Earth-Science Reviews 203 (2020) 103120

Contents lists available at ScienceDirect

Earth-Science Reviews
journal homepage: www.elsevier.com/locate/earscirev

A chronostratigraphic framework for the upper Stormberg Group: T

Implications for the Triassic-Jurassic boundary in southern Africa
Emese M. Bordya,⁎, Miengah Abrahamsa, Glenn R. Sharmanb, Pia A. Vigliettic,d,e,
Roger B.J. Bensonc,f, Blair W. McPheec, Paul M. Barrettc,g, Lara Sciscioa,h, Daniel Condoni,
Roland Mundilj, Zandri Rademank, Zubair Jinnahe, James M. Clarkl, Celina A. Suarezb,
Kimberley E.J. Chapellec,e, Jonah N. Choinierec
a
Department of Geological Sciences, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa
b
Department of Geosciences, University of Arkansas, 340 N. Campus Drive, Fayetteville, AR 72701, United States
c
Evolutionary Studies Institute, University of the Witwatersrand, Private Bag 3, Wits, 2050 Johannesburg, South Africa
d
Integrative Research Center, Field Museum of Natural History, 1400 South Lake Shore Drive, Chicago, IL 60605, United States
e
School of Geosciences, University of the Witwatersrand, Private Bag 3, Wits, 2050 Johannesburg, South Africa
f
Department of Earth Sciences, University of Oxford, Oxford OX1 3AN, United Kingdom
g
Department of Earth Sciences, Natural History Museum, Cromwell Road, London SW7 5BD, United Kingdom
h
Department of Geology, University of Johannesburg, Kingsway and Auckland Park, 2006 Johannesburg, South Africa
i
British Geological Survey (BGS), Keyworth, Nottingham NG12 5GG, United Kingdom
j
Berkeley Geochronology Center (BGC), 2455 Ridge Road, Berkeley, CA 94709, United States
k
Department of Earth Science, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa
l
Department of Biological Sciences, The George Washington University, 800 22nd St. NW Suite 6000, Washington, D.C., 20052, United States

ARTICLE INFO ABSTRACT

Keywords: The upper Stormberg Group (Elliot and Clarens formations) of the main Karoo Basin is well-known for its fossil
U–Pb detrital zircon geochronology vertebrate fauna, comprising early branching members of lineages including mammals, dinosaurs, and turtles.
Main Karoo Basin Despite 150 years of scientific study, the upper Stormberg Group lacks radioisotopic age constraints and remains
Elliot Formation coarsely dated via imprecise faunal correlations. Here we synthesise previous litho- and magnetostratigraphic studies,
Vertebrate biostratigraphy
and present a comprehensive biostratigraphic review of the upper Stormberg fauna. We also present the results of the
Early dinosaurs
first geochronological assessment of the unit across the basin, using U-Pb dates derived from detrital zircons obtained
Magnetostratigraphy
from tuffaceous sandstones and siltstones, the youngest of which are considered maximum depositional ages. Our
results confirm that the Elliot Formation contains the Triassic–Jurassic boundary, making it one of the few fossili-
ferous continental units that records the effects of the end-Triassic Mass Extinction event. Our work suggests a mid-
Norian–Rhaetian age for the lower Elliot Formation and a Hettangian–Sinemurian age for the upper Elliot Formation,
although the precise stratigraphic position of the Triassic/Jurassic (Rhaetian/Hettangian) boundary remains some-
what uncertain. A mainly Pliensbachian age is obtained for the Clarens Formation. The new dates allow direct
comparison with better-calibrated Triassic-Jurassic faunas of the Western Hemisphere (e.g., Chinle and Los Colorados
formations). We show that sauropodomorph, but not ornithischian or theropod, dinosaurs were well-established in
the main Karoo Basin ~220 million years ago, and that typical Norian faunas (e.g., aetosaurs, phytosaurs) are either
rare or absent in the lower Elliot Formation, which is paucispecific compared to the upper Elliot. While this is
unlikely the result of geographic sampling biases, it could be due to historical sampling intensity differences.

Motto: conclusions reached by workers of my and preceding generations

"Africa presents incomparable opportunities for geological studies; will demand constant revision in the light of new knowledge.”
the area is vast; the workers are comparatively few. The time has not Sidney Henry Haughton (1888—1982) [in Dunham, K.C., 1983.
yet arrived for extremely detailed studies […] Consequently, a Sidney Henry Haughton, 7 May 1888—24 May 1982. doi:10.1098/
temptation to generalize is ever present; and it is certain that rsbm.1983.0011]

⁎
Corresponding author.
E-mail address: [email protected] (E.M. Bordy).

https://doi.org/10.1016/j.earscirev.2020.103120
Received 25 August 2019; Received in revised form 2 February 2020; Accepted 10 February 2020
Available online 19 February 2020
0012-8252/ © 2020 Elsevier B.V. All rights reserved.
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

1. Introduction Early Jurassic depositional ages, respectively (e.g., Olsen and Galton,
1984; Lucas and Hancox, 2001). If these correlations are correct, then
The Upper Triassic to Lower Jurassic upper Stormberg Group (Elliot the Elliot Formation is indeed one of the few fossiliferous continental
and Clarens formations) in the main Karoo Basin (MKB) of southern deposits that spans the end-Triassic Mass Extinction event (ETME). This
Africa (Fig. 1) is rich in both vertebrate body and trace fossils and makes it crucially important for understanding shifts in the terrestrial
serves as a global standard for Triassic-Jurassic boundary (TJB) studies. ecosystems of southern Pangaea through the TJB interval. However,
This fossil record is critical for understanding early Mesozoic terrestrial these age assessments have not been tested using geochronological
vertebrate evolution, and it includes remains of dinosaurs, pseudo- methods and, instead, rely on low-precision ichno- and biostratigraphic
suchians, lepidosaurs, stem-group turtles, temnospondyl amphibians, correlations. This prevents the rich fossil record of the Elliot Formation
and later-branching therapsids, as well as their tracks and trackways from being fully used for studying the dynamics of extinction and re-
(e.g., Ellenberger et al., 1964; Ellenberger, 1970, 1972, 1974; Kitching covery during the ETME.
and Raath, 1984; Knoll, 2004, 2005). Faunal and sedimentary facies Determining the depositional age of clastic sedimentary rocks is
changes informally divide the Elliot Formation (EF) into lower (lEF) challenging, particularly in the absence of widespread and age-specific
and upper (uEF) sections (e.g., Ellenberger et al., 1964; Kitching and fossils or interbedded geochronologically datable primary volcanic tuff
Raath, 1984; Bordy et al., 2004a, 2004b, 2004c, 2004d). Historically, layers – the products of co-genetic volcanic events. Within the Karoo
the lEF and uEF were both considered to be Upper Triassic (e.g., Supergroup in the main Karoo Basin (Figs. 1 and 2), the Permo-Triassic
Haughton, 1924), but more recent work based on intercontinental Beaufort Group is a leading example of a radioisotopically well-dated,
faunal correlations has proposed that they represent Late Triassic and highly fossiliferous stratigraphic unit (e.g., Rubidge et al., 2013;

Fig. 1. (A) Overview geological map of the upper Karoo Supergroup showing the geographic distribution of the 16 detrital zircon samples presented in this study (see
Sections 3 and 4.1 for details). (B) Inset map of the MKB of South Africa and Lesotho showing the position of the Stormberg Group within it. (C) Karoo timewheel
showing the relative length of geological time represented by each main stratigraphic unit in the Karoo Supergroup.

2
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

(caption on next page)

3
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

Fig. 2. The chronostratigraphy of the Upper Triassic–Lower Jurassic Stormberg and Drakensberg groups (Karoo Supergroup), with special focus on the magneto- and
lithostratigraphy of the Elliot Formation, and the main body and trace fossil groups in the upper Stormberg Group. Radioisotopic dates are from this study for the
upper Stormberg Group and Moulin et al. (2017; K-Ar, 40Ar/39Ar methods) for the Drakensberg basalts. Unless otherwise marked, the maximum depositional ages are
based on the mean U-Pb detrital zircon dates of the youngest two or more grains with overlapping dates at 2σ (see Table 1). The less-than sign next to each MDA
indicates that the age of the sample could be younger (i.e., the MDAs are only maximum constraints on the age of deposition). Geological time scale based on the
International Chronostratigraphic Chart (v2018/08; ICS, 2018) and Cohen et al. (2013). The Triassic time scale shows both the ‘long Norian’ and ‘short Norian’
calibrations of the Late Triassic (see discussion in Lucas, 2018). Note that most lithostratigraphic boundaries are likely to be diachronous, i.e., they are time-
transgressive laterally across the basin from south to north and likely from west to east as well. Stormberg Group sample locations are shown in Fig. 1. For the
complete geochronologic dataset, see Supplemental Text S1, Table S1 and Fig. S1. For a key to the animal silhouettes and ichnites (mostly from Ellenberger, 1970),
see Supplemental Text S1.

Viglietti et al., 2018a). In contrast, radioisotopic dates have not been lithostratigraphic, sedimentologic, magnetostratigraphic, and bios-
used to constrain stratigraphic hypotheses for the overlying Stormberg tratigraphic (including ichnologic) perspectives. Therefore, we also
Group, which encompasses the Molteno, Elliot and Clarens formations. present a brief review of the unit on the basis of these multidisciplinary
Although both groups are globally recognised for their rich continental aspects and integrate the new dates into the overall stratigraphy of the
fossil assemblages and associated record of mass extinctions, the lack of Elliot and Clarens formations. While our dating results are circum-
a high-resolution chronology in the Stormberg Group limits its bios- scribed by the long duration and limited rock volume of the upper
tratigraphic utility. One reason for this is that the Stormberg Group Stormberg Group, this synthesis allows us to infer that the duration of
archives ~50 million years of geological history in <1.3 km of max- the Elliot depositional episode was middle Norian–Sinemurian, whereas
imum stratal thickness, whereas the Beaufort Group represents ~26 the Clarens depositional episode was mostly Pliensbachian, and thus to
million years of geological history in >4.5 km of thickness in the evaluate paleobiogeographical signals at the onset and recovery from
central MKB (see the Karoo timewheel: Fig. 1c). Regardless of the re- the ETME. Moreover, these new geochronological dates provide initial
solution of the geological archive provided by the Stormberg Group, steps towards a more quantitative understanding of the rates of sedi-
establishing a modern chronostratigraphic framework for the section ment preservation, basin evolution processes (e.g., sediment sources,
that encompasses the TJB in southern Africa is long overdue (Figs. 1 dispersal patterns), paleogeographic/climatic changes, and major pa-
and 2; e.g., Porro et al., 2010; Sciscio et al., 2017a; McPhee et al., leobiological events, including the tempo and mode of early dinosaur,
2017). crocodylomorph, turtle and mammalian evolution in southern Pangea
The depositional ages of the Stormberg Group formations (Fig. 2) during the transition from the Late Triassic to the Early Jurassic.
are constrained to some extent by biostratigraphy and, to a lesser ex-
tent, magnetostratigraphy. The oldest Stormberg Group unit, the Mol- 2. Stratigraphic background
teno Formation, is assumed to be Carnian, an assignment based on its
exceptionally well-preserved and diverse plant fossil assemblages that The ETME is one of the ‘Big Five’ biotic crises that are generally
are dominated by the seed fern Dicroidium (Figs. 2 and 3; Anderson and thought to have shaped large-scale patterns of Phanerozoic biodi-
Anderson, 1970; Anderson et al., 1998; Knoll, 2004; Labandeira et al., versity, fundamentally reorganising the taxonomic compositions of
2018). Based on biostratigraphic correlations with better-dated global both continental and marine biogeographic realms (e.g., Raup and
deposits, the two subdivisions of the unconformably overlying Elliot Sepkoski, 1982). Studying the effects of this global event in continental
Formation, the lEF and uEF, are believed to be Norian–Rhaetian and ecosystems is difficult because fossiliferous deposits spanning the Late
Hettangian–Sinemurian, respectively (for lEF see e.g., Hopson, 1984; Triassic–Early Jurassic (specifically the post-Carnian to pre-Toarcian
Gow and Hancox, 1993; Lucas and Hancox, 2001; Knoll, 2004; McPhee interval) are rare and generally poorly dated (e.g., Lucas, 2018). The
et al., 2017; for the uEF see e.g., Olsen and Galton, 1984; Smith and Elliot and Clarens formations of southern Africa have rich records of
Kitching, 1997; Lucas and Hancox, 2001; Knoll, 2005). The overall tetrapod body and trace fossils resulting from over a century and a half
Norian–Sinemurian age for the Elliot Formation has been confirmed of investigation (e.g., Owen, 1854; Haughton, 1924; Crompton and
recently via magnetostratigraphy (Sciscio et al., 2017a). The youngest Jenkins, 1968; Ellenberger, 1970, 1972, 1974; Kitching and Raath,
Stormberg Group unit, the Clarens Formation, contains a paucispecific 1984; Warren and Damiani, 1999; Yates and Kitching, 2003; Knoll,
assemblage composed of taxa similar or identical to those common in 2004, 2005; Butler et al., 2007; Yates, 2007a, 2007b; Yates et al., 2009;
the underlying uEF, and is inferred to be Sinemurian–Pliensbachian. McPhee et al., 2014, 2015a, 2015b, 2017, 2018; Dollman et al., 2019).
The age of the upper boundary of the Stormberg Group succession is However, the current state of its chronostratigraphic framework leaves
provided by the conformably overlying, Toarcian-aged continental the quantification of geological and biological processes uncertain. The
flood basalts, the outpouring of which terminated sedimentation of the lack of absolute age control in the upper Stormberg Group hampers our
Karoo Supergroup (Figs. 2 and 3). Radioisotopic dating of these basalts ability to accurately and precisely correlate the changes that have al-
indicates that the main pulse of this multi-peak volcanic event occurred ready been observed in basin development, ancient landscapes, climate,
between 181 and 183 Ma (Duncan et al., 1997; Svensen et al., 2012; and faunal assemblages with other global records, as well as with re-
Sell et al., 2014; Moulin et al., 2017). Moreover, field and radioisotopic gards to the position of the ETME. In the following sections, we review
evidence indicate that the outpouring of the first lava flows, at least in these and additional limitations of the currently weak chronostrati-
the southern MKB, started a few million years earlier in the late graphic framework, and highlight the recent advancements in the
Pliensbachian (at ~189 Ma; Fig. 2; Moulin et al., 2017). various stratigraphic studies of the Elliot Formation, the unit that
Here we present the first chronostratigraphic framework for the contains the Triassic-Jurassic boundary in southern Africa.
upper Stormberg Group, using new age constraints (maximum deposi-
tional ages, MDAs) obtained from detrital zircons in tuffaceous sand- 2.1. Lithostratigraphy
stones and siltstones via U-Pb geochronological methods (CA-ID-TIMS,
LA-ICP-MS). We regard this initial, basin-wide chronostratigraphic The first modern facies analysis study of the Elliot Formation con-
framework as an important independent test of previous age determi- firmed that the informal lithostratigraphic subdivisions of the unit (lEF
nation methods, while acknowledging that it is but a first step in solving and uEF, respectively) are recognizable on a basinal scale (Figs. 2 and 3;
a temporally and geographically enormous problem. Our efforts to ar- Bordy et al., 2004a, 2004b, 2004c, 2005). These studies also demon-
rive at the most accurate age interpretations of these newly obtained strated mappable regional spatiotemporal changes in facies distribu-
geochronological dates are informed by our collective tion, thickness, and sediment dispersal patterns (via provenance

4
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

Fig. 3. Sedimentary facies characteristics and main depositional conditions of the Molteno, lEF, uEF and Clarens formations (Stormberg Group) in the main Karoo
Basin of southern Africa. (Data sources: Ellenberger et al., 1964; Bordy et al., 2004a, 2004b, 2004c, 2004d, 2005; Bordy and Head, 2018; Rademan, 2018).

5
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

assessment of paleocurrents and source rock composition) in this fluvio- thickness trends at the lEF-uEF contact are explained by changes in
lacustrine succession. Although lithostratigraphic/sedimentary facies fluvial style and regional basin dynamics (Bordy et al., 2004a, 2004b,
characteristics (summarized in Fig. 3) assist in the straightforward se- 2004c, 2004d, 2005). The multi-storey, cliff-forming, asymmetrical
paration of the lEF and uEF, the lateral variability of the stratigraphic channel-fill sandstones of the lEF are interpreted as deposits of a per-
architecture at outcrop scale is so high throughout the basin that robust ennial fluvio-lacustrine system that formed under humid to semi-arid
correlation between facies associations is not feasible even in adjacent, climatic conditions. The moderately meandering channels in the lEF
high-quality outcrops. Although the correlation of fluvio-lacustrine were adorned by riparian forests separated from one another by ex-
strata is usually problematic (e.g., Miall, 2013, 2015, 2016), the ab- tensive overbank floodplain areas. In contrast to the lEF, the low-energy
sence of high-resolution subsurface data (e.g., seismic reflection studies, depositional system in the uEF was prone to cycles of sudden flash
core data) for the entire Stormberg Group further compound these flooding and prolonged desiccation under semi-arid climatic conditions.
correlation difficulties. The multi-storey, mostly tabular sandstones in the uEF formed in
The lEF is generally characterised by heterogeneous red-purple ephemeral watercourses, which intermittently flowed on vast flood-
(ranging from olive-grey to bluish-to-purplish-red) mudstone units with plains with abundant calcic paleosols and shallow, mostly ephemeral
rare colour mottling, and multi-storey, cliff-forming sandstone units lakes. A long-term trend of aridification that started in the uEF (e.g.,
with well-developed lateral accretion surfaces and irregular, erosive Bordy et al., 2004b; Bordy and Eriksson, 2015; Sciscio and Bordy, 2016)
basal bounding surfaces (Fig. 3; Bordy et al., 2004b, 2004d). The continued during the deposition of the Clarens Formation (Figs. 2 and
medium-grained, litho-quartzose sandstones in the lEF can be up to 3), which is dominated by massive to large-scale, cross-bedded sand-
~20 m thick, and have asymmetrical geometry in cross-section per- stones that formed in wet and dry deserts with large, down-wind and
pendicular to paleocurrent directions. The sandstones are often trough eastward migrating sand dunes (e.g., Beukes, 1970; Visser, 1984;
and planar cross-bedded and massive (structureless); low-angle cross- Eriksson, 1986; Bordy and Head, 2018).
bedding and planar stratification (horizontal lamination) are rare. The contrast in fluvial style, provenance and thickness trends across
Ripple cross-lamination, bioturbation structures and soft-sediment de- the contact of the lEF and uEF was interpreted by Bordy et al. (2004a,
formation features are all extremely rare. Well-defined, upward-fining 2004b, 2004c, 2004d, 2005) as a basin-wide unconformity. The dura-
successions often begin with mud-pebble conglomerate lags. The tion of this stratigraphic gap was sufficiently long to have allowed the
mudstones, which are 20–30 m thick on average, rarely display pedo- reorganization of regional fluvial depositional style and drainage pat-
genic alteration features (e.g., irregular mottles, desiccation cracks, terns. Neither the absolute time represented by this regionally map-
carbonate nodules) in contrast to uEF mudstones. pable paraconformity (essentially a sequence boundary) nor its date of
The majority of the uEF comprises very fine- to fine-grained, feld- occurrence relative to the TJB have been quantified, mainly because the
spatho-quartzose sandstones and pedogenically altered mudstones aforementioned lithostratigraphic and sedimentologic methods are
(mostly siltstones) (Fig. 3; Bordy et al., 2004b; McPhee et al., 2018). unsuitable for age assessments, except for coarse inferences of relative
The diagnostic sedimentary facies of the uEF are intraformational rates of sediment preservation (e.g., mature paleosols develop during
conglomerates, consisting mostly of reworked pedogenic nodules and prolonged low rates of clastic sediment accumulation vs flash flood
bone fragments, and clast-rich, massive, silty, very fine-grained sand- sediments representing sudden high rates of accumulation).
stones (e.g., Bordy et al., 2004b: pp. 393, 395, 397; Bordy et al., 2017a:
pp. 366, 369). The uEF sandstones are tabular, sheet-like bodies with 2.2. Magnetostratigraphy
thickness ranges of < 1–6 m that can extend laterally for several
hundred metres (Fig. 3). The uEF sandstones contain planar stratifica- Magnetostratigraphy is a stratigraphic correlation and relative
tion (horizontal lamination), ripple cross-lamination and, less com- dating tool that can provide an autonomous framework for delineating
monly, planar cross-bedding. Soft sediment deformation and bioturba- discrepancies between other stratigraphic correlation methods.
tion structures are common. In the uppermost uEF, sandstones are However, in fluvio-lacustrine units (such as the Elliot Formation),
slightly coarser, medium-grained and occur as channel-shaped bodies magnetostratigraphy is fraught with uncertainty due to the inherently
with rare lateral accretion beds that are up to 15 m in thickness, re- discontinuous nature and heterogenous stratigraphic architecture of
sulting in an overall upward-coarsening and upward-thickening char- such sedimentary rock successions (e.g., Miall, 2013). To minimize
acter. The uEF mudstones are brick-red, maroon to light pink in colour, analytical and stratigraphic uncertainties, modern magnetostrati-
and regularly show evidence for pedogenic overprinting (e.g., de- graphic approaches combine detailed litho- and biostratigraphic as-
siccation cracks, in-situ carbonate nodules, rootlets, colour mottling, sessments with robust age constraints, which serve as stratigraphic
bioturbation structures) in contrast to the lEF mudstones. Laminated calibration and anchoring points (e.g., Tauxe, 1998; Langereis et al.,
mudstones, appearing dark grey to black, are also present in the uEF. 2010).
They are rich in organic matter and some bear conchostracans in the Paleomagnetic studies in the Mesozoic of southern Africa have re-
uppermost uEF (e.g., Sciscio et al., 2017b; Rampersadh et al., 2018). latively low resolution and mostly focus on the Permo-Triassic
Sediment supply patterns prevailed from the south in the lEF but boundary and Toarcian continental flood basalts (e.g. De Kock and
were replaced by transportation directions mainly from west and south- Kirschvink, 2004; Lanci et al., 2013; Moulin et al., 2011, 2012, 2017).
west in the uEF (Fig. 3). Simultaneously, with the reorganization of the Magnetostratigraphic studies on the upper Stormberg Group are less
sediment supply patterns, the source of the sandstones also shifted from common (e.g., Opdyke, 1964; De Kock, 2003; Sciscio, 2016; Sciscio
a litho-quartzose provenance in the lEF to a feldspatho-quartzose one in et al., 2017a), and use magneto- and biostratigraphic correlations to
the uEF. Both the lEF and uEF show a decrease in thickness from south link this succession to its global counterparts in continental settings. For
to north within the MKB, especially over the southern margin of the example, Sciscio et al. (2017a) built a 280-m-thick composite magne-
Kaapvaal Craton. This south-to-north thinning is particularly distinct in tostratigraphic section for the Elliot Formation by combining nine
the lEF, which is ~300 m-thick near its type locality in the Barkly Pass, stratigraphic sections along a ~350-km-long transect in the MKB
and <10 m-thick in the northernmost region of the basin (e.g., Bordy (Figs. 1 and 2). This composite section comprises seven polarity pairs
et al., 2004b, 2004c; Bordy and Eriksson, 2015; McPhee et al., 2017). (EF2–EF8) and two single polarity intervals (EF1r and EF9n). The lEF
The uEF has a maximum thickness of ~255 m in the south and a contains four normal-reverse polarity intervals (EF2–EF5) and one re-
minimum thickness of <30 m in the north. verse polarity zone (EF1r). The uEF has three normal-reverse polarity
The marked differences in the architecture of the sedimentary facies intervals (EF6–EF8) and one normal polarity zone (EF9n). The basal
(i.e., contrasting sandstone body geometries and associated facies as- magnetostratigraphic tie is a single 40Ar-39Ar age of 215±3 Ma
semblages), sediment supply pattern, petrological composition and (Hälbich et al., 1983), which has been tentatively linked to a

6
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

deformation event in the Cape Fold Belt (Catuneanu et al., 1998) and to Eocursor, below and Section 5.4.3).
the unconformable contact between the Molteno and Elliot formations Therapsids in the lEF are represented by the traversodontid cyno-
(Bordy et al., 2005). This age has been debated by various authors (e.g., dont Scalenodontoides macrodontes (Crompton and Ellenberger, 1957),
Duane and Brown, 1992; Hansma et al., 2016; Blewett and Phillips, the youngest known traversodontid occurrence in Gondwana (Abdala
2016; Blewett et al., 2019), and is generally considered to be unreliable. and Gaetano, 2018). Recent work added the tritheledontid cynodont
The uppermost tie point is the radioisotopic age and magnetostrati- Elliotherium kersteni (Sidor and Hancox, 2006) and a possible diade-
graphy established for the Toarcian continental flood basalts (e.g., modontid cynodont (Abdala et al., 2007) to this list. However, the
Duncan et al., 1997; Moulin et al., 2011, 2012, 2017). Augmented with diademodontid site has been reassigned to the uEF (Bordy et al., 2017a)
biostratigraphic proxies, Sciscio et al. (2017a) also attempted a global and we consider it very likely that the Elliotherium kersteni is from a bed
correlation of the Elliot Formation to continental sections in the North in the lower uEF as this taxa is associated with other fossils and rock
American Newark APTS, Hartford Basin, Chinle and Moenave forma- types that are typical in the uEF (EMB and PAV, unpublished results).
tions of the Colorado Plateau (e.g., Nevada, Utah, northern Arizona, Although the ichnofossil record has hinted at the presence of large di-
western New Mexico), the Argentinian Ischigualasto–Villa Union Basin, cynodonts in the lEF (see Bordy et al., 2017b for a summary), this was
and the European St Audrie's Bay (UK) and Paris Basin (France) (e.g., only confirmed recently with the description of Pentasaurus goggai
Moreau et al., 2002; Donohoo-Hurley et al., 2010; Olsen et al., 2010; (Kammerer, 2018), and the discovery of new, currently unpublished
Zeigler and Geissman, 2011; Hüsing et al., 2014; Kent et al., 2014). fragmentary dicynodont material from Eastern Cape Province (JNC,
Sciscio et al. (2017a) emphasized that the magnetozones have an PAV, LS, unpublished data).
inconsistent thickness between sections given isopach changes across Pseudosuchian archosaurs have an enigmatic presence within the
the basin, which are due to the variable sedimentation rates and the lEF. Previous reports of aetosaurs (e.g., Kitching and Raath, 1984) have
erosional events typical in fluvio-lacustrine environments. Although the been shown to be spurious (Tolchard et al., 2019). Maxillary and
basin-wide unconformity between the lEF and uEF is of unknown dentary fragments of likely two species of ‘rauisuchians’ (i.e., non-
duration, it has been recognized to have impacted the magnetostrati- crocodylomorph pseudosuchians branching later than aetosaurs) were
graphic correlations. Moreover, the main caveat of Sciscio et al. (2017a) recently identified by Tolchard et al. (2019), but their exact provenance
is the need for reliable radioisotopic calibration points, and a revision of is unknown. It is possible that isolated occurrences of serrated, recurved
the current biostratigraphic framework of the Elliot Formation. There- teeth indicates the presence of carnivorous pseudosuchians, such as
fore, the reliability of the composite magnetostratigraphic section is poposauroids or non-crocodylomorph loricatans (e.g., ‘Basutodon’, von
expected to increase with the addition of new multi-disciplinary data- Huene, 1932; see also Tolchard et al., 2019) but, as noted above, a
sets. possible theropod identity cannot be discounted. Although never ade-
quately studied, some of the material collected as part of the ‘Aliwalia
2.3. Biostratigraphy rex’ (Galton, 1985) assemblage is possibly non-crocodylomorph pseu-
dosuchian in origin, and could explain why material otherwise refer-
The Elliot Formation preserves a series of exemplary Late able to Sauropodomorpha might have been incorrectly interpreted as a
Triassic–Early Jurassic continental faunas (Fig. 2) that have been the ‘herrerasaurid theropod’ (see Yates, 2007a).
subject of several in-depth reviews (e.g., Haughton, 1924; Ellenberger, The final major tetrapod group within the lEF is temnospondyl
1970; Kitching and Raath, 1984; Knoll, 2004, 2005; McPhee et al., amphibians. This fauna was most recently reviewed by Warren and
2017; Viglietti et al., 2020a, 2020b). We summarize these below and Damiani (1999) who noted the presence of several indeterminate ste-
present the preliminary findings of a quantitative investigation into the reospondyls all referable to chigutisaurids, which discounts evidence of
geospatial and stratigraphic distributions of upper Stormberg Group capitosaurids from the lEF (Kitching and Raath, 1984). Based on our
vertebrate taxa in Sections 3.2 and 4.2 (see also Viglietti et al., 2020a, recent field investigations and consultation with the original collector
2020b). (B. Battail, personal communication, 2018), we confirm that the stra-
The lEF is similarly fossiliferous to, but less taxonomically diverse tigraphic position of a gigantic brachyopid stereospondyl mentioned by
than, the uEF, and is dominated by early branching sauropodomorph Steyer and Damiani (2005) originated from the lower uEF near Alwyns
dinosaurs (Fig. 2). Following McPhee et al. (2017), this assemblage Kop in Lesotho.
includes the following valid sauropodomorph genera: Plateosauravus In contrast to the lEF, the uEF and Clarens Formation saur-
(Haughton, 1924; Yates, 2003), Eucnemesaurus (Van Hoepen, 1920; opodomorph assemblage is morphologically and taxonomically diverse
Yates, 2007a; McPhee et al., 2015a), and Blikanasaurus (Galton and Van (Fig. 2), and contains the following valid genera: Massospondylus
Heerden, 1985; Yates, 2008). Melanorosaurus is provisionally retained (Cooper, 1981; Gow et al., 1990; Yates and Barrett, 2010; Chapelle and
as a fourth valid sauropodomorph taxon on the basis of its syntype Choiniere, 2018; Barrett et al., 2019), Antetonitrus (Yates and Kitching,
material (Haughton, 1924; Galton et al., 2005; Yates, 2007b; PMB and 2003; McPhee et al., 2014), Aardonyx (Yates and Barrett, 2010), Pula-
JNC, unpublished results), although its taxonomic validity and the re- nesaura (McPhee et al., 2015b; McPhee and Choiniere, 2018), Ngwevu
ferral of key specimens have been questioned (McPhee et al., 2015a, (Chapelle et al., 2019) and Ledumahadi (McPhee et al., 2018). Together
2015b, 2017). It is clearly in need of substantial revision and this work these taxa comprise a diverse fauna ranging from gracile massos-
is currently in progress (PMB and JNC, unpublished results). Ad- pondylids to multi-tonne lessemsaurids, signalling disparate feeding
ditionally, although the recently named Sefapanosaurus (Otero et al., ecologies supported by distinct biomechanical strategies (McPhee et al.,
2015) and Meroktenos (Gauffre, 1993; de Fabrègues and Allain, 2016) 2015b, 2017, 2018). The discovery of Pulanesaura also indicates the
have poor provenance data, these might also represent valid lEF taxa. presence of the earliest branching sauropods (McPhee et al., 2015b).
As concluded by McPhee et al. (2017), determining clear morphological Two other genera, Arcusaurus (Yates et al., 2011) and Ignavusaurus
boundaries between lEF sauropodomorphs is far from straightforward, (Knoll, 2010) have also been named from the uEF in recent years, but
with all known taxa possessing medium-to-large bodied, variably robust their validity has been questioned (e.g., Yates et al., 2011; McPhee
phenotypes that are currently distinguished by subtle postcranial fea- et al., 2017).
tures (cranial remains are currently unknown for all named species). Theropod dinosaurs in the uEF and Clarens Formation are re-
Although tridactyl theropod trackways are well-known in the lEF (see presented by fragmentary remains attributed to the southern African
Section 2.4), the only body fossil evidence for theropods thus far comes coelophysid Megapnosaurus rhodesiensis (formerly Syntarsus rhodesiensis
from isolated teeth (e.g., Ray and Chinsamy, 2002), which are very and Coelophysis rhodesiensis, see below; Kitching and Raath, 1984;
difficult to distinguish from those of carnivorous pseudosuchians. There Smith and Kitching, 1997; Munyikwa and Raath, 1999; Bristowe and
are currently no ornithischian dinosaurs in the lEF (see comments on Raath, 2004). A larger theropod species, Dracovenator regenti, was

7
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

named from partial cranial material and is possibly related to the North deposits are approximately contemporaneous. These North American
American taxon Dilophosaurus (Yates, 2005). Several genera of basal strata are currently regarded as Hettangian in age on the basis of a
ornithischian dinosaur have been named from the uEF, including some radioisotopic date recovered from basalts underlying the McCoy Brook
of the earliest global occurrences of the group. Lesothosaurus diag- Formation (Sues and Olsen, 2015) and detrital zircon dating of the
nosticus (e.g., Galton, 1972; Butler, 2005; Porro et al., 2015; Barrett Moenave Formation (Suarez et al., 2017). Correlation with the McCoy
et al., 2016; Baron et al., 2017a; Sciscio et al., 2017c) represents the Brook Formation is further supported by the shared genera Pachygenelus
most abundant form (Stormbergia dangershoeki is currently thought to be (a trithelodontid cynodont: Shubin et al., 1991) and Clevosaurus (a
a junior synonym of this taxon: e.g., Baron et al., 2017a; Sciscio et al., rhynchocephalian lepidosauromorph: Sues and Reisz, 1995), both of
2017c). Several heterodontosaurids are also known (Porro et al., 2010), which are absent from the depauperate fauna of the Moenave Forma-
including: Heterodontosaurus tucki (e.g., Crompton and Charig, 1962; tion. The absence of these same genera from the diverse fauna of the
Norman et al., 2011; Sereno, 2012), Abrictosaurus consors (Thulborn, Kayenta Formation, which overlies the Moenave Formation, suggests
1974; Sereno, 2012), Lycorhinus angustidens and Pegomastax africanus that the fauna of the uEF is older than that of the Kayenta Formation,
(Butler et al., 2008; Sereno, 2012). Eocursor parvus, originally described which has been dated with detrital zircons as no older than late
as a lEF taxon (Butler et al., 2007), has recently been reinterpreted as Pliensbachian (Marsh and Rowe, 2018). Protosuchus has been reported
coming from the uEF (Olsen et al., 2010; McPhee et al., 2017). from the Hettangian of Poland (Gierliński and Potemska, 1985), pro-
The most abundant uEF taxa after dinosaurs are cynodonts, of which viding a potential European correlation, but the latter requires con-
both non-mammaliaforms and mammaliaforms are present. Following firmation given the similarities of this material to other ‘protosuchian’
the recent review of Abdala and Gaetano (2018), the former includes genera. More promisingly, Clevosaurus has been recorded from several
the tritheledontid taxa Tritheledon, Diarthrognathus and Pachygenelus, other Early Jurassic faunas, including the Hettangian of the UK (Evans
and the tritylodontids Tritylodontoideus and Tritylodon. Mammalia- and Kermack, 1994) and the Early Jurassic of China (Luo and Wu,
formes are represented by the morganucodontids Megazostrodon and 1994), extending its potential use in global correlations. However, the
Erythrotherium. stratigraphic range of Clevosaurus also extends into the Late Triassic,
The diverse uEF crocodylomorph fauna was recently reviewed by based on abundant material from the UK, Belgium, Luxembourg and
Dollman et al. (2019), with the following taxa recognised currently: the Brazil (Fraser, 1988; Godefroit and Sigogneau-Russell, 1995; Hsiou
non-crocodyliform crocodylomorphs Sphenosuchus acutus and Litargo- et al., 2015, 2019). Although these Late Triassic occurrences might
suchus leptorhynchus; and the crocodyliform ‘protosuchids’ Protosuchus offer some support for the proposed Rhaetian age inferred for the lower
haughtoni, Notochampsa istedana and Orthosuchus stormbergi. Dollman uEF on the basis of magnetostratigraphy (Sciscio et al., 2017a), it is
et al. (2019) noted that the majority of crocodylomorph occurrences are more likely that they represent species that are temporally distinct
restricted to the upper half of the uEF. Non-archosaurian reptiles are (Hsiou et al., 2019), although the monophyly of Clevosaurus has re-
represented by the early turtle Australochelys africanus (Gaffney and cently been questioned (Herrera-Flores et al., 2018; Hsiou et al., 2019).
Kitching, 1994) and a specimen of the rhynchocephalian Clevosaurus sp. Finally, coelophysoid theropods have been identified in both the Late
(Sues and Reisz, 1995). As with the lEF, the temnospondyl record of the Triassic (Norian) Chinle Formation of the USA (Coelophysis bauri;
uEF is restricted to mostly indeterminate remains of chigutisaurid ste- Colbert, 1989) and the uEF (Megapnosaurus rhodesiensis; Munyikwa and
reospondyls (Warren and Damiani, 1999), one of which is the largest Raath, 1999; as ‘Syntarsus’ rhodesiensis and as ‘Coelophysis’ rhodesiensis
brachyopid stereospondyl documented to-date (Steyer and Damiani, in previous works), although with distinct species in each region.
2005), as noted above. However, anatomically similar coelophysoids are also known from
other Lower Jurassic deposits such as the Lufeng Formation of China
2.3.1. Biostratigraphic correlations (You et al., 2014) and the Kayenta Formation of North America (e.g.,
Global biostratigraphic correlations for Stormberg Group vertebrate Rowe, 1989). The inclusion of M. rhodesiensis within Coelophysis was
faunas are based on weak evidence. Many rely on broad faunal simi- proposed by Bristowe and Raath (2004), and could potentially indicate
larities, or ‘stage of evolution’ arguments, rather than the presence of correlation with the Chinle Formation. However, recent phylogenetic
shared index taxa, leading to low precision, issues of replicability be- studies question that assignment (e.g., Ezcurra and Brusatte, 2011;
tween different workers, and decreased confidence in their results (e.g., Griffin and Nesbitt, 2019; Wang et al., 2017), and we regard the oc-
Olsen and Sues, 1986; Lucas, 1998; see Rayfield et al., 2009 for a cri- currence of coelophysoids in the uEF as consistent with an either Late
tique of this approach). No species-level taxa in the Stormberg Group Triassic or Early Jurassic age (also see Martínez and Apaldetti, 2017).
are shared with non-African faunas, only four genera are shared, and Biostratigraphic correlations between other Karoo-aged basins in
phylogenies including Stormberg taxa at the species level are either southern and eastern Africa are also weakly supported due to the ab-
labile, weakly supported, or have not been performed. sence of shared genus- and species-level taxa. Currently, no lEF verte-
The poorly known lEF fauna has been biostratigraphically corre- brate taxa are known outside the main Karoo Basin. However,
lated, often tentatively, with Norian–Rhaetian faunas such those from Megapnosaurus rhodesiensis has been identified in the uEF and the Forest
the ‘Middle Keuper’ of Germany, the Los Colorados Formation of Formation of Zimbabwe (e.g., Raath, 1969; Munyikwa and Raath,
Argentina, and the Chinle Formation of the USA (Lucas and Hancox, 1999), along with more taxonomically tentative records of Clevosaurus
2001; Knoll, 2004; McPhee et al., 2017). However, the absence of (Gow, 1977) and Notochampsa (Raath, 1981). Moreover, occurrences of
shared index taxa does not allow for a precise correlation with these the basal sauropodomorph dinosaur Massospondylus have been reported
deposits. The fauna of the uEF and Clarens Formation has been corre- in several Zimbabwean basins (e.g., the Mid-Zambezi, Mana Pools, and
lated with Early Jurassic faunas, such as the Lower Lufeng Formation of Tuli basins: Attridge, 1963; Raath et al., 1970; Bond, 1973; Cooper,
China, the Glen Canyon Group of the USA, and the McCoy Brook For- 1981; Munyikwa, 1997; Rogers et al., 2004) and might provide a direct
mation of Canada (Olsen and Galton, 1984; Smith and Kitching, 1997; faunal link, but the material reported from the Zimbabwean localities
Lucas and Hancox, 2001; Knoll, 2005) and, in this case, some poten- requires taxonomic reassessment to confirm these proposals (Barrett
tially useful shared taxa are present. For example, the close similarity of et al., 2019). A tentative link between the uEF and Karoo-aged strata in
species within the crocodyliform genus Protosuchus from the uEF the Luangwa Basin of Zambia has been posited on the basis of tax-
(P. haughtoni), the Moenave Formation of the Glen Canyon Group, Ar- onomically indeterminate sauropodomorph material (Choiniere and
izona (P. richardsoni: Clark, 1986), and the McCoy Brook Formation, Barrett, 2015), but requires additional support.
Nova Scotia (P. micmac: Sues et al., 1996) does suggest that these

8
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

2.4. Vertebrate ichnology 3. Methodology

The Stormberg Group, and in particular its upper part (Fig. 2), 3.1. Detrital zircon U-Pb geochronology
contains a diverse and globally important trackway record, with an
abundant assemblage of ichnites attributable to herbivorous and car- Given the dearth of prior radioisotopic constraints for the Stormberg
nivorous dinosaurs, dicynodonts, cynodonts, amphibians, crocodylo- Group, due to the lack of obvious primary volcanic lithologies amenable
morphs, and mammaliaforms (e.g., Ellenberger et al., 1964; to radioisotopic dating, we have taken the approach of seeking juvenile
Ellenberger, 1970, 1972, 1974; Raath et al., 1990; Smith et al., 2009; zircons from volcanoclastic lithologies. Juvenile zircons, those that
Wilson et al., 2009; Marsicano et al., 2014; Sciscio et al., 2016, 2017c; show morphological evidence for limited reworking within a sedi-
Abrahams et al., 2017; Bordy et al., 2017b; Rampersadh et al., 2018). mentary system, and return ages that are ‘close’ to the true depositional
Paul Ellenberger conducted the pioneering work on this record, age (TDA), have been used in similar continental sedimentary rock
erecting a great diversity of tetrapod ichnogenera and ichnospecies in (e.g., Ramezani et al., 2011) to provide useful maximum age constraints
addition to establishing an ichnostratigraphic subdivision for the on the timing of sediment accumulation, aiding the revision of the
southern African Upper Triassic to Lower Jurassic (e.g., Ellenberger, chronostratigraphic frameworks. Detrital zircons were successfully ex-
1970, 1972, 1974). This work introduced the first biozonation scheme tracted and dated from 16 rock samples taken from the Elliot and
for the Stormberg Group (e.g., lEF includes zones A1 – A6; uEF and the Clarens formations in South Africa and Lesotho (Supplemental Table
Clarens Formation zones A7, B1– B7) and underscored the division of S1). Because primary volcanic tuff layers (i.e., pyroclastics) were not
the regional ichnofaunas into two broad biozones that reflect a faunal identified in the studied succession, the sampled rocks are exclusively
change equivalent to the turnover in skeletal remains both locally and tuffaceous sandstones and siltstones.
globally during the Late Triassic to Early Jurassic. Ellenberger’s seminal Zircons were separated using a modified standard method described
work was revised some 35 years ago by Olsen and Galton (1984), who by Tucker et al. (2013), which includes rock crushing, panning, and
significantly reduced the number of ichnotaxa through synonymization. magnetic and heavy liquid separation of minerals. Zircons from each
While this ichnotaxonomic revision was valuable for pointing out sev- sample were inspected under optical microscopy and showed mixed
eral nomina dubia, it also lumped some valid ichnotaxa, which had the morphology ranging from rounded, indicating reworking in a sedi-
unfortunate corollary of decreasing confidence in the value of the mentary environment, to acicular and faceted, with medial melt in-
southern African ichnofossils for global correlation and regional bio- clusions, typical of volcanic zircon that has not undergone significant
diversity assessments. The shortcomings of this alternative ichnologic reworking. Zircons were then mounted in resin for laser ablation in-
framework have to some extent been rectified subsequently by, for ductively coupled plasma ionization mass spectrometry (LA-ICP-MS) U-
example, Lockley et al. (1996, 2004, 2006), Rainforth (2003), Lockley Pb dating, with a subset of samples undergoing LA-ICP-MS mounted on
and Gierlinski (2006), D’Orazi Porchetti and Nicosia (2007), and tape. The grains from the youngest population were removed for sub-
D'Orazi Porchetti et al. (2015, 2017, 2018). Based on these revisions, sequent high-precision chemical abrasion thermal ionization mass
the currently accepted tetrapod ichnofaunal list for the Upper Triassic spectrometry (CA-ID-TIMS) U-Pb dating (Supplemental Text S1 and
lEF includes Tetrasauropus, ?Lavinipes jaquesi, and Pseudotetrasauropus Table S1). Both of these geochronological methods have their ad-
(quadrupedal and bipedal sauropodomorph dinosaurs), Grallator vantages, as the LA-ICP-MS method allows for a large number of grains
(theropod dinosaur), Pentasauropus (dicynodont), Sauropodopus (prob- to be analysed, increasing the probability of finding the youngest
able rauisuchians, with similarities to Chirotherium), Paratetrasauropus grains, and the CA-ID-TIMS allows for high-precision analyses and
(crocodylomorph) and cf. Brachychirotherium (archosauromorph), treatment for open-system behavior. We have therefore attempted to
whereas for the Lower Jurassic uEF and Clarens Formation contains take advantage of both these methods. Laser ablation spots were se-
Episcopopus (amphibian), Batrachopus (crocodylomorph), Moyenisaur- lected after the careful consideration of inclusions, textural features,
opus and Trisauropodiscus (ornithischians), Ameghinichnus (tritylodontid and cracks within the imaged zircon grains. The geochronological
cynodonts), a great variety of ichnotaxa on the Grallator-Eubrontes dating procedures are further detailed in Supplemental Text 1.
plexus and Kayentapus (theropods) as well as tentative Brasilichnium-
like (mammaliaforms, ?synapsids) tracks. This list is far from complete,
and our ongoing ichnologic work is aimed at updating and refining the 3.2. Biostratigraphic methods
Stormberg Group ichnostratigraphic scheme to better reflect the true
ichnofaunal diversity and the degree of trackmaker endemism within We collated, to the best of our ability, all metadata relating to fossil
the Stormberg Group (EMB, unpublished results; Sciscio et al., 2016, collections of upper Stormberg tetrapods including both South African
2017c; Abrahams et al., 2017; Bordy et al., 2017b, 2020; Rampersadh (e.g., Iziko Museum, Evolutionary Studies Institute, Albany Museum,
et al., 2018). National Museum) and international repositories (e.g., London Natural
The vertebrate track record of the Stormberg Group, based on History Museum, Muséum National d’Histoire Naturelle Paris,
shared ichnofauna elements, has been linked to Late Triassic to Early Naturhistorisches Museum Wien Vienna) using a combination of pub-
Jurassic tracks on all continents except Antarctica (for relevant reviews lications, online databases, and collections records and archives of field
see Olsen and Galton, 1984; D’Orazi Porchetti and Nicosia, 2007; Lucas, notes obtained directly from museum collections. We cleaned these
2007; Klein and Lucas, 2010; D'Orazi Porchetti et al., 2015; Citton et al., data (e.g., removing spelling errors in taxon names) using automated
2018; Hunt et al., 2018 and references therein). However, the coarse means where possible. Geospatial positions and stratigraphic prove-
temporal and stratigraphic resolution of the tetrapod footprint record in nance of specimen occurrences were validated by cross-referencing
the upper Stormberg Group, in addition to an outdated regional ich- Google Earth, GIS software, field notes, personal measurements of
notaxonomy, limits its utility for fine-scale regional or global correla- stratigraphic sections at historic fossiliferous sites, and the collection of
tions, despite the great abundance of ichnites within the succession. As new, well-georeferenced fossil material. Collation of this large historic
mentioned above, this shortcoming is being addressed systematically by dataset resulted in over 1400 records, about 20% of which are geor-
our ongoing research program, and a major review of the upper eferenced with high credibility, and located across the northern,
Stormberg track assemblages, augmented by refined stratigraphic data eastern, southern, and western extent of the Elliot and Clarens forma-
on existing ichnofossil sites and additional collecting, will be forth- tions in the MKB (see Data Availability section).
coming. Because the current ichnozonation is in flux, this topic is not
discussed further herein.

9
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

Table 1
Summary of the results from the different age estimating methods used in constraining the maximum depositional age (MDA) of the samples from the Elliot and
Clarens formations in the upper Stormberg Group.
Sample ID Method Lithology & formation Preferred MDA YSG YC1σ(2+) YC2σ(2+) YC2σ(3+) YPP Concordia

Age 2σ Age 2σ Age 2σ Age 2σ Age 2σ Age Age 2σ

BH-15 ID-TIMS tsa lEF 204.9 0.88 204.02 0.59 205 1.2 204.9 0.17 205 205 0.64
GV-14 ID-TIMS tsa lEF 212.1 0,19 212.1 0,19
HAF LA-ICP-MS tsa lEF 209.6 1.4 208.2 2.9 209 1.6 209.6 1.4 209.6 1.4 209 209.7 2.4
HB-15 LA-ICP-MS tsi lEF 218.2 2 216.5 4.7 518 9.9 218.2 2 518 9.7 520 218.9 10
LEP LA-ICP-MS tsa uEF 197.3 2.3 195.7 3.1 239.6 2.7 197.3 2.3 243 1.5 196 197.4 2.3
LGT ID-TIMS tsi lEF 209.11 0.20 209.11 0.2 208.01 0.86 208.01 0.86 209 208 0.9
LK-17 LA-ICP-MS tsi lEF 207 5 207.6 5.1 207 5.1 467 8 207 208.3 4.7
LMO LA-ICP-MS tsa uEF 199.9 2.3 193.9 2.9 199.8 2.1 199.9 2.3 232 2 194 200 8.6
MAF LA-ICP-MS tsa uEF 201 2.3 199 3 254 2.8 201 2.3 256.4 1.4 256
MAP LA-ICP-MS tsa lEF 215.4 1.9 215.4 1.1
PHU LA-ICP-MS tsa lEF 219.6 2.5 217.6 3.4 242.6 2.2 219.6 2.5 243.1 2 243 219.7 2.5
PS-15 ID-TIMS tsa uEF 202.33 0.19 202.33 0.19 217.14 0.14 217.21 0.11 217.21 0.11 217 217.2 0.54
Q6 LA-ICP-MS tsa um uEF 191.1 1.5 190 2.4 193.4 1.2 191.9 1.5 467.4 2.5 193
QSS1 LA-ICP-MS tsa lEF 211.5 2.8 207.5 9 211.5 2.9 211.5 2.9 211.5 2.9 213
SUB LA-ICP-MS tsa lEF 216.4 2.4 215.4 3.4 216.4 2.4 216.4 2.4 252 2 216
UMC LA-ICP-MS tsa Clarens 187.5 1.6 185.9 2.2 186.7 1.2 187.5 1 187.5 1 187

For spatiotemporal distribution of the samples, see Figs. 1, 2, 4 and 5. Abbreviations: lEF – lower Elliot Formation; uEF – upper Elliot Formation; tsa – tuffaceous
sandstone; tsi – tuffaceous siltstone; um – uppermost; σ – internal error. Ages and errors are in millions of years (Ma). Strike-through indicates that the MDA is too
young based on stratigraphic constraints. For justification of the preferred MDAs, see Supplemental Text 1.

4. Results anticipated from other stratigraphic considerations (e.g., samples BH-

15, LGT). This is likely due to Pb-loss as explained further in the Sup-
4.1. Appraisal of the detrital zircon U-Pb geochronology methods plementary Text 1 (as well as in e.g., Dickinson and Gehrels, 2009;
Corfu, 2013; Andersen et al., 2019; Gehrels et al., 2019; Herriott et al.,
To constrain the maximum depositional age (MDA) of the samples 2019; Rossignol et al., 2019). To put it differently, the YSG method is
from the Elliot and Clarens formations (Table 1), we explore a range of statistically the least robust, because it focuses on a single data point
different age calculations for the data obtained from the analysed det- that is often not reproducible in detrital zircon samples. The MDA
rital zircons in order to assess the sensitivity of the age constraints to calculations based on age clusters are more robust than the YSG be-
our chosen interpretive framework. The complete dataset of geochro- cause they incorporate multiple data points and account for internal
nological measurements is provided in Supplemental Table S1. The analytical error and the external error of the population (e.g., Dickinson
relevant measurements for each sample are illustrated in Figs. 2, 4 and and Gehrels, 2009; Gehrels et al., 2019; Herriott et al., 2019). Although
5, and in Supplemental Fig. S1. The latter provides, for each sample, the YC2σ(3+) is the second most quoted MDA metric in the literature (e.g.,
relative-age-probability plots (or probability density plots; PDPs) for Coutts et al., 2019), our preferred method for interpreting MDA is its
zircons < 500 Ma, concordance diagrams for zircons younger than modified version, the YC2σ(2+) metric as explained above. Contrary to
~250 Ma and plots of the weighted mean dates (at 95% confidence) for the YSG metric, the weighted mean calculated from the youngest grain
zircons younger than ~250 Ma. The sample-by-sample appraisal of the cluster composed of three or more grains that overlap at 2σ uncertainty
detrital zircon U-Pb geochronology dates and the justification of the (YC2σ[3+]) often appears to be too conservative, yielding older dates
preferred MDAs and true depositional ages (TDAs) are in Supplemental (early Middle Triassic – Ordovician) than the currently accepted age
Text 1. estimates of Late Triassic – Early Jurassic for the upper Stormberg
Our preferred method for estimating the MDA (Table 1) is the Group (e.g., samples HB-15, LK-17). In these cases, it is likely that the
weighted mean of the youngest cluster comprised of two or more grains measured dates are significantly older than the TDA, because near-de-
overlapping with the youngest one at 2σ internal error (YC2σ[2+]). positional-age zircons were not captured in the sediment during de-
Using two or more analyses instead of the single youngest analysis re- position or are too few in the sample for this date-calculation method
duces the chance that the MDA will be younger than the TDA (e.g., as a (e.g., Andersen et al., 2016). For more comprehensive discussions on
consequence of Pb-loss; see below). Furthermore, our use of CA-ID- the various MDA calculations used, including their relative merits, the
TIMS analyses for a subset of the samples helps deal with this issue – the readers are referred to Dickinson and Gehrels (2009), Andersen et al.
use of chemical abrasion pre-treatment means that Pb-loss is effectively (2019), Coutts et al. (2019), Gehrels et al. (2019), Herriott et al. (2019),
eliminated, and the higher-precision of the data allow for a more robust Johnstone et al. (2019), and Rossignol et al. (2019).
assessment of Pb-loss via discordance. Although somewhat subjective,
the use of the YC2σ(2+) metric resulted in the MDA calculations being
more consistently compatible with other stratigraphic considerations in 4.2. Biostratigraphic results
the study area. The considerations that are used to constrain our age
interpretations include: the stratigraphic context of the sample (i.e., its Several clear patterns emerge from our database of upper Stormberg
position relative to key stratigraphic contacts and other dated samples tetrapod occurrences (Fig. 6). There is a pronounced turnover across the
in the same section to which a sample must obey the law of super- lEF/uEF contact. No currently valid tetrapod genera cross this
position); the location of the sample within the Karoo foreland basin; boundary, and although it is possible that the dipnoan Ceratodus is
lithostratigraphic characters of the host rocks (i.e., lEF, uEF); and the present in the lEF (EMB, unpublished results) and uEF (Kitching and
body and ichnofossil content of the adjacent strata. In several samples, Raath, 1984), this genus is temporally wide-ranging and the upper
the youngest single grain method (YSG) seems to be the least re- Stormberg material is in need of revision. Family level or higher
presentative of the TDA, because the measured dates of the youngest boundary crossers include Sauropodomorpha, Cynodontia, and Chigu-
grain often seem younger than the expected TDA than what would be tisauridae.
Lower Elliot Formation tetrapod occurrences, in general, are as

10
 E.M. Bordy, et al.

11
Fig. 4. Distribution of the calculated weighted mean dates of the youngest detrital zircon populations (at 95% confidence level) in the analysed samples from the Elliot (15) and Clarens formations (1). Unless otherwise
stated on the figure, all ages shown are our preferred MDAs (see Table 1). The arrow pointing forward in time on each MDA line indicates that the age of the sample could be younger (i.e., the MDAs are only maximum
constraints on the age of deposition). Bar heights depict 2σ internal analytical uncertainty for each sample. Only detrital zircon dates used in the MDA calculation are depicted; older zircon grain dates are not shown,
Earth-Science Reviews 203 (2020) 103120

however, each sample is further illustrated in Supplemental Fig. S1. For the complete analytical dataset, see Supplemental Text S1, Table S1 and Fig. S1. Sample locations are shown in Fig. 1 and corresponding
stratigraphic logs are in Fig. 5. (Rademan, 2018).
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

(caption on next page)

12
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

Fig. 5. Simplified stratigraphic sections showing the detrital zircon sample position within the local succession. Sections are arranged from north to south (see inset
map for locations). Question marks at the stratigraphic contact of the lEF-uEF indicate a <10 m thick zone where lack of outcrops obscure the position of the contact.
The uncertain position of the TJB is emphasised by the grey band. The less-than sign next to each MDA indicates that the age of the sample could be younger (i.e., the
MDAs are only maximum constraints on the age of deposition). For the complete analytical dataset, see Supplemental Text S1, Table S1 and Fig. S1. (Rademan,
2018).

geographically widespread as those in the uEF (Fig. 7), but the lEF has 2019; Gehrels et al., 2019; Herriott et al., 2019; Rossignol et al., 2019).
many fewer well-georeferenced records for valid taxa. For example, in Given that both assumptions cannot be independently constrained, we
the lEF, only Scalenodontoides, Eucnemesaurus, and possibly Melanor- assume that all zircon is recycled and that MDA calculations provide
osaurus have representative occurrences approximating the geographic only a maximum constraint on the TDA (i.e., the TDA is younger or
range of the overall fossil sample, whereas the uEF has at least five such equal to the MDA). By integrating MDA calculations with existing age
taxa (Heterodontosaurus, Lesothosaurus, Massospondylus, Protosuchus, constraints from bio-, litho-, and magnetostratigraphic considerations,
Tritylodon). In addition, poor taxonomic data for many lEF tetrapod we show that, in many cases, the youngest detrital zircons provide
occurrences lead to insufficient ability to infer stratigraphic ranges for meaningful age constraints on the age of the upper Stormberg Group. In
most lEF taxa, with the notable exception of Scalenodontoides (Viglietti the following sections, we discuss the implications of the new detrital
et al., 2020a). This traversodontid cynodont is the only lEF tetrapod zircon U-Pb ages for the geological history of the TJB-bearing Elliot
taxon that is currently taxonomically valid, easy to identify in the field, Formation in southern Africa from various from litho-, magneto- and
and known from multiple specimens with well-constrained provenance biostratigraphic aspects.
from several lEF localities across the basin (unlike any of the lEF di-
nosaur taxa – see e.g., McPhee et al., 2018). In contrast, the uEF is rich 5.2. New radioisotopic dates and basin evolution
in well-georeferenced fossil taxa, providing detailed information for the
stratigraphic ranges of most uEF fossil vertebrate lineages (Fig. 7). We 5.2.1. Lithostratigraphic considerations
also observe a similar trend to that noted by Kitching and Raath (1984), Based on our new geochronological dataset, obtained from U-Pb
where a sudden increase in the relative abundance of fossils occurs radioisotopic dates of detrital zircons in the Elliot and Clarens forma-
above the lower quarter of the uEF and at the contact of the uEF and tions (Figs. 1, 2 and 4), the upper Stormberg Group probably spans ~40
Clarens Formation. Ma, and we infer that this represents a sedimentation record from the
Dinosaur lineage diversity increases markedly in the uEF (Fig. 6). middle Norian to early Toarcian. In particular, the TJB-bearing Elliot
Sauropodomorpha are present and relatively abundant in both the lEF Formation represents a ~30-million-year-long depositional episode
and uEF (Fig. 8), although more species are currently known from the (from ~220 to ~190 Ma ago), of which the lEF and uEF account for
uEF. In contrast, ornithischians and non-dental theropod remains are ~20 and ~10 Ma, respectively.
absent from the lEF, indicating that they were either rare or absent. These new age-constraints help us understand the resolution and
Both groups have their first definite occurrences low in the uEF. Many completeness of the geological history captured in the late stages of
other tetrapod lineages also make their first appearance in the uEF, MKB evolution. The completeness of the depositional record is variable
with the first crocodylomorphs, probainagnathians (including trithelo- across the MKB, as evidenced by the fact that the Elliot Formation,
dontids and tritylodontids), and turtles appearing relatively low in the which was deposited in ~30 million years, is ~500-m-thick in the south
uEF and the first mammaliaforms definitively appearing closer to the and less than 30-m-thick in the north. Assuming that the unit’s strati-
contact of the uEF and Clarens Formation (Fig. 6). graphic contacts are isochronous across the MKB, the ~30 Ma worth of
rock record of the Elliot Formation can imply that this unit: (1) was
5. Discussion deposited under low sediment preservation conditions (i.e., accumula-
tion rates of ~1.5-27.5 m/Ma), (2) contains cryptic stratigraphic gaps,
5.1. Detrital zircon U-Pb geochronology which are possibly long but rarer in the south, and short but more
abundant in the north of the basin, and (3) is separated from the un-
In continental rock successions that are lacking pyroclastics (i.e., derlying Carnian Molteno Formation by a regional stratigraphic gap of
primary volcaniclastic rocks), utilizing the U-Pb radioisotopic dates ~6–7 Ma. In foreland basins, low sediment preservation conditions are
from detrital zircons is an effective approach for inferring the maximum normally favourable for mature paleosol development (e.g., DeCelles,
depositional ages (MDAs) of strata (e.g., Dickinson and Gehrels, 2009; 2012; Miall, 2016), whereas regional unconformities are typically as-
Ramezani et al., 2011; Coutts et al., 2019; Rossignol et al., 2019). It sociated with flexural tectonics driven largely by mountain building
should be emphasized that the ages of the youngest detrital zircons in events (e.g., Catuneanu et al., 1998; Catuneanu, 2004; Bordy et al.,
any given sedimentary rock sample only provides a maximum con- 2004a, 2005). Moreover, the newly obtained dates also suggest that: (1)
straint on the depositional age; the sample will be younger than the the relative resolution of the rock record is higher in the uEF compared
MDA if the detrital zircons are recycled from older sediment sources. to the lEF (Fig. 1c); and (2) the overall resolution of the Elliot rock
The extent to which the youngest detrital zircons provide precise esti- record is modest to very low, especially in the northern part of the
mates of the true depositional age (TDA) depends on the availability of basin, where the thickness of the Elliot Formation is rarely >100 m.
syn-depositional zircons in the depositional system, typically supplied Our radioisotopic age estimates are an important critical test of
via windblown volcanic ash or from erosion of active volcanic sources. previous assessments based on relative chronostratigraphy.
In some cases, MDA calculations can be older than the TDA even in Nevertheless, they largely uphold those previous assertions, albeit with
volcaniclastic sediment due to the presence of antecrystic, epicrystic, more nuance. While the results currently lack the spatial resolution
and xenocrystic zircon that formed prior to the most recent eruption required to precisely identify the position of the TJB at the sampling
(Rossignol et al., 2019). sites and for the entire basin, both vertically and laterally, these new
Our detrital zircons were recovered from sandstones and tuffaceous results nonetheless support previous proposals that the TJB is close to
siltstones, and both lithologies show evidence of reworking. In this the geological transition from lEF to uEF facies, and lay the foundation
context, the extent to which detrital zircons provide meaningful age for quantifying the vertical distance between these key chrono- and
constraints depends on: (1) the extent to which autocrystic (syn-erup- lithostratigraphic boundaries (Fig. 3).
tive) zircon is present, and (2) the time lag between the zircon crys- Determining the position of the TJB relative to this major strati-
tallisation in the original magma chamber and final deposition in the graphic contact will require much denser sampling of radioisotopic ages
upper Stormberg Group (e.g., Spencer et al., 2016; Andersen et al., in multiple vertical sections throughout the basin. In addition to

13
 E.M. Bordy, et al.

14
Fig. 6. Vertebrate biostratigraphy of the Elliot and Clarens formations (upper Stormberg Group) in the MKB. Stratigraphic positions of tetrapod taxa are shown using relative position notation (%) to account for the
variable thicknesses of the lower and upper Elliot and Clarens formations. Where poor provenance data is available, the stratigraphic distribution is shaded and identified with a question mark.
Earth-Science Reviews 203 (2020) 103120
 E.M. Bordy, et al.

15
Fig. 7. Distribution maps of the dominant vertebrate genera in the lEF (lower four panels), and uEF and Clarens formations (upper four panels) in the MKB. Grey circles show the distribution of all known fossil taxa for
each unit. Grid system and scale shown for Massospondylus applies to all maps.
Earth-Science Reviews 203 (2020) 103120
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

Fig. 8. Proportional representation of the key vertebrate higher taxa in the lEF (left), and uEF and Clarens formations (right) in the MKB. These relative abundances
only show occurrences that are identified at generic level or lower. Silhouettes not to scale.

maximum depositional ages from juvenile detrital zircons, radioisotopic lakes, similar to those in the extensional Fundy (Canada) and Newark
dating of authigenic mineral phases (e.g., feldspars, phosphates, car- (USA) basins, are lacking; and (2) a potential volcanic ash fall gap in the
bonates, especially from the abundant pedogenic carbonates in the uEF) rocks due to either a) relative inactivity or dormancy of the regional
may provide useful minimum depositional ages (e.g., Wang et al., 1998; volcanic arcs in the Late Triassic to Early Jurassic or b) the site of
Tabor and Myers, 2015). Such high-density geochronological sampling continental deposition in the final stages of the MKB evolution was
campaign at selected sites could also assist in reliably answering key distal to major, contemporaneous volcanic eruption centres, i.e., the
geological questions, such as: (1) how much time is represented by the primary source of datable, syn-sedimentary zircon grains were located
unconformity currently thought to separate the lEF from the uEF; (2) in faraway sectors of the incipiently fragmenting Pangean super-
whether this stratigraphic gap is diachronous or isochronous across the continent (e.g., Pankhurst et al., 2006, 2014; Schiuma and Llambias,
basin (i.e., are there lateral changes in the size of the gap at regional 2008; Muravchik et al., 2011; Spikings et al., 2016).
and subregional scales); (3) what is the frequency and magnitude of Notwithstanding the lack of primary ash fall deposits (i.e., pyr-
smaller stratigraphic gaps in the lEF and uEF; and (4) what were the oclastics), the newly acquired dataset of the detrital zircons, especially
regional depositional dynamics in the late foreland basin history of the the older part of the age spectra in each sample, provides new avenues
MKB (i.e., differential sedimentation/preservation rates from site-to- for interpreting the difference in provenance characteristics of the lEF
site across the basin; residence time of zircons in the dynamic fluvial vs uEF, and thus the depositional history of the Elliot Formation in the
setting of the Elliot depositional episode). Establishing a higher re- MKB. By tapping into this dataset, a deeper understanding of the se-
solution age framework in the Elliot Formation could refine the de- dimentary recycling regime and detrital zircon budget of southern
positional history of this unit, placing its high lateral facies variability Africa in the Mesozoic may be achieved, especially if this is thought-
into context. It would also facilitate judicious short-range, local corre- fully combined with documented (and newly measured) sediment
lation of strata (see Section 2.1, above). The latter, in turn, could also transport directions within the Stormberg Group (e.g., Bordy et al.,
assist with robust answers to key paleobiological questions, many of 2004c, 2004d, 2005). The most notable difference (Supplemental Table
which have global relevance (see Section 5.4, below). S1) between the detrital zircon populations of the lEF and uEF is that
Neoproterozoic grains (~35%) are the dominant single population in
the lEF, whereas Palaeozoic grains are the dominant single population
5.2.2. Sediment provenance considerations
in the uEF (49%), and the single Clarens Formation sample included in
Primary volcanic tuff layers (i.e., pyroclastics) that are suitable for
this study (59%). The lEF detrital zircon signature is expected from the
robust chronostratigraphic age determinations remain elusive in the
recycling of rocks in the south from the Cape Fold Belt as well as the
studied part of the MKB upper Stormberg succession. Such non-recycled
lower Karoo rocks (Dwyka and Ecca groups; Fig. 1). The uEF detrital
air-fall tuffs are also commonly lacking in other Upper Triassic–Lower
zircon signature, sourced from the west and south and dominated by
Jurassic continental deposits worldwide (e.g., Lucas, 2018). In the
Palaeozoic grains, suggests a proximal source area of mostly pre-
MKB, we tentatively attribute the lack of primary volcaniclastics to a
Stormberg Karoo and Cape Fold Belt rocks (Fig. 1), with little or no
combination of geological processes: (1) sediment recycling and other
input from older Precambrian terranes (e.g., western Kaapvaal Craton).
dynamics of continental sedimentation in foreland basins, where deep

16
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

5.3. New radioisotopic dates and implications for magnetostratigraphy current maximum depositional ages for the uEF do not improve upon
previous evidence from bio- and magnetostratigraphy. Sciscio et al.
Using paleomagnetism and the close phylogenetic affinities of sev- (2017a) considered it plausible that the TJB may lie within the lower
eral prosauropod dinosaurs, Sciscio et al. (2017a) correlated the low- uEF, either within the normal polarity chron EF6.2n or higher in the
ermost lEF to the Los Colorados Formation (Kent et al., 2014). Given Elliot magnetostratigraphic composite sequence (i.e., within EF7n;
the detrital zircon ages herein (Figs. 2, 4 and 5), the basal magnetozone Fig. 2). The Upper Triassic-Lower Jurassic Moenave Formation was
EF1r in the lEF (Fig. 2) can now be more firmly tied to the LC5r mag- previously correlated with the Elliot EF4–EF8 magnetozones based on
netozone in the Los Colorados Formation. Maximum depositional ages bio- and magnetostratigraphy, and suggests an Early Jurassic age for
from the lowermost lEF, specifically samples PHU, HB-15, SUB the EF7 magnetozone (EF7 = M3 magnetozone; Donohoo-Hurley et al.,
(<219.6, <218.2, and <216.4, respectively; Figs. 2, 4 and 5), increase 2010; Whiteside et al., 2011; Kirkland et al., 2014; Martz and Parker,
the credibility of tying the lEF (EF1r) to the Los Colorados Formation 2017). Magnetozones EF5–EF6 have not been sampled for detrital
(LC5r). By extension, these new maximum depositional ages further zircon-bearing samples in this study, and therefore the radioscopic age
support correlation to the radiometrically dated upper Blue Mesa and of this interval remains uncertain. However, because of the occurrence
lower Sonsela members in the Chinle Formation in the USA (Ramezani of Protosuchus in the lower-middle uEF (Dollman et al., 2019) shared
et al., 2011, 2014; Kent et al., 2018), although recycling of older det- with the lower Dinosaur Canyon Member (Moenave Formation, USA;
rital zircons in these members has been reported by Gehrels et al. Suarez et al., 2017) and the Rhaetian McCoy Brook Formation (Nova
(2019) and Kent et al. (2019). The correlation does still serve to reduce Scotia, Canada) could suggest that magnetozones EF5–EF6 may be
the uncertainty of the most likely basal age of the lEF to within ~3 latest Triassic in age, could straddle the ETE, and be a reflection of the
million years, and places it into the early middle Norian (i.e., within the Apachean holochron (Martz and Parker, 2017).Thus, the first lower
Adamanian holochron, which has an estimated age range between Jurassic rocks of the Elliot Formation may be, potentially, within the
~224–215 Ma in the USA; Martz and Parker, 2017). Correlation of the EF7 magnetozone, and therefore above the lEF/uEF boundary. How-
Newark-Hartford APTS E11r–E12r, lower Chinle PF7r–PF6r and asso- ever, this correlation is only weakly supported by a single uEF sample
ciated relative ages (~221.47 and 219.29 Ma, respectively; Kent et al., (MAF) with a maximum depositional age of <201 Ma (Figs. 2, 4 and 5).
2017, 2019) with the magnetozone LC5 of the Los Colorados Formation The pattern and number of the polarity pairs in the uppermost uEF
(Kent et al., 2014) and therefore EF1r of the lEF complements the above (EF7, EF8 and EF9n; Fig. 2) are not distinctive enough for firm mag-
statements. netostratigraphic correlations to other basins. Based on biostratigraphic
Biostratigraphically, magnetozones EF2 and EF3 of the middle lEF argument alone, Sciscio et al. (2017a) tied these magnetochrons to the
(Fig. 2) can be correlated to the upper third of the Los Colorados For- Hartford H24r–H27 and the St Audrie’s Bay/East Quantoxhead com-
mation (~La Esquina assemblage), with magnetozone EF2n tied to posite AQ1r–AQ3r providing a Hettangian-Sinemurian age for the
LC8n (Sciscio et al., 2017a). This, in turn, links EF2 to Newark APTS upper half of the uEF. The new maximum depositional ages in this study
E15/E16 at ~213–212 Ma (Kent et al., 2017, 2018). Furthermore, do not assist in validating this correlation, because of the limited
correlation of the upper Chinle Formation chronostratigraphy number of productive detrital zircon-bearing samples in the uEF (e.g.,
(Ramezani et al., 2011, 2014; Gehrels et al., 2019; Kent et al., 2018, Q6: <191.9 Ma; LEP: <197.3 Ma, LMO: <199.9 Ma, MAF: <201 Ma;
2019) to the middle lEF and the upper Los Colorados Formation suggest Figs. 2, 4 and 5).
that the EF2 may be correlated with upper Sonsela and lower Petrified
Forest members. This is corroborated by the middle lEF maximum de- 5.4. New radioisotopic dates and implications for biostratigraphy
positional ages, which range from <215.4 to <211.5 Ma (samples
MAP, GV-14 and QSS1 Figs. 2, 4 and 5). In contextualising the EF2 5.4.1. Biostratigraphic discussion
magnetozone, EF3 is likely younger than ~210 Ma, and the lEF max- The striking differences in tetrapod diversity between the uEF and
imum depositional ages suggest an age range between <209.6 and lEF have long been noted but have been relatively understudied. The
<207 Ma (samples HAF, LGT, LK-17) for EF3. Thus, a more meaningful geospatial studies undertaken here show that species-area effects (Close
approximation can be made between the magnetostratigraphic and et al., 2017) are a poor explanatory model for this difference in di-
depositional ages of the Petrified Forest/Painted Desert and Owl Rock versity, as the geographic spread of lEF and uEF localities is nearly
members (particularly PF2r, PF1r: 210.08±0.22 Ma -<207.8 Ma; identical (Fig. 7). Sampling intensity may provide a partial explanation,
Ramezani et al., 2011, 2014; Kent et al., 2018, 2019) relative to the lEF because the uEF has many more collections records than the lEF. Our
EF2 and EF3 magnetozones. While the maximum depositional ages database currently contains 128 generically-determined records of lEF
across this lEF interval improve and support previous assertions of vertebrate body fossils, of which 42 have precise stratigraphic occur-
Sciscio et al. (2017a), they do not firmly constrain this interval beyond rence data, compared to 618 from the uEF (131 with precise strati-
<210–~205 Ma, which approximates the age of the Revueltian holo- graphic data). In the case of sauropodomorphs, a number of well-pro-
chron estimated by Martz and Parker (2017). venanced records in the lEF are confounded by taxonomic uncertainty,
Elliot Formation magnetochrons EF4 –EF9n (Fig. 2) are more diffi- and continued revision of lEF taxa will undoubtedly improve our
cult to correlate globally based on: (i) poorly constrained taxon range knowledge of the stratigraphic distributions of those dinosaurs. How-
zone(s), (ii) lack of shared and/or diagnostic taxa with global dis- ever, a sufficient sample exists in the lEF to infer that theropod and
tributions, and (iii) limited radioisotopic ages (see Section 5.2). The ornithischian dinosaurs, as well as crocodylomorphs, later-branching
interval of the lEF represented by the EF5–EF6.2n magnetozones cynodonts, and turtles were either rare components of Norian-Rhaetian
yielded one datable sample (BH-15) with a maximum depositional age ecosystems in the MKB or absent. This contrasts with the presence and
of <204.9 Ma (Figs. 2, 4 and 5). Based on this age and the magne- abundance of these groups in the uEF, and is suggestive of a marked
tostratigraphic correlation of Sciscio et al. (2017a), it is possible that faunal change during the Triassic–Jurassic transition.
lEF EF4–EF6.1r magnetozones straddle the Norian-Rhaetian boundary Our chronostratigraphic data show that the thicker lEF represents a
at ~205.7 Ma (Wotzlaw et al., 2014; Maron et al., 2015), and the longer period of time than the thinner uEF (Fig. 1, 2 and 4; see Section
Newark magnetozones E20r.2r, E21n and E21r.1r (and their relative 5.1), meaning that it had longer periods of non-deposition or more
ages at ~206.03 Ma, 204.65 Ma, ~204.12 Ma, respectively; Kent et al., frequent/longer periods of erosion. This likely has the effect of com-
2017). It should be noted that the lEF sample BH-15 is from a single pressing the vertical range of lEF taxa, confounding our efforts to es-
locality in the northern part of the basin, and the Newark ages are in- tablish meaningful biozonation. However, recent biostratigraphic in-
ferred from astrochronology cyclostratigraphy. vestigations (Viglietti et al., 2020a) suggest that lEF taxa, particularly
In refining the placement of the TJB in the Elliot Formation, the Scalenodontoides, have a fairly wide stratigraphic distribution at least in

17
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

the lower parts of the lEF, especially in the southern part of the basin dinosaurs from anywhere in the world (Kammerer, 2018). This occur-
(Fig. 7). This may reflect heterogeneity in depositional rates or in ba- rence provides additional evidence that placeriine dicynodonts had a
sinal subsidence patterns, or possibly an abnormally long duration of Pangean distribution (Kammerer, 2018), contrary to earlier paleobio-
that genus. Targeted future collection efforts are necessary to test these geographic hypotheses (Kammerer et al., 2013).
hypotheses. In addition, the strong thickness variation of the lEF (see
Section 5.1; Bordy et al., 2004a, 2004b, 2004d; Bordy and Eriksson, 5.4.3. Implications for the early evolution of Dinosauria
2015) may indicate that significant parts of the section are missing in Revisions to the stratigraphy and age of key upper Stormberg Group
different portions of the basin, providing fewer options to sample the vertebrate localities have major implications for our understanding of
entire lEF record for its tetrapod diversity. early dinosaur evolution, and particularly for understanding the di-
versification of Ornithischia. The early ornithischian dinosaur Eocursor
5.4.2. Taxon sampling in the Norian–Rhaetian lEF was originally reported as from the lEF and thus inferred to be of Late
Our new age assessment of the Elliot and Clarens formations pro- Triassic (Norian) age (Butler et al., 2007). Late Triassic ornithischian
vides more robust temporal correlations with increasingly well-dated material has been considered to be very rare (Irmis et al., 2007) and this
vertebrate faunas in the Western Hemisphere (e.g., Langer et al., 2018). taxon was therefore important in calibrating divergence times within
In particular, we can more confidently state that the lEF overlaps in ornithischian phylogeny and in elucidating the nature of the early or-
time with the well-studied faunas of the Chinle and Los Colorados nithischian bauplan. However, more recent work at the Eocursor type
formations, potentially including those from famous localities like the locality demonstrated that this material pertains to the uEF and is more
Placerias and Hayden Quarries and the La Esquina fauna. likely earliest Jurassic in age (Olsen et al., 2010; McPhee et al., 2017).
Given that we can now confirm a Late Triassic (Norian–Rhaetian) This revision, alongside stratigraphic and taxonomic revisions of other
age for the lEF, it is surprising that this unit has not yielded body fossils purported Late Triassic ornithischians, has now removed all known
of the many diverse terrestrial tetrapod taxa present in other Late evidence for Late Triassic ornithischians from the global record (e.g.
Triassic faunas worldwide. Representatives of small-bodied lineages Irmis et al., 2007; Olsen et al., 2010; Irmis, 2011; Agnolín and
that are known from Late Triassic deposits globally are absent, in- Rozadilla, 2018; Baron et al., 2017b; Baron, 2019). This leaves an ex-
cluding mammaliaforms, rhynchocephalians, drepanosauromorphs, tensive ghost lineage between the earliest confirmed ornithischians
recumbirostrans, and pterosaurs. These absences are most likely ex- from the earliest Jurassic and their Late Triassic sister-group (either
plained by taphonomic or collector biases, which are currently un- Theropoda or Saurischia; compare phylogenies of Baron et al., 2017b,
characterised and unexplained. However, it is unlikely that small- 2017c; Langer et al., 2017), which suggests that their early history took
bodied vertebrates were actually absent from the fauna. Absences of place in a currently unsampled area. Alternatively, this ‘ornithischian
larger-bodied groups that are widespread in other Late Triassic as- gap’ might represent a genuine absence that could indicate a relatively
semblages are more likely to reflect genuine absence, or rarity, in- late derivation of ornithischians from within Theropoda (Padian, 2013;
cluding those of phytosaurs and several pseudosuchian lineages (such Baron, 2019), although this idea has not gained wide acceptance.
as poposauroids; see review in Knoll, 2004). Nevertheless, the uEF currently provides the best available window on
In the case of the aquatic phytosaurs, it had been suggested that early ornithischian evolution, possessing not only a diversity of taxa
their absence from southern Africa might be due to a clade-specific from across the tree (Abrictosaurus, Eocursor, Heterodontosaurus, Le-
paleotropical preference, as all known localities were situated between sothosaurus, Lycorhinus), but an abundance of material for functional
~30°N and ~30°S (Olsen and Galton, 1984; Shubin and Sues, 1991). and paleoecological, as well as phylogenetic, analysis (e.g., Butler,
However, discoveries of phytosaur material in the Baltic region of 2005, 2010; Butler et al., 2007, 2008; Knoll, 2008; Knoll et al., 2010;
Europe (approximately 45° N during the Late Triassic: Brusatte et al., Maidment and Barrett, 2011; Porro et al., 2010, 2015; Norman et al.,
2013) and in the Tashinga Formation of the Mid-Zambezi Basin, Zim- 2011; Sereno, 2012; Galton, 2014; Barrett et al., 2016; Baron et al.,
babwe (paleolatitude of approximately 40°S: Barrett et al., 2020) de- 2017b, 2017c; Sciscio et al., 2017c). The presence of a diverse or-
monstrate conclusively that they were not constrained to the paleo- nithischian fauna in the uEF coincides with their widespread appear-
tropics. Although phytosaurs would have been confined to regions with ance in the global fossil record at this time (e.g., Irmis, 2011; Barrett
perennial lakes or rivers, placing limits on their geographic range et al., 2014; Raven et al., 2019; earliest Jurassic of China, Venezuela,
(Buffetaut, 1993), lEF paleoenvironments were seasonally wet and the UK, and USA) and clear evidence of phenotypic divergence in body
supported large, permanent rivers (e.g., Smith et al., 1993; Bordy et al., size, stance, and dental morphology (Benson et al., 2014; Benson,
2004b, 2004d), so their absence from the MKB remains unexplained. 2018). Taken together, these observations demonstrate that or-
Interestingly, phytosaurs are also absent from the La Esquina fauna of nithischians radiated rapidly in the wake of the ETME.
the Los Colorados Formation. Our results also enhance the understanding of the early evolution of
A general explanation for the scarcity of pseudosuchians in the sauropodomorph dinosaurs. Abundant fossil discoveries from the Elliot
Elliot Formation might relate to latitudinal temperature gradients. This Formation have been central to the development of knowledge on early
group, which includes extant crocodylians, has well-defined pattern of sauropodomorph evolution, and complement the rich record found
high abundance and diversity at low latitudes, and low abundance or elsewhere, particularly that from South America. Sauropodomorphs are
absence at high latitudes, and has done since its origin in the Triassic by far the dominant vertebrate taxa in our collection records for the
(Markwick, 1998; Mannion et al., 2015). The MKB represents a rela- Norian–Rhaetian lEF (Fig. 8), echoing their relative abundances in the
tively high paleolatitude assemblage at approximately 50° south, and penecontemporaneous and latitudinally nearly equivalent Los Color-
this potentially explains both the relatively low abundance of pseudo- ados Formation of South America (Martínez et al., 2015). During this
suchians in the lEF, and the under-representation of pseudosuchian time, sauropodomorphs underwent key morphological transitions, in-
clades (other than crocodylomorphs). Nevertheless, this does not pro- cluding the evolution of giant body size (Apaldetti et al., 2018) and the
vide a good explanation of the abundance of crocodylomorphs in the first instances of quadrupedalism (McPhee et al., 2018) and much of
uEF. our understanding of these transitions results from the systematic study
Until recently, dicynodont synapsids were considered to be absent of the uEF body fossil record (e.g., Huxley, 1867; Haughton, 1924;
from the lEF fauna (although possible trackways had been reported: Galton and Van Heerden, 1985; Yates and Kitching, 2003; Bonnan and
Ellenberger, 1955, 1970, 1972; Ellenberger and Ellenberger, 1958; Yates, 2007; Yates, 2007b; Yates and Barrett, 2010; McPhee et al.,
Bordy et al., 2017b), but their presence has now been demonstrated 2015a, 2015b, 2017, 2018). Body fossil specimens of sauropodomorphs
with the description of Pentasaurus, which also records the first known from the lEF are generally less complete, and have yielded fewer direct
association of large-bodied dicynodonts with sauropodomorph insights into sauropodomorph evolution. For example, although limb

18
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

proportions within the problematic Plateosauravus syntype (Haughton, number of analysed samples and applied radioisotopic methods, which
1924; Yates, 2003; McPhee et al., 2017) are suggestive of a quadrupedal were dictated by the rock types suitable for age dating and resources
locomotor habit for that taxon, unequivocally quadrupedal saur- available to us as a group of multidisciplinary researchers, mostly based
opodomorphs have yet to be identified from the lEF—despite their in the Global South. Nonetheless, our basin-wide study of the upper
presence in the Norian deposits of South America (Riojasaurus, Stormberg Group gives a clear exposition of the status quo, whilst it also
Bonaparte, 1969; Lessemsaurus and Ingentia, Apaldetti et al., 2018). substantially advances knowledge on the age, litho-, magneto- and
Nonetheless, quadrupedal trackways (i.e., Paratetrasauropus seakensis, biostratigraphy of this important unit. Future efforts must, therefore,
Sauropodopus antiquus, ?Lavinipes jaquesi, Tetrasauropos unguiferus, focus on obtaining high-precision dates for stratigraphic sections that
Ellenberger, 1970, 1972; D’Orazi Porchetti and Nicosia, 2007) are traverse the lEF–uEF boundary, using an intensive, densely spaced
common in the lEF (Fig. 2). Some of these trackways are attributed to sampling regime and high-precision geochronological methods like CA-
quadrupedal sauropodomorphs, and are present within the first 15 m of ID-TIMS. Ideally, these focus areas should be in places with abundant
the lEF (e.g., at our PHU sample site), and continue throughout that vertebrate fossils.
section (for example, at our MAP and HAF sample sites with MDAs By expanding our chronostratigraphic research to potentially cor-
215.4 and 209.6 Ma, respectively; Figs. 4 and 5). Moreover, some relatable continental successions in southern Africa (e.g., the Mid-
quadrupedal trackways (e.g., ?Lavinipes jaquesi – D’Orazi Porchetti and Zambezi Basin), we will gain a better understanding of geographic
Nicosia, 2007 occur alongside the bipedal sauropodomorph trackway controls on Triassic–Jurassic biodiversity in southern Africa. Indeed,
Pseudotetrasauropous bipedoida, Ellenberger, 1970, 1972; D’Orazi work in that area is already underway (Viglietti et al., 2018b; Barrett
Porchetti and Nicosia, 2007) at our PHU sample site (Fig. 5), which has et al., 2020), but more comprehensive surveys, with integrated sedi-
a MDA of 219.6 Ma. Therefore, we confidently infer the presence of mentology, biostratigraphy, magnetostratigraphy, and absolute dating
quadrupedal sauropodomorphs in southern Africa by between ~220 methods, are needed of all upper Karoo units across southern Africa.
and ~216 Ma. The first known multi-tonne Elliot sauropodomorphs Finally, by integrating stable isotope geochemistry in this research
only appear in the lower part of the uEF (Ledumahadi; McPhee et al., (e.g., C-isotope excursions), we can potentially link major climatic
2018), postdating their first appearance in South America (Apaldetti changes with faunal differences on an absolute timescale – and in doing
et al., 2018). However, undescribed specimens of lEF sauropodomorphs so help set a world standard for understanding landscape changes and
in the South African collections record are of enormous size. Future basin development in southwestern Gondwana during the Late Triassic
research will likely show that both gigantic body size and quad- and Early Jurassic.
rupedalism appeared at similar times in sauropodomorphs of the Supplementary data to this article can be found online at https://
Norian of South America and southern Africa. doi.org/10.1016/j.earscirev.2020.103120.
Mass accumulations of vertebrate fossil material have great poten-
tial to improve our knowledge of lEF faunas, especially those in the Declaration of competing interest
middle Norian. To-date, two major bone beds have been documented
from the lower lEF, both occurring within 50 m of the lower contact of None.
the unit. The first one, containing abundant sauropodomorph remains,
was discovered in the 1950s at Maphutseng (Lesotho) (e.g., Ellenberger, Acknowledgements
1955; Ellenberger and Ginsburg, 1966), and is located <20 m below
our sample MAP (with an MDA of 215.4 Ma; Figs. 4 and 5). The second, We would like to extend our sincere gratitude to Akhil Rampersadh,
found <20 km south of the Maphutseng site, is a recent discovery made Maposholi Mokhethi, Howard Head, Robert Muir, Mhairi Reid, and
by the local community near Qhemegha, a village in Eastern Cape T’Nielle Haupt who assisted us during geological fieldwork. Staff at the
Province, South Africa and is a multi-taxic assemblage including di- U-(Th)-Pb Geochronology Division in the Central Analytical Facilities of
nosaurs, dicynodonts, cynodonts, and possibly other taxa (JNC, PMB, Stellenbosch University are acknowledged for guidance with detrital
RBJB, PAV, LS, EMB, unpublished results). Moreover, with one excep- zircon data processing and reduction. We thank Barry Shaulis for as-
tion, all other significant lEF sauropodomorph taxa that were prove- sistance with laser ablation at the University of Arkansas TRAIL facility.
nanced by McPhee et al. (2017; see their fig. 5) occur within the first Assistance with the GIS components of the research was provided by
~50 m of the lEF. Considering the MDAs of strata from adjacent beds in Michelle Clack and Simon Wills. Members of various field parties as-
the lEF, the two major lEF bone beds and other sauropodomorph- sisting with paleontological field work in South Africa, especially
bearing strata in the lowermost lEF appear to have formed in the middle Kathleen Dollman, Casey Staunton, Katherine Clayton, Simon Wills,
Norian (i.e., between ~216 and ~220 Ma), in the first 6–7 Ma after the Cory Dinter, Cebisa Mdekazi, Gilbert Mokgethoa, James Neenan, Rick
end of the Carnian (Fig. 2). Tolchard, Wilfred Bilankulu and Matt Baron, are also acknowledged.
Bernhard Zipfel, Sifelani Jirah, Zaituna Skosan, Claire Browning, Elize
6. Concluding remarks Butler, Jennifer Botha-Brink, Rose Prevec, William J. de Klerk, Heidi
Fourie provided access to specimens and specimen records in their care
The increasing sophistication (i.e., accuracy, precision) and afford- in South Africa. We also thank staff at the London Natural History
ability of modern dating techniques, combined with more systematic Museum, and Muséum National d’Histoire Naturelle Paris, and
and higher density sampling efforts, will profoundly improve our un- Naturhistorisches Museum Wien Vienna for assistance with the re-
derstanding of the geological history of the upper Stormberg Group, a spective specimen records.
key Mesozoic terrestrial succession in the upper Karoo Supergroup of
South Africa. In turn, this will increase its utility for decoding not only Funding
southern Pangean paleogeographic and ecological changes during the
final stages of MKB development, but will also enhance its impact on We gratefully acknowledge the following funding sources: National
studies of the ETME and ensuing global biodiversity changes. Research Foundation (NRF) of South Africa [Competitive Programme
Our study, which contains the first-ever and long-overdue radio- for Rated Researches (CPRR); African Origins Programme (AOP); re-
isotopic assessment of the upper Stormberg Group, provides a solid search funding 93544, 113394, 98825 to
initial framework for these future efforts. However, some important EMB; 98906, 98800 and 118794 to JNC; postgraduate funding to MA,
questions remain unanswered at this time, such as the length of the LS, ZR]; Department of Science and Technology (DST) and NRF of
depositional hiatus between the lEF and the uEF and the stratigraphic South Africa [Centre of Excellence in Palaeosciences (COE PAL); op-
position of the TJB and ETME. This is largely due to our relatively low erational support funding in 2015, 2019 to EB; 2015–2018 to JNC;

19
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

postgraduate and postdoctoral funding to MA, LS, PAV, BWM, KEJC diagnosticus (Dinosauria: Ornithischia) from the Lower Jurassic of southern Africa:
(2015–2019)]; National Research Foundation (NRF) of South Africa implications for basal ornithischian taxonomy and systematics. Zool. J. Linnean Soc.
179 (1), 125–168. https://doi.org/10.1111/zoj.12434.
[Incentive funding; grants to EB and JNC from 2015–2018]; Baron, M.G., Norman, D.B., Barrett, P.M., 2017b. A new hypothesis of dinosaur re-
Palaeontological Scientific Trust (PAST) of South Africa [research lationships and early dinosaur evolution. Nature 543 (7646), 501. https://doi.org/
funding to JNC from 2013–2015, 2017; postgraduate funding to ZR 10.1038/nature21700.
Baron, M.G., Norman, D.B., Barrett, P.M., 2017c. Untangling the dinosaur family tree
(2016–2018), BWM (2015), KEJC (2016-2019)]; National Science reply. Nature 551 (7678), E4–E5. https://doi.org/10.1038/nature24012.
Foundation (NSF) of the USA [Division of Earth Sciences (EAR) Barrett, P.M., Butler, R.J., Mundil, R., Scheyer, T.M., Irmis, R.B., Sánchez-Villagra, M.R.,
Sedimentary Geology and Paleobiology (SGP); research fund 2014. A palaeoequatorial ornithischian and new constraints on early dinosaur di-
versification. Proc. R. Soc. B Biol. Sci. 281 (1791), 20141147. https://doi.org/10.
1761576 to CAS and GRS]; Claude Leon Foundation of South Africa 1098/rspb.2014.1147.
[postdoctoral fellowship funding to LS (2019)]; Royal Society of Barrett, P.M., Butler, R.J., Yates, A.M., Baron, M.G., Choiniere, J.N., 2016. New speci-
London of UK [research funding to PMB]; Natural History Museum, mens of the basal ornithischian dinosaur Lesothosaurus diagnosticus Galton, 1978 from
the Early Jurassic of South Africa. Palaeontol. Afr. 50, 48–63.
London of UK [Earth Sciences Departmental Investment Fund; research
Barrett, P.M., Chapelle, K.E., Staunton, C.K., Botha, J., Choiniere, J.N., 2019. Postcranial
funding to PMB]; NERC Facilities of UK [research funding IP-1607- osteology of the neotype specimen of Massospondylus carinatus Owen, 1854
1115 to PMB]; National Geographic Society of the USA [research fund (Dinosauria: Sauropodomorpha) from the upper Elliot formation of South Africa.
CP-033R-2017 to JMC (PI), EMB and JNC]. Palaeontol. Afr. 53, 114–178.
Barrett, P.M., Sciscio, L., Viglietti, P.A., Broderick, T.J., Suarez, C.A., Sharman, G.A.,
The funders had no role in study design, data collection and ana- Jones, A.S., Munyikwa, D., Edwards, S.F., Chapelle, K.E.J., Dollman, K.N., Zondo, M.,
lysis, decision to publish, or preparation of the manuscript. Opinions Choiniere, J.N., 2020. The age of the Tashinga Formation (Karoo Supergroup) in the
expressed and conclusions arrived at are those of the authors and are Mid-Zambezi Basin, Zimbabwe and the first phytosaur from sub-Saharan Africa.
Gondwana Res. 81, 445–460.
not necessarily to be attributed to the CoE in Palaeosciences, NRF, PAST Benson, R.B.J., 2018. Dinosaur macroevolution and macroecology. Annu. Rev. Ecol. Evol.
or any other sponsor listed above. The authors have declared that no Syst. 49, 379–408.
competing interests exist. Benson, R.B.J., Campione, N.E., Carrano, M.T., Mannion, P.D., Sullivan, C., Upchurch, P.,
Evans, D.C., 2014. Rates of dinosaur body mass evolution indicate 170 million years
of sustained ecological innovation on the avian stem lineage. PLoS Biol. 12 (5)
Data availability e1001853.
Beukes, N.J., 1970. Stratigraphy and sedimentology of the Cave Sandstone stage, Karoo
System. In: Haughton, S.H. (Ed.), Proceedings 2nd IUGS Symposium on Gondwana
South African Heritage regulations stipulate that the biostrati- Stratigraphy and Palaeontology. CSIR, Pretoria, pp. 321–341.
graphic database must reside with a collections manager or curator. The Blewett, S.C., Phillips, D., 2016. An overview of Cape Fold Belt Geochronology: im-
database used in this contribution is available for all bonafide requests plications for sediment provenance and the timing of Orogenesis. In: Origin and
Evolution of the Cape Mountains and Karoo Basin. Springer, Verlag, pp. 45–55.
by emailing either [email protected] or sifelani.jirah@wit-
https://doi.org/10.1007/978-3-319-40859-0_5.
s.ac.za, or the current University Curator of Collections at the Blewett, S.C., Phillips, D., Matchan, E.L., 2019. Provenance of Cape Supergroup sediments
University of the Witwatersrand. and timing of Cape Fold Belt orogenesis: constraints from high-precision 40Ar/39Ar
All other datasets mentioned as Supplementary Materials in the text dating of muscovite. Gondwana Res. 70, 201–221. https://doi.org/10.1016/j.gr.
2019.01.009.
of this article can be found at doi:10.6084/m9.figshare.9730100, an Bonaparte, J.F., 1969. Dos nuevas “faunas” de reptiles Triásicos de Argentina.
open-source online data repository hosted at Figshare (Bordy et al., Ameghiniana 10 (1), 89–102.
2019). Bond, G., 1973. The palaeontology of Rhodesia. Rhod. Geol. Surv. Bull. 70, 1–121.
Bonnan, M.F., Yates, A.M., 2007. A new description of the forelimb of the basal saur-
[dataset] Bordy et al., Dating of the upper Stormberg Group: opodomorph Melanorosaurus: implications for the evolution of pronation, manus
Datasets and Supplementary Materials, 2019, Figshare, v1, https://dx. shape and quadrupedalism in sauropod dinosaurs. Spec. Pap. Palaeontol. 77,
doi.org/10.6084/m9.figshare.9730100 157–168.
Bordy, et al., 2019. Dating of the upper Stormberg Group: Datasets and Supplementary
Materials, Figshare v1. https://doi.org/10.6084/m9.figshare.9730100.
References Bordy, E.M., Eriksson, P., 2015. Lithostratigraphy of the Elliot Formation (Karoo
Supergroup), South Africa. S. Afr. J. Geol. 118 (3), 311–316. https://doi.org/10.
2113/gssajg.118.3.311.
Abdala, F., Gaetano, L.C., 2018. The Late Triassic record of cynodonts: time of innova-
Bordy, E.M., Head, H.V., 2018. Lithostratigraphy of the Clarens Formation (Stormberg
tions in the mammalian lineage. In: The Late Triassic World. Springer, Cham, pp.
Group, Karoo Supergroup), South Africa. S. Afr. J. Geol. 121 (1), 119–130.
407–445.
Bordy, E.M., Hancox, P.J., Rubidge, B.S., 2004a. Basin development during the deposition
Abdala, F., Damiani, R., Yates, A., Neveling, J., 2007. A non-mammaliaform cynodont
of the Elliot Formation (Late Triassic – Early Jurassic), Karoo Supergroup, South
from the Upper Triassic of South Africa: a therapsid Lazarus taxon. Palaeontol. Afr.
Africa. S. Afr. J. Geol. 107, 395–410.
42, 17–23.
Bordy, E.M., Hancox, P.J., Rubidge, B.S., 2004b. Fluvial style variations in the Late
Abrahams, M., Bordy, E.M., Sciscio, L., Knoll, F., 2017. Scampering, trotting, walking
Triassic – Early Jurassic Elliot Formation, main Karoo Basin, South Africa. J. Afr.
tridactyl bipedal dinosaurs in southern Africa: ichnological account of a Lower
Earth Sci. 38, 383–400.
Jurassic palaeosurface (upper Elliot Formation, Roma Valley) in Lesotho. Hist. Biol.
Bordy, E.M., Hancox, P.J., Rubidge, B.S., 2004c. Provenance Study of the Late Triassic –
29 (7), 958–975.
Early Jurassic Elliot Formation, main Karoo Basin, South Africa. S. Afr. J. Geol. 107,
Agnolín, F.L., Rozadilla, S., 2018. Phylogenetic reassessment of Pisanosaurus mertii
587–602.
Casamiquela, 1967, a basal dinosauriform from the Late Triassic of Argentina. J. Syst.
Bordy, E.M., Hancox, P.J., Rubidge, B.S., 2004d. A description of the sedimentology and
Palaeontol. 16 (10), 853–879.
palaeontology of the Late Triassic–Early Jurassic Elliot Formation in Lesotho.
Andersen, T., Elburg, M., Cawthorn-Blazeby, A., 2016. U–Pb and Lu–Hf zircon data in
Palaeontol. Afr. 40, 43–58.
young sediments reflect sedimentary recycling in eastern South Africa. J. Geol. Soc.
Bordy, E.M., Hancox, P.J., Rubidge, B.S., 2005. The contact of the Molteno and Elliot
173 (2), 337–351. https://doi.org/10.1144/jgs2015‐006.
formations through the main Karoo Basin, South Africa: a second-order sequence
Andersen, T., Elburg, M.A., Magwaza, B.N., 2019. Sources of bias in detrital zircon
boundary. S. Afr. J. Geol. 108, 349–362.
geochronology: discordance, concealed lead loss and common lead correction. Earth
Bordy, E.M., Sciscio, L., Abdala, F., McPhee, B.W., Choiniere, J.N., 2017a. First Lower
Sci. Rev. 197, 102899.
Jurassic vertebrate burrow from southern Africa (upper Elliot Formation, Karoo
Anderson, H.M., Anderson, J.M., 1970. A preliminary review of the biostratigraphy of the
Basin, South Africa). Palaeogeogr. Palaeoclimatol. Palaeoecol. 468, 362–372.
uppermost Permian, Triassic and lowermost Jurassic of Gondwanaland. Palaeontol.
Bordy, E.M., Abrahams, M., Sciscio, L., 2017b. The Subeng vertebrate tracks: strati-
Afr. 13, 1–22.
graphy, sedimentology and a digital archive of a historic Upper Triassic palaeosurface
Anderson, J.M., Anderson, H.M., Cruickshank, A.R., 1998. Late Triassic ecosystems of the
(lower Elliot Formation), Leribe, Lesotho (southern Africa). Boll. Soc. Paleontol. Ital.
Molteno/Lower Elliot biome of southern Africa. Palaeontology 41 (3), 387–421.
56 (2), 181–198. https://doi.org/10.4435/BSPI.2017.12.
Apaldetti, C., Martínez, R.N., Cerda, I.A., Pol, D., Alcober, O., 2018. An early trend to-
Bordy, E.M., Rampersadh, A., Abrahams, M., Lockley, M.G., Head, H.V., 2020. Tracking
wards gigantism in Triassic sauropodomorph dinosaurs. Nat. Ecol. Evol. 2,
the Pliensbachian–Toarcian Karoo firewalkers: trackways of quadruped and biped
1227–1232.
dinosaurs and mammaliaforms. PLoS One 15 (1). https://doi.org/10.1371/journal.
Attridge, J., 1963. The Upper Triassic Karroo deposits and fauna of southern Rhodesia. S.
pone.0226847. e0226847.
Afr. J. Sci. 59 (5), 242–247.
Bristowe, A., Raath, M.A., 2004. A juvenile coelophysoid skull from the Early Jurassic of
Baron, M.G., 2019. Pisanosaurus mertii and the Triassic ornithischian crisis: could phy-
Zimbabwe, and the synonymy of Coelophysis and Syntarsus (USA). Palaeontol. Afr. 40,
logeny offer a solution? Hist. Biol. 31, 967–981. https://doi.org/10.1080/08912963.
31–41.
2017.1410705.
Brusatte, S.L., Butler, R.J., Niedźwiedzki, G., Sulej, T., Bronowicz, R., Satkūnas, J., 2013.
Baron, M.G., Norman, D.B., Barrett, P.M., 2017a. Postcranial anatomy of Lesothosaurus
First record of Mesozoic terrestrial vertebrates from Lithuania: phytosaurs (Diapsida:

20
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

Archosauriformes) of probable Late Triassic age, with a review of phytosaur bio- analysis of gait variability and locomotor skills in Brasilichnium elusivum Leonardi,
geography. Geol. Mag. 150 (1), 110–122. with inferences on trackmaker identification. Palaeogeogr. Palaeoclimatol.
Buffetaut, E., 1993. Phytosaurs in time and space. Paleontol. Lombarda Nuova Ser. 2, Palaeoecol. 465, 14–29.
39–44. D'Orazi Porchetti, S., Bertini, R.J., Langer, M.C., 2018. Proposal for ichnotaxonomic al-
Butler, R.J., 2005. The ‘fabrosaurid’ornithischian dinosaurs of the upper Elliot Formation location of therapsid footprints from the Botucatu Formation (Brazil). Ichnos 25
(Lower Jurassic) of South Africa and Lesotho. Zool. J. Linnean Soc. 145 (2), 175–218. (2–3), 192–207.
Butler, R.J., 2010. The anatomy of the basal ornithischian dinosaur Eocursor parvus from Duane, M.J., Brown, R.W., 1992. Geochemical open-system behaviour related to fluid
the lower Elliot Formation (Late Triassic) of South Africa. Zool. J. Linnean Soc. 160 flow and metamorphism in the Karoo Basin. In: Inversion Tectonics of the Cape Fold
(4), 648–684. Belt, Karoo and Cretaceous Basins of Southern Africa. 127. AA Balkema, Rotterdam,
Butler, R.J., Smith, R.M., Norman, D.B., 2007. A primitive ornithischian dinosaur from The Netherlands, pp. 137.
the Late Triassic of South Africa, and the early evolution and diversification of Duncan, R.A., Hooper, P.R., Rehacek, J., Marsh, J., Duncan, A.R., 1997. The timing and
Ornithischia. Proc. R. Soc. B Biol. Sci. 274 (1621), 2041–2046. https://doi.org/10. duration of the Karoo igneous event, southern Gondwana. J. Geophys. Res. Solid
1098/rspb.2007.0367. Earth 102 (B8), 18127–18138. https://doi.org/10.1029/97JB00972.
Butler, R.J., Porro, L.B., Norman, D.B., 2008. A juvenile skull of the primitive or- Ellenberger, P., 1955. Note preliminaire sur les pistes et les restes osseux de vertebres du
nithischian dinosaur Heterodontosaurus tucki from the ‘Stormberg’ of southern Africa. Basutoland (Afrique-Du-Sud). C. R. Seances Acad. Sci. 240 (8), 889–891.
J. Vertebr. Paleontol. 28 (3), 702–711. https://doi.org/10.1671/0272-4634(2008) Ellenberger, P., 1970. The fossil-bearing strata associated with the earliest appearance of
28[702:AJSOTP]2.0.CO;2. mammals in South Africa and their ichnology: establishment of detailed stratigraphic
Catuneanu, O., 2004. Basement control on flexural profiles and the distribution of fore- zones in the Stormberg of Lesotho, South Africa, Upper Triassic to Jurassic. In:
land facies: the Dwyka Group of the Karoo Basin, South Africa. Geology 32 (6), Proceedings and Papers of the Second Gondwanaland Symposium on Gondwana
517–520. Stratigraphy and Palaeontology. Council for Scientific & Industrial Research,
Catuneanu, O., Hancox, P.J., Rubidge, B.S., 1998. Reciprocal flexural behaviour and Pretoria, pp. 343–370.
contrasting stratigraphies: a new basin development model for the Karoo retroarc Ellenberger, P., 1972. Contribution to the classification of Triassic vertebrate trackways:
foreland system, South Africa. Basin Res. 10 (4), 417–439. types found in the Stormberg Series of South Africa (I). In: Palaeo-vertebrata,
Chapelle, K.E., Choiniere, J.N., 2018. A revised cranial description of Massospondylus Memoire Extraordinaire, Montpellier, pp. 134.
carinatus Owen (Dinosauria: Sauropodomorpha) based on computed tomographic Ellenberger, P., 1974. Contribution à la classification des pistes de vertébrés du Trias: les
scans and a review of cranial characters for basal Sauropodomorpha. PeerJ 6, e4224. types du Stormberg d'Afrique du Sud (IIème partie: Le Stormberg superieur).
Chapelle, K.E.J., Barrett, P.M., Botha, J., Choiniere, J.N., 2019. Ngwevu intloko: a new Laboratoire de paléontologie des vertébrés. In: Palaeo–vertebrata, Memoire
early sauropodomorph dinosaur from the Lower Jurassic Elliot Formation of South Extraordinaire, Montpellier, pp. 142.
Africa and comments on cranial ontogeny in Massospondylus carinatus. PeerJ 7 (e7240 Ellenberger, F., Ellenberger, P., 1958. Main types of vertebrate footprints in the
https://doi.org/1.7717/peerj.7240). Stormberg beds of Basutoland (South Africa). (Preliminary note). CR Somm. Soc.
Choiniere, J.N., Barrett, P.M., 2015. A sauropodomorph dinosaur from the Early Jurassic Géol. Fr. 65–67.
of Lusitu, Zambia. Palaeontol. Afr. 49, 42–52. Ellenberger, F., Ginsburg, L., 1966. Le gisement de dinosauriens triasiques de Maphutseng
Citton, P., Díaz-Martínez, I., de Valais, S., Cónsole-Gonella, C., 2018. Triassic pentadactyl (Basutoland) et lorigine des sauropodes. C. R. Seances Acad. Sci. Ser. D 262 (4),
tracks from the Los Menucos Group (Río Negro province, Patagonia Argentina): 444–447.
possible constraints on the autopodial posture of Gondwanan trackmakers. PeerJ 6 Ellenberger, F., Ellenberger, P., Fabre, J., Ginsburg, L., Mendrez, C., 1964. The Stormberg
e5358. https://doi.org/10.7717/peerj.5358. Series of Basutoland (South Africa). In: Reports of the 22nd International Geological
Clark, J.M., 1986. Phylogenetic Relationships of the Crocodylomorph Archosaurs. [Ph. Congress. 9. pp. 320–330.
D. dissertation]. University of Chicago, Chicago, Illinois. Eriksson, P.G., 1986. Aeolian dune and alluvial fan deposits in the Clarens Formation of
Close, R.A., Benson, R.B.J., Upchurch, P., Butler, R.J., 2017. Controlling for the species- the Natal Drakensberg. Trans. Geol. Soc. S. Afr. 89, 389–394.
area effect supports constrained long-term Mesozoic terrestrial vertebrate diversifi- Evans, S.E., Kermack, K.A., 1994. Assemblages of small tetrapods from the Early Jurassic
cation. Nat. Commun. 8, 15381. of Britain. In: Fraser, N.C., Sues, H.-D. (Eds.), In the Shadow of the Dinosaurs: Early
Cohen, K.M., Finney, S.C., Gibbard, P.L., Fan, J.-X., 2013. International chronostrati- Mesozoic Tetrapods. Cambridge University Press, pp. 271–283.
graphic chart of the international commission on stratigraphy (v2018/08). Episodes Ezcurra, M.D., Brusatte, S.L., 2011. Taxonomic and phylogenetic reassessment of the
36, 199–204. (updated). http://www.stratigraphy.org/ICSchart/ early neotheropod dinosaur Camposaurus arizonensis from the Late Triassic of North
ChronostratChart2018-08.jpg. America. Palaeontology 54 (4), 763–772.
Colbert, E.H., 1989. The Triassic dinosaur Coelophysis. Mus. Nort. Ariz. 57, 1–160. Fraser, N.C., 1988. The osteology and relationships of Clevosaurus (Reptilia:
Cooper, M.R., 1981. The prosauropod dinosaur Massospondylus carinatus Owen from Sphenodontida). Philos. Trans. R. Soc. Lond. B Biol. Sci. 321 (1204), 125–178.
Zimbabwe: its biology, mode of life and phylogenetic significance. Occas. Pap. Nat. Gaffney, E.S., Kitching, J.W., 1994. The most ancient African turtle. Nature 369, 55–58.
Mus. Monum. Rhod. Ser. B Nat. Sci. 6 (10), 690–840. Galton, P.M., 1972. Classification and evolution of ornithopod dinosaurs. Nature 239
Corfu, F., 2013. A century of U-Pb geochronology: the long quest towards concordance. (5373), 464.
Bulletin 125 (1–2), 33–47. Galton, P.M., 1985. The poposaurid thecodontian Teratosaurus suevicus v. Meyer, plus
Coutts, D.S., Matthews, W.A., Hubbard, S.M., 2019. Assessment of widely used methods referred specimens mostly based on prosauropod dinosaurs, from the Middle
to derive depositional ages from detrital zircon populations. Geosci. Front. https:// Stubensandstein (Upper Triassic) of Nordwürttemberg. Stuttg. Beitr. Naturkunde Ser.
doi.org/10.1016/j.gsf.2018.11.002. B Geol. Paläontol. 116, 1–29.
Crompton, A.W., Charig, A.J., 1962. A new ornithischian from the Upper Triassic of South Galton, P.M., 2014. Notes on the postcranial anatomy of the heterodontosaurid dinosaur
Africa. Nature 196 (4859), 1074. Heterodontosaurus tucki, a basal ornithischian from the Lower Jurassic of South Africa.
Crompton, A.W., Ellenberger, F., 1957. On a new cynodont from the Molteno Beds and Rev. Paléobiol. 33 (1), 97–141.
the origin of the tritylodontids. Ann. S. Afr. Mus. 44 (1), 1–13. Galton, P.M., Van Heerden, J., 1985. Partial hindlimb of Blikanasaurus cromptoni n. gen.
Crompton, A.W., Jenkins, F.A., 1968. Molar occlusion in Late Triassic mammals. Biol. and n. sp., representing a new family of prosauropod dinosaurs from the Upper
Rev. 43 (4), 427–458. Triassic of South Africa. Geobios 18 (4), 509–516.
D’Orazi Porchetti, S., Nicosia, U., 2007. Re-examination of some large early Mesozoic Galton, P.M., Van Heerden, J., Yates, A.M., 2005. Postcranial anatomy of referred spe-
tetrapod footprints from the African collection of Paul Ellenberger. Ichnos 14, cimens of the sauropodomorph dinosaur Melanorosaurus from the Upper Triassic of
219–245. South Africa. In: Tidwell, V., Carpenter, K. (Eds.), Thunder-Lizards: The
de Fabrègues, C.P., Allain, R., 2016. New material and revision of Melanorosaurus tha- Sauropodomorph Dinosaurs. Indiana University Press, Bloomington, pp. 1–37.
banensis, a basal sauropodomorph from the Upper Triassic of Lesotho. PeerJ 4, e1639. Gauffre, F.-X., 1993. The most recent Melanorosauridae (Saurischia, Prosauropoda),
De Kock, M.O., 2003. Selected Magnetostratigraphic Studies in the Main Karoo Basin Lower Jurassic of Lesotho, with remarks on the prosauropod phylogeny. N. Jb. Geol.
(South Africa): Implications for Mass Extinction Events and the Supercontinent of Paläont. 11, 648–654.
Pangea. Masters dissertation. University of Johannesburg. Gehrels, G., Giesler, D., Olsen, P., Kent, D., Marsh, A., Parker, W., Rasmussen, C., Mundil,
De Kock, M.O., Kirschvink, J.L., 2004. Paleomagnetic constraints on the Permian-Triassic R., Irmis, R., Geissman, J., Lepre, C., 2019. LA-ICPMS U-Pb geochronology of detrital
boundary in terrestrial strata of the Karoo Supergroup, South Africa: implications for zircon grains from the Coconino, Moenkopi, and Chinle Formations in the Petrified
causes of the end-Permian extinction event. Gondwana Res. 7 (1), 175–183. Forest National Park (Arizona). Geochronol. Discuss. https://doi.org/10.5194/
DeCelles, P.G., 2012. Foreland basin systems revisited: Variations in response to tectonic gchron-2019-12. (in review).
settings. Tectonics Sediment. Basins Recent Adv. 405–426. Gierliński, G., Potemska, A., 1985. Protosuchus sp. z dolnej jury północnego obrzeżenia
Dickinson, W.R., Gehrels, G.E., 2009. Use of U–Pb ages of detrital zircons to infer max- Gór Świętokrzyskich. Prz. Geol. 33 (10), 567–570.
imum depositional ages of strata: a test against a Colorado Plateau Mesozoic data- Godefroit, P., Sigogneau-Russell, D., 1995. Cynodontes et Mammiferes primitifs du Trias
base. Earth Planet. Sci. Lett. 288 (1–2), 115–125. Superieur, en region Lorraine et Luxembourgeoise. Bull. Soc. Géol. 104, 9–21.
Dollman, K.N., Viglietti, P.A., Choiniere, J.N., 2019. A new specimen of Orthosuchus Gow, C.E., 1977. Fossil vertebrate studies in Rhodesia: sphenodontid remains from the
stormbergi (Nash 1968) and a review of the distribution of Southern African Lower Upper Trias of Rhodesia. Palaeontol. Afr. 20, 121–122.
Jurassic crocodylomorphs. Hist. Biol. 31 (5), 653–664. Gow, C.E., Hancox, P.J., 1993. First complete skull of the Late Triassic Scalenodontoides
Donohoo-Hurley, L.L., Geissman, J.W., Lucas, S.G., 2010. Magnetostratigraphy of the (Reptilia, Cynodontia) from southern Africa. N. M. Mus. Nat. Hist. Sci. Bull. 3,
uppermost Triassic and lowermost Jurassic Moenave Formation, western United 161–168.
States: correlation with strata in the United Kingdom, Morocco, Turkey, Italy, and Gow, C.E., Kitching, J.W., Raath, M.A., 1990. Skulls of the prosauropod dinosaur
eastern United States. Geol. Soc. Am. Bull. 122 (11–12), 2005–2019. Massospondylus carinatus Owen in the collections of the Bernard Price Institute for
D'Orazi Porchetti, S., Mocke, H.B., Latiano, M., Wagensommer, A., 2015. First record of Palaeontological Research. Palaeontol. Afr. 27, 45–58.
Otozoum from Namibia. Lethaia 48 (1), 72–82. Griffin, C.T., Nesbitt, S.J., 2019. Does the maximum body size of theropods increase
D'Orazi Porchetti, S., Bertini, R.J., Langer, M.C., 2017. Walking, running, hopping: across the Triassic–Jurassic boundary? Integrating ontogeny, phylogeny, and body

21
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

size. Anat. Rec. https://doi.org/10.1002/ar.24130. Knoll, F., Padian, K., de Ricqlès, A., 2010. Ontogenetic change and adult body size of the
Hälbich, I.W., Fitch, F.J., Miller, J.A., 1983. Dating the Cape orogeny. In: Sohnge, A.P.G., early ornithischian dinosaur Lesothosaurus diagnosticus: implications for basal or-
Hälbich, I.W. (Eds.), Geodynamics of the Cape Fold Belt. 12. pp. 149–164 Special nithischian taxonomy. Gondwana Res. 17 (1), 171–179.
Publication of the Geological Society of South Africa. Labandeira, C.C., Anderson, J.M., Anderson, H.M., 2018. Expansion of arthropod her-
Hansma, J., Tohver, E., Schrank, C., Jourdan, F., Adams, D., 2016. The timing of the Cape bivory in Late Triassic South Africa: the Molteno Biota, Aasvoëlberg 411 site and
Orogeny: new 40Ar/39Ar age constraints on deformation and cooling of the Cape developmental biology of a gall. In: The Late Triassic World. Springer, Cham, pp.
Fold Belt, South Africa. Gondwana Res. 32, 122–137. https://doi.org/10.1016/j.gr. 623–719.
2015.02.005. Lanci, L., Tohver, E., Wilson, A., Flint, S., 2013. Upper Permian magnetic stratigraphy of
Haughton, S.H., 1924. The fauna and stratigraphy of the Stormberg Series. Ann. S. Afr. the lower Beaufort Group, Karoo Basin. Earth Planet. Sci. Lett. 375, 123–134.
Mus. 12 (8), 323–497. Langer, M.C., Ezcurra, M.D., Rauhut, O.W., Benton, M.J., Knoll, F., McPhee, B.W., Novas,
Herrera-Flores, J.A., Stubbs, T.L., Elsler, A., Benton, M.J., 2018. Taxonomic reassessment F.E., Pol, D., Brusatte, S.L., 2017. Untangling the dinosaur family tree. Nature 551
of Clevosaurus latidens Fraser, 1993 (Lepidosauria, Rhynchocephalia) and rhyncho- (7678), E1.
cephalian phylogeny based on parsimony and Bayesian inference. J. Paleontol. 92 Langer, M.C., Ramezani, J., Da Rosa, Á.A., 2018. U-Pb age constraints on dinosaur rise
(4), 734–742. https://doi.org/10.1017/jpa.2017.136. from south Brazil. Gondwana Res. 57, 133–140.
Herriott, T.M., Crowley, J.L., Schmitz, M.D., Wartes, M.A., Gillis, R.J., 2019. Exploring Langereis, C.G., Krijgsman, W., Muttoni, G., Menning, M., 2010.
the law of detrital zircon: LA-ICP-MS and CA-TIMS geochronology of Jurassic forearc Magnetostratigraphy–concepts, definitions, and applications. Newsl. Stratigr. 43 (3),
strata, Cook Inlet, Alaska, USA. Geology 47 (11), 1044–1048. 207–233.
Hopson, J.A., 1984. Late Triassic traversodont cynodonts from Nova Scotia and southern Lockley, M.G., Gierlinski, G.D., 2006. Diverse vertebrate ichnofaunas containing
Africa. Palaeontol. Afr. 25, 181–201. Anomoepus and other unusual trace fossils from the Lower Jurassic of the western
Hsiou, A.S., De França, M.A.G., Ferigolo, J., 2015. New data on the Clevosaurus United States: implications for paleoecology palichnostratigraphy. The
(Sphenodontia: Clevosauridae) from the Upper Triassic of southern Brazil. PLoS One Triassic–Jurassic Terrestrial Transition. N. M. Mus. Nat. Hist. Sci. Bull. 37, 176–191.
10 (9). https://doi.org/10.1371/journal.pone.0137523. e0137523. Lockley, M.G., Meyer, C.A., dos Santos, V.F., 1996. Megalosauripus, Megalosauropus and
Hsiou, A.S., Nydam, R.L., Simões, T.R., Pretto, F.A., Onary, S., Martinelli, A.G., Liparini, the concept of megalosaur footprints. In: The Continental Jurassic: Symposium
A., Martínez, P.R.R.D.V., Soares, M.B., Schultz, C.L., Caldwell, M.W., 2019. A new Volume: Museum of Northern Arizona Bulletin. vol. 60. pp. 113–118.
clevosaurid from the triassic (carnian) of Brazil and the Rise of Sphenodontians in Lockley, M.G., Lucas, S.G., Gaston, R., Hunt, A.P., 2004. Ichnofaunas from the Triassic-
Gondwana. Sci. Rep. 9, 11821. Jurassic boundary sequences of the Gateway area, Western Colorado: implications for
Hunt, A.P., Lucas, S.G., Klein, H., 2018. Late Triassic nonmarine vertebrate and in- faunal composition and correlations with other areas. Ichnos 11, 89–102.
vertebrate trace fossils and the pattern of the Phanerozoic record of vertebrate trace Lockley, M.G., Lucas, S.G., Hunt, A.P., 2006. Eosauropus, a new name for a Late Triassic
fossils. In: The Late Triassic World. Springer, Cham, pp. 447–544. track: further observations on the Late Triassic ichnogenus Tetrasauropus and related
Hüsing, S.K., Beniest, A., van der Boon, A., Abels, H.A., Deenen, M.H.L., Ruhl, M., forms, with notes on the limits of interpretation. The Triassic–Jurassic Terrestrial
Krijgsman, W., 2014. Astronomically-calibrated magnetostratigraphy of the Lower Transition. N. M. Mus. Nat. Hist. Sci. Bull. 37, 192–198.
Jurassic marine successions at St. Audrie's Bay and East Quantoxhead Lucas, S.G., 1998. Global Triassic tetrapod biostratigraphy and biochronology.
(Hettangian–Sinemurian; Somerset, UK). Palaeogeogr. Palaeoclimatol. Palaeoecol. Palaeogeogr. Palaeoclimatol. Palaeoecol. 143 (4), 347–384.
403, 43–56. https://doi.org/10.1016/j.palaeo.2014.03.022. Lucas, S.G., 2007. Tetrapod footprint biostratigraphy and biochronology. Ichnos 14 (1–2),
Huxley, T.H., 1867. On the remains of large dinosaurian reptiles from the Stormberg 5–38.
Mountains, South Africa. Q. J. Geol. Soc. 23, 1–6. Lucas, S.G., 2018. The late triassic world: earth in a time of transition. In: Tanner, L.H.
Irmis, R.B., 2011. Evaluating hypotheses for the early diversification of dinosaurs. Earth (Ed.), The Late Triassic World: Earth in a Time of Transition. Springer International
Environ. Trans. R. Soc. Edinb. 101, 397–426. Publishing, Cham, pp. 1–25.
Irmis, R.B., Parker, W.G., Nesbitt, S.J., Liu, J., 2007. Early ornithischian dinosaurs: the Lucas, S.G., Hancox, P.J., 2001. Tetrapod-based correlation of the nonmarine Upper
Triassic record. Hist. Biol. 19 (1), 3–22. Triassic of southern Africa. Albertiana 25, 5–9.
Johnstone, S.A., Schwartz, T.M., Holm-Denoma, C.S., 2019. A stratigraphic approach to Luo, Z.X., Wu, X.C., Fraser, N.C., Sues, H.-D., 1994. The small tetrapods of the lower
inferring depositional ages from detrital geochronology data. Front. Earth Sci. 7, 57. Lufeng Formation, Yunnan, China. In: The Shadow of the Dinosaurs: Early Mesozoic
https://doi.org/10.3389/feart.2019.00057. Tetrapods. Cambridge University Press, pp. 251–270.
Kammerer, C.F., 2018. The first skeletal evidence of a dicynodont from the lower Elliot Maidment, S.C., Barrett, P.M., 2011. The locomotor musculature of basal ornithischian
Formation of South Africa. Palaeontol. Afr. 52, 102–128. dinosaurs. J. Vertebr. Paleontol. 31 (6), 1265–1291.
Kammerer, C.F., Fröbisch, J., Angielczyk, K.D., 2013. On the validity and phylogenetic Mannion, P.D., Benson, R.B.J., Carrano, M.T., Tennant, J.P., Judd, J., Butler, R.J., 2015.
position of Eubrachiosaurus browni, a kannemeyeriiform dicynodont (Anomodontia) Climate constrains the evolutionary history and biodiversity of crocodilians. Nat.
from Triassic North America. PLoS One 8 (5). https://doi.org/10.1371/journal.pone. Commun. 6, 8438.
0064203. e64203. Markwick, P.J., 1998. Crocodilian diversity in space and time: the role of climate in
Kent, D.V., Malnis, P.S., Colombi, C.E., Alcober, O.A., Martínez, R.N., 2014. Age con- paleoecology and its implication for understanding K/T extinctions. Paleobiology 24,
straints on the dispersal of dinosaurs in the Late Triassic from magnetochronology of 470–497.
the Los Colorados Formation (Argentina). Proc. Natl. Acad. Sci. 111 (22), 7958–7963. Maron, M., Rigo, M., Bertinelli, A., Katz, M.E., Godfrey, L., Zaffani, M., Muttoni, G., 2015.
https://doi.org/10.1073/pnas.1402369111. Magnetostratigraphy, biostratigraphy, and chemostratigraphy of the Pignola-Abriola
Kent, D.V., Olsen, P.E., Muttoni, G., 2017. Astrochronostratigraphic polarity time scale section: new constraints for the Norian-Rhaetian boundary. Bulletin 127 (7-8),
(APTS) for the Late Triassic and Early Jurassic from continental sediments and cor- 962–974.
relation with standard marine stages. Earth Sci. Rev. 166, 153–180. https://doi.org/ Marsh, A.D., Rowe, T.B., 2018. Anatomy and systematics of the sauropodomorph
10.1016/j.earscirev.2016.12.014. Sarahsaurus aurifontanalis from the Early Jurassic Kayenta Formation. PLoS One 3
Kent, D.V., Olsen, P.E., Rasmussen, C., Lepre, C., Mundil, R., Irmis, R.B., Gehrels, G.E., (10). https://doi.org/10.1371/journal.pone.0204007. e0204007.
Giesler, D., Geissman, J.W., Parker, W.G., 2018. Empirical evidence for stability of Marsicano, C.A., Wilson, J.A., Smith, R.M., 2014. A temnospondyl trackway from the
the 405-kiloyear Jupiter–Venus eccentricity cycle over hundreds of millions of years. early mesozoic of western Gondwana and its implications for basal tetrapod loco-
Proc. Natl. Acad. Sci. 115 (24), 6153–6158. https://doi.org/10.1073/pnas. motion. PLoS One 9 (8) e103255.
1800891115. Martínez, R.N., Apaldetti, C., 2017. A Late Norian—Rhaetian Coelophysid Neotheropod
Kent, D.V., Olsen, P.E., Lepre, C., Rasmussen, C., Mundil, R., Gehrels, G.E., Giesler, D., (Dinosauria, Saurischia) from the Quebrada Del Barro Formation, Northwestern
Irmis, R.B., Geissman, J.W., Parker, W.G., 2019. Magnetochronology of the entire Argentina. Ameghiniana 54 (5), 488–506.
Chinle Formation (Norian age) in a scientific drill core from Petrified Forest National Martínez, R.N., Apaldetti, C., Correa, G., Colombi, C.E., Fernández, E., Santi Malnis, P.,
Park (Arizona, USA) and implications for regional and global correlations in the Late Praderio, A., Abelín, D., Benegas, L.G., Aguilar-Cameo, A., Alcober, O.A., 2015. A
Triassic. Geochem. Geophys. Geosyst. 20, 4654–4664. https://doi.org/10.1029/ new Late Triassic vertebrate assemblage from northwestern Argentina. Ameghiniana
2019GC008474. 52, 379–390.
Kirkland, J.I., Milner, A.R.C., Olsen, P.E., Hargrave, J.E., 2014. The Whitmore Point Martz, J.W., Parker, W.G., 2017. Revised formulation of the Late Triassic Land Vertebrate
Member of the Moenave Formation in its type area in Northern Arizona and its age “Faunachrons” of western North America: recommendations for codifying nascent
and correlation with the section in St. George, Utah: evidence for two major lacus- systems of vertebrate biochronology. In: Zeigler, K.E., Parker, W. (Eds.), Terrestrial
trine sequences. Geol. Utah Far South Utah Geol. Assoc. Publ. 43, 321–356. Depositional Systems: Deciphering Complexities Through Multiple Stratigraphic
Kitching, J.W., Raath, M.A., 1984. Fossils from the Elliot and Clarens Formations (Karoo Methods. Elsevier, pp. 39–125. https://doi.org/10.1016/B978-0-12-803243-5.
Sequence) of the North-eastern Cape, Orange Free State and Lesotho, and a suggested 00002-9.
biozonation based on tetrapods. Palaeontol. Afr. 25, 111–125. McPhee, B.W., Choiniere, J.N., 2018. The osteology of Pulanesaura eocollum: implica-
Klein, H., Lucas, S.G., 2010. Tetrapod footprints-their use in biostratigraphy and bio- tions for the inclusivity of Sauropoda (Dinosauria). Zoological Journal of the Linnean
chronology of the Triassic. Geol. Soc. Lond., Spec. Publ. 334 (1), 419–446. Society 182 (4), 830–861.
Knoll, F., 2004. Review of the tetrapod fauna of the “Lower Stormberg Group” of the main McPhee, B.W., Yates, A.M., Choiniere, J.N., Abdala, F., 2014. The complete anatomy and
Karoo Basin (southern Africa): implication for the age of the Lower Elliot Formation. phylogenetic relationships of Antetonitrus ingenipes (Sauropodiformes, Dinosauria):
Bull. Soc. Géol. Fr. 175 (1), 73–83. implications for the origins of Sauropoda. Zool. J. Linnean Soc. 171 (1), 151–205.
Knoll, F., 2005. The tetrapod fauna of the Upper Elliot and Clarens formations in the main McPhee, B.W., Choiniere, J.N., Yates, A.M., Viglietti, P.A., 2015a. A second species of
Karoo Basin (South Africa and Lesotho). Bull. Soc. Géol. Fr. 176 (1), 81–91. Eucnemesaurus Van Hoepen, 1920 (Dinosauria, Sauropodomorpha): new information
Knoll, F., 2008. Buccal soft anatomy in Lesothosaurus (Dinosauria: Ornithischia). N. Jb. on the diversity and evolution of the sauropodomorph fauna of South Africa's lower
Geol. Paläont. (Abh.) 248 (3), 355–364. Elliot Formation (latest Triassic). J. Vertebr. Paleontol. 35 (5) p.e980504.
Knoll, F., 2010. A primitive sauropodomorph from the upper Elliot Formation of Lesotho. McPhee, B.W., Bonnan, M.F., Yates, A.M., Neveling, J., Choiniere, J.N., 2015b. A new
Geol. Mag. 147 (6), 814–829. basal sauropod from the pre-Toarcian Jurassic of South Africa: evidence of niche-

22
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

partitioning at the sauropodomorph–sauropod boundary? Sci. Rep. 5, 13224. Ramezani, J., Hoke, G.D., Fastovsky, D.E., Bowring, S.A., Therrien, F., Dworkin, S.I.,
McPhee, B.W., Bordy, E.M., Sciscio, L., Choiniere, J.N., 2017. The sauropodomorph Atchley, S.C., Nordt, L.C., 2011. High-precision U-Pb zircon geochronology of the
biostratigraphy of the Elliot Formation of southern Africa: tracking the evolution of Late Triassic Chinle Formation, Petrified Forest National Park (Arizona, USA): tem-
Sauropodomorpha across the Triassic–Jurassic boundary. Acta Palaeontol. Pol. 62 poral constraints on the early evolution of dinosaurs. Bulletin 123 (11-12),
(3), 441–465. https://doi.org/10.4202/app.00377.2017. 2142–2159.
McPhee, B.W., Benson, R.B., Botha-Brink, J., Bordy, E.M., Choiniere, J.N., 2018. A giant Ramezani, J., Fastovsky, D.E., Bowring, S.A., 2014. Revised chronostratigraphy of the
dinosaur from the earliest Jurassic of South Africa and the transition to quad- lower Chinle Formation strata in Arizona and New Mexico (USA): high-precision U-
rupedality in early sauropodomorphs. Curr. Biol. 28 (19), 3143–3151. Pb geochronological constraints on the Late Triassic evolution of dinosaurs. Am. J.
Miall, A.D., 2013. Sophisticated stratigraphy.The web of geological sciences: advances, Sci. 314 (6), 981–1008.
impacts and interactions. Geol. Soc. Am. Spec. Pap. 500, 169–190. Rampersadh, A., Bordy, E.M., Sciscio, L., Abrahams, M., 2018. Dinosaur behaviour in an
Miall, A.D., 2015. Updating uniformitarianism: stratigraphy as just a set of ‘frozen acci- Early Jurassic palaeoecosystem–uppermost Elliot Formation, Ha Nohana, Lesotho.
dents’ Geol. Soc. Lond., Spec. Publ. 404 (1), 11–36. Ann. Soc. Geol. Pol. 88 (2), 163–179.
Miall, A.D., 2016. The valuation of unconformities. Earth Sci. Rev. 163, 22–71. Raup, D.M., Sepkoski, J.J., 1982. Mass extinctions in the marine fossil record. Science 215
Moreau, M.G., Bucher, H., Bodergat, A.M., Guex, J., 2002. Pliensbachian magnetos- (4539), 1501–1503.
tratigraphy: new data from Paris Basin (France). Earth Planet. Sci. Lett. 203 (2), Raven, T.J., Barrett, P.M., Xu, X., Maidment, S.C.R., 2019. A reassessment of the pur-
755–767. ported ankylosaurian dinosaur Bienosaurus lufengensis from the Lower Lufeng
Moulin, M., Fluteau, F., Courtillot, V., Marsh, J., Delpech, G., Quidelleur, X., Gérard, M., Formation of Yunnan, People’s Republic of China. Acta Palaeontol. Pol. 64, 335–342.
Jay, A.E., 2011. An attempt to constrain the age, duration, and eruptive history of the https://doi.org/10.4202/app.00577.2018.
Karoo flood basalt: Naude's Nek section (South Africa). J. Geophys. Res. Solid Earth Ray, S., Chinsamy, A., 2002. A theropod tooth from the Late Triassic of southern Africa. J.
116 (B7). https://doi.org/10.1029/2011JB008210. Biosci. 27 (3), 295–298.
Moulin, M., Courtillot, V., Fluteau, F., Valet, J.P., 2012. The “van Zijl” Jurassic geo- Rayfield, E.J., Barrett, P.M., Milner, A.R., 2009. Utility and validity of Middle and Late
magnetic reversal revisited. Geochem. Geophys. Geosyst. 13 (3), Q03010. https:// Triassic ‘land vertebrate faunachrons’ J. Vertebr. Paleontol. 29 (1), 80–87.
doi.org/10.1029/2011GC003910. Rogers, R.R., Rogers, K.C., Munyikwa, D., Terry, R.C., Singer, B.S., 2004. Sedimentology
Moulin, M., Fluteau, F., Courtillot, V., Marsh, J., Delpech, G., Quidelleur, X., Gérard, M., and taphonomy of the upper Karoo-equivalent Mpandi Formation in the Tuli Basin of
2017. Eruptive history of the Karoo lava flows and their impact on early Jurassic Zimbabwe, with a new 40Ar/39Ar age for the Tuli basalts. J. Afr. Earth Sci. 40 (3-4),
environmental change. J. Geophys. Res. Solid Earth 122 (2), 738–772. 147–161.
Munyikwa, D., 1997. Faunal analysis of Karoo-aged sediments in the northern Limpopo Rossignol, C., Hallot, E., Bourquin, S., Poujol, M., Jolivet, M., Pellenard, P., Ducassou, C.,
Valley, Zimbabwe. Arnoldia Zimb. 10 (13), 129–139. Nalpas, T., Heilbronn, G., Yu, J., Dabard, M.P., 2019. Using volcaniclastic rocks to
Munyikwa, D., Raath, M.A., 1999. Further material of the ceratosaurian dinosaur constrain sedimentation ages: to what extent are volcanism and sedimentation syn-
Syntarsus from the Elliot Formation (Early Jurassic) of South Africa. Palaeontol. Afr. chronous? Sediment. Geol. 381, 46–64. https://doi.org/10.1016/j.sedgeo.2018.12.
35, 55–59. 010.
Muravchik, M., D'Elia, L., Bilmes, A., Franzese, J.R., 2011. Syn-eruptive/inter-eruptive Rowe, T., 1989. A new species of the theropod dinosaur Syntarsus from the Early Jurassic
relations in the syn-rift deposits of the Precuyano Cycle, Sierra de Chacaico, Neuquén Kayenta Formation of Arizona. J. Vertebr. Palaeontol. 9, 125–136.
Basin, Argentina. Sediment. Geol. 238 (1-2), 132–144. Rubidge, B.S., Erwin, D.H., Ramezani, J., Bowring, S.A., de Klerk, W.J., 2013. High-
Norman, D.B., Crompton, A.W., Butler, R.J., Porro, L.B., Charig, A.J., 2011. The Lower precision temporal calibration of Late Permian vertebrate biostratigraphy: U-Pb
Jurassic ornithischian dinosaur Heterodontosaurus tucki Crompton & Charig, 1962: zircon constraints from the Karoo Supergroup, South Africa. Geology 41 (3),
cranial anatomy, functional morphology, taxonomy, and relationships. Zool. J. 363–366.
Linnean Soc. 163 (1), 182–276. Schiuma, M., Llambias, E.J., 2008. New ages and chemical analysis on Lower Jurassic
Olsen, P.E., Galton, P.M., 1984. A review of the reptile and amphibian assemblages from volcanism close to the dorsal de Huincul, Neuquen. Rev. Asoc. Geol. Argent. 63,
the Stormberg of southern Africa, with special emphasis on the footprints and the age 644–652.
of the Stormberg. Palaeontol. Afr. 25, 87–110. Sciscio, L., 2016. Position of the Triassic-Jurassic Boundary in South Africa and Lesotho: A
Olsen, P.E., Sues, H.D., 1986. Correlation of continental Late Triassic and Early Jurassic Multidisciplinary Approach Aimed at Improving the Chronostratigraphy and
sediments, and patterns of the Triassic–Jurassic tetrapod transition. In: The Beginning Biostratigraphy of the Elliot Formation, Stormberg Group. Doctoral dissertation.
of the Age of Dinosaurs. Cambridge University Press, Cambridge, pp. 321–351. University of Cape Town. http://hdl.handle.net/11427/20847.
Olsen, P.E., Kent, D.V., Whiteside, J.H., 2010. Implications of the Newark Supergroup- Sciscio, L., Bordy, E.M., 2016. Palaeoclimatic conditions in the Late Triassic-Early
based astrochronology and geomagnetic polarity time scale (Newark-APTS) for the Jurassic of southern Africa: a geochemical assessment of the Elliot Formation. J. Afr.
tempo and mode of the early diversification of the Dinosauria. Earth Environ. Sci. Earth Sci. 119, 102–119.
Trans. R. Soc. Edinb. 101 (3-4), 201–229. https://doi.org/10.1017/ Sciscio, L., Bordy, E.M., Reid, M., Abrahams, M., 2016. Sedimentology and ichnology of
S1755691011020032. the Mafube dinosaur track site (Lower Jurassic, eastern Free State, South Africa): a
Opdyke, N.D., 1964. The paleomagnetism of some Triassic red beds from Northern report on footprint preservation and palaeoenvironment. PeerJ 4 p.e2285.
Rhodesia. J. Geophys. Res. 69 (12), 2495–2497. Sciscio, L., de Kock, M., Bordy, E., Knoll, F., 2017a. Magnetostratigraphy across the
Otero, A., Krupandan, E., Pol, D., Chinsamy, A., Choiniere, J.N., 2015. A new basal Triassic-Jurassic boundary in the main Karoo Basin. Gondwana Res. 51, 177–192.
sauropodiform from South Africa and the phylogenetic relationships of basal saur- https://doi.org/10.1016/j.gr.2017.07.009.
opodomorphs. Zool. J. Linnean Soc. 174, 589–634. Sciscio, L., Bordy, E.M., Abrahams, M., Knoll, F., McPhee, B.W., 2017b. The first mega-
Owen, R., 1854. Descriptive Catalogue of the Fossil Organic Remains of Reptilia and theropod tracks from the Lower Jurassic upper Elliot Formation, Karoo Basin,
Pisces Contained in the Museum of the Royal College of Surgeons of England. p. 184. Lesotho. PLoS One 12 (10), e0185941. https://doi.org/10.1371/journal.pone.
Taylor and Francis, London. 0185941.
Padian, K., 2013. The problem of dinosaur origins: integrating three approaches to the Sciscio, L., Knoll, F., Bordy, E.M., de Kock, M.O., Redelstorff, R., 2017c. Digital re-
rise of Dinosauria. Earth Environ. Sci. Trans. R. Soc. Edinb. 103 (3-4), 423–442. construction of the mandible of an adult Lesothosaurus diagnosticus with insight into
https://doi.org/10.1017/S1755691013000431. the tooth replacement process and diet. PeerJ 5 p.e3054.
Pankhurst, R.J., Rapela, C.W., Fanning, C.M., Márquez, M., 2006. Gondwanide con- Sell, B., Ovtcharova, M., Guex, J., Bartolini, A., Jourdan, F., Spangenberg, J.E., Vicente,
tinental collision and the origin of Patagonia. Earth Sci. Rev. 76 (3-4), 235–257. J.C., Schaltegger, U., 2014. Evaluating the temporal link between the Karoo LIP and
Pankhurst, R.J., Rapela, C.W., De Luchi, M.L., Rapalini, A.E., Fanning, C.M., Galindo, C., climatic–biologic events of the Toarcian Stage with high-precision U–Pb geochro-
2014. The Gondwana connections of northern Patagonia. J. Geol. Soc. 171 (3), nology. Earth Planet. Sci. Lett. 408, 48–56.
313–328. Sereno, P.C., 2012. Taxonomy, morphology, masticatory function and phylogeny of
Porro, L.B., Butler, R.J., Barrett, P.M., Moore-Fay, S., Abel, R.L., 2010. New hetero- heterodontosaurid dinosaurs. ZooKeys 226, 1–225.
dontosaurid specimens from the Lower Jurassic of southern Africa and the early Shubin, N.H., Sues, H.D., 1991. Biogeography of early Mesozoic continental tetrapods:
ornithischian dinosaur radiation. Earth Environ. Sci. Trans. R. Soc. Edinb. 101 (3-4), patterns and implications. Paleobiology 17 (3), 214–230.
351–366. https://doi.org/10.1017/S175569101102010X. Shubin, N.H., Crompton, A.W., Sues, H.D., Olsen, P.E., 1991. New fossil evidence on the
Porro, L.B., Witmer, L.M., Barrett, P.M., 2015. Digital preparation and osteology of the sister-group of mammals and early Mesozoic faunal distributions. Science 251
skull of Lesothosaurus diagnosticus (Ornithischia: Dinosauria). PeerJ 3 p.e1494. (4997), 1063–1065. https://doi.org/10.1126/science.251.4997.1063.
Raath, M.A., 1969. A new coelurosaurian dinosaur from the Forest Sandstone of PMID:17802092.
Rhodesia. Arnoldia 4 (28), 1–25. Sidor, C.A., Hancox, P.J., 2006. Elliotherium kersteni, a new tritheledontid from the lower
Raath, M.A., 1981. A protosuchid crocodilian from the Forest Sandstone Formation Elliot Formation (Upper Triassic) of South Africa. J. Paleontol. 80 (2), 333–342.
(upper Karoo) of Zimbabwe. Palaeontol. Afr. 24, 69–174. Smith, R., Kitching, J., 1997. Sedimentology and vertebrate taphonomy of the Tritylodon
Raath, M.A., Smith, C.C., Bond, G., 1970. A new Upper Karroo dinosaur fossil locality on Acme Zone: a reworked palaeosol in the Lower Jurassic Elliot Formation, Karoo
the lower Angwa River, Sipolilo District, Rhodesia. Arnoldia 4 (35), 1–10. Supergroup, South Africa. Palaeogeogr. Palaeoclimatol. Palaeoecol. 131 (1-2), 29–50.
Raath, M.A., Kitching, J.W., Shone, R.W., Rossouw, G.J., 1990. Dinosaur tracks in Triassic Smith, R.M.H., Eriksson, P.G., Botha, W.J., 1993. A review of the stratigraphy and sedi-
Molteno sediments: the earliest evidence of dinosaurs in South Africa? Palaeontol. mentary environments of the Karoo-aged basins of Southern Africa. J. Afr. Earth Sci.
Afr. 27, 89–95. 16 (1-2), 143–169.
Rademan, Z., 2018. Radiometric Dating and Stratigraphic Reassessment of the Elliot and Smith, R.M., Marsicano, C.A., Wilson, J.A., 2009. Sedimentology and paleoecology of a
Clarens Formations; Near Maphutseng and Moyeni, Kingdom of Lesotho, Southern diverse Early Jurassic tetrapod tracksite in Lesotho, southern Africa. Palaios 24 (10),
Africa. MSc thesis. Stellenbosch University. http://hdl.handle.net/10019.1/ 672–684.
104888. Spencer, C.J., Kirkland, C.L., Taylor, R.J., 2016. Strategies towards statistically robust
Rainforth, E.C., 2003. Revision and re–evaluation of the Early Jurassic dinosaurian ich- interpretations of in situ U–Pb zircon geochronology. Geosci. Front. 7, 581–589.
nogenus Otozoum. Palaeontology 46, 803–838. Spikings, R., Reitsma, M.J., Boekhout, F., Mišković, A., Ulianov, A., Chiaradia, M., Gerdes,

23
E.M. Bordy, et al. Earth-Science Reviews 203 (2020) 103120

A., Schaltegger, U., 2016. Characterisation of Triassic rifting in Peru and implications Biostratigraphy of the Massospondylus Assemblage Zone (Stormberg Group, Karoo
for the early disassembly of western Pangaea. Gondwana Res. 35, 124–143. Supergroup), South Africa. S. Afr. J. Geol in press.
Steyer, J.S., Damiani, R., 2005. A giant brachyopoid temnospondyl from the Upper Visser, J.N.J., 1984. A review of the Stormberg Group and Drakensberg volcanics in
Triassic or Lower Jurassic of Lesotho. Bulletin de la Société Géologique de France 176 southern Africa. Palaeontol. Afr. 25, 5–27.
(3), 243–248. von Huene, F., 1932. Die fossile Reptil-Ordung Saurischia, ihre Entwicklung und
Suarez, C.A., Knobbe, T.K., Crowley, J.L., Kirkland, J.I., Milner, A.R., 2017. A chronos- Geschichte. Monographien zur Geologie und Palaeontologie Leipzig, Verlag von
tratigraphic assessment of the Moenave Formation, USA using C-isotope chemos- Gebrüder Borntraeger. Serie I Heft 4 Teil I 361.
tratigraphy and detrital zircon geochronology: implications for the terrestrial end Wang, Z.S., Rasbury, E.T., Hanson, G.N., Meyers, W.J., 1998. Using the U-Pb system of
Triassic extinction. Earth Planet. Sci. Lett. 475, 83–93. calcretes to date the time of sedimentation of clastic sedimentary rocks. Geochim.
Sues, H.D., Olsen, P.E., 2015. Stratigraphic and temporal context and faunal diversity of Cosmochim. Acta 62 (16), 2823–2835.
Permian-Jurassic continental tetrapod assemblages from the Fundy rift basin, eastern Wang, S., Stiegler, J., Amiot, R., Wang, X., Du, G.H., Clark, J.M., Xu, X., 2017. Extreme
Canada. Atl. Geol. 51, 139–205. https://doi.org/10.4138/atlgeol.2015.006. ontogenetic changes in a ceratosaurian theropod. Curr. Biol. 27 (1), 144–148.
Sues, H.D., Reisz, R.R., 1995. First record of the early Mesozoic sphenodontian https://doi.org/10.1016/j.cub.2016.10.043.
Clevosaurus (Lepidosauria: Rhynchocephalia) from the Southern Hemisphere. J. Warren, A., Damiani, R., 1999. Stereospondyl amphibians from the Elliot Formation of
Paleontol. 69 (1), 123–126. https://doi.org/10.1017/s0022336000026974. South Africa. Palaeontol. Afr. 35, 45–54. http://hdl.handle.net/10539/1646.
Sues, H.D., Shubin, N.H., Olsen, P.E., Amaral, W.W., 1996. On the cranial structure of a Whiteside, J.H., Olsen, P.E., Eglinton, T., Brookfield, M.E., Sambrotto, R.N., 2011.
new protosuchid (Archosauria: Crocodyliformes) from the McCoy Brook Formation Compound-specific carbon isotopes from EarthTs largest flood basalt eruptions di-
(Lower Jurassic) of Nova Scotia, Canada. J. Vertebr. Paleontol. 16 (1), 34–41. rectly linked to the end-Triassic mass extinction. PNAS 107 (15), 6721–6725. https://
https://doi.org/10.1080/02724634.1996.10011281. doi.org/10.1073/pnas.1001706107.
Svensen, H., Corfu, F., Polteau, S., Hammer, Ø., Planke, S., 2012. Rapid magma empla- Wilson, J.A., Marsicano, C.A., Smith, R.M., 2009. Dynamic locomotor capabilities re-
cement in the Karoo large igneous province. Earth Planet. Sci. Lett. 325, 1–9. https:// vealed by early dinosaur trackmakers from Southern Africa. PLoS One 4 (10), e7331.
doi.org/10.1016/j.epsl.2012.01.015. Wotzlaw, J.F., Guex, J., Bartolini, A., Gallet, Y., Krystyn, L., McRoberts, C.A., Taylor, D.,
Tabor, N.J., Myers, T.S., 2015. Paleosols as indicators of paleoenvironment and paleo- Schoene, B., Schaltegger, U., 2014. Towards accurate numerical calibration of the
climate. Annu. Rev. Earth Planet. Sci. 43, 333–361. Late Triassic: High-precision U-Pb geochronology constraints on the duration of the
Tauxe, L., 1998. Paleomagnetic Principles and Practice – Modern Approaches in Rhaetian. Geology 42 (7), 571–574. https://doi.org/10.1130/G35612.1.
Geophysics. Kluwer Academic Publisher, Dordrecht, Netherlands, pp. 309. Yates, A.M., 2003. A definite prosauropod dinosaur from the lower Elliot Formation
Thulborn, R.A., 1974. A new heterodontosaurid dinosaur (Reptilia: Ornithischia) from the (Norian: Upper Triassic) of South Africa. Palaeontol. Afr. 39, 63–68.
Upper Triassic Red Beds of Lesotho. Zool. J. Linn. Soc. Lond. 55, 151–175. Yates, A.M., 2005. A new theropod dinosaur from the Early Jurassic of South Africa and
Tolchard, F., Nesbitt, S.J., Desojo, J., Viglietti, P., Butler, R.J., Choiniere, J.N., 2019. its implications for the early evolution of theropods. Palaeontol. Afr. 41, 105–122.
‘Rauisuchian’ material from the lower Elliot Formation of South Africa: implications Yates, A.M., 2007a. Solving a dinosaurian puzzle: the identity of Aliwalia rex Galton. Hist.
for Late Triassic biogeography and biostratigraphy. J. Afr. Earth Sci. 160, 103610. Biol. 19, 93–123.
Tucker, R.T., Roberts, E.M., Hu, Y., Kemp, A.I., Salisbury, S.W., 2013. Detrital zircon age Yates, A.M., 2007b. The first complete skull of the Triassic dinosaur Melanorosaurus
constraints for the Winton Formation, Queensland: contextualizing Australia's Late Haughton (Sauropodomorpha: Anchisauria). Spec. Pap. Paleontol. 77, 9–55.
Cretaceous dinosaur faunas. Gondwana Res. 24 (2), 767–779. https://doi.org/10. Yates, A.M., 2008. A second specimen of Blikanasaurus (Dinosauria: Sauropoda) and the
1016/j.gr.2012.12.009. biostratigraphy of the lower Elliot Formation. Palaeontol. Afr. 43, 39–43.
Van Hoepen, E.C.N., 1920. Contributions to the knowledge of the reptiles of the Karoo Yates, A.M., Barrett, P.M., 2010. Massospondylus carinatus Owen 1854 (Dinosauria:
Formation. 6. Further dinosaurian material in the Transvaal Museum. Ann. Transv. Sauropodomorpha) from the Lower Jurassic of South Africa: proposed conservation
Mus. 7, 93–140. of the usage by designation of a neotype. Palaeontol. Afr. 45, 7–10.
Viglietti, P.A., Frei, D., Rubidge, B.S., Smith, R.M., 2018a. U-Pb detrital zircon dates and Yates, A.M., Kitching, J.W., 2003. The earliest known sauropod dinosaur and the first
provenance data from the Beaufort Group (Karoo Supergroup) reflect sedimentary steps towards sauropod locomotion. Proc. R. Soc. Lond. Ser. B Biol. Sci. 270 (1525),
recycling and air-fall tuff deposition in the Permo-Triassic Karoo foreland basin. J. 1753–1758.
Afr. Earth Sci. 143, 59–66. Yates, A.M., Bonnan, M.F., Neveling, J., Chinsamy, A., Blackbeard, M.G., 2009. A new
Viglietti, P.A., Barrett, P.M., Broderick, T.J., Munyikwa, D., MacNiven, R., Broderick, L., transitional sauropodomorph dinosaur from the Early Jurassic of South Africa and the
Chapelle, K., Glynn, D., Edwards, S., Zondo, M., Broderick, P., Choiniere, J.N., 2018b. evolution of sauropod feeding and quadrupedalism. Proc. R. Soc. B Biol. Sci. 277
Stratigraphy of the Vulcanodon type locality and its implications for regional corre- (1682), 787–794.
lations within the Karoo Supergroup. J. Afr. Earth Sci. 137, 149–156. https://doi.org/ Yates, A.M., Bonnan, M.F., Neveling, J., 2011. A new basal sauropodomorph dinosaur
10.1016/j.jafrearsci.2017.10.015. from the Early Jurassic of South Africa. J. Vertebr. Paleontol. 31, 610–625.
Viglietti, P.A., McPhee, B.W., Bordy, E.M., Sciscio, L., Barrett, P.M., Benson, R.B., Wills, You, H.-L., Azuma, Y., Wang, T., Wang, Y.-M., Don, Z.-M., 2014. The first well-preserved
S., Tolchard, F., Choiniere, J.N., 2020a. Biostratigraphy of coelophysoid theropod dinosaur from Asia. Zootaxa 3873, 233–249.
the Scalenodontoides Assemblage Zone (Stormberg Group, Karoo Supergroup), South Zeigler, K.E., Geissman, J.W., 2011. Magnetostratigraphy of the Upper Triassic Chinle
Africa. S. Afr. J. Geol in press. Group of New Mexico: Implications for regional and global correlations among Upper
Viglietti, P.A., McPhee, B.W., Bordy, E.M., Sciscio, L., Barrett, P.M., Benson, R.B., Wills, Triassic sequences. Geosphere 7 (3), 802–829.
S., Chapelle, K.J.E., Dollman, K.N., Mdekazi, C., Choiniere, J.N., 2020b.

24