Content uploaded by Guang-Xu Wang
Author content
All content in this area was uploaded by Guang-Xu Wang on Mar 12, 2019
Content may be subject to copyright.
Content uploaded by Guang-Xu Wang
Author content
All content in this area was uploaded by Guang-Xu Wang on Feb 22, 2019
Content may be subject to copyright.
Contents lists available at ScienceDirect
Earth-Science Reviews
journal homepage: www.elsevier.com/locate/earscirev
Invited review
The end-Ordovician mass extinction: A single-pulse event?
Guangxu Wang
a,⁎
, Renbin Zhan
a,b
, Ian G. Percival
c
a
State Key Laboratory of Palaeobiology and Stratigraphy, Center for Excellence in Life and Paleoenvironment, Nanjing Institute of Geology and Palaeontology, Chinese
Academy of Sciences, 39 East Beijing Road, Nanjing 210008, China
b
University of Chinese Academy of Sciences, Beijing, China
c
Geological Survey of New South Wales, 947-953 Londonderry Road, Londonderry, NSW 2753, Australia
ARTICLE INFO
Keywords:
End-Ordovician
Hirnantian
Mass extinction
Recovery
Stratigraphy
Glaciation
ABSTRACT
The end-Ordovician mass extinction (EOME) is widely interpreted as consisting of two pulses associated with the
onset and demise of the Gondwana glaciation, respectively, with the second pulse eradicating the distinctive,
glacially related Hirnantian benthic biota (HBB). A global review of occurrence data of latest Ordovician benthic
marine organisms reveals that virtually all warm-water benthic assemblages previously assigned to the HBB
comprise two distinct and clearly postglacial faunas, both younger (middle and late Hirnantian, respectively)
than the cool-water Hirnantia fauna (latest Katian to early Hirnantian). The newly recognised three Transitional
Benthic Faunas (i.e., TBFs 1–3) can be closely tied to graptolite, conodont, and chitinozoan biozonations, the
Hirnantian Isotope Carbon Excursion (HICE), and the glaciation, thereby providing an integrated, much higher-
resolution timescale for understanding the tempo and nature of the EOME. At this finer resolution, we postulate
a more profound impact of the first pulse of the EOME than hitherto envisaged, as evidenced by opportunistic
expansion of the Hirnantia fauna globally and the complete absence of metazoan reefs in its immediate after-
math. We also argue, based on high-quality data from well-documented benthic groups in South China (i.e.,
brachiopods, tabulate and rugose corals, trilobites, and sponges), that the magnitude of the second pulse of the
EOME caused by the deglaciation has been overestimated because the two postglacial faunas (i.e., TBFs 2–3)
were part of a subsequent recovery phase of marine ecosystems rather than contributing to biodiversity decline.
Thus, it is more plausible to reinterpret the EOME as a single-pulse, rapid event that was followed by a prolonged
initial recovery intermittently impeded by climatic shocks through the Hirnantian, prior to the onset of a pro-
gressive reestablishment of marine ecosystems during the early Silurian (Rhuddanian and Aeronian) associated
with an overall amelioration of climatic conditions.
1. Introduction
The end-Ordovician mass extinction (EOME) was the first of the
“Big Five”extinctions of the Phanerozoic (Raup and Sepkoski, 1982;
Stanley, 2016). Since being proposed by Brenchley and Newall (1984)
the EOME has traditionally been depicted as consisting of two pulses,
the first linked to the onset of rapid, extensive glaciation near the Ka-
tian/Hirnantian boundary followed by the introduction and flourishing
of a distinctive Hirnantian benthic biota (HBB). This biota was inter-
preted to have been substantially extinguished during the second pulse,
in the middle Hirnantian, due to widespread anoxia resulting from
postglacial sea level rise, and was succeeded by a survival interval prior
to broad biotic recovery since the beginning of the Silurian (Sheehan,
2001;Rong and Zhan, 2004a,b;Brenchley et al., 2003;Rong et al.,
2013;Harper et al., 2014). This two-pulse model has become a widely
accepted paradigm within which tempo, magnitude, pattern and dy-
namics of the EOME and the subsequent recovery have been variously
interpreted (Finnegan et al., 2011, 2012, 2016, 2017;Hammarlund
et al., 2012;Melchin et al., 2013;Rong et al., 2013;Bergström et al.,
2014;Ghienne et al., 2014;Harper et al., 2014;Jones et al., 2017;Zou
et al., 2018;Bergström and Goldman, 2019).
Recent investigations in South China, however, strongly suggest a
temporally composite nature of benthic faunas previously assigned to
the HBB, which has been shown to consist of three successive benthic
faunas (Wang et al., 2015, 2016, 2017, 2018). Only the oldest of these
is a cool-water type and therefore linked to the major glaciation,
whereas the two younger faunas, both being of warm-water nature, are
associated with postglacial warm-water environments. Contrary to
previous studies, these findings imply a far more complex evolutionary
scenario of benthic faunas after the first pulse of the EOME. If verified
https://doi.org/10.1016/j.earscirev.2019.01.023
Received 31 August 2018; Received in revised form 25 January 2019; Accepted 25 January 2019
⁎
Corresponding author.
E-mail address: gxwang@nigpas.ac.cn (G. Wang).
Earth-Science Reviews 192 (2019) 15–33
Available online 21 February 2019
0012-8252/ © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
T
globally, such a faunal succession will potentially provide a more pre-
cise and accurate timeline for the sequence of biotic and environmental
changes, and hence underpin a critical reevaluation of the tempo and
nature of the EOME. To further elucidate this key issue, our research
combines palaeontological and stratigraphic data primarily on two
major benthic groups, i.e., brachiopods and corals from several pa-
laeocontinents, which indicate that the three-phase evolution of the
Hirnantian benthic faunas recognised in South China probably re-
presents a global signal. This, along with published high-quality mac-
roevolutionary data about planktic graptolites, necessitates a re-
evaluation and reinterpretation of the timing and evolutionary pattern
of the EOME.
Early, middle and late subdivisions of the Hirnantian Age used
herein approximately correspond to the Metabolograptus extraordinarius,
the lower M. persculptus, and the upper M. persculptus graptolite bio-
zones, respectively, which align with the three distinctive benthic
faunas recognised. The glacial interval referred to herein represents the
major glaciation during the early Hirnantian, which possibly includes
small-scale interglacials; similarly, the postglacial interval commencing
in the middle Hirnantian follows the substantial demise of the
Gondwana ice sheet, though this interval may also contain minor gla-
cial episodes as shown in Fig. 1.
2. Recognition of three successive, globally identifiable benthic
faunas across the O–S transition
A global review of occurrence data of marine benthic organisms
across the Ordovician and Silurian transition enables the recognition of
three distinct, globally identifiable, successive benthic faunas between
the pre-extinction faunas and the typical Silurian faunas, termed herein
Transitional Benthic Faunas 1–3 (TBFs 1–3), respectively in ascending
order (Figs. 1, 2). In contrast to previous and succeeding faunas, these
TBFs are all relatively low in diversity, characterised by a mixture of
typical Ordovician and Silurian forms, and associated with
environmental perturbations. Although discerned principally on bra-
chiopods and corals, each TBF comprises all benthic marine groups in a
particular time interval, thus representing a distinctive evolutionary,
rather than environmental, entity with worldwide implications.
2.1. Concepts of the Transitional Benthic Faunas 1–3
2.1.1. TBF 1
This fauna thrived in the immediate aftermath of the EOME. It is
characterised by the cool/cold-water Hirnantia brachiopod fauna sensu
Rong and Harper (1988) including both typical and atypical types, but
excluding those we assign to TBF 2. Corals in TBF 1 are extremely rare.
At a few low-latitude sites, they are of fairly low diversity, consisting
exclusively of solitary rugosans. TBF 1 inhabited a wide variety of water
depths from near-shore to more offshore settings, and occupied nearly
all latitudes from the South Pole to the tropics during the early Hir-
nantian (Sections 3 and 5;Rong and Harper, 1988;Rong et al., 2002,
2010;Wang et al., 2017, 2018).
2.1.2. TBF 2
This fauna mainly inhabited warm-water seas of nearly all latitudes
immediately following the major glaciation, but is most readily re-
cognised in warm-water conditions at low-latitude regions. Although
traditionally interpreted as a contemporary ecological variant of TBF 1
based on brachiopods (Brenchley and Cocks, 1982;Brenchley and
Cullen, 1984;Rong and Li, 1999), its stratigraphically higher occur-
rence (i.e. lower M. persculptus Biozone) and faunal distinctiveness
support the recognition of TBF 2 as a separate evolutionary unit. Bra-
chiopods in TBF 2 are often dominated by Hindella or Dalmanella,and
generally lack key elements of the Hirnantia fauna, particularly Hir-
nantia itself. Examples of brachiopod-dominated TBF 2 include the
Dalmanella testudinaria–Dorytreta longicrura community in northern
Guizhou (Rong and Li, 1999), the Hindella–Cliftonia and the Dalmanella
associations in the Oslo–Asker district (Brenchley and Cocks, 1982),
NAINA
DDUHR
Y
REVODNALL
NAIT
NAN
R
I
H
NAI
CIVODRO E
TAL
NAITAK
susnecsa .A
-ar
txe.M
suiranidro s
u
t
plucsrep
.
M
Wufeng
Fm.
Lungma
-chi Fm.
Ulunda
Fm.
Jonstorp
Fm.
Glisstjärn Fm.
unit A
283 79
unit B
unit C
Shiqian
Fm.
unit A
unit B
1
Wufeng
Fm.
Husber-
gøya Fm.
.
bM
n
e
r
e
uakS
Høyerhol
-men Mb.
Pilodden
&Kalvøya
mbs.
6
.mF
ako
L
.
mF akoL
Lower
Mb.
Boda Fm.
.mFt
egrebsd
n
u
ms
O
Hindella
coquinas
B. kjerulfi
coquinas
Skultorp
Mb.
Skultorp
Mb.
Yuqian
Fm. Yuqian
Fm.
Yankou
Fm. Yan kou
Fm.
Anji
Fm. Anji
Fm.
4
Upper
Mb. Upper
Mb.
5
.mFene
yøg
na
L
Saldus
Fm.
.
m
F
a
g
idl
uK
Edole
Mb.
Adila
Fm.
10
Bernati
Mb.
?
SOUTH CHINA BALTICA
suci
f
icap
.P
.
m
FCYK
.mFCYK
no
itai
cal
ge
D
rojaM
n
oita
ica
lG
TBF 2
TBF 1
TBF 3
BENTHOS
GLACI-
ATI ON
?
?
Warm Cold oolitic limestones
Sta iūnaič&
Õhne fms. Varbola
Fm.
Solvik Fm.
Typical
Silurian
faunas
Pre-
extinction
faunas
???
?
.
m
F a
n
i
r
Ä
Röa
Mb.
Vohilaid
Siuge &
Torevere
mbs
Kama-
riku Mb.
11
Jonstorp
Fm.
Fig. 1. Stratigraphic evidence of the temporal structure of the Transitional Benthic Faunas (TBFs) 1–3 across the EOME recorded on the palaeocontinents of South
China and Baltica. 1—Shichang, Renhuai, NW Guizhou (Wang et al., 2018); 2—Dongkala South, Fenggang, N Guizhou (Wang et al., 2018); 3—Tunping, Shiqian, NE
Guizhou (Wang et al., 2018); 4—Yuqian, Lin'an, W Zhejiang (Rong et al., 2013); 5—Yuhang, Hangzhou, W Zhejiang (Rong et al., 2008a;Rong et al., 2013); 6—Oslo-
Asker district, southern Norway (Brenchley and Cocks, 1982;Bockelie et al., 2017); 7—Ållebergsände and Kinnekulle, Västergötland, southern Sweden (Bergström,
1968;Bergström et al., 2011a); 8—Borenshult drillcore, Östergötland, south Sweden (Bergström et al., 2011a, 2014); 9—Osmundsberget 4 and 5, Siljan area, central
Sweden (Kröger et al., 2015, 2016); 10—Ruhnu (500) drillcore, southern Estonia (Harper and Hints, 2016); 11—Northern Estonia (Kaljo et al., 2001, 2008;Ainsaar
et al., 2015). Note that some modifications are made in this correlation to incorporate our refined Hirnantian timescale (Section 3).
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
16
and Hindella coquinas in the Siljan area (Kröger et al., 2015). Corals in
TBF 2 are strikingly distinctive, consisting predominantly of diverse
solitary streptelasmatid rugosans, such as those from units B and C of
the Kuanyinchiao Formation at several localities of northern Guizhou
(He et al., 2007;Wang et al., 2018) and from the Skultorp Member of
the Loka Formation at Råssnäsudden of Östergötland, southern Sweden
(Neuman, 1969, 1975, 2003;Bergström and Bergström, 1996). In off-
shore facies, however, TBF 2 appears to have been replaced by sponge-
dominated benthic associations (Section 5;Botting et al., 2018). This
fauna is of middle Hirnantian age (Figs. 1, 2; for further discussion see
Section 3).
2.1.3. TBF 3
The succeeding TBF 3 inhabited similar geographic and climatic
conditions to those of TBF 2, and was subsequently replaced by typical
Silurian faunas around the O–S boundary. In shallow-water environ-
ments, it was initially recognised as part of TBF 1 (Amsden, 1974;
Amsden and Barrick, 1986) but was later regarded as distinct from, but
more or less coeval with, that fauna (Rong and Harper, 1988;Bergström
et al., 2014;Harper et al., 2014). More recently it was interpreted as
younger than both TBF 1 and TBF 2 (Wang et al., 2017, 2018). Bra-
chiopods in TBF 3 are represented by the pioneers of the succeeding
diverse Cathaysiorthis fauna in South China (Rong et al., 2013;Wang
et al., 2015; this paper) and the Edgewood fauna in east-central Laur-
entia and the Oslo–Asker District of Baltica (Brenchley and Cocks, 1982;
Rong and Harper, 1988;Harper and Hints, 2016;Wang et al., 2017).
The Edgewood coral fauna in TBF 3 consists of poorly diversified ru-
gosans and tabulates (McAuley and Elias, 1990;Young and Elias, 1995;
Elias et al., 2013;Wang et al., 2017). In deeper-water settings, similar
to TBF 2, this fauna is dominated by sponges (Section 5;Botting et al.,
2018), although brachiopod-dominated associations may also occur
(Rong et al., 2008a). The age of TBF 3 is probably restricted to within
the late Hirnantian (Section 3).
2.2. Stratigraphic evidence of the faunal succession of TBFs 1–3: a global
review
The latest Ordovician witnessed intense glacioeustatic sea-level
fluctuations, producing globally widespread hiatuses as well as marine
anoxic conditions, resulting in a spatially and temporally fragmentary
fossil record, especially of marine benthic organisms. This study focuses
on low-latitude, near-shore sections across the O–S transition in South
China and Baltica, where complete benthic faunal successions are well
preserved and best documented, thus providing direct stratigraphic
evidence for the relative ages of TBFs 1–3(Fig. 1). In establishing a
reliable temporal pattern for benthic organisms across this transition,
we mainly rely on distribution data of brachiopods and corals which are
relatively well-documented, supplemented by information from trilo-
bites and conodonts. We also review occurrences of TBF 2 and/or TBF 3
from other palaeocontinents, including Laurentia, Kolyma, Avalonia,
Altai-Sayan, Alxa, and Gondwana; those with good age constraints
provide additional support for such a globally-recognisable faunal
succession.
2.2.1. South China
TBF 1 inhabited a wide variety of water depths on the Yangtze
Platform of the South China palaeoplate, with a temporal span ranging
from the latest Katian to early Hirnantian (Rong et al., 2002, 2010;
Fig. 2. Arefined integrated Hirnantian stratigraphic framework, against which biotic and environmental changes across the EOME are also shown. The δ
13
C
carb
curve
is from the Ruhnu (500) drillcore of southern Estonia (modified from Harper and Hints, 2016). Data on biodiversity changes of the major benthic groups are given in
the Supplementary materials.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
17
Zhan et al., 2010;Wang et al., 2018; this paper). In contrast, bra-
chiopod-dominated TBF 2 and TBF 3 are hitherto known in near-shore
areas of northern Guizhou and Zhejiang, South China.
2.2.1.1. Northern Guizhou. At a number of near-shore localities, the
typical Hirnantia fauna is restricted to the lowest part (i.e., unit A) of the
Kuanyinchiao Formation (lower-middle Hirnantian), accompanied
occasionally by only a few small solitary rugose corals represented by
Pycnactoides (Wang et al., 2018). The remainder of this formation (units
B and C) yields brachiopods of the Dalmanella testudinaria–Dorytreta
longicrura community (Rong and Li, 1999) and rugosans of a diverse
fauna typified by Paramplexoides (He et al., 2007;Wang et al., 2018).
In the Shiqian area, abundant shelly fossils occur in the upper
Hirnantian Shiqian Formation, including the brachiopod
Cathaysiorthis?, and corals comparable to the Edgewood fauna (Wang
et al., 2015). A similar fauna has also been reported from the Wulipo
Bed probably of the same age in the Meitan area, northern Guizhou
(Wang et al., 2017; see Section 3 for a brief discussion on the age of the
Wulipo Bed).
2.2.1.2. Zhejiang. At most localities with O–S boundary successions in
Jiangxi and Zhejiang provinces, East China, the Cathaysiorthis fauna is
associated with graptolites of the Akidograptus ascensus Biozone (Rong
et al., 2013). Slightly older Hirnantian shelly faunas are known in
Zhejiang.
At Tangjia of Lin'an, western Zhejiang, postulated pioneers of the
Cathaysiorthis fauna (including Brevilamnulella,Eoplectodonta,
Eospirigerina,Fardenia,Leptaena,Levenea,Mendacella,Paracraniops,
Sulcatospira and Triplesia) appear in the M. persculptus Biozone, about
150 m above a level in which occurs a deep-water assemblage of the
typical Hirnantia fauna accompanied by graptolites of the M. extra-
ordinarius Biozone (Rong et al., 2013).
At Shizishan of Yuhang near Hangzhou, eastern Zhejiang, brachio-
pods related to the Cathaysiorthis fauna occur just above a level yielding
the Leangella–Mucronaspis (Songxites)Assemblage, both of which are
regarded as late Hirnantian in age (Rong et al., 2008a, 2013). Despite
the presence of some common constituents of TBF 1, the L.–M. As-
semblage also includes brachiopods Brevilamnulella and Eospirigerina?,
and the trilobite Niuchangella, indicating an affinity with TBF 3.
2.2.2. Baltica
Hirnantian benthic faunas are best developed in the epicontinental
seas of western Baltica, including much of present-day southern
Norway, central and southern Sweden, Estonia, Latvia and northern
Poland (Harper and Hints, 2016). Of particular interest are well-studied
near-shore sections where a relatively complete faunal sequence of
TBFs 1–3 is recorded.
2.2.2.1. Oslo-Asker District of southern Norway. In this area, Hirnantian
brachiopods were differentiated into several associations as ecological
variants of the Hirnantia fauna (Brenchley and Cocks, 1982). They can
actually be grouped into three categories in terms of their stratigraphic
positions, each containing a distinct coral assemblage. The Hirnantia
Association occurs in the lowest part of the succession, accompanied by
only a few corals. Directly above appears the Hindella–Cliftonia or
Dalmanella brachiopod association, with co-occurring corals dominated
by solitary streptelasmatids including Helicelasma,Grewingkia?,
Leolasma, and Streptelasma. The top of this succession yields
brachiopods referred to the Brevilamnulella and Thebesia associations
together with numerous corals such as heliotitids, favositids, and
streptelasmatids, both benthic groups being typical of the Edgewood
fauna. Higher up in the basal part of the Solvik Formation occurs a
diverse brachiopod fauna of clearly Silurian aspect (Baarli, 2014),
which was dated as latest Hirnantian based on the falling limb of the
δ
13
C
carb
excursion and, from a higher horizon, the first appearance of
conodont Ozarkodina oldhamensis, a common element of the Ozarkodina
hassi Biozone (Baarli, 2014;Bockelie et al., 2017); neither indicator,
however, is conclusive for such an age determination (see also Section
3).
2.2.2.2. Västergötland and Östergötland of southern Sweden. Hirnantian
rocks in the region, initially known as the “Dalmanitina Beds”, are now
referred to the Loka Formation, which consists of the Lower, the
Skultorp and the Upper members (Bergström and Bergström, 1996;
Bergström et al., 2011a). The complete succession of the formation is
developed at a number of localities in Västergötland; whereas in
Östergötland, the Lower Member is virtually missing (Bergström,
1968;Stridsberg, 1980;Bergström and Bergström, 1996;Bergström
et al., 2011a). The Lower Member is predominantly composed of
mudstone, occasionally with siltstone and limestone, containing rich
shelly fossils at some horizons; the Skultorp Member has a lithology of
locally oolitic conglomerate limestone yielding numerous rugose corals.
The Upper Member shows a similar lithology to the Lower Member but
is much thinner and less fossiliferous. Although brachiopods and corals
from the Loka Formation were previously treated as one
contemporaneous fauna, the following information provides some key
clues about their vertical successions.
The typical Hirnantia brachiopod fauna was documented at many
localities in Västergötland, such as Vrågården, Bestorp, Mt. Gisseberg,
Ållebergsände, Ekebacken and Kullatorp (Bergström, 1968), where the
richly fossiliferous Lower Member is developed (Stridsberg, 1980).
Bergström (1968) suggested that this fauna “seems to be restricted to a
muddy-silty environment”, indicating an occurrence probably from the
Lower Member, rather than coral-bearing limestones of the Skultorp
Member. It is also unlikely from the Upper Member because this hor-
izon “contains fewer fossils”(Bergström, 1968), and because brachio-
pods from the same horizon at Råssnäsudden of Östergötland are
probably of the Edgewood fauna, as presented below. This interpreta-
tion also plausibly explains the puzzling absence of the typical Hirnantia
brachiopod fauna in Östergötland (e.g., Borenshult) and part of Väs-
tergötland (e.g., Skultorp) where the Lower Member is completely
missing (Bergström and Bergström, 1996;Rong et al., 2008b;Bergström
et al., 2011a). At Borenshult, for example, Hirnantian benthic fossils,
which are famously known as the Borenshult fauna, include the corals
Borelasma,Fosselasma,Helicelasma and Streptelasma (Neuman, 1969,
1975, 2003), and the brachiopods “Atrypa”,Coolinia,Dalmanella,Epi-
tomyonia,Hindella, Hirnantia,Leptaena,Pholidops,“Rhynchonella”,Ske-
nidioides,Stegerhynchus,Sulevorthis and Triplesia (Rong et al., 2008b).
We suggest that the corals are typical of TBF 2; brachiopods, although
possibly including Hirnantia, appear to be of TBF 3 affinity, or possibly
are a mixture of TBF 2 and TBF 3.
Brachiopods and corals from the Skultorp and the Upper members
are well documented at Råssnäsudden of Östergötland. There, bra-
chiopods from the Skultorp Member, including Platystrophia sp.,
Dolerorthis? sp. and an indeterminate triplesiid, are extremely rare
whereas co-occurring corals are rich and diverse including Borelasma
crassitangens,Crassilasma sp., Fosselasma unicum and Streptelasma os-
trogothicum (Neuman, 1969, 1975, 2003;Bergström and Bergström,
1996), and they are clearly of TBF 2 affinity. Not surprisingly, bra-
chiopods from the overlying Upper Member contain some typical forms
of the Edgewood fauna, such as Leangella,Dolerorthis?, Cryptothyrella,
Aphanomena and Stegerhynchus (Bergström and Bergström, 1996). Note
that typical tabulate coral elements of the Edgewood fauna like Pa-
laeofavosites forbesiformes (?) and Plasmopora conferta were questionably
considered to occur in the Skultorp Member or much younger strata.
However, these corals have never been reported from the Skultorp
Member elsewhere in southern Sweden, and thus more likely come
from the Upper Member.
2.2.2.3. Siljan area of central Sweden. Hirnantian rocks in the area
include the Osmundsberget and the slightly higher Glisstjärn
formations (Kröger et al., 2016), with the former comprising a lower
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
18
succession dominated by Hindella terebratulina and an upper succession
characterised by Brevilamnulella kjerulfi(Rasmussen et al., 2010;Kröger
et al., 2015, 2016). These two brachiopods are common constituents of
TBF 2 and TBF 3, respectively, indicating their temporal correlations.
Rugose corals typical of TBF 2, such as Fosselasma unicum and Borelasma
sp., were described from the Osmundsberget Formation at the localities
Kullsberg and Nittsjö but without detailed stratigraphic information
(Neuman, 1969, 1975, 2003). Thus, we argue that TBF 1 is probably
missing in this region, which is consistent with the prominent
disconformity between the Osmundsberget and the pre-Hirnantian
Boda formations (Kröger et al., 2015, 2016).
2.2.2.4. Estonia and western Latvia. In southern Estonia and western
Latvia, the typical Hirnantia brachiopod fauna, lacking associated
corals, is known only from the lower Kuldiga Formation (i.e. the
Bernati Member and possibly the lower Edole Member) of early
Hirnantian age in offshore successions (Harper and Hints, 2016). The
upper part of the Kuldiga Formation yields abundant brachiopods
including Dalmanella and Cliftonia (Hints and Harper, 2015;Harper
and Hints, 2016), probably of TBF 2 affinity. This observation is
consistent with co-occurring corals from onshore sections in southern
Estonia (e.g., Ruhnu, Ikla, and Taagepera drill cores), where rugosans
are predominant among the shelly fossils (Harper and Hints, 2016). The
upper Hirnantian Saldus Formation, of siliciclastic and oolitic lithology,
is virtually barren or yields only very few brachiopods and no corals
(Hints and Harper, 2015;Harper and Hints, 2016). Although the faunal
affinity of this horizon is still uncertain, a TBF 3 affinity is more likely
due to its stratigraphic position, as well as the existence of the O. hassi
conodont fauna and the HICE (Kaljo et al., 2008;Harper and Hints,
2016). The overlying Stačiūnai and Õhne formations contain typical
Silurian faunas, but were recently dated as latest Hirnantian based on
the existence of the HICE (Ainsaar et al., 2015); this chemostratigraphic
indication, however, is problematic (see Section 3).
In northern Estonia, Hirnantian rocks were replaced by the Ärina
Formation of more near-shore facies, consisting of the lower dolomites
(the Röa Member), a middle reef complex (the Vohilaid, Siuge and
Torevere members) and the upper sandy carbonates (the Kamariku
Member) (Kaljo et al., 2001;Ainsaar et al., 2015). This formation was
traditionally correlated with the Kuldiga Formation in Latvia and
southern Estonia, and assigned an early-middle Hirnantian age (Kaljo
et al., 2001, 2008). Recently, however, Ainsaar et al. (2015) suggested
that the upper part of the formation (i.e., the Kamariku Member), which
recorded the falling limb of the HICE, is probably equivalent to the
upper Hirnantian Saldus Formation of the latter regions. Despite their
shelly faunas lacking clear signals of TBFs, we suggest that the lower
and middle parts of the Ärina Formation are correlative with the Ber-
nati and Edole members of offshore facies, and are therefore of early
and middle Hirnantian ages, respectively. This conclusion is based on
the fact that chitinozoans of the Spinachitina taugourdeaui Biozone only
occur in the lower dolomites but not in the reef complex, and that the
HICE reaches its maximum in the latter (Kaljo et al., 2008;Ainsaar
et al., 2015)(Section 3). Above the Ärina Formation rests the Varbola
Formation, which was recently assigned to the latest Hirnantian based
on evidence from δ
13
C
carb
chemostratigraphy (Ainsaar et al., 2015), but
this dating is inconclusive (Section 3).
2.2.3. Laurentia
In the interior of Laurentia, Hirnantian shelly fossils are typified by
the Edgewood fauna, which is known from the Edgewood Group of the
east-central USA (Amsden, 1974;Rong and Harper, 1988;Elias et al.,
2013;Harper et al., 2014), the upper Stonewall Formation of the Grand
Rapids Uplands of Manitoba (Demski et al., 2015) and the Manitoulin
Formation of southern Ontario (Stott and Jin, 2007). Associated grap-
tolites belong to the M. persculptus Biozone, and conodonts represent
the Ozarkodina hassi Biozone, both indicative of a late Hirnantian age
(Bergström et al., 2011b,2014;Demski et al., 2015).
Along the cratonic margin of Laurentia, the typical Hirnantia fauna
of TBF 1 is only known in the Percé area of Quebec, eastern Canada
(Lespérance et al., 1987). Intriguingly, rather different Hirnantian
shelly faunas are documented on nearby Anticosti Island and in the
Girvan district of southwest Scotland.
On Anticosti, despite intensive studies, debate is unresolved over
whether the lower and the middle part of the Ellis Bay Formation is of
Hirnantian or pre-Hirnantian age, primarily owing to somewhat am-
biguous biostratigraphic signals from various fossil groups (Kaljo et al.,
2008;Copper et al. 2013;Bergström et al., 2014). Since typical Hir-
nantian brachiopods, such as Hindella and Eospirigerina, were docu-
mented from near the base to the top of the formation, Copper et al.
(2013) favored a Hirnantian age for the entire formation, which is
tentatively followed herein. However, the brachiopod assemblage
seems more likely to be of TBF 2 and/or TBF 3 affinity rather than TBF
1, which means that TBF 1 is missing in this region. This is because
Eospirigerina is common in TBFs 2 and 3 but virtually never present in
TBF 1. Moreover, despite the occasional presence of Hirnantia itself (Jin
and Zhan, 2008), some other core elements of the Hirnantia fauna are
absent from this formation, such as Kinnella,Paromalomena, and Plec-
tothyrella. Corals are present throughout but it is difficult to clearly
assign these to TBFs (Elias et al., 2013), although the incoming of ta-
bulates like heliolitids and favositids in the upper two members of the
Ellis Bay Formation (particularly the uppermost Laframboise Member),
may mark a faunal shift from TBF 2 to TBF 3. Typical Silurian bra-
chiopods began to appear in the basal part of the succeeding Becscie
Formation (Copper and Jin, 2014), with a likely latest Hirnantian age
indicated by the presence of the HICE (Mauviel and Desrochers, 2016);
however, this HICE-based age assignment is not conclusive (Section 3).
In the Girvan district of Scotland, two successive associations of
Hirnantian brachiopods were documented from the High Mains
Formation (Harper, 1979, 1981) which appears to disconformably
overlie the pre-Hirnantian Drummuck Group. While the exact nature of
this boundary is unclear due to lack of exposure, Harper (1988) inter-
preted the High Mains Formation as channel fills cut into the Drum-
muck Group, based on the thickness, geometry, and sandstone lithology
of the former. If so, the lower Hirnantian is missing, which agrees with
our interpretation of the brachiopod faunas as TBF 2. The older asso-
ciation is dominated by Hindella and Eostropheodonta, along with Eo-
chonetes and Fardenia. In the younger association, Hindella continues its
strong dominance, with species of Rostricellula,Hypsiptycha,Hirnantia
and Eospirigerina being less common. Although Hirnantia is present in
the latter association, the other brachiopods most likely indicate a TBF
2affinity. The High Mains Formation is unconformably overlain by the
Mulloch Hill Conglomerate possibly of early Rhuddanian age, which
implies that TBF 3 is absent in the Girvan region.
2.2.4. Kolyma
Hirnantian shelly fossils, including brachiopods and corals that are
part of the Edgewood fauna of Laurentia, are best known at Mirny
Creek section, in the Omulev Mountains of Northeast Russia (Rong and
Harper, 1988;Koren and Sobolevskaya, 2008;Wang et al., 2017). Co-
occurring graptolites are indicative of the M. persculptus Biozone (Koren
and Sobolevskaya, 2008).
2.2.5. Avalonia
Reliable stratigraphic evidence of a faunal succession of the TBFs is
only known in North Wales, where the Hindella and Hirnantia bra-
chiopod associations were initially thought to be coeval, inhabiting
near-shore and more offshore environments, respectively (Brenchley
and Cullen, 1984), but this interpretation is speculative since they do
not occur in the same vertical succession. Given the almost complete
absence of corals in the Hirnantia Association compared with the pre-
sence of numerous corals in the Hindella Association (Brenchley and
Cullen, 1984), it is more likely that the Hindella Association is strati-
graphically higher than the Hirnantia Association, as shown in South
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
19
China and Baltica. It should be noted that, although not very common,
some key taxa of the typical Hirnantia fauna (e.g., Hirnantia and Kin-
nella) are also recorded in the Hindella Association around Meifod and
Llanfyllin, although information on the occurrence of these brachiopods
is lacking (Brenchley and Cullen, 1984;Brenchley et al., 2006). It is
possible that the typical Hirnantia Association does occur in these areas,
but from a stratigraphic level beneath that of the typical Hindella As-
sociation. Further investigation is needed to clarify their stratigraphic
relationship.
2.2.6. Altai-Sayan (Peri-Siberia)
Hirnantian benthic fossils are only well known in the Charysh-Inya
area of the Gorny Altai (Yolkin et al., 1988;Sennikov et al., 2014). In
the upper reaches of the Inya River, Kulkov and Severgina (1987, 1989)
documented abundant brachiopods from the Dorzhninskii Formation,
although without reliable age constraints due to a lack of graptolites.
Among five intervals of the formation, only intervals 1, 3 and 5 are
fossiliferous, yielding Trucizetina,Dedzetina,Ravozetina,Rostricellula
and Stegerhynchus from the oldest Interval 1, Dedzetina,Ravozetina,
Giraldibella,Streptis,Leptaena,Brevilamnulella,Rostricellula,Eospir-
igerina,Epitomyonia,Cliftonia,Alispira and Zygospirella from Interval 3,
and Kinnella,Giraldibella,Streptis,Eostropheodonta,Brevilamnulella,
Plectatrypa,Leangella,Zygospirella and Stegerhynchus from Interval 5.
Rong and Harper (1988) suggested that these brachiopods contain
elements of both the Edgewood (e.g., Brevilamnulella,Rostricellula, and
Eospirigerina)and the typical Hirnantia faunas (e.g., Kinnella,Eos-
tropheodonta, and Cliftonia). Because the identification of the key taxa of
the Hirnantia fauna (i.e., Kinnella) is problematic (Rong et al., 2018),
and the forms characteristic of the Edgewood fauna are present through
the formation, it is more likely that the fauna is entirely of TBF 3 af-
finity.
Yolkin et al. (1988) listed fossils from the Hirnantian limestones
near Ust'-Chagyrka village on the left bank of Chagyrka Creek of the
Gorny Altai area, including the brachiopods Thebesia and Brevilamnu-
lella, and the tabulate coral Catenipora, all indicative of TBF 3. This
shelly fauna is overlain by strata containing graptolites of the M. pers-
culptus Biozone (Yolkin et al., 1988).
2.2.7. Alxa
Latest Ordovician shelly fossils are only known in the present-day
Danmianshan area, Ejin Banner, western Inner Mongolia, palaeogeo-
graphically on the northern margin of the Alxa block (Zheng et al.,
1987). There, Rong et al. (2003) documented a poorly diversified
Hirnantian brachiopod assemblage from a 1 m thick limestone of the
Danmianshan Bed, which is dominated by Cliftonia and Dalmanella,
associated with less common Leptaena and Triplesia. The lack of char-
acteristic forms of the typical Hirnantia fauna led them to conclude that
these brachiopods do not belong to, but are coeval with, the Hirnantia
fauna, and to attribute the faunal difference to the tropical position of
the Alxa Block. Alternatively, we interpret the Danmianshan fauna as
belonging to the younger TBF 2 in view of its generic composition.
Additional support comes from the associated trilobite Niuchangella,
which has been demonstrated to have first appeared in TBF 2 in the
well-documented South China region (Zhou et al., 2004;Wang et al.,
2018;Wei and Zhan, 2018; Supplementary materials).
2.2.8. East Gondwana
Hirnantian shelly fossils are relatively poorly represented in the
tropical (eastern) part of the main Gondwana continent, largely as a
result of significant Devonian and younger tectonic disruption of the
region. Documented faunas are known from central southern Tasmania
(Laurie, 1991), and from the Wangapeka Valley on the South Island of
New Zealand (Cocks and Cooper, 2004).
2.2.8.1. Central southern Tasmania.Laurie (1991) described
brachiopods from the Westfield (now Arndell) Sandstone, the
uppermost formation of the Gordon Group, in the Florentine Valley in
central southern Tasmania. Two levels yielding brachiopods were
sampled in the Westfield Quarry; Hirnantia and Isorthis (Ovalella)
appear in the lower part of the formation at locality 6 of Laurie
(1991, Fig. 10).Onniella?, Eospirifer?, Isorthis (Ovalella), Hirnantia and
an indeterminate leptaenid were collected 40 m stratigraphically higher
(locality 11) associated with graptolites indicative of the M. persculptus
Biozone (Banks, 1988;Rong et al., 1994). The older fauna may belong
to TBF 1, while the younger one is a possible representative of TBF 2 or
TBF 3, as implied by the appearance of Eospirifer?.
Additionally, Laurie (1991) reported Hirnantia from the Westfield
Sandstone in a road cut near the turnoffto Westfield Quarry (locality
2), and Hirnantia and Kinnella in a low road cut just west of the quarry
(locality 3). Unfortunately, their exact stratigraphic positions are un-
certain due to structural complexity and thick vegetation overgrowth
(Banks, 1988;Laurie, 1991), but Banks (1988) suggested that one or
both may occur below the fauna recorded from locality 11 in the
quarry. These two impoverished brachiopod faunas are probably of TBF
1affinity.
2.2.8.2. Wangapeka Valley of New Zealand.Cocks and Cooper (2004)
documented the brachiopods Cliftonia,Eostropheodonta,Leptaena, and
Plectothyrella, along with forms referred tentatively to Hindella and
Cryptothyrella, from the uppermost Wangapeka Formation of the
northern part of the South Island, with Hirnantia apparently lacking.
The trilobite Mucronaspis and two rugose corals co-occur with these
brachiopods (Cocks and Cooper, 2004). Although age control of this
shelly fauna is weak, we tentatively regard it as TBF 2.
2.2.9. West Gondwana
Hirnantian shelly fossils in the high-latitude (western) part of
Gondwana or peri-Gondwanan terranes consist of brachiopods and a
few trilobites, but lack corals. Brachiopods of TBF 1 (i.e., the Hirnantia
fauna) occur just below (e.g., Mergl 2011), associated with (e.g.,
Colmenar et al., 2019), or immediately above (e.g., Benedetto, 1986),
the glacial diamictites. Interestingly, as discussed below, these shelly
faunas found in postglacial sediments are characterised by the absence
of core elements of the Hirnantia fauna (e.g., Hirnantia and Kinnella),
and, when well dated, are of middle-late Hirnantian age, thus belonging
to TBF 2 or TBF 3.
2.2.9.1. Argentine Precordillera. At Mogotes Azules of San Juan
Province, Benedetto and Cocks (2009) described a brachiopod fauna
from the basal La Chilca Formation. This fauna includes dominant
Hindella, as well as (less commonly) an indeterminate pseudolinguloid,
Leangella,Eoplectodonta,Fardenia,Glyptorthis,Dolerorthis,an
indeterminate giraldiellid, Dalmanella, and Stegerhynchus, which we
regard as TBF 3 on the basis of its generic composition. Benedetto and
Cocks (2009) assigned this fauna to early or middle Rhuddanian in age
because “there are no Hindella crassa dominated assemblages known
from anywhere in rocks older? than Rhuddanian”. Considering that the
presence of the M.persculptus Biozone has been confirmed from the
same level at Cerro del Fuerte and Talacasto in this region (Benedetto
and Cocks, 2009), a late Hirnantian age for this fauna is more likely,
although the base of the La Chilca Formation has proven diachronous.
Around the Cerro Le Chilca, Cerro Cumillango and Villicum Ranges of
the same region, the La Chilca Formation overlies the Don Braulio
Formation yielding the Hirnantia fauna (Benedetto, 1986;Halpern
et al., 2014).
2.2.9.2. Northwestern Argentina, northern Chile and Asunción area of
Paraguay. In the western Puna region of northwestern Argentina,
brachiopods found in the postglacial sediments of the Lower Member
of the Salar del Rincón Formation include Hindella,Fardenia, and
Heterorthella (Isaacson et al., 1976;Benedetto and Sánchez, 1990;
Benedetto et al., 2015). The overlying strata (lower part of the Upper
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
20
Member of this formation) contain chitinozoans of the Spinachitina
oulebsiri Biozone with an indication of middle-late Hirnantian age (de la
Puente et al., 2015). The absence of typical elements of both the
Hirnantia fauna and TBF 3 in these post-glacial sediments and the age
constraint supports a TBF 2 affinity for this brachiopod fauna.
In the southern Salar de Atacama of northern Chile, some 150 km
west of the previous locality, a similar Hindella-dominated brachiopod
fauna was reported from Unit 2 of the Quebrada Ancha Formation,
which was assigned an early Llandovery age “based on the presence of
abundant Hindella”(Niemeyer et al., 2010). This fauna may also be TBF
2, or, less likely, TBF 3.
In the southern part of the Sierra de Santa Bárbara of northwestern
Argentina, Benedetto et al. (2015) documented a poorly diversified
brachiopod assemblage from the dark gray mudstones at the top of the
glacial-related Zapla Formation, consisting of Orbiculoidea,Dalmanella,
and Paromalomena. This fauna was assigned a broadly Hirnantian age,
mainly based on the combination of similarity to the Hirnantia fauna
and the occurrence of chitinozoans of the Spinachitina oulebsiri Biozone
from the same stratigraphic level at the Río Capillas section. We re-
assess it as a possible example of TBF 2.
In the Asunción area of Paraguay, the Eusebio Ayala Formation
yields the brachiopods Arenorthis,Plectothyrella?, Eostropheodonta, and
Hindella (Benedetto et al., 2013), with associated graptolites from the
same level indicative of the M.persculptus Biozone (Cingolani et al.,
2011;Alfaro et al., 2013). Although not exposed in the area, glacial
deposits were regarded as probably older than the Eusebio Ayala For-
mation based on a regional correlation (Benedetto et al., 2013). This
fauna is probably referable to TBF 2 based on its age and faunal com-
position.
2.2.9.3. Western Cape of South Africa. Hirnantian brachiopods have
hitherto been documented from the Soom Shale and overlying Disa
Siltstone members of the Cedarberg Formation in the region (Cocks
et al., 1970;Cocks and Fortey, 1986;Bassett et al., 2009;Gabbott et al.,
2017). The Soom Shale yields Kosoidea,Plectothyrella,Trematis and
Palaeoglossa (Bassett et al., 2009) while the overlying Disa Siltstone
Member has a similar generic composition, including Plectoglossa,
Orbiculoidea,Trematis,Heterorthella,Eostropheodonta and Plectothyrella
(Cocks et al., 1970;Cocks and Fortey, 1986). Due to a lack of key taxa
of the Hirnantia fauna, Rong and Harper (1988) concluded that the Disa
assemblage is not a typical member of the fauna, but “may be a related
assemblage of Hirnantian age”. Since both brachiopod-bearing
members occur above glaciomarine diamictites of the Pakhuis
Formation and have been considered as postglacial deposits (Gabbott
et al., 2017), and the Soom Member contains chitinozoans of
Spinachitina oulebsiri indicative of a middle-late Hirnantian age
(Vandenbroucke et al., 2009), these brachiopods are obviously
younger than the Hirnantia fauna. A TBF 2 affinity is more likely
owing to the absence of common members of TBF 3.
2.2.9.4. Southern France. In the Mouthoumet massif, Álvaro et al.
(2016) recognised five allegedly coeval brachiopod associations of the
Hirnantia fauna in the diamictite-free, shale-dominated Marmairane
Formation. Among these, however, only Association 5 actually belongs
Fig. 3. Three biostratigraphically well-constrained Hirnantian successions in South China, particularly showing an integration of TBFs 1–3 with graptolite biozo-
nation. Thicknesses of rock units are not at same scale. See Section 3.4 for detailed information concerning carbon isotope excursions from these successions.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
21
to the Hirnantia fauna, whereas the other four apparently lack key forms
of the fauna (e.g., Hirnantia and Kinnella). Association 5 (potentially
equivalent to TBF 1) comes from a different locality to the others, and
their stratigraphic relationship is unclear. It is likely that Associations
1–4 (lacking Hirnantia) belong to the slightly younger TBF 2.
3. Temporal patterns of TBFs 1–3 and a refined Hirnantian
stratigraphic framework
The surprisingly similar faunal succession shared among widely
separated regions at both low and high latitudes indicates the globally
near-synchronous nature of TBFs 1–3. The three faunas can be closely
tied to graptolite, conodont and chitinozoan biozones, and to the HICE,
leading to a refined, higher-resolution Hirnantian timescale (Fig. 2).
3.1. Integration with graptolite biozonation
TBF 1 has been securely dated by graptolites in successions world-
wide as substantially within the M. extraordinarius Biozone, although in
places it may extend downwards into the uppermost Paraorthograptus
pacificus, and upwards into the basal M. persculptus biozones (Rong
et al., 2002, 2010;Wang et al., 2018). At the Hirnantian GSSP section of
South China, the lower M. persculptus Biozone graptolites occur im-
mediately above the TBF 1-bearing Kuanyinchiao Formation (Wang-
jiawan North section; Fig. 3;Chen et al., 2006).
The ages of TBFs 2 and 3 are less precisely constrained because both
faunas often occur in carbonate successions of low latitudes where
graptolites are absent. A few exceptions include some near-shore lo-
calities of northern Guizhou, where TBFs 2 and 3 can be reliably tied to
the graptolite biozones (Fig. 3;Wang et al., 2015, 2018). Graptolites
overlying TBF 2 include, among others, Neodiplograptus shanchongensis,
N. minor and N. rhizinus, around Shichang of Renhuai (Chen et al.,
2005a), indicating the upper M. persculptus Biozone. Given its inter-
vening position between TBF 1 and TBF 3, TBF 2 here most likely
corresponds to the lower M. persculptus Biozone.
TBF 3 from the Shiqian Formation, in the Shiqian area of northeast
Guizhou, has been reliably dated as late Hirnantian (upper M. pers-
culptus Biozone) by overlying graptolites of the Akidograptus ascensus
Biozone and by the presence of the decreasing limb of the HICE (Figs. 3,
5;Wang et al., 2015, 2018), consistent with the occurrence of this fauna
from elsewhere around the world (Wang et al., 2017). Note that our
earlier view that this fauna could range into the middle Rhuddanian in
South China mostly relied on its occurrence in the Wulipo Bed in
Meitan of northern Guizhou (Wang et al., 2017, 2018). The age as-
signment of this rock unit was in fact initially deduced from graptolites
of the Cystograptus vesiculosus Biozone occurring in the immediately
overlying strata (Chen et al., 2001;Rong and Zhan, 2004b). Our un-
published chemostratigraphic data, however, show the existence of the
HICE in the Wulipo Bed, and therefore indicate a Hirnantian age.
Considering also the combination of their striking faunal similarity and
close geographic proximity, the age of the Wulipo Bed is most likely late
Hirnantian, similar to that of the Shiqian Formation.
3.2. Integration with conodont biozonation
Conodonts associated with TBF 1 are poorly represented, with
sparse faunas and low diversity. At the Hirnantian GSSP section near
Yichang in South China, for example, conodont sampling has been
unsuccessful in the TBF 1-bearing Kuanyinchiao Formation. However,
the underlying carbonate lens within the M. extraordinarius Biozone,
which is within the temporal span of TBF 1, yields elements of the
Amorphognathus ordovicicus Biozone (Fig. 3;Chen et al., 2006). A few
conodonts from the Bernati Member of the Kuldiga Formation in East
Baltica that yields the typical Hirnantia fauna are also indicative of this
biozone (Kaljo et al., 2008).
Conodonts associated with TBF 2 are also very poor in abundance
and diversity, but limited data show a mixture of typical Ordovician (A.
ordovicicus Biozone) and Silurian (Ozarkodina hassi Biozone) forms, for
example from the Skultorp Member of the Loka Formation in
Östergötland, Sweden. At Råssnäsudden of Östergötland, conodonts
from this horizon include not only elements of the A. ordovicicus
Biozone, like A. ordovicicus,B.circumplicata,P. liripipus, and Dapsilodus
mutatus, but also those of the O. hassi Biozone such as Walliserodus
curvatus and Ozarkodina? sp. (Bergström and Bergström, 1996). In-
triguingly, TBF 2 from the Edole Member of the Kuldiga Formation in
East Baltica contains the “Noixodontus”fauna, a sparse fauna char-
acterised by Noixodontus girardeauensis (Kaljo et al., 2008).
Conodonts in TBF 3 typically align with the O. hassi Biozone, as
demonstrated in the Shiqian Formation of northeastern Guizhou (Fig. 3;
Wang and Aldridge, 2010), the upper Stonewall Formation of the Grand
Rapids Uplands of Manitoba (Demski et al., 2015), the Manitoulin
Formation of southern Ontario (Bergström et al., 2011b), the Mosalem
and Wilhelmi formations of eastern Iowa and northeastern Illinois
(Bergström et al., 2012a), the Laframboise Member of the Ellis Bay
Formation on Anticosti of Canada (Zhang and Barnes, 2002;Bergström
et al., 2014), and the Leemon Formation of Cape Girardeau region of
Illinois and Missouri (Bergström et al., 2014).
3.3. Integration with chitinozoan biozonation
Hirnantian chitinozoans occur at both low and high palaeolatitudes
but differ conspicuously in terms of faunal composition (Figs. 2, 4). In
low-latitude regions, samples from the Hirnant Limestone Member of
the Foel-y-Ddinas Mudstone Formation at the Cwm Hirnant Quarry of
Wales that contains typical Hirnantia fauna yield chitinozoans of the
Spinachitina taugourdeaui Biozone (Vandenbroucke et al., 2008). This is
also the case in the TBF 1-bearing Bernati Member of the Kuldiga
Formation in the East Baltica (Kaljo et al., 2008). Chitinozoans co-oc-
curring with TBF 2, which are only known in the succeeding Edole
Member of the Kuldiga Formation, belong to the Conochitina scabra
Biozone (Kaljo et al., 2008). Unfortunately, no chitinozoan biozone has
thus far been established in TBF 3 bearing-strata at low latitudes.
In high-latitude regions of Gondwana, TBF 1-associated chit-
inozoans are best known in the Bou Ingarf section in the Central Anti-
Atlas of Morocco (Fig. 4). There, the Hirnantia fauna occurs through
much of the Lower Second Bani Formation (LSBF) and the entire Upper
Second Bani Formation (USBF) (Sutcliffe et al., 2001;Colmenar et al.,
2019). Chitinozoans from LSBF and USBF were recognised as the Ta-
nuchitina elongata and “Spinachitina oulebsiri”biozones, respectively
(Paris, 1990;Bourahrouh et al., 2004). However, as noted by
Bourahrouh et al. (2004), the zonal index S. oulebsiri (identified with
some uncertainty) actually coexists with species of the preceding T.
elongata Biozone in the USBF. Elsewhere in North Gondwana, S. ou-
lebsiri occurs without the T. elongata fauna in younger postglacial de-
posits, as is the case in Saudi Arabia (Paris et al., 1995), Mauritania
(Paris et al., 1998), South Africa (Vandenbroucke et al., 2009), and
northwestern Argentina (de la Puente et al., 2015). Such an inter-
mediate fauna in this region indicates a lower portion of the S. oulebsiri
Biozone. It also suggests that TBF 1 at high latitudes may have persisted
during the melting interval of the Gondwana ice sheet, and so occurs
later than its counterpart at low latitudes.
Chitinozoans of the S. oulebsiri Biozone associated with TBF 2 were
recently documented from the Soom Shale Member of the Cedarberg
Formation in Western Cape Province, South Africa (Fig. 4;
Vandenbroucke et al., 2009). In contrast, chitinozoans from TBF 3-
bearing rocks are currently unknown, but can be indirectly tied to the
same biozone, which has been shown to correlate with the M. pers-
culptus Biozone, as observed in the Hodh area of Mauritania (Fig. 4;
Paris et al., 1998;Underwood et al., 1998), and the Zagros Mountains of
Iran (Ghavidel-Syooki et al., 2011).
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
22
3.4. The calibrated carbon isotope chemostratigraphy and its relation with
the glaciation
Our recognition of TBFs 1–3 contributes to a refined stratigraphic
framework of Hirnantian carbonate rocks, thereby necessitating a re-
calibration in time of the HICE and a reconsideration of its relations
with biotic events and glaciation (Figs. 2, 5, 6).
3.4.1. δ
13
C
carb
excursion
To explore its general pattern, we compiled TBF-associated δ
13
C
carb
curves from Tunping of Shiqian, Shichang of Renhuai and
Honghuayuan of Tongzi in north Guizhou (Munnecke et al., 2011;
Supplementary materials), from the Borenshult and Rhunu (500)
drillcores in Baltica (Bergström et al., 2012b;Harper and Hints, 2016),
and from Short Farm of Cape Girardeau in North America (Bergström
et al., 2006), as shown in Fig. 5. Additionally, δ
13
C
carb
data from TBF-
bearing rocks elsewhere around the world are also included in the
following discussion.
Carbon isotope values from TBF 1-bearing rocks are least known
due to their siliciclastic-dominated lithology, but positive shifts with an
increasing trend generally occur, for example, at Shichang,
Honghuayuan, and in the Ruhnu (500) drillcore (Fig. 5). Note that
negative δ
13
C
carb
shifts were often observed in TBF 1-bearing rocks, for
example, at the Hirnantian GSSP section and other offshore localities in
South China (Fig. 5;Munnecke et al., 2011;Gorjan et al., 2012;Tu
et al., 2012;Chen et al., 2017), but these negative values most likely
resulted from strong diagenetic influence by
13
C-depleted carbon from
underlying and overlying black shales, and therefore did not represent a
primary marine δ
13
C
carb
signal (Chen et al., 2017).
Available data indicate that rocks yielding typical TBF 2 constantly
produce peak values, as analyzed in the Honghuayuan, Shichang,
Borenshult and Ruhnu (500) successions (Fig. 5). This is also the case in
the Hindella coquinas of the Osmundsberget Formation in the Siljan
area of Sweden (Ebbestad et al., 2015;Kröger et al., 2015).
Strata containing typical TBF 3 also exhibit high values indicative of
the HICE but in most cases with a decreasing trend, as described in well-
studied successions from northern Guizhou, east-central USA, and East
Baltica (Fig. 5). Similar δ
13
C
carb
curves are also observed in TBF 3-
bearing rocks from the Williston and Hudson Bay basins (Demski et al.,
2015) and southern Ontario of Canada (Bergström et al., 2011b), from
the Oslo region of Norway (Calner et al., 2017), and from the Mirny
Creek area (Kaljo et al., 2012).
Importantly, a similar δ
13
C
carb
curve has also been documented in
well-dated successions lacking evidence of TBFs at a few localities, such
as Vinini Creek and the Monitor Range of the western USA (LaPorte
et al., 2009), suggesting a general pattern of the HICE development.
3.4.2. Comparison of δ
13
C
carb
and δ
13
C
org
curves
For comparison with the newly calibrated δ
13
C
carb
development, we
also compiled high-quality δ
13
C
org
curves in several continuous, well-
dated successions (Fig. 6), including the Wangjiawan (Riverside) and
Honghuayuan sections of South China (Zhang et al., 2009;Gorjan et al.,
2012), the Röstånga-1 drillcore of southern Sweden (Bergström et al.,
2014), and the Roberts Mountains and Monitor Range of central Ne-
vada (Laporte et al., 2009).
Clearly, these δ
13
C
org
curves show a general resemblance with the
δ
13
C
carb
ones in having a gradually increasing trend from the start,
reaching the peak in the middle, and developing positive values toward
the end. However, it is worth noting that variations do occur in δ
13
C
org
values from different successions. For example, at Wangjiawan
(Riverside) and Monitor Ranges (Fig. 6), peak δ
13
C
org
values seem to
appear just prior to the middle Hirnantian. Another major variation
concerns the magnitude of the upper portion of the HICE; the δ
13
C
org
values of this interval at the Monitor Ranges are apparently higher than
Fig. 4. Three biostratigraphically well-constrained Hirnantian successions at high latitudes of Gondwana, particularly showing an integration of TBFs 1–3 with
chitinozoan biozonation. The correlation of these successions with anoxic and glacial events is also presented. Thicknesses of rock units are not to scale. Part of legend
as in Fig. 2. SSM = Soom Shale Member; DSM = Disa Siltstone Member.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
23
Fig. 5. Comparison of Hirnantian δ
13
C
carb
curves from representative sections in South China, Baltica, and Laurentia where TBFs 1–3 are excellently recorded. New δ
13
C
carb
data from Tunping of Shiqian and Shichang of
Renhuai in South China are given in the Supplementary materials. Stratigraphic correlation of the successions is based on the refined Hirnantian timescale in the present paper. Thicknesses of rock units are not to scale.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
24
Fig. 6. Comparison of Hirnantian δ
13
C
org
curves from stratigraphically continuous, well-dated sections in South China, Baltica and Laurentia. Stratigraphic correlation of the successions is based on the refined Hirnantian
timescale in the present paper. Thicknesses of rock units are not to scale. KYC Fm. =Kuanyinchiao Formation.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
25
those in successions of the Wangjiawan Riverside section and Röstånga-
1 drillcore (Fig. 6).
3.4.3. Discussion
A brief discussion regarding the duration and structure of the HICE
and its relation with the glaciation is provided below.
As to its duration, as suggested by previous authors (e.g., Bergström
and Goldman, 2019), the HICE began near the end of the Katian, and
substantially fell within the Hirnantian. However, it could continue
well into the earliest Silurian though with approximately baseline va-
lues, as recorded in well-dated successions in terms of both δ
13
C
carb
(Kaljo et al., 2012) and δ
13
C
org
(Melchin and Holmden, 2006;Gorjan
et al., 2012;Bergström et al., 2014). Thus, using the HICE to precisely
determine the Ordovician-Silurian boundary as advocated by some
authors (e.g., Baarli, 2014;Ainsaar et al., 2015;Mauviel and
Desrochers, 2016;Bockelie et al., 2017) seems somewhat questionable.
The structure of the HICE has been summarised by Bergström and
Goldman (2019) as consisting of two broad divisions, “the lower-middle
Hirnantian distinctive positive excursion (HICE) and the upper Hir-
nantian-lowest Silurian (lower Rhuddanian) interval with rather uni-
form isotope values of essentially baseline magnitude”. This is generally
true in most δ
13
C
org
profiles, although an exception is seen at Monitor
Range (LaPorte et al., 2009)(Fig. 6). For the δ
13
C
carb
development,
despite a general decreasing trend, the positive shifts in the late Hir-
nantian could be about 4‰or more (Fig. 5;Kaljo et al., 2012;Harper
and Hints, 2016;Mauviel and Desrochers, 2016). However, because of
the widespread stratigraphic gap present between rocks containing TBF
2 and TBF 3, it is difficult to determine whether the positive values of
δ
13
C
carb
from the two levels represent two distinct peaks or just dif-
ferent portions of a single excursion (δ
13
C
org
curves from continuous
successions seem to be in favor of the latter).
We also argue for a new interpretation of the temporal relationship
between the HICE and the glaciation based on the refined time fra-
mework, regardless of its origin. Following a gradual increase during
the major glaciation, the HICE reaches a peak in the initial postglacial
warming interval, which is in turn followed by another small-scale
positive shift connected to further (or another) postglacial warming
during the late Hirnantian.
4. Relations between TBFs 1–3 and Gondwanan glaciation
We provide palaeoclimatic interpretations of TBFs 1–3, and further
relate them to Gondwanan glaciation, as discussed below and sum-
marised in Fig. 2.
4.1. Cool-water or warm-water?
The cool-water nature of TBF 1 is widely accepted with the most
compelling evidence being that it occurs within or associated with
glaciomarine deposits at high palaeolatitudes (Sutcliffe et al., 2001;
Mergl 2011;Halpern et al., 2014;Colmenar et al., 2019). This is further
strengthened by observations that other associated fossil groups (e.g.,
corals, trilobites, and conodonts) are extremely rare to absent even at
low latitudes and that this fauna is generally restricted to siliciclastic
rocks or, less commonly, impure limestones with significant siliciclastic
input probably as a result of glacioeustatic sea-level fall.
In contrast, TBFs 2 and 3 are clearly of warm-water nature, as in-
dicated by the presence of numerous corals and of abundant carbonate
grains (ooids, peloids and/or oncoids) in host rocks in low-latitude
regions of Baltica (Stridsberg, 1980;Bergström et al., 2011a), Laurentia
(Amsden and Barrick, 1986) and South China (Wang et al., 2018)
(Fig. 1). Also understandably, both faunas in high-latitude regions of
Gondwana consistently occur above, rather than being associated with,
the glacial diamictites (see Section 2.2.9). This agrees well with the fact
that rocks of this age (e.g., middle-late Hirnantian) in these regions are
often iron-rich, which are interpreted as having formed during the early
stages of the extensive postglacial transgression (Van Houten, 1985;
Young, 1989;Benedetto et al., 2015).
4.2. Glacial, interglacial or postglacial?
4.2.1. Timing and duration of Gondwanan glaciation
We argue that the major glaciation as evidenced by glaciogenic
sediments was substantially restricted to within the early Hirnantian.
No reliable pre-Hirnantian glaciogenic deposits have been confirmed,
with the only possible exception from the Chirfa area of Djado, Niger,
where the glacial sediments were inconclusively dated as latest Katian
(P. pacificus Biozone) by the graptolite Metabolograptus ojsuensis in as-
sociation with a piece of trinucleid trilobite (Legrand, 2011).
The most reliable dating of the glaciation comes from graptolites in
strata overlying the glaciomarine deposits at high latitudes of
Gondwana, which broadly indicate that glaciation ended as early as the
M. persculptus Biozone, and that postglacial transgression continued
well into the Silurian. Examples come from the Hodh area of Mauritania
(Underwood et al., 1998), the Murzuq Basin of Libya (Sachanski et al.,
2018), the Prague Basin of the Czech Republic (Štorch, 1990), western
Bulgaria (Sachanski, 1993;Chatalov, 2017), southern Jordan (Loydell,
2007), northern Saudi Arabia (Hayton et al., 2017), the Zagros Moun-
tains of Iran (Ghavidel-Syooki et al., 2011), and the San Juan area of
Argentina (Halpern et al., 2014). This is in excellent agreement with
indirect evidence from low latitudes, where, in continuous sections,
black shales above TBF 1-bearing rocks contain graptolites ranging
from the lower M. persculptus Biozone through the A. ascensus Biozone
to much younger Silurian biozones, for example, at the Hirnantian
GSSP section in South China (Chen et al., 2006) (see also Melchin et al.,
2013 for a review). Altogether, the global distribution of black shales
ranging in age from middle Hirnantian to early Silurian has been in-
terpreted as representing an oceanic anoxic event resulting from post-
glacial sea-level rise (Page et al., 2007;Melchin et al., 2013), indicating
that the major glaciation had substantially ended no later than the start
of the middle Hirnantian.
Some authors (e.g., Bourahrouh et al., 2004;Loi et al., 2010;
Ghienne et al., 2014;Le Heron et al., 2018;Colmenar et al., 2019)
suggested that the glacial maximum possibly persisted through into the
late Hirnantian, citing evidence from the Anti-Atlas of Morocco, where
chitinozoans of the Spinachitina oulebsiri Biozone (widely regarded as
middle-late Hirnantian in age) co-occur with the Hirnantia fauna in the
upper beds of glacial-related sedimentation. In fact, as stated in Section
3.3, these chitinozoans, coexisting with species of the preceding T.
elongata Biozone, indicate the lower part of the S. oulebsiri Biozone to be
of middle Hirnantian age, and can more reasonably be interpreted as
representing the initial melting phase of the Gondwanan ice cap. Hence,
this Moroccan example is consistent with the temporal span of the
major Late Ordovician glaciation identified from elsewhere around the
world such as those regions in high southern palaeolatitudes listed
earlier in this section.
Claims that local glacial episodes may have continued into the early
Silurian solely rely on purported glaciomarine records of this age from
South America (e.g., Díaz-Martínez and Grahn, 2007). This interpreta-
tion, however, has recently been challenged by new biostratigraphic
studies (e.g., de la Puente and Rubinstein, 2013;Benedetto et al., 2015;
de la Puente et al., 2015;Rubinstein et al., 2016). Benedetto et al.
(2015) proposed that these records probably derived from postglacial
gravity flows and that the age of the glacial deposits constrained by in
situ faunas is still within the Hirnantian, also in accordance with the
timing of maximum glaciation globally as discussed above. Even if the
local glacial episodes were confirmed, we believe that they cannot be
compared to the global one ranging from latest Katian to early Hir-
nantian in terms of its duration, magnitude, and impact on marine life.
4.2.2. Interpretations
Given the remarkable temporal coincidence with glacial deposits,
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
26
we argue that, of the three TBFs, TBF 1 is the only one tied to the
Gondwanan glaciation. Furthermore, contrary to previous studies, TBFs
2 and 3 are readily interpreted as recovery, indeed flourishing, of the
biota after the major glaciation due to their younger age and warm-
water attributes. Those previous studies have variously ascribed the
distribution of what we now recognise as TBFs 2 and 3 to latitudinal
differences (Rong and Harper, 1988;Sheehan, 2001;Bergström et al.,
2014;Harper et al., 2014;Bergström and Goldman, 2019), geological
isolation (Cocks and Fortey, 2002), or near-shore ecological variations
(Brenchley and Cocks, 1982;Rong and Li, 1999). In these studies, the
middle and late Hirnantian faunas, together with TBF 1, were inter-
preted to have coincided with a prolonged glacial interval. An alter-
native possibility –that all these faunas may have inhabited an inter-
glacial period –has recently been raised to explain the presence of
warm-water “Hirnantia fauna”and associated sedimentation
(Bergström et al., 2014;Bergström and Goldman, 2019). However,
none of these claims explain why TBF 1 is also present in near-shore
localities of low latitudes, and why it occurs with TBFs 2 and/or 3 in the
same succession yet consistently occupies the lowest stratigraphic po-
sition at these particular sections, as mentioned above.
Widespread unconformities are present at both the base and top of
strata containing TBF 3 (Figs. 1, 2, 3), possibly indicating sea-level falls
in shallow-water regions on a global scale. Such unconformities coin-
cide with parts of the HA and HB lowstands originally defined by
Bergström et al. (2006) as two major stratigraphic gaps beneath and
above the TBF 3-bearing Leemon Formation in the east-central USA.
These authors interpreted the lowstands as possibly corresponding to
major Gondwanan glacial episodes during the early and late Hirnan-
tian, respectively. Contrary to this interpretation, we postulate that
both stratigraphic hiatuses may be related to minor glacial episodes
within the middle-late Hirnantian postglacial phase, based on the ab-
sence of coeval high-latitude glacial deposits and glacially related cool-
water faunas worldwide.
5. Spatial patterns of TBFs 1–3 across the O–S transition
The bathymetric patterns of TBFs recognised herein are mainly
based on faunal data from well-documented low-latitude regions, in
particular South China, where TBFs occupied a wide range of habitats
from near-shore to offshore settings through the Hirnantian (Botting
et al., 2018;Rong et al., 2018;Wang et al., 2018;Wei and Zhan, 2018).
Our interpretations of latitudinal distribution patterns (Fig. 7) and on-
shore-offshore depth profiles characteristic of the low-latitude settings
(Fig. 8) endeavor to depict the differing spatial relationships among the
three successive Hirnantian faunas and how these patterns were con-
trolled by glacial and postglacial sea levels.
5.1. Early Hirnantian
TBF 1 featuring this interval is dominated by brachiopods and tri-
lobites known as the Hirnantia and Mucronaspis faunas, respectively
(Owen et al., 1991;Harper et al., 2014). The globally widespread Hir-
nantia–Mucronaspis community indicates an extremely low degree of
provincialism (Owen et al., 1991;Harper et al., 2014), with only two
biogeographic provinces, i.e., the high-latitude Bani and middle-low
latitude Kosov, recognised in brachiopods (Rong and Harper, 1988;
Harper et al., 2013;Wang et al., 2017;Fig. 7).
Both the Hirnantia and Mucronaspis faunas occupied a wide range of
water depths and were differentiated into distinctive onshore-offshore
associations (Fig. 8), as typically known in South China (Rong and
Harper, 1988;Zhou et al., 2004;Zhan et al., 2010;Rong et al., 2018;
Wei and Zhan, 2018). In low-latitude regions, like South China,
shallow-water associations may include rare small solitary rugose corals
(Wang et al., 2018), but no extensive reefal development has thus far
been confirmed. Note that previous reef records of this interval, as re-
viewed in Copper (2001), are all either pre-Hirnantian or middle-late
Hirnantian in age (see discussion below).
5.2. Middle Hirnantian
This interval witnessed the flourishing of TBF 2, which was char-
acterised by a Hindella (or Dalmanella)-dominated brachiopod fauna
lacking Hirnantia, and by the Mucronaspis trilobite fauna. Like TBF 1, it
had a near-global distribution (see Section 2.2), therefore indicating a
similarly weak provincialism (Fig. 7), although uncertainties con-
cerning its biogeographic pattern still need clarification.
TBF 2 exhibits striking onshore-offshore changes as a result of ob-
vious facies differentiation (Fig. 8). In shallow-water environments at
low latitudes, apart from the low-diversity Hindella (or Dalmanella)–-
Mucronaspis assemblage, TBF 2 also contains abundant solitary rugose
corals, e.g. in South China and Baltica (Neuman, 1969, 1975, 2003;
Bergström and Bergström, 1996;He et al., 2007;Wang et al., 2018).
Reefs may locally develop, with the only known record from the middle
part of the Ärina Formation in northern Estonia (Hints et al., 2000;
Kaljo et al., 2001) (see Section 2.2.2.4 for discussion concerning its
age); the shelly fauna from this reef shows a composition distinct from
that documented elsewhere in other low-latitude regions. In contrast,
no corals or reefs have been confirmed from this interval in similar
environments at high latitudes. TBF 2 was replaced by a sponge-
dominated type in offshore settings (Fig. 8), as demonstrated in vast
areas across South China (Li et al., 2015;Botting et al., 2017, 2018).
Although no contemporary sponges of comparable offshore environ-
ments have been found in other palaeocontinents, Botting et al. (2018)
predicted that such a sponge proliferation was likely to represent a
global signal.
5.3. Late Hirnantian
The preceding TBF 2 was replaced by the distinctive TBF 3 during
this interval, with the latter typified by brachiopods of the Edgewood
fauna or pioneers of the Cathaysiorthis fauna, and by Edgewood coral
fauna, as summarised above. The associated trilobites still contain
Mucronaspis, but appear to have been dominated by other forms, for
example Niuchangella, in South China (Wei and Zhan, 2018). Due to a
scarcity of faunal data especially in high-latitude regions (Section
2.2.9), the biogeographic pattern of this fauna is less known, but the
widespread distribution of the TBF 3 across low-latitude regions, as well
as one locality (i.e., Mogotes Azules of San Juan Province, Argentina,
see Section 2.2.9.1) at high latitudes, probably suggests a similarly
weak provincialism (Fig. 7).
The onshore-offshore pattern of TBF 3 is remarkably similar to that
of TBF 2 (Fig. 8). In near-shore well‑oxygenated conditions of low-la-
titude regions, this fauna made a rapid expansion, with reefs spor-
adically occurring in the tropics, as recorded in the Laframboise
Member of the Ellis Bay Formation on Anticosti Island (Copper, 2001),
and in the Leemon Formation of Missouri (Amsden, 1974;Amsden and
Barrick, 1986). In more offshore settings, sponges continued their
strong dominance in benthic ecosystems probably on a global scale
(Botting et al., 2018).
6. Synchronous evolution of benthos and plankton across the O–S
transition
The evolutionary history of the benthic faunas (i.e. TBFs 1–3) dis-
cussed above shows a good consistency with that of planktic groups. An
excellent example comes from extensively studied graptolites, which
experienced a replacement of the Diplograptina by the Neograptina
across the EOME through multiple-phased radiation (Chen et al.,
2005b;Finney et al., 2007;Melchin et al., 2011). According to these
studies, the first phase of the radiation was typified by a modest di-
versification of the Retiolitoidea, coinciding with the expansion of TBF
1 and with glaciation during the early Hirnantian. It was followed by an
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
27
expansion of the Monograptoidea starting at the post-glacial sea level
rise, at the time when TBF 2 began to flourish. The disappearance of the
Diplograptina lineages occurs near the end of the M. persculptus Bio-
zone, which appears to coincide with the faunal turnover between TBF
2 and TBF 3. Further work is needed to confirm whether this evolu-
tionary pattern is applicable to other planktic groups.
7. The EOME: a single-pulse event?
7.1. The two-pulse extinction model: a critical comment
The prevailing view of the second pulse of EOME has been based on
the substantial extinction of the temporally composite Hirnantian
benthic faunas near the base of middle Hirnantian, and the presence of
a subsequent gap in the benthic fossil record during the remainder of
the Hirnantian, this gap being interpreted as a survival interval after the
two-pulse extinction (e.g., Brenchley et al., 2001;Sheehan, 2001;Rong
and Zhan, 2004b;Rong et al., 2006;Harper et al., 2014). Our re-
cognition of TBFs 2–3 as middle-late Hirnantian postglacial faunas in-
dicates that the previous model results in an overestimate of the bio-
diversity losses for the second pulse of this extinction, and an
underestimate of its subsequent biotic recovery. The two-pulse model
has been further challenged by recent discoveries of diverse ex-
ceptionally well-preserved benthic faunas in the immediate aftermath
of the second pulse, including the Soom Shale Lagerstätte from South
Africa (Gabbott et al., 2017), and the sponge-dominated Anji fauna and
its equivalents from South China (Li et al., 2015;Botting et al., 2017,
2018). The Soom Shale fauna displays a surprising biodiversity, in-
cluding conodonts, arthropods, mollusks, annelids, brachiopods, and
some enigmatic taxa, providing an exceptional window into marine life
at high-latitudes immediately following the major glaciation (Gabbott
et al., 2017).
Another prominent flaw of the two-pulse model lies in an un-
balanced focus on benthos, because pelagic graptolites actually suffered
no prominent biodiversity loss after the initial pulse of the EOME (Chen
et al., 2005b;Finney et al., 2007;Melchin et al., 2011). Even for ben-
thos, evidence for the model mostly comes from the dominant bra-
chiopods characterised by the Hirnantia fauna and its purported con-
temporaries (Brenchley et al., 2001;Sheehan, 2001;Rasmussen and
Harper, 2011;Harper et al., 2014;Finnegan et al., 2016, 2017). Other
benthos, like trilobites, corals and conodonts, were also believed to
have suffered a marked two-phase reduction in diversity, however, to
varying degrees (Sepkoski, 1995;Kaljo, 1996;Brenchley et al., 2001;
Sheehan, 2001;Harper et al., 2014). Again, we regard all these previous
estimates of biodiversity losses of benthos to be based on an erroneous
conflation of TBFs.
7.2. A new extinction-recovery pattern of the EOME
The calibrated evolutionary history of marine organisms using our
refined time framework for the Hirnantian necessitates a significant
reassessment of the extinction-recovery pattern of the EOME. However,
Fig. 7. Models of latitudinal patterns of TBFs 1–3 through the Hirnantian.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
28
owing to varying degrees of documentation for different regions, esti-
mates of origination and extinction rates on a global scale are currently
unavailable, even for most of the major fossil groups with a substantial
fossil record. For this reason, we here focus primarily on faunal data
across the O–S boundary from South China in an attempt to understand
the EOME in a comprehensive and balanced way, since in this region
major marine groups, including brachiopods, corals, trilobites, sponges,
and graptolites, are well-recorded and documented in detail. The bio-
diversity changes of these groups in the region are evaluated on the
genus level based on data from 10 extensively studied sections with
TBFs 1–3 (Supplementary materials).
The results suggest that the EOME comprises just one pulse of mass
extinction, followed by a prolonged, initial recovery prior to the pro-
gressive return of a stable marine ecosystem with overall amelioration
of physical conditions since the beginning of the Silurian (Fig. 2). The
initial recovery consists of three phases, with each manifested by the
flourishing and replacement of TBFs 1–3, respectively, and punctuated
by dramatic environmental perturbations as indicated by the HICE and
widespread stratigraphic gaps.
7.2.1. Initial recovery phase 1 (latest Katian to early Hirnantian) =TBF 1
This phase immediately follows the so-called “first pulse”of the
EOME associated with the onset of the Gondwanan ice cap and coin-
cides with the development of the glaciation. As noted above, unlike
pre-extinction faunas, the contemporary cosmopolitan cool-water TBF 1
was widely distributed, covering nearly all latitudes and all water-
depths, and signifies the disappearance of provincialism. It implies that
the first pulse of extinction had a much more serious impact on the
marine ecosystem than was previously believed.
During this phase, some lineages, like trilobites, were consistently
Fig. 8. Postulated onshore-offshore pattern of TBFs 1–3 through the Hirnantian at low latitudes.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
29
dominated by poorly diversified opportunistic forms (Owen et al.,
1991;Wei and Zhan, 2018). Some groups, such as corals (particularly
tabulates) and stromatoporoid sponges, were even almost absent
(Nestor and Webby, 2013;Wang et al., 2017, 2018). In contrast, other
groups including planktic graptolites (Chen et al., 2005b;Finney et al.,
2007;Melchin et al., 2011), acritarchs (Vecoli, 2008), and the benthic
brachiopods (Rong and Harper, 1999;Rong and Zhan, 2004a;Harper
et al., 2014), experienced adaptive morphological radiations. Varia-
tions of recovery patterns observed in these marine groups reflect their
different responses to the difficult environmental conditions.
7.2.2. Initial recovery phase 2 (middle Hirnantian) = TBF 2
This phase commenced with the termination of the glaciation and
coincided with subsequent global warming and widespread anoxia.
Recognition of TBF 2 as the immediate postglacial fauna allows re-
evaluation of the magnitude of the second extinction pulse. Although
biodiversity losses still did occur, particularly in dominant brachiopods,
their magnitude was notably smaller than hitherto envisaged, with only
several key elements of the Hirnantia fauna becoming extinct. Trilobites
seem almost not to have been affected by this climatic change (Fig. 2;
Supplementary materials). In contrast, like the planktic graptolites
(Chen et al., 2005b;Finney et al., 2007;Melchin et al., 2011), other
groups of TBF 2 underwent a modest radiation (rather than an extinc-
tion) during this time interval. For example, in South China, rugose
corals showed a striking increase in generic diversity from 1 (in TBF 1)
to nearly 20 (in TBF 2) in shallow warm-water environments, and
sponges experienced a similar expansion with more than 75 species by
taking advantage of empty ecospaces in deeper, somewhat anoxic
conditions (Fig. 2;Wang et al., 2017, 2018;Botting et al., 2017, 2018;
Supplementary materials). In addition, this interval was also marked by
the return of local development of reefs in the tropics after a con-
spicuous gap during the early Hirnantian (Section 5). These observa-
tions suggest that the demise of the major glaciation appears to have
resulted in dramatic changes in onshore-offshore pattern of benthic
faunas, without remarkable biodiversity losses. Thus, we argue that the
so-called second pulse of the EOME should not be regarded as a mass
extinction.
7.2.3. Initial recovery phase 3 (late Hirnantian) = TBF 3
Similar climatic and oceanic conditions persisted into the late
Hirnantian, during which the distinctive TBF 3 flourished, typified by
further introduction of forms with Silurian affinities. Most groups (with
the notable exception of the rugose corals) underwent a modest in-
crease in biodiversity (Fig. 2; Supplementary materials). The biotic
turnover from TBF 2 to TBF 3 may have resulted from eustatic sea-level
fall as indicated by widespread hiatuses that are thought to be closely
related to climatic perturbations (e.g., Bergström et al., 2014).
The proliferation of TBF 3 was possibly interrupted by another
minor glacial episode, as suggested by a widespread stratigraphic gap
around the O–S boundary interval. Succeeding Silurian faunas typically
show a generally steady increase in biodiversity, with a progressive
addition of new forms in every major group, particularly graptolites
(Chen et al., 2005b;Melchin et al., 2011) and brachiopods (e.g., Cocks
and Rong, 2008;Baarli, 2014;Huang et al., 2018). This is consistent
with an overall amelioration of physical conditions, as evidenced by the
absence of prominent geochemical anomalies through much of the
Llandovery (Cramer et al., 2011;Trotter et al., 2016).
8. Concluding remarks
We attribute the traditional two-pulse model of the EOME to an
erroneous temporal and causal link between the “Hirnantian benthic
biota”and the Gondwanan glaciation. Instead, we recognise the
worldwide development of three successive, globally near-synchronous
benthic faunas with only the oldest one, i.e. the Hirnantia fauna and its
equivalents, of cool-water and glacial affinity. This finding leads to a
calibrated, finer-resolution evolutionary history of marine organisms
through the Late Ordovician, thereby providing a new framework for
understanding the tempo and nature of the EOME.
We reassess the EOME as most likely a single-pulse mass extinction
event that was followed by a protracted, stepwise recovery punctuated
by intermittent climatic shocks throughout the Hirnantian. With an
overall amelioration of physical conditions coinciding approximately
with the beginning of the Silurian, complete biotic recovery ensued as
faunas diversified and were distributed globally.
For Supplementary data to this article see https://doi.org/10.1016/
j.earscirev.2019.01.023.
Acknowledgments
We are grateful to Jiayu Rong, Bing Huang, Xin Wei and Lixia Li for
discussions, and to Xiaocong Luan and Jing Liu for help with prepara-
tion of new δ
13
C
carb
data. Thanks also to two anonymous reviewers for
constructive comments. This study is funded by the Strategic Priority
Research Program of Chinese Academy of Sciences (XDB26000000),
National Natural Science Foundation of China (41602008) and State
Key Laboratory of Palaeobiology and Stratigraphy. Ian Percival pub-
lishes with permission of the Executive Director of the Geological
Survey of New South Wales. This is a contribution to the IGCP Project
653 “The onset of the Great Ordovician Biodiversification Event”.
References
Ainsaar, L., Truumees, J., Meidla, T., 2015. The position of the Ordovician–Silurian
boundary in Estonia tested by high-resolution δ
13
C chemostratigraphic correlation.
In: Ramkumar, M. (Ed.), Chemostratigraphy: Concepts, Techniques and Applications.
Elsevier, Amsterdam, pp. 395–412.
Alfaro, M.B., Uriz, N.J., Cingolani, C.A., Tortello, M.F., Bidone, A.R., Galeano Inchausti,
J.C., 2013. Graptolites and trilobites from the Eusebio Ayala Formation
(Hirnantian?–early Llandovery), Paraná Basin, eastern Paraguay. Geol. J. 48,
236–247.
Álvaro, J.J., Colmenar, J., Monceret, É., Pouclet, A., Vizcaïno, D., 2016. Late Ordovician
(post–Sardic) rifting branches in the North Gondwanan Montagne Noire and
Mouthoumet massifs of southern France. Tectonophysics 681, 111–123.
Amsden, T.W., 1974. Late Ordovician and Early Silurian articulate brachiopods from
Oklahoma, southwestern Illinois, and eastern Missouri. Okla. Geol. Surv., Bull. 119,
1–154.
Amsden, T.W., Barrick, J.E., 1986. Late Ordovician-Early Silurian strata in the Central
United States and the Hirnantian Stage. Okla. Geol. Surv., Bull. 139, 1–95.
Baarli, B.G., 2014. The early Rhuddanian survival interval in the Lower Silurian of the
Oslo Region: a third pulse of the end-Ordovician extinction. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 395, 29–41.
Banks, M.R., 1988. The base of the Silurian System in Tasmania. Bull. Brit. Mus. Nat. Hist.
(Geol.) 43, 191–194.
Bassett, M.G., Popov, L.E., Aldridge, R.J., Gabbott, S.E., Theron, J.N., 2009. Brachiopoda
from the Soom Shale Lagerstätte (Upper Ordovician, South Africa). J. Paleontol. 83,
614–623.
Benedetto, J.L., 1986. The first typical Hirnantia fauna from South America (San Juan
Province, Argentine Precordillera). Biostratigraphie Du Paleozoïque 4, 439–447.
Benedetto, J., Cocks, L.R.M., 2009. Early Silurian (Rhuddanian) brachiopods from the
Argentine Precordillera and their biogeographic affinities. Ameghiniana 46,
241–253.
Benedetto, J.L., Sánchez, T.M., 1990. Fauna y edad del estratotipo de la Formación Salar
del Rincón (Eopaleozoico, Puna Argentina). Ameghiniana 27, 317–326.
Benedetto, J.L., Halpern, K., Galeano Inchausti, J.C., 2013. High-latitude Hirnantian
(latest Ordovician) brachiopods from the Eusebio Ayala Formation of Paraguay,
Paraná Basin. Palaeontology 56, 61–78.
Benedetto, J.L., Halpern, K., Puente, G.S.D.L., Monaldi, C.R., 2015. An in situ shelly fauna
from the lower Paleozoic Zapla diamictite of northwestern Argentina: implications
for the age of glacial events across Gondwana. J. S. Am. Earth Sci. 64, 166–182.
Bergström, J., 1968. Upper Ordovician brachiopods from Västergötland, Sweden. Geol.
Palaeontol. 2, 1–35.
Bergström, S.M., Bergström, J., 1996. The Ordovician-Silurian boundary successions in
Östergötland and Västergötland, S. Sweden. GFF 118, 25–42.
Bergström, S.M., Goldman, D., 2019. δ
13
C chemostratigraphy of the Ordovician-Silurian
boundary interval. In: Sial, A.N., Gaucher, C., Ramkumar, M., Ferreira, V.P. (Eds.),
Chemostratigraphy Across Major Chronological Boundaries. John Wiley & Sons, Inc,
pp. 143–158.
Bergström, S.M., Saltzman, M.M., Schmitz, B., 2006. First record of the Hirnantian (Upper
Ordovician) δ
13
C excursion in the North American Midcontinent and its regional
implications. Geol. Mag. 143, 657–678.
Bergström, S.M., Calner, M., Lehnert, O., Noor, A., 2011a. A new upper Middle
Ordovician–Lower Silurian drillcore standard succession from Borenshult in
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
30
Östergötland, southern Sweden: 1. Stratigraphical review with regional comparisons.
GFF 133, 149–171.
Bergström, S.M., Kleffner, M., Schmitz, B., Cramer, B.D., Dix, G., 2011b. Revision of the
position of the Ordovician-Silurian boundary in southern Ontario: regional chron-
ostratigraphic implications of δ
13
C chemostratigraphy of the Manitoulin Formation
and associated strata. Can. J. Earth Sci. 48, 1447–1470.
Bergström, S.M., Kleffner, M., Schmitz, B., 2012a. Late Ordovician-Early Silurian δ
13
C
chemostratigraphy in the Upper Mississippi Valley: implications for chronostrati-
graphy and depositional interpretations. Earth Environ. Sci. Trans. R. Soc. Edinb.
102, 159–178.
Bergström, S.M., Lehnert, O., Calner, M., Joachimski, M.M., 2012b. A new upper Middle
Ordovician–Lower Silurian drillcore standard succession from Borenshult in
Östergötland, southern Sweden: 2. Significance of δ
13
C chemostratigraphy. GFF 134,
39–63.
Bergström, S.M., Eriksson, M.E., Young, S.A., Ahlberg, P., Schmitz, B., 2014. Hirnantian
(latest Ordovician) δ
13
C chemostratigraphy in southern Sweden and globally: a re-
fined integration with the graptolite and conodont zone successions. GFF 136,
355–386.
Bockelie, J.F., Baarli, B.G., Johnson, M.E., 2017. Hirnantian (latest Ordovician) glacia-
tions and their consequences for the Oslo Region, Norway, with a revised lithos-
tratigraphy for the Langøyene Formation in the inner Oslofjorden area. Nor. J. Geol.
97, 119–143.
Botting, J.P., Muir, L.A., Zhang, Y.D., Ma, X., Ma, J.Y., Wang, L.W., Zhang, J.F., Song,
Y.Y., Fang, X., 2017. Flourishing sponge-based ecosystems after the dnd-Ordovician
mass extinction. Curr. Biol. 27, 556–562.
Botting, J.P., Muir, L.A., Wang, W.H., Qie, W.K., Tan, J.Q., Zhang, L.N., Zhang, Y.D.,
2018. Sponge-dominated offshore benthic ecosystems across South China in the
aftermath of the end-Ordovician mass extinction. Gondwana Res. 61, 150–171.
Bourahrouh, A., Paris, F., Elaouad-Debbaj, Z., 2004. Biostratigraphy, biodiversity and
palaeoenvironments of the chitinozoans and associated palynomorphs from the
Upper Ordovician of the Central Anti-Atlas, Morocco. Rev. Palaeobot. Palynol. 130,
17–40.
Brenchley, P.J., Cocks, L.R.M., 1982. Ecological associations in a regressive sequence: the
latest Ordovician of the Oslo-Asker district, Norway. Palaeontology 25, 783–815.
Brenchley, P.J., Cullen, B., 1984. The environmental distribution of associations be-
longing to the Hirnantia fauna–evidence from North Wales and Norway. Palaeontol.
Contrib. Univ. Oslo 295, 113–125.
Brenchley, P.J., Newall, G., 1984. Late Ordovician environmental changes and their effect
on faunas. Palaeontol. Contrib. Univ. Oslo 295, 65–79.
Brenchley, P.J., Marshall, J.D., Underwood, C.J., 2001. Do all mass extinctions represent
an ecological crisis? Evidence from the Late Ordovician. Geol. J. 36, 329–340.
Brenchley, P.J., Carden, G.A., Hints, L., Kaljo, D., Marshall, J.D., Martma, T., Meidla, T.,
Nõlvak, J., 2003. High-resolution stable isotope stratigraphy of Upper Ordovician
sequences: constraints on the timing of bioevents and environmental changes asso-
ciated with mass extinction and glaciation. Geol. Soc. Am. Bull. 115, 89–104.
Brenchley, P.J., Marshall, J.D., Harper, D.A.T., Buttler, C.J., Underwood, C.J., 2006. A
late Ordovician (Hirnantian) karstic surface in a submarine channel, recording glacio-
eustatic sea-level changes: Meifod, Central Wales. Geol. J. 41, 1–22.
Calner, M., Calner, H., Lehnert, O., 2017. The Hirnantian isotope carbon excursion (HICE)
in southern Norway. Geol. Soc. Am. Abstr. Programs 49, 175.
Chatalov, A., 2017. Sedimentology of Hirnantian glaciomarine deposits in the Balkan
Terrane, western Bulgaria: fixing a piece of the north peri-Gondwana jigsaw puzzle.
Sediment. Geol. 350, 1–22.
Chen, X., Rong, J.Y., Zhou, Z.Y., Zhang, Y.D., Zhan, R.B., Liu, J.B., Fan, J.X., 2001. The
Central Guizhou and Yichang uplifts, Upper Yangtze region, between Ordovician and
Silurian. Chin. Sci. Bull. 46, 1580–1584.
Chen, X., Fan, J.X., Melchin, M.J., Mitchell, C.E., 2005a. Hirnantian (Latest Ordovician)
graptolites from the Upper Yangtze region, China. Palaeontology 48, 235–280.
Chen, X., Melchin, M.J., Sheets, H.D., Mitchell, C.E., Fan, J.X., 2005b. Patterns and
processes of latest Ordovician graptolites extinction and recovery based on data from
South China. J. Paleontol. 79, 842–861.
Chen, X., Rong, J.Y., Fan, J.X., Zhan, R.B., Mitchell, C.E., Harper, D.A.T., Melchin, M.J.,
Peng, P.A., Finney, S.C., Wang, X.F., 2006. The Global Boundary Stratotype Section
and Point (GSSP) for the base of the Hirnantian Stage (the uppermost of the
Ordovician System). Episodes 29, 183–196.
Chen, C., Wang, J.S., Algeo, T.J., Wang, Z., Tu, S., Wang, G.H., Yang, J.X., 2017. Negative
δ
13
C
carb
shifts in Upper Ordovician (Hirnantian) Guanyinqiao Bed of South China
linked to diagenetic carbon fluxes. Palaeogeogr. Palaeoclimatol. Palaeoecol. 487,
430–446.
Cingolani, C.A., Uriz, N.J., Alfaro, M.B., Tortello, F., Bidone, A.R., Galeano Inchausti, J.C.,
2011. The Hirnantian–early Llandovery transition sequence in the Paraná Basin,
eastern Paraguay. In: Gutiérrez-Marco, J.C., Rábano, I., García-Bellido, D. (Eds.),
Ordovician of the World. Instituto Geológico y Minero de España, Madrid, pp.
103–108.
Cocks, L.R.M., Cooper, R.A., 2004. Late Ordovician (Hirnantian) shelly fossils from New
Zealand and their significance. N. Z. J. Geol. Geophys. 47, 71–80.
Cocks, L.R.M., Fortey, R.A., 1986. New evidence on the South African Lower Palaeozoic:
age and fossils reviewed. Geol. Mag. 123, 437–444.
Cocks, L.R.M., Fortey, R.A., 2002. The palaeogeographical significance of the latest
Ordovician fauna from the Panghsa-pye Formation of Burma. Spec. Pap. Palaeontol.
67, 57–76.
Cocks, L.R.M., Rong, J.Y., 2008. Earliest Silurian faunal survival and recovery after the
end Ordovician glaciation: evidence from the brachiopods. Earth Environ. Sci. Trans.
R. Soc. Edinb. 98, 291–301.
Cocks, L.R.M., Brunton, C.H.C., Rowell, A.J., Rust, I.C., 1970. The first Lower Paleozoic
fauna proved from South Africa. J. Geol. Soc. Lond. 125, 583–603.
Colmenar, J., Villas, E., Rasmussen, C., 2019. A synopsis of Late Ordovician brachiopod
diversity in the Anti-Atlas, Morocco. Geol. Soc. Lond. Spec. Publ. 485 (485). https://
doi.org/10.1144/SP485.3.
Copper, P., 2001. Reefs during the multiple crises towards the Ordovician-Silurian
boundary: Anticosti Island, eastern Canada, and worldwide. Can. J. Earth Sci. 38,
153–171.
Copper, P., Jin, J.S., 2014. The revised Lower Silurian (Rhuddanian) Becscie Formation,
Anticosti Island, eastern Canada records the tropical marine faunal recovery from the
end-Ordovician Mass Extinction. Newsl. Stratigr. 47, 61–83.
Copper, P., Jin, J., Desrochers, A., 2013. The Ordovician-Silurian boundary (late Katian-
Hirnantian) of western Anticosti Island: revised stratigraphy and benthic megafaunal
correlations. Stratigraphy 10, 213–228.
Cramer, B.D., Brett, C.E., Melchin, M.J., Männik, P., Kleffner, M.A., McLaughlin, P.I.,
Loydell, D.K., Munnecke, A., Jeppsson, L., Corradini, C., Brunton, F.R., Saltzman,
M.R., 2011. Revised correlation of Silurian Provincial Series of North America with
global and regional chronostratigraphic units and δ
13
C
carb
chemostratigraphy.
Lethaia 44, 185–202.
de la Puente, G.S., Rubinstein, C.V., 2013. Ordovician chitinozoans and marine phyto-
plankton of the Central Andean Basin, northwestern Argentina: a biostratigraphic and
paleobiogeographic approach. Rev. Palaeobot. Palynol. 198, 14–26.
de la Puente, G.S., Rubinstein, C., Vaccari, N., Paris, F., 2015. Latest Ordovician-earliest
Silurian chitinozoans from Puna, western Gondwana. Stratigraphy 12, 127–128.
Demski, M.W., Wheadon, B.J., Stewart, L.A., Elias, R.J., Young, G.A., Nowlan, G.S.,
Dobrzanski, E.P., 2015. Hirnantian strata identified in major intracratonic basins of
Central North America: implications for uppermost Ordovician stratigraphy. Can. J.
Earth Sci. 52, 68–76.
Díaz-Martínez, E., Grahn, Y., 2007. Early Silurian glaciation along the western margin of
Gondwana (Peru, Bolivia and northern Argentina): palaeogeographic and geody-
namic setting. Palaeogeogr. Palaeoclimatol. Palaeoecol. 245, 62–81.
Ebbestad, J.O.R., Högström, A.E.S., Frisk, Å.M., Martma, T., Kaljo, D., Kröger, B.,
Pärnaste, H., 2015. Terminal Ordovician stratigraphy of the Siljan district, Sweden.
GFF 137, 36–56.
Elias, R.J., Young, G.A., Lee, D.J., Bae, B.Y., 2013. Chapter 9 Coral biogeography in the
Late Ordovician (Cincinnatian) of Laurentia. Geol. Soc. Lond. Mem. 38, 97–115.
Finnegan, S., Bergmann, K., Eiler, J.M., Jones, D.S., Fike, D.A., Eisenman, I., Hughes, N.C.,
Tripati, A.K., Fischer, W.W., 2011. The magnitude and duration of Late
Ordovician–Early Silurian glaciation. Science 331, 903–906.
Finnegan, S., Heim, N.A., Peters, S.E., Fischer, W.W., 2012. Climate change and the se-
lective signature of the Late Ordovician mass extinction. Proc. Natl. Acad. Sci. 109,
6829–6834.
Finnegan, S., Rasmussen, C.M.Ø., Harper, D.A.T., 2016. Biogeographic and bathymetric
determinants of brachiopod extinction and survival during the Late Ordovician mass
extinction. Proc. R. Soc. B 283, 20160007.
Finnegan, S., Rasmussen, C.M.Ø., Harper, D.A.T., 2017. Identifying the most surprising
victims of mass extinction events: an example using Late Ordovician brachiopods.
Biol. Lett. 13, 20170400.
Finney, S.C., Berry, W.B.N., Cooper, J.D., 2007. The influence of denitrifying seawater on
graptolite extinction and diversification during the Hirnantian (latest Ordovician)
mass extinction event. Lethaia 40, 281–291.
Gabbott, S.E., Browning, C., Theron, J.N., Whittle, R.J., 2017. The late Ordovician Soom
Shale Lagerstätte: an extraordinary post-glacial fossil and sedimentary record. J.
Geol. Soc. 174, 1–9.
Ghavidel-Syooki, M., Álvaro, J.J., Popov, L., Pour, M.G., Ehsani, M.H., Suyarkova, A.,
2011. Stratigraphic evidence for the Hirnantian (latest Ordovician) glaciation in the
Zagros Mountains, Iran. Palaeogeogr. Palaeoclimatol. Palaeoecol. 307, 1–16.
Ghienne, J.F., Desrochers, A., Vandenbroucke, T.R.A., Achab, A., Asselin, E., Dabard,
M.P., Farley, C., Loi, A., Paris, F., Wickson, S., 2014. A Cenozoic-style scenario for the
end-Ordovician glaciation. Nat. Commun. 5, 4485.
Gorjan, P., Kaiho, K., Fike, D.A., Xu, C., 2012. Carbon- and sulfur-isotope geochemistry of
the Hirnantian (Late Ordovician) Wangjiawan (Riverside) section, South China:
global correlation and environmental event interpretation. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 337, 14–22.
Halpern, K., Rustan, J.J., Arcerito, F.R.M., 2014. The first Hirnantian (uppermost
Ordovician) odontopleurid trilobite from western Gondwana (Argentina). Rev. Bras.
Paleontol. 17, 3–10.
Hammarlund, E.U., Dahl, T.W., Harper, D.A.T., Bond, D.P.G., Nielsen, A.T., Bjerrum, C.J.,
Schovsbo, N.H., Schonlaub, H.P., Zalasiewicz, J.A., Canfield, D.E., 2012. A sulfidic
driver for the end-Ordovician mass extinction. Earth Planet. Sci. Lett. 331, 128–139.
Harper, D.A.T., 1979. The environmental significance of some faunal changes in the
Upper Ardmillan succession (upper Ordovician), Girvan, Scotland. Geol. Soc. Lond.
Spec. Publ. 8, 439–445.
Harper, D.A.T., 1981. The stratigraphy and faunas of the Upper Ordovician High Mains
Formation of the Girvan district. Scott. J. Geol. 17, 247–255.
Harper, D.A.T., 1988. Ordovician-Silurian junctions in the Girvan district. S.W. Scotland.
Bull. Brit. Mus. Nat. Hist. (Geol.) 43, 45–52.
Harper, D.A.T., Hints, L., 2016. Hirnantian (Late Ordovician) brachiopod faunas across
Baltoscandia: a global and regional context. Palaeogeogr. Palaeoclimatol. Palaeoecol.
444, 71–83.
Harper, D.A.T., Rasmussen, C.M.Ø., Liljeroth, M., Blodgett, R.B., Candela, Y., Jin, J.S.,
Percival, I.G., Rong, J.Y., Villas, E., Zhan, R.B., 2013. Biodiversity, biogeography and
phylogeography of Ordovician rhynchonelliform brachiopods. Geol. Soc. Lond. Mem.
38, 127–144.
Harper, D.A.T., Hammarlund, E.U., Rasmussen, C.M.Ø., 2014. End Ordovician extinc-
tions: a coincidence of causes. Gondwana Res. 25, 1294–1307.
Hayton, S., Rees, A.J., Vecoli, M., 2017. A punctuated Late Ordovician and early Silurian
deglaciation and transgression: evidence from the subsurface of northern Saudi
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
31
Arabia. Am. Assoc. Pet. Geol. Bull. 101, 863–886.
He, X.Y., Chen, J.Q., Xiao, J.Y., 2007. Combination features, paleobiogeographic affinity
and mass extinction of the latest Ordovician (Hirnantian) rugosan fauna from
northern Guizhou, China. Acta Geol. Sin. 81, 23–41.
Hints, L., Harper, D.A.T., 2015. The Hirnantian (Late Ordovician) brachiopod fauna of the
East Baltic: taxonomy of the key species. Acta Palaeontol. Pol. 60, 395–420.
Hints, L., Oraspõld, A., Kaljo, D., 2000. Stratotype of the Porkuni Stage with comments on
the Röa Member (uppermost Ordovician, Estonia). Proc. Est. Acad. Sci., Geol. 49,
177–199.
Huang, B., Jin, J., Rong, J.Y., 2018. Post-extinction diversification patterns of brachio-
pods in the early–middle Llandovery, Silurian. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 493, 11–19.
Isaacson, P.E., Antelo, B., Boucot, A.J., 1976. Implications of a Llandovery (early Silurian)
brachiopod fauna from Salta Province, Argentina. J. Paleontol. 50, 1103–1112.
Jin, J.S., Zhan, R.B., 2008. Late Ordovician Orthide and Billingsellide Brachiopods from
Anticosti Island, Eastern Canada: Diversity Change through Mass Extinction. NRC
Research Press, Ottawa, pp. 159.
Jones, D.S., Martini, A.M., Fike, D.A., Kaiho, K., 2017. A volcanic trigger for the Late
Ordovician mass extinction? Mercury data from South China and Laurentia. Geology
45, 631–634.
Kaljo, D., 1996. Diachronous recovery patterns in Early Silurian corals, graptolites and
acritarchs. Geol. Soc. Spec. Publ. 102, 127–133.
Kaljo, D., Hints, L., Martma, T., Nõlvak, J., 2001. Carbon isotope stratigraphy in the latest
Ordovician of Estonia. Chem. Geol. 175, 49–59.
Kaljo, D., Hints, L., Männik, P., Nõlvak, J., 2008. The succession of Hirnantian events
based on data from Baltica: brachiopods, chitinozoans, conodonts, and carbon iso-
topes. Est. J. Earth Sci. 57, 197–218.
Kaljo, D., Männik, P., Martma, T., Nőlvak, J., 2012. More about the Ordovician–Silurian
transition beds at Mirny Creek, Omulev Mountains, NE Russia: carbon isotopes and
conodonts. Est. J. Earth Sci. 61, 277–294.
Koren, T.N., Sobolevskaya, R.F., 2008. The regional stratotype section and point for the
base of the Hirnantian Stage (the uppermost Ordovician) at Mirny Creek, Omulev
Mountains, Northeast Russia. Est. J. Earth Sci. 57 (1), 10.
Kröger, B., Ebbestad, J.O.R., Lehnert, O., Ullmann, C.V., Korte, C., Frei, R., Rasmussen,
C.M.Ø., 2015. Subaerial speleothems and deep karst in Central Sweden linked to
Hirnantian glaciations. J. Geol. Soc. 172, 349–356.
Kröger, B., Ebbestad, J.O.R., Lehnert, O., 2016. Accretionary Mechanisms and Temporal
Sequence of Formation of the Boda Limestone Mud-Mounds (Upper Ordovician),
Siljan District, Sweden. J. Sediment. Res. 86, 363–379.
Kulkov, N.P., Severgina, L.G., 1987. Ordovician and Silurian boundary in Gornoi Altai.
Seriya Geologicheskaya 9, 69–74 (in Russian).
Kulkov, N.P., Severgina, L.G., 1989. Ordovician and Silurian Stratigraphy and
Brachiopods from Gorny Altai. Nauka, Moscow (in Russian).
LaPorte, D.F., Holmden, C., Patterson, W.P., Loxton, J.D., Melchin, M.J., Mitchell, C.E.,
Finney, S.C., Sheets, H.D., 2009. Local and global perspectives on carbon and ni-
trogen cycling during the Hirnantian glaciation. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 276, 182–195.
Laurie, J.R., 1991. Articulate brachiopods from the Ordovician and Lower Silurian of
Tasmania. Mem. Assoc. Australas. Palaeontol. 11, 1–106.
Le Heron, D.P., Tofaif, S., Melvin, J., 2018. The Early Palaeozoic glacial deposits of
Gondwana: overview, chronology, and controversies. In: Menzies, J., van der Meer,
J.J.M. (Eds.), Past Glacial Environments, 2nd ed. Elsevier, pp. 47–73.
Legrand, P., 2011. On the Katian/Hirnantian boundary. Application on the north African
border of Gondwana. In: Gutiérrez-Marco, J.C., Rábano, I., García-Bellido, D. (Eds.),
Ordovician of the World. Instituto Geológico y Minero de España, Madrid, pp.
287–294.
Lespérance, P.J., Malo, M., Sheehan, P.M., Skidmore, W.B., 1987. A stratigraphical and
faunal revision of the Ordovician–Silurian strata of the Percé area, Quebec. Can. J.
Earth Sci. 24, 117–134.
Li, L.X., Feng, H.Z., Dorte, J., Joachim, R., 2015. Unusual Deep Water sponge assemblage
in South China—Witness of the end-Ordovician mass extinction. Sci. Rep. 5, 16060.
Loi, A., Ghienne, J.F., Dabard, M.P., Paris, F., Botquelen, A., Christ, N., Elaouad-Debbaj,
Z., Gorini, A., Vidal, M., Videt, B., 2010. The Late Ordovician glacio-eustatic record
from a high-latitude storm-dominated shelf succession: the Bou Ingarf section (Anti-
Atlas, Southern Morocco). Palaeogeogr. Palaeoclimatol. Palaeoecol. 296, 332–358.
Loydell, D.K., 2007. Graptolites from the Upper Ordovician and Lower Silurian of Jordon.
Spec. Pap. Palaeont. 78, 1–78.
Mauviel, A., Desrochers, A., 2016. A high-resolution, continuous δ
13
C record spanning the
Ordovician–Silurian boundary on Anticosti Island, eastern Canada. Can. J. Earth Sci.
53, 795–801.
McAuley, R.J., Elias, R.J., 1990. Latest Ordovician to earliest Silurian solitary rugose
corals of the east-central United States. Bull. Am. Paleoenviron. 98, 1–82.
Melchin, M.J., Holmden, C., 2006. Carbon isotope chemostratigraphy in Arctic Canada:
sea-level forcing of carbonate platform weathering and implications for Hirnantian
global correlation. Palaeogeogr. Palaeoclimatol. Palaeoecol. 234, 186–200.
Melchin, M.J., Mitchell, C.E., Naczk-Cameron, A., Fan, J.X., Loxton, J., 2011. Phylogeny
and adaptive radiation of the Neograptina (Graptoloida) during the Hirnantian mass
extinction and Silurian recovery. Proc. Yorks. Geol. Soc. 58, 281–309.
Melchin, M.J., Mitchell, C.E., Holmden, C., Štorch, P., 2013. Environmental changes in
the Late Ordovician–early Silurian: review and new insights from black shales and
nitrogen isotopes. Geol. Soc. Am. Bull. 125, 1635–1670.
Mergl, M., 2011. Earliest occurrence of the Hirnantia Fauna in the Prague Basin (Czech
Republic). Bull. Geosci. 86, 63–70.
Munnecke, A., Zhang, Y., Liu, X., Cheng, J., 2011. Stable carbon isotope stratigraphy in
the Ordovician of South China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 307, 17–43.
Nestor, H., Webby, B.D., 2013. Biogeography of the Ordovician and Silurian
Stromatoporoidea. Geol. Soc. Lond. Mem. 38, 67–79.
Neuman, B.E.E., 1969. Upper Ordovician streptelasmatid corals from Scandinavia. Bull.
Geol. Inst. Univ. Uppsala, New Ser. 1, 1–73.
Neuman, B.E.E., 1975. New lower Paleozoic streptelasmatid corals from Scandinavia.
Nor. Geol. Tidsskr. 55, 335–359.
Neuman, B.E.E., 2003. The new early Palaeozoic rugose coral genera Eurogrewingkia gen.
nov. and Fosselasma gen. nov. Proc. Est. Acad. Sci., Geol. 52, 199–212.
Niemeyer, H., Alvarez, F., Boucot, A.J., Bruna, M., 2010. Brachiopods from Cordon de
Lila, Lower Silurian (Llandovery) Quebrada Ancha Formation, Antofagasta region.
Chile. Spec. Pap. Palaeontol. 84, 287–307.
Owen, A.W., Harper, D.A.T., Rong, J.Y., 1991. Hirnantian trilobites and brachiopods in
space and time. Geol. Surv. Can., Bull. 90–9, 179–190.
Page, A.A., Zalasiewicz, J.A., Popov, L.E., 2007. Were transgressive black shales a ne-
gative feedback modulating glacioeustasy in the early Palaeozoic icehouse? In:
Williams, M., Haywood, A.M., Gregory, F.J., Schmidt, D.N. (Eds.), Deep-Time
Perspectives on Climate Change: Marrying the Signal from Computer Models and
Biological Proxies. Geological Society of London, London, pp. 123–156.
Paris, F., 1990. The Ordovician chitinozoan biozones of the Northern Gondwana Domain.
Rev. Palaeobot. Palynol. 66, 181–209.
Paris, F., Verniers, J., Al-Hajri, S., Al-Tayyar, H., 1995. Biostratigraphy and palaeogeo-
graphic affinities of Early Silurian chitinozoans from central Saudi Arabia. Rev.
Palaeobot. Palynol. 89, 75–90.
Paris, F., Deynoux, M., Ghienne, J.F., 1998. First record of chitinozoans from the
Ordovician–Silurian boundary beds of Mauritania; palaeogeographic implications. C.
R. Acad. Sci. Paris, Sciences de la Terre et des Planètes 326, 499–504.
Rasmussen, C.M.Ø., Harper, D.A.T., 2011. Interrogation of distributional data for the End
Ordovician crisis interval: where did disaster strike? Geol. J. 46, 478–500.
Rasmussen, C.M.Ø., Ebbestad, J.O.R., Harper, D.A.T., 2010. Unravelling a Late
Ordovician pentameride (Brachiopoda) hotspot from the Boda Limestone, Siljan
district, central Sweden. GFF 132, 133–152.
Raup, D.M., Sepkoski Jr., J.J., 1982. Mass extinctions in the marine fossil record. Science
215, 1501–1503.
Rong, J.Y., Harper, D.A.T., 1988. A global synthesis of the latest Ordovician Hirnantian
brachiopod faunas. Trans. R. Soc. Edinb. Earth Sci. 79, 383–402.
Rong, J.Y., Harper, D.A.T., 1999. Brachiopod survival and recovery from the latest
Ordovician mass extinctions in South China. Geol. J. 34, 321–348.
Rong, J.Y., Li, R.Y., 1999. A silicified Hirnantia fauna (latest Ordovician brachiopods)
from Guizhou, southwest China. J. Paleontol. 75, 831–849.
Rong, J.Y., Zhan, R.B., 2004a. Late Ordovician brachiopod mass extinction of South
China. In: Rong, J.Y., Fang, Z.J. (Eds.), Mass Extinction and Recovery—Evidences
from the Palaeozoic and Triassic of South China. University of Science and
Technology of China Press, Hefei, pp. 71–96 (in Chinese with English summary).
Rong, J.Y., Zhan, R.B., 2004b. Survival and recovery of brachiopods in Early Silurian of
South China. In: Rong, J.Y., Fang, Z.J. (Eds.), Mass Extinction and
Recovery—Evidences from the Palaeozoic and Triassic of South China. University of
Science and Technology of China Press, Hefei, pp. 97–126 (in Chinese with English
summary).
Rong, J.Y., Zhan, R.B., Han, N.R., 1994. The oldest known Eospirifer (Brachiopoda) in the
Changwu Formation (Late Ordovician) of western Zhejiang, East China, with a review
of the earliest spiriferoids. J. Paleontol. 68, 763–776.
Rong, J.Y., Chen, X., Harper, D.A.T., 2002. The latest Ordovician Hirnantia Fauna
(Brachiopoda) in time and space. Lethaia 35, 231–249.
Rong, J.Y., Chen, X., Zhan, R.B., Zhou, Z.Q., Zheng, Z.C., Wang, Y., 2003. Late Ordovician
biogeography and Ordovician-Silurian boundary in the Zhusilenhaierhan area, Ejin,
western Inner Mongolia. Acta Palaeontol. Sin. 42, 149–167 (in Chinese with English
summary).
Rong, J.Y., Boucot, A.J., Harper, D.A.T., Zhan, R.B., Neuman, R.B., 2006. Global analyses
of brachiopod faunas through the Ordovician and Silurian transition: reducing the
role of the Lazarus effect. Can. J. Earth Sci. 43, 23–39.
Rong, J.Y., Huang, B., Zhan, R.B., Harper, D.A.T., 2008a. Latest Ordovician brachiopod
and trilobite assemblage from Yuhang, northern Zhejiang, East China: a window on
Hirnantian deep-water benthos. Hist. Biol. 20, 137–148.
Rong, J.Y., Jin, J.S., Zhan, R.B., Bergström, J., 2008b. The earliest known Stegerhynchus
(Rhynchonellida, Brachiopoda) from the Hirnantian strata (uppermost Ordovician) at
Borenshult, Östergötland, Sweden. GFF 130, 21–30.
Rong, J.Y., Chen, X., Zhan, R.B., Fan, J.X., Wang, Y., Zhang, Y.D., Li, Y., Huang, B., Wu,
R.C., Wang, G.X., Liu, J.B., 2010. New observation on Ordovician-Silurian boundary
strata of Southern Tongzi County, northern Guizhou, Southwest China. J. Stratigr. 34,
337–348 (in Chinese with English abstract).
Rong, J.Y., Huang, B., Zhan, R.B., Harper, D.A.T., 2013. Latest Ordovician and earliest
Silurian Brachiopods succeeding the Hirnantia Fauna in Southeast China. Spec. Pap.
Palaeontol. 90, 1–142.
Rong, J.Y., Wei, X., Zhan, R.B., Wang, Y., 2018. A deep water shelly fauna from the
uppermost Ordovician in northwestern Hunan, South China and its paleoecological
implications. Sci. China Earth Sci. 61, 730–744.
Rubinstein, C.V., de La Puente, G.S., Delabroye, A., Astini, R.A., 2016. The palynological
record across the Ordovician/Silurian boundary in the Cordillera Oriental, Central
Andean Basin, northwestern Argentina. Rev. Palaeobot. Palynol. 224, 14–25.
Sachanski, V., 1993. Boundaries of the Silurian System in Bulgaria defined by graptolites.
Geol. Balc. 23, 25–33.
Sachanski, V., Loydell, D., Chatalov, A., 2018. New data from graptolites and quartz
arenites for the Ordovician–Silurian boundary in the Murzuq Basin. Libya. Compt.
Rend. Acad. Bulg. Sci. 71, 506–515.
Sennikov, N.V., Lykova, E.V., Obut, O.T., Tolmacheva, T.Y., Izokh, N.G., 2014. The new
Ordovician stage standard as applied to the stratigraphic units of the western
Altai–Sayan Folded Area. Russ. Geol. Geophys. 55, 971–988.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
32
Sepkoski Jr., J.J., 1995. The Ordovician radiations: diversification and extinction shown
by global genus-level taxonomic data. In: Cooper, J.D., Droser, M.L., Finney, S.C.
(Eds.), Ordovician Odyssey: Short Papers for the Seventh International Symposium on
the Ordovician System. Pacific Section, Society for Economic Paleontologists and
Mineralogists, Fullerton, California, pp. 393–396.
Sheehan, P.M., 2001. The Late Ordovician mass extinction. Annu. Rev. Earth Planet. Sci.
29, 331–364.
Stanley, S.M., 2016. Estimates of the magnitudes of major marine mass extinctions in
earth history. Proc. Natl. Acad. Sci. 113, E6325–E6334.
Štorch, P., 1990. Upper Ordovician –lower Silurian sequences in the Bohemian Massif,
central Europe. Geol. Mag. 127, 225–239.
Stott, C.A., Jin, J.S., 2007. Rhynchonelliformean brachiopods from the Manitoulin
Formation of Ontario, Canada: potential implications for the position of the
Ordovician–Silurian boundary in cratonic North America. Acta Palaeontol. Sin. 46,
449–459.
Stridsberg, S., 1980. Sedimentology of Upper Ordovician regressive strata in
Västergötland. GFF 102, 213–221.
Sutcliffe, O.E., Harper, D.A.T., Salem, A.A., Whittington, R.J., Craig, J., 2001. The de-
velopment of an atypical Hirnantia-brachiopod Fauna and the onset of glaciation in
the late Ordovician of Gondwana. Earth Environ. Sci. Trans. R. Soc. Edinb. 92, 1–14.
Trotter, J.A., Williams, I.S., Barnes, C.R., Männik, P., Simpson, A., 2016. New conodont
δ
18
O records of Silurian climate change: implications for environmental and biolo-
gical events. Palaeogeogr. Palaeoclimatol. Palaeoecol. 443, 34–48.
Tu, S., Wang, Z., Wang, J.S., 2012. Interpretation for high resolution stable carbon and
oxygen isotope records across Ordovician-Silurian boundary from Wangjiawan. Earth
Sci. –J. China Univ. Geosci. 37, 165–174 (in Chinese with English abstract).
Underwood, C.J., Deynoux, M., Ghienne, J.F., 1998. High palaeolatitude (Hodh,
Mauritania) recovery of graptolite faunas after the Hirnantian (end Ordovician) ex-
tinction event. Palaeogeogr. Palaeoclimatol. Palaeoecol. 142, 91–105.
Van Houten, F.B., 1985. Oolitic ironstones and contrasting Ordovician and Jurassic pa-
leogeography. Geology 13, 722–724.
Vandenbroucke, T.R.A., Hennissen, J., Zalasiewicz, J.A., Verniers, J., 2008. New chit-
inozoans from the historical type area of the Hirnantian Stage and additional key
sections in the Wye Valley, Wales, UK. Geol. J. 43, 397–414.
Vandenbroucke, T.R.A., Gabbott, S.E., Aldridge, R.J., Theron, J.N., 2009. Chitinozoans
and the age of the Soom Shale, an Ordovician black shale Lagerstätte, South Africa. J.
Micropalaeontol. 28, 53–66.
Vecoli, M., 2008. Fossil microphytoplankton dynamics across the Ordovician–Silurian
boundary. Rev. Palaeobot. Palynol. 148, 91–107.
Wang, C.Y., Aldridge, R.J., 2010. Silurian conodonts from the Yangtze Platform, South
China. Spec. Pap. Palaeontol. 83, 1–136.
Wang, G.X., Zhan, R.B., Percival, I.G., Huang, B., Li, Y., Wu, R.C., 2015. Late Hirnantian
(latest Ordovician) carbonate rocks and shelly fossils in Shiqian, northeastern
Guizhou, Southwest China. Newsl. Stratigr. 48, 241–252.
Wang, G.X., Zhan, R.B., Percival, I.G., 2016. New data on Hirnantian (latest Ordovician)
postglacial carbonate rocks and fossils in northern Guizhou, Southwest China. Can. J.
Earth Sci. 53, 660–665.
Wang, G.X., Zhan, R.B., Huang, B., Percival, I.G., 2017. Coral faunal turnover through the
Ordovician-Silurian transition in South China and its global implications for carbo-
nate stratigraphy and macroevolution. Geol. Mag. 154, 829–836.
Wang, G.X., Zhan, R.B., Rong, J.Y., Huang, B., Percival, I.G., Luan, X.C., Wei, X., 2018.
Exploring the end-Ordovician extinctions in Hirnantian near-shore carbonate rocks of
northern Guizhou, SW China: a refined stratigraphy and regional correlation. Geol. J.
53, 3019–3029.
Wei, X., Zhan, R.B., 2018. A late Rhuddanian (early Llandovery, Silurian) trilobite as-
sociation from South China and its implications. Palaeoworld 27, 42–52.
Yolkin, E.A., Obut, A.M., Sennikov, N.V., 1988. The Ordovician-Silurian boundary in the
Altai Mountains, USSR. Bull. Brit. Mus. Nat. Hist. (Geol.) 43, 139–143.
Young, T.P., 1989. Phanerozoic ironstones: an introduction and review. Geol. Soc. Spec.
Publ. 46, ix–xxv.
Young, G.A., Elias, R.J., 1995. Latest Ordovician to earliest Silurian colonial corals of the
east-central United States. Bull. Am. Paleo. 108, 1–153.
Zhan, R.B., Liu, J.B., Percival, I.G., Jin, J.S., Li, G.P., 2010. Biodiversification of Late
Ordovician Hirnantia fauna on the Upper Yangtze Platform, South China. Sci. China
Earth Sci. 53, 1800–1810.
Zhang, S.X., Barnes, C.R., 2002. A new Llandovery (early Silurian) conodont biozonation
and conodonts from the Becscie, Merrimack, and Gun River formations, Anticosti
Island, Québec. J. Paleontol. 76, 1–46.
Zhang, T.G., Shen, Y.N., Zhan, R.B., Shen, S.Z., Chen, X., 2009. Large perturbations of the
carbon and sulfur cycle associated with the Late Ordovician mass extinction in South
China. Geology 37, 299–302.
Zheng, Z.C., Zhou, Z.Q., He, Z.X., Zhu, H., 1987. Late Ordovician-early Silurian strati-
graphy in Badanjilin area, western Nei Monggol. In: Zhu, H., Zheng, S.C., He, X.Y.
(Eds.), Paleozoic Biostratigraphy and Tectonic Evolution of the Alxa Massif Margin.
Wuhan College of Geology Press, Wuhan, pp. 43–66 (in Chinese with English ab-
stract).
Zhou, Z.Y., Yuan, W.W., Han, N.R., Zhou, Z.Q., 2004. Survival and recovery of brachio-
pods in early Silurian of South China. In: Rong, J.Y., Fang, Z.J. (Eds.), Mass Extinction
and Recovery—Evidences from the Palaeozoic and Triassic of South China.
University of Science and Technology of China Press, Hefei, pp. 127–152 (in Chinese
with English summary).
Zou, C.N., Qiu, Z., Poulton, S.W., Dong, D.Z., Wang, H.Y., Chen, D.Z., Lu, B., Shi, Z.S.,
Tao, H.F., 2018. Ocean euxinia and climate change “double whammy”drove the Late
Ordovician mass extinction. Geology 46, 535–538.
G. Wang, et al. Earth-Science Reviews 192 (2019) 15–33
33
