a block of strata indicate approximate duration of the strata (often constrained by magnetostratigraphy), and absence of such lines indicates uncertainty of the duration of sedimentation. We adopt the Neogene-Quaternary (Pliocene/Pleistocene) boundary at 2.6 Ma, as formally defined by the International Commission on Stratigraphy (Mascarelli 2009), and many of the faunas falling within the 1.8–2.6 Ma interval and formerly considered late Pliocene are not treated here. Locality numbers correspond to those in figure I.1. Major Neogene faunas and strata of western Asia: (1) Pas¸alar, Gönen Basin, Turkey (Andrews and Alpagut 1990); (2) Sinap, Turkey (Kappelman et al. 2003; numbers indicate select fossil localities); (3) central and western Anatolia, Turkey (Sen 1996; -F indicates fossil horizons); (4) Maragheh, Iran (Mirzaie Ataabadi et al., chapter 25, this volume); (5) Bazaleti, Georgia (Vekua and Lordkipanidze 2008; Vangengeim and Tesakov, chapter 23, this volume); (6) Eldari, Georgia (Vangengeim, Lungu, and Tesakov 2006; Vekua and Lordkipanidze 2008); (7) Al Gharbia, United Arab Emirates (Bibi et al., chapter 27, this volume); (8) Al Jadidah (Hofuf Formation), Saudi Arabia (Thomas 1983; Whybrow, McClure, and Elliott 1987; Whybrow and Clemens 1999; Flynn and Wessels, chapter 18, this volume); (9) Jabal Midra ash-Shamali (Hadrukh Formation), Saudi Arabia (Whybrow, McClure, and Elliott 1987; Whybrow and Clemens 1999; Flynn and Wessels, chapter 18, this volume); (10) Ad Dabtiyah (Dam Formation), Saudi Arabia (Whybrow, McClure, and Elliott 1987; Whybrow and Clemens 1999); (11) As Sarrar (Dam Formation), Saudi Arabia (Whybrow, McClure, and Elliott 1987; Whybrow and Clemens 1999; Flynn and Wessels, chapter 18, this volume).

Figure I.2b Major Neogene faunas and strata of South and Southeast Asia: (12) Khurdkabul Basin, Afghanistan (Sen 2001); (13) Kabul Basin, Afghanistan (Brandy 1981; Sen 1983, 2001); (14) Bugti and (15) Zinda Pir, Pakistan (Antoine et al., chapter 16, this volume; Flynn et al., chapter 14, this volume); (16) Potwar Plateau (Siwaliks), Pakistan (Barry et al., chapter 15, this volume; Flynn et al., chapter 14, this volume); (17) Ramnagar, India (Patnaik, chapter 17, this volume); (18) Nurpur, India (Patnaik, chapter 17, this volume); (19) Haritalyangar, India (Patnaik, chapter 17, this volume); (20) Chandigarh (including Patiali Rao, Ghaggar, and Nadah sections) and Haripur Khol areas, India (Patnaik, chapter 17, this volume); (21) Kalagarh, India (Patnaik, chapter 17, this volume); (22) Dang Valley, Nepal (Patnaik, chapter 17, this volume); (23) Chaungtha, Myanmar (Chavasseau et al., chapter 19, this volume); (24) Yenangyaung, Myanmar (Chavasseau et al., chapter 19, this volume); (25) Magway, Myanmar (Chavasseau et al., chapter 19, this volume); (26) Li Mae Long Basin, Thailand (Mein and Ginsburg 1997; Ratanasthien 2002; Chaimanee et al. 2007); (27) Mae Moh Basin, Thailand (Chaimanee et al. 2007; Coster et al. 2010); (28) Chiang Muan Basin, Thailand (Coster et al. 2010); (29) Mun River Sand Pits, Thailand (Chaimanee et al. 2006; Hanta et al. 2008).

Figure I.2c Major Neogene faunas and strata of Central Asia: (30) Daraispon, Tajik Basin, Tajikistan (Sotnikova, Dodonov, and Pen’kov 1997; Vislobokova, Sotnikova, and Dodonov 2001); (31) Magian and Pedjikent, northwestern Tajikistan (Sotnikova, Dodonov, and Pen’kov 1997; Vislobokova, Sotnikova, and Dodonov 2001); (32) Kairakkum, Fergana Basin, Uzbekistan (Sotnikova, Dodonov, and Pen’kov 1997; Vislobokova, Sotnikova, and Dodonov 2001); (33) Ortok, Kochkor Basin, Kyrgyzstan (Sotnikova, Dodonov, and Pen’kov 1997; Vislobokova, Sotnikova, and Dodonov 2001); (34) Djilgyndykoo and Akterek, Issyk Kul Lake, Kyrgyzstan (Sotnikova, Dodonov, and Pen’kov 1997; Vislobokova, Sotnikova, and Dodonov 2001); (35) Aktau Mountain area (Kordikova and Mavrin 1996; Lucas et al. 1997; Kordikova, Heizmann, and Marvin 2000; stratigraphic nomenclature and faunal contents cannot be easily reconciled among cited authors), Esekartkan and Adyrgan (Sotnikova, Dodonov, and Pen’kov 1997; Vislobokova, Sotnikova, and Dodonov 2001), Ili Basin, Kazakhstan; (36) Kalmakpay, Zaysan Basin, Kazakhstan (Vangengeim et al. 1993; Sotnikova, Dodonov, and Pen’kov 1997; Vislobokova, Sotnikova, and Dodonov 2001; Lucas et al. 2009); (37) Pavlodar, Pavlodar region, Irtysh River, Kazakhstan (Gnibidenko 1990; Vislobokova, Sotnikova, and Dodonov 2001; Zykin, Zykina, and Zazhigin 2007); (38) Novaya Stanitsa and Isakovka, Omsk region, Irtysh River, Russia (Zykin and Zazhigin 2004; Zykin, Zykina, and Zazhigin 2007); (39) North Aral regions, Kazakhstan (Lopatin 2004); (40) northern Ustyurt region, Kazakhstan (Lopatin 2004).

Figure I.2d Major Neogene faunas and strata of Mongolia and eastern Russia: (41) Altan-Teli and Hyargas (Khyargas, Khirgis) Nor, Mongolia (Pevzner et al. 1982; but see Tedford et al. 1991 for an alternative interpretation; Sotnikova 2006); (42) Valley of Lakes, Mongolia (Höck et al. 1999; Daxner-Höck et al., chapter 20, this volume); (43) Kholobolchi Nor and Hung Kureh, Mongolia (Flynn and Bernor 1987); (44) Shamar, Mongolia (Vislobokova, Sotnikova, and Dodonov 2001); (45) Taralyk-Cher, Tuva, Russia (Vislobokova 2009); (46) Tagay (Tagai) and Sarayskoe (Saray), Olkhon Island, Lake Baikal, Russia (arrows indicate widely divergent interpretations of the Tagay Fauna) (Daxner-Höck et al., chapter 22, this volume; Erbajeva and Alexeeva, chapter 21, this volume); (47) Aya Cave, western shore of Lake Baikal, Russia (Erbajeva and Filippov 1997; Sen and Erbajeva 2011); (48) Tologoi 1, (49) Udunga, and (50) Beregovaya of Transbaikal area, east of Lake Baikal, Russia (Erbajeva and Alexeeva, chapter 21, this volume).

Figure I.2e Major Neogene faunas and strata of Xinjiang and the Tibetan Plateau: (51) Botamoyin (XJ99005) section, Junggar Basin, Xinjiang Autonomous Region (Meng et al. 2006; Meng et al., chapter 3, this volume); (52) Chibaerwoyi section, Junggar Basin, Xinjiang Autonomous Region (Meng et al. 2006; Meng et al., chapter 3, this volume); (53) Dingshanyanchi section, Junggar Basin, Xinjiang Autonomous Region (Meng et al. 2008); (54) Xishuigou Fauna, Tabenbuluk (Danghe) Basin, Gansu Province (Wang, Qiu, and Opdyke 2003; Wang et al., chapter 10, this volume); (55) Olongbuluk Fauna, Qaidam Basin, Qinghai Province (Wang et al. 2007; Wang et al. 2011; Wang et al., chapter 10, this volume); (56) Tuosu Fauna, Qaidam Basin, Qinghai Province (Wang et al. 2007; Wang et al. 2011; Wang et al., chapter 10, this volume); (57) Shengou Fauna, Qaidam Basin, Qinghai Province (Wang et al. 2007; Qiu and Li 2008; Wang et al., chapter 10, this volume); (58) Huaitoutala Fauna, Qaidam Basin, Qinghai Province (Wang et al. 2007; Wang et al. 2011; Wang et al., chapter 10, this volume); (59) Bulong (Biru) Fauna, Bulong Basin, Tibetan Autonomous Region (Huang et al. 1980; Zheng 1980; Wang et al., chapter 10, this volume); (60) Yuzhu Fauna, Kunlun Pass Basin, Qinghai Province (Song et al. 2005; Wang et al., chapter 10, this volume); (61) Woma Fauna, Gyirong Basin, Tibetan Autonomous Region (Huang et al. 1980; Yue et al. 2004; Wang et al., chapter 10, this volume); (62) Zanda Fauna, Zanda Basin, Tibetan Autonomous Region (Deng et al. 2011; Wang et al., chapter 10, this volume); (63) Guide Fauna, Guide Basin, Qinghai Province (Zheng, Wu, and Li 1985; Fang et al. 2005; Wang et al., chapter 10, this volume); (64) Xiejia and Chetougou faunas, Xining Basin, Qinghai Province (Li and Qiu 1980; Li, Qiu, and Wang 1981; Qiu et al., chapter 1, this volume); (65) Linxia Basin, Gansu Province (Deng et al., chapter 9, this volume; Qiu et al., chapter 1, this volume).

Figure I.2f Major Neogene faunas and strata of the Loess Plateau: (66) Zhangjiaping and Duitinggou faunas, Lanzhou Basin, Gansu Province (Qiu et al. 2001; Qiu et al., chapter 1, this volume); (67) Quantougou Fauna, Lanzhou Basin, Gansu Province (Qiu 2001; Qiu et al., chapter 1 this volume); (68) Dingjiaergou Fauna, Tongxin Basin, Ningxia Autonomous Region (Qiu et al., chapter 1, this volume); (69) Leijiahe biozones I–V, Lingtai, Gansu Province (Zheng and Zhang 2001; Qiu et al., chapter 1, this volume); (70) Renjiagou Fauna, Lingtai, Gansu Province (Zhang et al. 1999); (71) Bahe Fauna, Lantian Basin, Shaanxi Province (Zhang et al. 2002; Kaakinen and Lunkka 2003; Zhang et al., chapter 6, this volume); (72) Baode Fauna, Shanxi Province (Zhu et al. 2008; Kaakinen et al., chapter 7, this volume); (73) Laogaochuan section, Fugu area, Shaanxi Province (Xue, Zhang, and Yue 1995; Zhang et al. 1995; Xue, Zhang, and Yue 2006); (74) Mahui, Nanzhuanggou, and Mazegou faunas, Yushe Basin, Shanxi Province (Tedford et al. 1991; Flynn, Wu, and Downs 1997); (75) Hefeng Fauna, Jingle Basin, Shanxi Province (Chen 1994; Yue and Zhang 1998); (76) Daodi Fauna, Nihewan Basin, Hebei Province (Cai et al., chapter 8, this volume).

Figure I.2g Major Neogene faunas and strata of Inner Mongolia: (77) Damiao section, Siziwang Qi, Inner Mongolia (Zhang et al. 2011); (78) upper and lower Aoerban, Balunhalagen, and Bilutu faunas, Inner Mongolia Autonomous Region (Wang et al. 2009; Qiu, Wang, and Li, chapter 5, this volume); (79) Gashunyinadege Fauna, Inner Mongolia (Meng, Wang, and Bai 1996; Qiu, Wang, and Li, chapter 5, this volume); (80) Tairum Nor, Moergen, and Tamuqin faunas, Inner Mongolia (Qiu 1996; Wang, Qiu, and Opdyke 2003; Qiu, Wang, and Li, chapter 5, this volume); (81) Ulan Hushuyin Nur and Baogeda Ula faunas, Inner Mongolia (Qiu, Wang, and Li, chapter 5, this volume); (82) Shala and Amuwusu faunas, Inner Mongolia (Qiu, Wang, and Li, chapter 5, this volume); (83) Bilike, Ertemte, Harr Obo, and Tuchengzi faunas, Inner Mongolia (Fahlbusch, Qiu, and Storch 1983; Qiu and Storch 2000; Qiu, Wang, and Li, chapter 5, this volume); (84) Gaotege and Huitenghe faunas, Inner Mongolia (Li, Wang, and Qiu 2003; Xu et al. 2007; Qiu, Wang, and Li, chapter 5, this volume).

Figure I.2h Major Neogene faunas and strata of eastern China and Yunnan: (85) Xiejiahe Fauna, Shandong Province (Deng, Wang, and Yue 2008; Qiu and Qiu, chapter 4, this volume); (86) Xiacaowan Fauna, Jiangsu Province (Li et al. 1983; Qiu and Qiu, chapter 4, this volume); (87) Fangshan and Liuhe faunas, Jiangsu Province (Bi, Yu, and Qiu 1977; Qiu et al., chapter 1, this volume); (88) Laodong, Xindong, Tiesiju fissure faunas, Anhui Province (Jin, Kawamura, and Tatuno 1999; Jin 2004; Tomida and Jin 2009); (89) Xiaolongtan Fauna, Xiaolongtan Basin, Yunnan Province (Dong 2001; Dong and Qi, chapter 11, this volume); (90) Shihuiba and Miaoshanpo faunas, Lufeng Basin, Yunnan Province (Qi 1985; Chen 1986; Yue and Zhang 2006; Dong and Qi, chapter 11, this volume); (91) Xiaohe and Shagou faunas, Yuanmou Basin, Yunnan Province (Zhu et al. 2005; Dong and Qi, chapter 11, this volume); (92) Shihuiba Fauna, Zhaotong (Chaotung) Basin, Yunnan Province (Chow and Zhai 1962; Zhang et al. 1989; Denise Su, pers. comm.).

Figure I.2i Major Neogene faunas and strata of Japan: (93) Dota locality and other fossil sites (marked with an -F), Kani Basin (Tomida et al., chapter 12, this volume); (94) terrestrial vertebrate fossil sites (marked with an -F), Mizunami Basin (Tomida et al., chapter 12, this volume); (95) Diatomys locality (-F), Sasebo area (Tomida et al., chapter 12, this volume); (96) Sendai area (Tomida et al., chapter 12, this volume); (97) Parailurus locality (-F), Tochio area (Sasagawa et al. 2003; Nakagawa, Kawamura, and Taruno, chapter 13, this volume); (98) Dolichopithecus locality, Aikawa area (Nakagawa, Kawamura, and Taruno, chapter 13, this volume); (99) Iga-Omi Basin (Nakagawa, Kawamura, and Taruno, chapter 13, this volume); (100) Awaji Island, Hyogo Prefecture (Nakagawa, Kawamura, and Taruno, chapter 13, this volume).

It is also immediately clear that there is much unevenness in concepts and in practices. At the conceptual level, countries in the former Soviet Union (such as Georgia, Tajikistan, Uzbekistan, Kyrgyzstan, Kazakhstan, and Russia) or those influenced by the Soviet Union (Mongolia, and China to a lesser extent) have used stratigraphic schemes combining various notions of litho- or biochronology. The stratigraphic term svita (here translated as suite), is not only defined by lithologic characteristics (as in formations in western countries), but is also laden with a connotation of time, often indicated by fossil content. When western concepts of separation of lithoand biostratigraphy are applied to some of the areas, major discrepancies can occur that are difficult to reconcile, such as in the Aktau Mountain area (Kordikova and Mavrin 1996; Lucas et al. 1997) and in the Zaysan Basin (Sotnikova, Dodonov, and Pen’kov 1997; Lucas et al. 2000). As a result, our summaries for these countries are often not strictly comparable to those found elsewhere (see figure I.2c). These are areas that can benefit greatly by applications of consistent criteria to evaluate the stratigraphic schemes.

In practical correlations, the European MN system continues to exert influences in many areas. In some cases, such as countries in the former Soviet Union and western Asian countries, the MN units sometimes have been directly projected to the local strata. The European influence can also be felt as far as Southeast Asia, sometimes with disparate results, such as in the Li Mae Long Basin in northern Thailand (Ginsburg 1984; Mein and Ginsburg 1997; Chaimanee et al. 2007). This basin was considered to have a late Early Miocene small mammal fauna or even as earlier Miocene (MN 4), but ongoing work has benefitted from paleomagnetic data (Chaimanee et al. 2007) that, together with continued systematic studies, place the assemblage in the Middle Miocene. Examples like these highlight the perilous nature of long-distance correlation to a European system that is itself full of uncertainties and ambiguities and the importance of establishment of an indigenous biostratigraphy.

In stratigraphic resolution, existing Asian frameworks span a full spectrum of resolving power of continental biostratigraphy. At the finest scale, the Siwalik sequences in Pakistan, well constrained by 47 magnetic sections, boast consistent resolution of up to 200,000 years or less for 80% of the more than 1,000 fossil localities, and 100,000 years or less for 50% of the localities. Such a remarkable precision is pushing the resolving power in terrestrial sedimentation to the limits (Barry et al., chapter 15, this volume) and can rival the resolution of any basin of continental deposits in the world. At the other end of the spectrum, however, crude biochronologic characterizations, oft en without any independent calibrations, are still widely practiced, which is the norm in many Asian countries.

Although a true sense of biostratigraphy for individual taxa within reasonably fossiliferous spans is emerging for a number of basins (e.g., Sinap, Maragheh, Siwalik, Junggar Basin, Valley of Lakes, Qaidam Basin, Lingtai, Bahe, Yushe Basin), in the majority of regions in Asia, nominal local faunas or faunas are still widely used as a traditional way to communicate an aggregate of taxa oft en spanning a certain stratigraphic thickness representing a certain amount of time. More inclusive terms, such as chronofauna (or faunistic complex) can be useful concepts to construct ideas of larger biota that span greater geographic and temporal ranges.

From the perspective of geologic time represented in various Asian regions, our charts show that Early Miocene has the largest gaps in the fossil records of almost every region in Asia. This is especially true for the beginning of the Miocene (23–20 Ma), during which preciously few localities have any records and those that have some data are represented by mostly small mammals.

Another conspicuous gap in the Chinese coverage has somewhat unexpectedly turned out to be the early part of the Late Miocene, the temporal equivalent of the European Vallesian (11.2–9.5 Ma). The reason why this was not at first realized has to do in part with lack of stratigraphic control and in part with the monsoon-driven climate history of East Asia. Recent fieldwork has revealed that several key Chinese Late Miocene localities are significantly younger than previously assumed (Zhang et al., chapter 6, this volume; Kaakinen et al., chapter 7, this volume), leaving the Vallesian time equivalent of China remarkably poorly sampled. Recent rediscovery and correlation of Bohlin’s Tsaidamotherium locality (now called Quanshuiliang section) in the Qaidam Basin reveal that much of the Quanshuiliang section corresponds to the magnetically calibrated Tuosu Fauna of the early Bahean (~11–10 Ma; Wang et al. 2011). However, while rich in large mammals, much of the fauna represents an endemic Tibetan Plateau assemblage that is not easily related to faunas elsewhere. It has also recently become clear that the general trend of climate change in China during the Late Miocene is the opposite of the global trend of increasing aridity seen in Europe and North America and that faunal evolution in China accordingly follows a different path (e.g., the reappearance in the record and survival of the anchitherine horse Sinohippus far into the Late Miocene). Evidence from multiple sources now shows that China instead became gradually more humid during this interval, most probably as a result of a strengthened summer monsoon (Fortelius et al. 2002; Passey et al. 2009). The Chinese mammal fauna of the early, dry part of the late Miocene is characterized by low diversity and endemism, and it is only the more humid part of the late Miocene, from about 8 Ma onward, that sees the proper, pancontinental Hipparion fauna established in China (Fortelius and Zhang 2006; Mirzaie Ataabadi et al., chapter 29, this volume). In this perspective, the potential for establishing a long stratigraphic sequence from Lingtai takes on a special importance, and it will therefore be of considerable interest and importance to see whether future fieldwork will verify the tentative Vallesian correlations suggested by Deng et al. (chapter 9, this volume).

ZOOGEOGRAPHIC COMPLEXITY

As the largest continent on Earth, Asia defies easy categorization. With vast latitudinal, longitudinal, and altitudinal spans, as well as the attendant climatic zonations, zoogeographic differentiations are profound (figure I.3). Indeed, in Alfred Russel Wallace’s (1876:map 1) classic map of zoogeographic provinces, the boundary between Palearctic and Oriental provinces was drawn within Asia, mostly along the southern slopes of the Himalaya Range and its lateral extensions. In other words, northern Asia and Europe are zoogeographically more similar to each other than either is to southern Asia. Plate tectonics along the India-Asia collision zone and its resulting uplift of the Himalayan Range and Tibetan Plateau are thus critical factors imposing a first-order organization of the Asian continent. This pattern of modern zoogeographic division can be traced back in deep time at least to the early Neogene, if not earlier, based on fossil mammals (Qiu and Li 2003, 2005; Flynn 2008), consistent with the early attainment of a high Tibet (e.g., Rowley and Currie 2006; Quade et al. 2011). Climatic differentiations are similarly recorded by Neogene mammal records (Fortelius et al. 2002; Fortelius et al. 2003; Fortelius et al. 2006; Liu et al. 2009). Given such complexity in geography and climate, questions naturally arose during the workshops as to the feasibility of devising an Asia-wide land mammal age system that can work across major zoogeographic boundaries. If Europe and East Asia within the Palearctic Province are to have a separate chronologic system, shouldn’t South Asia in the Oriental Province have its own?

The East–West Divide Between Europe and Asia

The Eurasian continent spans the entire eastern hemisphere and beyond. Since the disappearance of the epicontinental Turgai Sea by about early Oligocene, Europe and Asia have been a single connected landmass. Despite this continuity during the Neogene, however, distant faunas from the extreme ends in western Europe and eastern Asia show marked Early Miocene differences, although there is a tendency for increased similarity through time (Mirzaie Ataabadi et al., chapter 29, this volume). A climatic gradient is likely, since sheer distance alone probably cannot fully account for such faunal differences.

Diamond (1997) advanced the thesis that organismic (including human) migrations are easily achieved along the east–west axis because Earth’s atmospheric variances are oft en organized latitudinally; that is, organisms can readily adapt to habitats of similar climatic zones of similar latitudes. This is in contrast to the north–south axis, which entails the crossing of climatic zones. By this argument, western Europe and eastern China, both of similar latitudes, should share more faunal similarities despite their vast geographic distance. Existence of distinct faunas from Europe and eastern Asia thus indicates climatic differentiations (wetter Europe vs. drier central and eastern Asia) or distinct environmental barriers, such as deserts in central Asia and the Tibetan Plateau. Faunal distinction through much of the Neogene (few species in common) is the strongest rationale for an Asian land mammal age system independent of the European MN units. This is in contrast to North America, which has a much narrower longitudinal span, and its paleofaunas have even narrower distributions within the western half of North America (eastern North America is poorly fossiliferous). As a result, faunal differences between Pacific coastal states and the Great Plains are small enough to be subsumed within a single NALMA system.

Figure I.3 Schematics of inter- and intracontinental faunal interchanges and dispersals centered around Asia. Europe–East Asia faunal interchange entails largely the same latitudes in the east–west direction; the main barrier is the Tibetan Plateau and adjacent arid regions of Central Asia. Except for mammals adapted to Arctic regions, Asia–North America dispersals include a large vertical axis component along longitude, and mammals must cross different climate zones in order to reach to the other side. Thin air and high mountains present formidable barriers along the southern slopes of the Himalaya, which form a sharp zoogeographic boundary; to the east along the east coast of China, however, the boundary becomes fuzzy and a transitional zone shifts through time along with climate changes. Africa–Asia connection is intermittent during the Neogene. Gray tones in continents roughly reflect the amount of vegetation: white or light gray indicates desert or dry environments and darker gray indicates more vegetation coverage. Width of arrows is suggestive of magnitude of terrestrial dispersals.

Despite east–west faunal differentiations, however, broad faunal similarities can be recognized in much of western and Central Asia at select time periods. For example, the notion of a Pikermian paleobiome recognizes a wide swath of Eurasia during the late Miocene that is dominated by dry climate, increasingly open environments, and seasonally adapted mammals (Bernor et al. 1996). Such a widespread biome of long duration has been termed the Pikermian chronofauna (Eronen et al. 2009), which lends support for Asian land mammal ages spanning at least northern Asia. As demonstrated by Mirzaie Ataabadi et al. (chapter 29, this volume), such a concept may also be applicable in parts of Eurasia in the Middle Miocene, as represented by the Tunggurian chronofauna, although the evolving nature of this chronofauna from an earlier appearance in western Europe and migrating east to eastern China near the latest Middle Miocene implies diachrony. Such diachrony has obvious implications for correlation, a case in point being the carnivore genus Dinocrocuta, which in Europe and western Asia is a good indicator of early Late Miocene age and has been used to support a Vallesian correlation of Bahean age localities in China. Recent studies suggest, however, that Dinocrocuta has a primarily (or even exclusively?) Turolian age range in China, with the best-dated records so far clustering around 8 Ma (Zhang et al., chapter 6, this volume).

The North–South Divide Between North and South Asia

A first-order zoogeographic division between the Pale-arctic Province to the north and Oriental Province to the south was long recognized to be the result of Earth’s surface processes (Wallace 1876). Such a clear distinction is rooted in the following two interrelated processes: erection of a formidable geographic barrier in Tibet-Himalaya and its lateral extensions, and formation of summer monsoons in South and East Asia and winter westerlies and northwesterlies in northern China and central Asia. This factor, coupled with major west–east river systems, distinguishes much of China. A Palearctic/Oriental-style provinciality can be recognized since the early to middle Miocene based on small-mammal records in eastern China (Qiu and Li 2003, 2005), large mammals from the northern rim of the Tibetan Plateau (Qiu et al. 2001), and small mammals from South Asia (Flynn 2008). Furthermore, in contrast to increasing faunal homogeneity between east and west Eurasia during the Neogene, the north–south faunal division became progressively more clearly delineated through time as Tibet was being uplifted and its climatic effects became more pronounced. The above process thus presents the biggest obstacle in the establishment of a truly Asia-wide land mammal age system.

Intermittent Connections Between Africa and South Asia

Faunal exchanges between Africa and South Asia, either by direct migration through the Arabian Peninsula or by indirect routes of western Europe (across the Strait of Gibraltar), are evidenced by records from the Siwaliks and equivalent deposits of Dera Bugti and Sulaiman areas (Barry et al. 1991; Flynn et al. 1995; Antoine et al., chapter 16, this volume). Being in similar latitudes and warm climates, the main control of Africa–South Asia dispersal was by intermittent land corridors. It is thus not surprising that South Asia often has the largest number of African elements outside of Africa, and an Ethiopian-Oriental connection seems to be recognizable (Flynn and Wessels, chapter 18, this volume), featuring occasional dispersals in both directions, notably among rodents and primates.

Connection of North America and Asia

Since the early Miocene, immigrants to North America from Asia seem to suggest a closed Bering Strait for much of the time (Woodburne and Swisher 1995). The Bering Land Bridge undoubtedly acts as a filter that allows faunas in the Arctic realm to pass freely but severely limits those from middle or lower latitudes. Because of this limited faunal exchange, correlations of Asian and North American land mammal ages, which are entirely based on mid-latitude faunas, are not easy, and the NALMA did not have much influence on the developments of the Asian mammal system.

Contributions of Asia–North America faunal exchange are often asymmetrical; a large number of immigrant events have been recorded in North America, but far fewer mammals made it to Asia. Tedford et al. (2004:fig. 6.3) counted 88 allochthonous genera of Old World origin during the Miocene Epoch (Arikareean through Hemphillian); many of these became significant components of local communities in North America. With the exception of horses (Anchitherium, Hipparion, Equus), camels (Paracamelus), dogs (Eucyon, Nyctereutes, Vulpes), and several small mammals (the rabbit Alilepus, squirrels like Marmota, beavers, and birch mice [see Kimura, chapter 30, this volume]), mammals that dispersed from North America did not exert a corresponding presence in Asia. Although such a discrepancy potentially may be accounted for by sampling effects, at least in the Pliocene (Flynn et al. 1991), it is also possible that a larger Eurasian continent presented a more competitive environment for North American newcomers. One striking example is an early Pliocene Arctic North American fauna that shares close similarity with contemporaneous faunas from North China (Tedford and Harington 2003).

Embedded within the overall balance of exchange favoring entry of Asian forms into North America is a striking, apparently climate-driven exception. The dispersal of Eurasian ungulates into North America was discontinuous, greatly declining during the later Miocene. Between 15 and 5 million years ago only four ungulate genera of Eurasian origin are known from North America: Pseudoceras, Neotragoceros, Platybelodon, and Tapirus (Tedford et al. 2004). In contrast to the successful dispersal of horses and camels in the opposite direction during this interval, none of the new arrivals diversified after dispersal, and, with the exception of Tapirus, all had a short duration in the fossil record. Eronen et al. (in press) attribute this imbalance to the fact that during this time North America was significantly more arid than Eurasia, creating a situation where North American ungulates were literally pre-adapted to the conditions yet to appear in Asia, while Asian ungulates correspondingly lagged behind the environmental conditions already in place in North America. That the result is not due to sampling error is testified by continued successful dispersal of Eurasian carnivores into North America during the same interval, and by the fact that ungulate dispersal into North America resumed when the climatic imbalance disappeared in the Plio-Pleistocene.

CRITICAL TRANSITIONS

This book is the result of collaborative efforts in Beijing and a follow-up Los Angeles workshop, which are part of a Sino-U.S. collaborative research agenda on critical transitions in the history of life. The goal is to address critical transitions in geologic history that profoundly affect biological and environmental evolution on global scales. Once again, Asia, by its unique geographic position and geologic history, has much to offer in our understanding of global environmental changes. Mammal distributions in space (zoogeography) and time (biostratigraphy and geochronology) are two key components in any attempt to formulate ideas about paleoenvironmental change. In many ways, mammal biostratigraphy by itself offers evidences of critical transitions. In that sense, we hope this volume will provide the initial dataset and encouragement to stimulate further research on the various critical transitions.

Looming large among Asian Cenozoic geologic events is the rise of the Himalayan and Tibetan highlands and effects on the initiation of Indian and East Asian monsoon climates. Without doubt, Himalaya-Tibet, as an imposing physical entity in central Asia, is a first-order climate maker. Much debate, however, is centered on the timing and process of the coupling of mountain uplift and climate change and their feedback on erosion and weathering (e.g., Molnar 2005). From a paleontological perspective, mammals as a biological component and a chronological marker have much to offer in this debate.

The emergence of Himalaya-Tibet and the ensuing zoogeographic division of Palearctic and Oriental provinces affects mammal distributions in two ways. The rising Himalaya coupled with drastic changes in climatic zonation form an effective barrier for all but high-flying birds. Once again, fossil mammals offer direct evidence for this profound change. Furthermore, as consumers of vegetation, mammalian ungulates are also invaluable for assessing plant compositions. Isotope ratios of dental enamels, hypsodonty indices, microwear, and mesowear have become critical means to deduce plant coverage, paleotemperature, and precipitation. As the field matures, such ecometrics (Eronen et al. 2010) are likely to become welded into an increasingly quantitative paleoenvironmental framework that can be used in conjunction with paleoclimate modeling to constrain and refine reconstructions of past conditions and processes (Eronen et al. 2009).

ACKNOWLEDGMENTS

It goes without saying that a volume such as this is not possible without the contributions from all authors—we express our gratitude to all who took the time to undertake this worthy project. The Institute of Vertebrate Paleontology and Paleoanthropology provided logistic support in the two Beijing workshops, and we thank the numerous graduate students from the IVPP for their help and participation. We thank our publisher Patrick Fitzgerald, senior manuscript editor Irene Pavitt, assistant editor Bridget Flannery-McCoy, copyeditors Karen Victoria Brown and Richard Camp, production editor Edward Wade, designer Milenda Lee, and indexer Maria Coughlin for their tireless work to ensure that this book will come to fruition. We also appreciate the valuable comments and suggestions by two anonymous volume reviewers. We are grateful to Alexey Tesakov for his review of a late draft of this chapter and for providing important Russian literature on several localities. This book and its companion workshops in Beijing and Los Angeles benefited from financial support from the Sedimentary Geology and Paleobiology program of the National Science Foundation (U.S.) and its counterpart in the Chinese National Natural Science Foundation. In connection to these financial assistances, we would like to acknowledge Raymond L. Bernor, H. Richard Lane, and Lisa Boush, whose sustained support are keys to our success in putting together the largest gathering of mammalian paleontologists working on Asian continental biostratigraphy. The Society of Vertebrate Paleontology made it possible for three young scholars to attend. Finally, but certainly not least emphatically, we are greatly indebted to Zhan-xiang Qiu, who not only produced the key summary chapter on Chinese land mammal ages/stages but was more than generous in his financial support of the production of this volume through his Special Researches Program of Basic Science and Technology grant (No. 2006FY120400) from the Ministry of Science and Technology.

REFERENCES

All-China Stratigraphic Commission. 2001. Regional Stratigraphic Guide of China and Its Explanation. Revised ed. Beijing: Geological Publishing House.

Alroy, J. 1992. Conjunction among taxonomic distributions and the Miocene mammalian biochronology of the Great Plains. Paleobiology 18:326–343.

Alroy, J. 1994. Appearance event ordination: A new biochronologic method. Paleobiology 20:191–207.

Alroy, J. 1998. Diachrony of mammalian appearance events: Implications for biochronology. Geology 26:23–26.

Alroy, J. 2000. New methods for quantifying macroevolutionary patterns and processes. Paleobiology 26:707–733.

Alroy, J., R. L. Bernor, M. Fortelius, L. Werdelin. 1998. The MN System: Regional or Continental? Mitteilungen der Bayerischen Staatssammllung für Paläontologie und historische Geologie 38: 243–258.

Andrews, P. and B. Alpagut. 1990. Description of the fossiliferous units at Paşalar, Turkey. Journal of Human Evolution 19:343–361.

Barry, J. C., M. E. Morgan, A. J. Winkler, L. J. Flynn, E. H. Lindsay, L. L. Jacobs, and D. Pilbeam. 1991. Faunal interchange and Miocene terrestrial vertebrates of southern Asia. Paleobiology 17:231–245.

Bernor, R. L., N. Solounias, C. C. Swisher III, and J. A. Van Couvering. 1996. The correlation of three classical Pikermian mammal faunas—Maragheh, Samos, and Pikermi—with the European MN unit system. In The Evolution of Western Eurasian Neogene Mammal Faunas, ed. R. L. Bernor, V. Fahlbusch, and H.-W. Mittmann, pp. 137–154. New York: Columbia University Press.

Bi, Z.-g., Z.-j. Yu, and Z.-x. Qiu. 1977. First discovery of mammal remains from upper Tertiary deposits near Nanking. Vertebrata PalAsiatica 15:126–138.

Brandy, L. D. 1981. Rongeurs