ArticlePDF Available

Morphology of the temporal skull region in tetrapods: research history, functional explanations, and a new comprehensive classification scheme

September 2021
Biological reviews of the Cambridge Philosophical Society 96(5):2229-2257

DOI:10.1111/brv.12751

Authors:

Pascal Abel

Ingmar Werneburg

Senckenberg Research Institute

The morphology of the temporal region in the tetrapod skull traditionally has been a widely discussed feature of vertebrate anatomy. The evolution of different temporal openings in Amniota (mammals, birds, and reptiles), Lissamphibia (frogs, salamanders, and caecilians), and several extinct tetrapod groups has sparked debates on the phylogenetic, developmental, and functional background of this region in the tetrapod skull. This led most famously to the erection of different amniote taxa based on the number and position of temporal fenestrae in their skulls. However, most of these taxa are no longer recognised to represent natural groupings and the morphology of the temporal region is not necessarily an adequate trait for use in the reconstruction of amniote phylogenies. Yet, new fossil finds, most notably of parareptiles and stem-turtles, as well as modern embryological and biomechanical studies continue to provide new insights into the morphological diversity of the temporal region. Here, we introduce a novel comprehensive classification scheme for the various temporal morphotypes in all Tetrapoda that is independent of phylogeny and previous terminology and may facilitate morphological comparisons in future studies. We then review the history of research on the temporal region in the tetrapod skull. We document how, from the early 19th century with the first recognition of differences in the temporal region to the first proposals of phylogenetic relationships and their assessment over the centuries, the phylogenetic perspective on the temporal region has developed, and we highlight the controversies that still remain. We also compare the different functional and developmental drivers proposed for the observed morphological diversity and how the effects of internal and external factors on the structure of the tetrapod skull have been interpreted.

Content uploaded by Ingmar Werneburg

Content may be subject to copyright.

Morphology of the temporal skull region in

tetrapods: research history, functional

explanations, and a new comprehensive

classiﬁcation scheme

Pascal Abel

1,2

and Ingmar Werneburg

1,2

Senckenberg Centre for Human Evolution and Palaeoenvironment (SHEP) at Eberhard Karls Universität, Sigwartstraße 10, Tübingen, 72076,

Germany

Fachbereich Geowissenschaften der Eberhard-Karls-Universität Tübingen, Hölderlinstraße 12, Tübingen, 72074, Germany

ABSTRACT

The morphology of the temporal region in the tetrapod skull traditionally has been a widely discussed feature of verte-

brate anatomy. The evolution of different temporal openings in Amniota (mammals, birds, and reptiles), Lissamphibia

(frogs, salamanders, and caecilians), and several extinct tetrapod groups has sparked debates on the phylogenetic, devel-

opmental, and functional background of this region in the tetrapod skull. This led most famously to the erection of dif-

ferent amniote taxa based on the number and position of temporal fenestrae in their skulls. However, most of these taxa

are no longer recognised to represent natural groupings and the morphology of the temporal region is not necessarily an

adequate trait for use in the reconstruction of amniote phylogenies. Yet, new fossil ﬁnds, most notably of parareptiles and

stem-turtles, as well as modern embryological and biomechanical studies continue to provide new insights into the mor-

phological diversity of the temporal region. Here, we introduce a novel comprehensive classiﬁcation scheme for the var-

ious temporal morphotypes in all Tetrapoda that is independent of phylogeny and previous terminology and may

facilitate morphological comparisons in future studies. We then review the history of research on the temporal region

in the tetrapod skull. We document how, from the early 19th century with the ﬁrst recognition of differences in the tem-

poral region to the ﬁrst proposals of phylogenetic relationships and their assessment over the centuries, the phylogenetic

perspective on the temporal region has developed, and we highlight the controversies that still remain. We also compare

the different functional and developmental drivers proposed for the observed morphological diversity and how the effects

of internal and external factors on the structure of the tetrapod skull have been interpreted.

Key words: fenestration, macroevolution, functional morphology, Tetrapoda, Amniota, Lissamphibia, skull anatomy,

biomechanics

CONTENTS

I. Introduction ......................................................................2230

(1) Morphological and taxonomic deﬁnitions ............................................2231

(a) Morphology ............................................................... 2231

(b) Taxonomy ................................................................ 2231

II. An updated morphological classiﬁcation scheme ...........................................2231

(1) Scutal ........................................................................2232

(2) Nudital ......................................................................2235

(3) Infrafenestral ..................................................................2235

(4) Infrafossal ....................................................................2235

*Address for correspondence (Tel: +49 7071 29 73068; E-mail: ingmar.werneburg@senckenberg.de)

Society.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction

in any medium, provided the original work is properly cited and is not used for commercial purposes.

Biol. Rev. (2021), 96, pp. 2229–2257. 2229

doi: 10.1111/brv.12751

(5) Suprafenestral .................................................................2235

(6) Suprafossal ...................................................................2235

(7) Bifenestral ....................................................................2235

(8) Bifossal .......................................................................2236

(9) Additofenestral ................................................................2236

(10) Fossafenestral ..................................................................2236

III. Research history of the temporal skull region .............................................2236

(1) Temporal skull diversity as a classiﬁcation tool ........................................2236

(a) Early phylogenetic inferences .................................................. 2236

(b) Handling morphological diversity .............................................. 2238

tradition ..................................................................... 2239

(d) Modern phylogeny and morphological patterns ................................... 2241

(e) Unresolved controversies ..................................................... 2244

(2) Understanding the functional origins of temporal openings ...............................2246

(a) Comprehending diversity (1880s–1900s) ......................................... 2246

(b) The interplay between muscle and bone (1910s–1950s) .............................. 2246

(d) Quantitative modelling and other recent approaches (1990s–present) ................... 2249

(e) Developmental studies ....................................................... 2250

(3) Summary on the origins of temporal openings ........................................2251

IV. Conclusions .......................................................................2252

V. Acknowledgements .................................................................2253

VI. References ........................................................................2253

I. INTRODUCTION

The temporal region, the part of the dermatocranium

between the orbits and the occiput serves multiple purposes

in the vertebrate skull: it incorporates the cranial origin sites

of the jaw adductor musculature, the jaw hinge, and covers

the ear and ultimately also the braincase. Accordingly, the

temporal region exhibits a vast array of morphotypes, reﬂect-

ing evolutionary differences in lifestyle and developmental

patterns in the respective taxa. The high morphological

diversity of the temporal region is most evident in the tetra-

pod crown-groups Lissamphibia and Amniota. While most,

if not all, Paleozoic stem-tetrapods had a fully closed tempo-

ral dermatocranium (e.g. Clack, 1997, 2002; Blom, 2005;

Daeschler, Shubin & Jenkins, 2006), most lissamphibians

and amniotes show either a greatly reduced temporal derma-

tocranium or have developed a variety of temporal openings

(e.g. Kleinteich et al., 2012; Werneburg, 2019; Ford &

Benson, 2020; Paluh, Stanley & Blackburn, 2020). Even in

radiations of extinct crown-tetrapods, like Lepospondyli, a

variety of temporal morphotypes, such as large ventral

excavations or overall reductions of the dermatocranium,

can be observed (e.g. Bolt & Rieppel, 2009; Pardo &

Anderson, 2016). The morphology of the temporal region

in tetrapods has thus attracted much interest over the last

200 years, from palaeontologists and neontologists alike

(e.g. Hallmann, 1837; Cope, 1892; Baur, 1894; Gaupp,

1895a,b; Williston, 1904; Fuchs, 1909; Jaekel, 1913; Boas,

1915; Versluys, 1919; Broom, 1922; Fox, 1964; Frazzetta,

1968; Tarsitano et al., 2001; Werneburg, 2019). Apart from

purely morphological comparisons (Hallmann, 1837),

previous workers focussed on the functional (e.g. Dollo,

1884; Gaupp, 1895b; Gregory & Adams, 1915; Case, 1924;

Frazzetta, 1968), developmental (e.g. Tarsitano et al., 2001;

Schoch, 2014b; Ford, 2018; Werneburg, 2019), and phyloge-

netic background (e.g. Baur, 1894; Osborn, 1903; Williston,

1917; Broom, 1922; Kuhn-Schnyder, 1980; Müller, 2003;

MacDougall & Reisz, 2014; Ford & Benson, 2020) of the

structural diversity in the temporal skull region. Much

emphasis has been put on the phylogenetic value of temporal

morphology, inspiring the naming of several higher taxa,

such as ‘Synapsida’,‘Diapsida’,‘Anapsida’,‘Euryapsida’,

‘Stegokrotaphia’, and ‘Stegocephali’(e.g. Cope, 1868;

Osborn, 1903; Fuchs, 1909; Williston, 1917; Broom, 1922;

Colbert, 1945; Boettger 1952; Cannatella & Hillis 1993),

some of which remain in use to the present day –although

often with a different deﬁnition.

Yet it has been demonstrated that some morphological

traits, such as an infratemporal fenestra or a fully closed tem-

poral dermatocranium, appeared independently and repeat-

edly in distantly related taxa (Müller, 2003; MacDougall &

Reisz, 2014; Ford & Benson, 2020), probably not always in

response to the same selective pressure (Carroll, 1982;

Werneburg, 2012, 2015, 2019). Additionally, the temporal

region can vary distinctly in morphology among closely

related taxa (Gow, 1972; Tsuji, Müller & Reisz, 2012), spec-

imens of the same species (Cisneros, 2008; Ezcurra, Butler &

Benson, 2015), or throughout ontogeny (Gow, 1972; Haridy

et al., 2016). This highlights the complex evolution of the tem-

poral region in tetrapods and casts doubt on a classiﬁcation of

various tetrapod clades that emphasizes their temporal

morphology, especially within Amniota (e.g. ‘Synapsida’,

Society.

2230 Pascal Abel and Ingmar Werneburg

‘Diapsida’,‘Anapsida’,‘Parapsida’,‘Euryapsida’). It seems

likely that similar temporal morphotypes evolved in different

clades, either because of similar selective pressures or because

different selective regimes favoured the convergent evolution

of similar morphotypes. Thus, to establish an understanding

of the diversity of the tetrapod temporal region, a holistic

approach involving phylogenetic, functional, developmental,

and ecological considerations is needed.

Here, we provide a completely new classiﬁcation scheme for

temporal morphology in both tetrapod groups (amphibians

and reptiliomorphs), enabling us to discuss the diversity of

the temporal skull region without adding confusion by

expanding or modifying the vast number of previous perspec-

tives. We then provide an overview of the research history of

temporal morphology in Tetrapoda and their extinctrelatives.

We illustrate the different approaches used by previous, some-

times rarely cited, researchers to investigate the high disparity

of the tetrapod temporal region, and we discuss which func-

tional, developmental, and evolutionary factors they consid-

ered as fundamental to changes in temporal morphology.

(1) Morphological and taxonomic deﬁnitions

(a)Morphology

Morphological terms for the temporal region have been

used differently by previous researchers. This applies espe-

cially to the question of whether excavations in the temporal

region (e.g. ventrally in the ‘cheek’of most squamates or the

posterodorsal and ventral excavations observed in most

turtles) should be described as temporal fenestrae or not.

For clariﬁcation on morphotypes as used herein, we provide

deﬁnitions in Table 1 that are mostly based on Werneburg

(2013b, 2019). [Correction added on 8 June 2021 after ﬁrst

publication: Werneburg (2013a) has been corrected to

Werneburg (2013b) in the previous sentence]

(b)Taxonomy

In the works discussed herein, taxonomic names often were

used in a non-cladistic sense or differed in their taxonomic

content from their modern usage [e.g. ‘Synapsida’, which

also incorporated ‘cotylosaurs’, placodonts, plesiosaurs, and

turtles when Osborn (1903) ﬁrst introduced the name].

Additionally, even in recent decades, the deﬁnitions of some

commonly used taxa have been extensively discussed (e.g.

Laurin & Anderson, 2004; Modesto & Anderson, 2004).

Hence, to avoid confusion, we deﬁne how we use some of

these names in Table 2.

As we show in this review, an apomorphy-based deﬁni-

tion of clades using temporal morphology can be ambigu-

ous, thus we avoid the application of such deﬁnitions here.

This applies especially to Diapsida, which generally has

been used as the amniote clade possessing both an infratem-

poral and a supratemporal fenestra; a presumed apomor-

phy inherited by the extant tuatara (Gauthier & de

Queiroz, 2020). However, whether this trait appeared only

once or arose several times independently in early reptiles is

hard to determine (Ford & Benson, 2020; see Section III.1e).

In fact, the morphotype of the tuatara may have evolved

secondarily (e.g. Müller, 2003; Evans, 2008). We therefore

use Diapsida herein with a node-based deﬁnition as has

been applied in previous publications (e.g. Laurin, 1991;

Laurin & Reisz, 1995).

II. AN UPDATED MORPHOLOGICAL

CLASSIFICATION SCHEME

The history of research on the tetrapod temporal region

involved the introduction of several classiﬁcation schemes

for the different temporal morphotypes and sometimes the

Table 1. Morphological terms used in this review

Term Deﬁnition

Temporal

opening

Temporal openings are reductions of the temporal dermatocranium that are either formed within the suture of two or

more bones (temporal fenestra) or by excavations in the dermal armour in a ventrolateral or posterodorsal direction.

Temporal

fenestra

Temporal fenestrae are temporal openings that are completely surrounded by bone. They always form within the

sutural contact of two or more temporal bones. An infratemporal fenestra forms in the ‘cheek’region of the skull and

is ventrally always bordered by a lower temporal bar (i.e. zygomatic arch). A supratemporal fenestra forms in the skull

roof and is medially always bordered by the parietal.

In some cases the temporal fenestra is conﬂuent with the orbit (e.g. many Mammalia), or the orbit extends distinctly into

the temporal region (e.g. Procolophonidae). Both conditions are referred to as an orbitotemporal opening, however,

they have different developmental and evolutionary origins.

Temporal

excavation

Temporal excavations are ventrolateral or posterodorsal excavations of the dermal armour, formed either by the

reduction of a temporal bar (squamates, birds), loss of bones (lissamphibians), or by an embayment of the dermal

armour (mammals, turtles). The otic notch is not a temporal excavation.

Temporal bar Temporal bars are bony arches that border temporal fenestrae. The lower temporal bar (or zygomatic arch) is the bony

arch that ventrally borders the infratemporal fenestra. The upper temporal bar borders the infratemporal fenestra

dorsally, or the supratemporal fenestra laterally.

The arch bordering the infratemporal fenestra anteriorly is the postorbital bar. The posterior equivalent is the

posttemporal bar.

Temporal bridge Temporal bridges are remnants of the temporal dermal armour in-between two temporal excavations. They are mostly

found in turtles but are also present in other taxa such as Recumbirostra (early Tetrapoda).

Society.

Morphology of the tetrapod temporal skull region 2231

erection of new taxa based on these morphotypes. While sys-

tems following Osborn (1903; see Section III.1c) and

Gaupp (1895b; see Section III.1b) are in common use in com-

parative anatomy, we currently face two challenges.

First, amniote and amphibian researchers often apply two

different classiﬁcation schemes to describe the temporal mor-

phology of their studied species. Amniote researchers usually

prefer the scheme initiated by Osborn (1903) (which uses

‘synapsid’,‘anapsid’,‘diapsid’,‘euryapsid’, and so forth),

whereas in the amphibian literature the terms of

Gaupp (1895b)(‘stegokrotaphic’,‘zygokrotaphic’,‘gymnok-

rotaphic’) prevail. It may be argued that this can be justiﬁed

by the evolutionary distance and developmental/morpho-

logical differences between these groups. Yet, as demon-

strated herein, developmental or functional reasons for

temporal openings may even vary within amniotes and lis-

samphibians, and similar factors might invoke their conver-

gent development.

Second, in the scheme used for amniotes, the commonly

used terms are based on formally erected taxa. However, of

these taxa only Synapsida and Diapsida are still considered

to be monophyletic (e.g. Benton, 1985; Gauthier & de

Queiroz, 2020; Laurin & Reisz, 2020), although with ongoing

controversies on their actual taxonomic composition

(e.g. Berman, 2013; Schoch & Sues, 2015; Laurin & Piñeiro,

2017, 2018; Ford & Benson, 2020). In fact, even the current

consensus differs markedly from the point of view when these

taxa were ﬁrst introduced (Osborn, 1903; Williston, 1917)

and the original terms are now often used to describe the mor-

phology of species that were previously not considered to

belong to one of these taxa (e.g. Nussbaum, 1983; Tarsitano,

1983; Carroll, 1988; Heckert, Lucas & Spielmann, 2012).

Hence, instead of further modifying one of the traditional

schemes, especially by adding a new array of ‘-apsid’types,

we introduce here a new and comprehensive classiﬁcation

scheme for the tetrapod temporal region that is neither based

on phylogeny nor on assumed functional backgrounds or

homology criteria (Fig. 1).

Our main goal for the new terms to be (i) short and coher-

ent in their phrasing, and (ii) descriptively distinct but still

generally applicable to a high number of clades. Our classiﬁ-

cation scheme introduces 10 distinct morphotypes applicable

to Tetrapoda and other Stegocephali (Fig. 2). Some skulls

may not be unambiguously assignable to a single one of these

morphotypes. This is as expected: naturally occurring mor-

phological variation may never ﬁt an artiﬁcial scheme per-

fectly. Yet, we believe the majority of tetrapod skulls can be

assigned to one of these types. Our scheme is not based on

homology criteria, but rather on the presence, conﬁguration,

and number of temporal openings. Hence, it does not con-

sider the exact suturing between the temporal bones, the

presence or absence of a speciﬁc bone, the presence of a post-

temporal fenestra, skull proportions, or muscle arrange-

ments. The relative size of a temporal opening is only

considered for temporal excavations, whereas the relative

size of a temporal fenestra has no implications for morpho-

logical assignment herein. This approach thus carries the risk

of integrating only superﬁcially similar temporal skull condi-

tions within the same morphotype. Nevertheless, this scheme

also retains independence from taxonomic and functional

interpretations. Ordering the tetrapod temporal region by

a simple morphological scheme in our opinion ensures a bet-

ter comprehension of the complexity and functional adapta-

tions in this large area of the cranium and provides a basis for

future quantitative studies that examine the underlying

developmental patterns and structural homology of the tem-

poral dermatocranium.

(1) Scutal

From Latin scutum =‘shield’, scutal describes skulls with a

fully roofed temporal region, i.e. a temporal dermatocra-

nium lacking fenestrations or distinct temporal excavations

(Fig. 1A). It corresponds, among others, to ‘anapsid’(after

Williston, 1917), ‘stegokrotaphic’(after Gaupp, 1895b), and

‘stegal’(Jaekel, 1909a; see also Fig. 1A for further synonyms).

It is the ancestral condition in Tetrapodomorpha and is

mostly retained, with a reduced number of dermal bones,

in the stem-groups of Lissamphibia and Amniota, and likely

as a symplesiomorphy in early Amniota and Gymnophiona

(for further details of the taxa and literature mentioned in

the skull type descriptions, see Section III.1d). It reappeared

likely secondarily in Pareiasauromorpha, early Testudinata

and a number of Testudines, as well as in some hyperossiﬁed

Anura. The skulls of some Mammalia may be referred to

this type.

Table 2. Deﬁnitions of taxonomic names used in this review

Taxon Deﬁntion

Stegocephali Used mostly sensu Laurin (2020a) as the clade

containing Tetrapoda, but not Panderichthys

rhombolepis and Tiktaalik roseae.

Tetrapoda Used sensu Laurin (2020b) as the least inclusive

clade containing Amniota and Lissamphibia.

Amphibia Used sensu Laurin et al. (2020a) as the clade

containing Lissamphibia and all taxa closer to

Lissamphibia than to Amniota.

Lissamphibia Used sensu Laurin et al. (2020b) as the least-inclusive

clade containing Anura (frogs), Caudata

(salamanders), and Gymnophiona (caecilians).

Reptilia Used mostly sensu Modesto & Anderson (2004) as

the clade containing Lepidosauria (squamates

and tuatara), Archosauria (crocodiles and birds),

Testudines (turtles), and all taxa closer to these

than to Mammalia.

Diapsida Used mostly sensu Laurin (1991) and Laurin &

Reisz (1995) as the least-inclusive clade

containing Lepidosauria, Archosauria, Youngina

capensis, and Araeoscelis gracilis.

Synapsida Used mostly sensu Laurin & Reisz (2020) as the

clade containing Therapsida (mammals and

their relatives) but not Reptilia after Modesto &

Anderson (2004).

Society.

2232 Pascal Abel and Ingmar Werneburg

Fig. 1. Novel comprehensive classiﬁcation scheme for the arrangement of temporal openings in Tetrapoda. (A) Scutal, (B) nudital,

(C) infrafenestral, (D) infrafossal, (E) suprafenestral, (F) suprafossal, (G) bifenestral, (H) bifossal, (I) additofenestral, (J) fossafenestral.

Red =temporal fenestra; light pink =temporal excavation; dark pink =other skull openings. Skull outlines are generalized after

the early reptile Captorhinus from Fox & Bowman (1966). Widely used synonyms are highlighted in blue.

Society.

Morphology of the tetrapod temporal skull region 2233

Fig. 2. The ten morphotypes proposed by our novel classiﬁcation scheme (see Fig. 1), with examples of taxa to show the potential

variation. (I) Acanthostega gunnari (early Stegocephali; after Clack, 2002); (II) Cacops morrisi (Temnospondyli; after Reisz et al., 2009);

(III) Microcaecilia iwokramae (Gymnophiona; after Wake & Donnelly, 2012); (IV) Captorhinus aguti (early Reptilia; after Fox &

Bowman, 1966); (V) Kapes bentoni (Parareptilia; after Zaher et al., 2019); (VI) Proganochelys quenstedti (early Testudinata; after

Gaffney, 1990); (VII) Cryptobranchus alleganiensis (Caudata; after Carroll & Holmes, 1980); (VIII) Bombina orientalis (Anura; after

AmphibiaTree, 2004); (IX) Argyrogena fasciolata (Squamata; after Das et al., 2019); (X) Terrapene ornata (Testudines; after

Gaffney, 1979); (XI) Anthracosarus russelli (Embolomeri; after Panchen, 1977); (XII) Calyptocephalella gayi (Anura; after Boas, 1915);

(XIII) Cotylorhynchus romeri (early Synapsida; after Romer & Price, 1940); (XIV) Ophiacodon uniformis (early Synapsida; after Romer &

Price, 1940); (XV) Syodon biarmicum (Dinocephalia; after Kammerer, 2011); (XVI) Bolosaurus striatus (early Parareptilia; after

(Figure legend continues on next page.)

Society.

2234 Pascal Abel and Ingmar Werneburg

(2) Nudital

From Latin nuditas =‘bareness’, nudital describes skulls that

lack temporal bars or bridges due to a heavily reduced tem-

poral dermatocranium (Fig. 1B). Usually only the parietal

region still retains dermal bones. This morphotype corre-

sponds with ‘gymnokrotaphic’(after Gaupp, 1895b). It is

widespread in Batrachia and Ophidia, but occurs also in a

number of other Squamata, as well as some Testudines.

(3) Infrafenestral

From Latin infra =‘below’and fenestra =‘window’, infrafenes-

tral describes skulls that possess an infratemporal fenestra as

their only temporal opening (Fig. 1C). This morphotype corre-

sponds mainly to ‘synapsid’(after Osborn, 1903), but also partly

to ‘zygokrotaphic’(after Gaupp, 1895b)or‘zygal’(Jaekel,

1909a). It occurs most notably in the majority of non-

neotherapsidan Synapsida, Gorgonopsia, and many Pararepti-

lia. It can be also found in the embolomere Anthracosaurus russelli

and the caiman Paleosuchus, and a number of Anura. The size of

the fenestra varies distinctly among taxa, ranging from minis-

cule to large openings that occupy most of the ‘cheek’.The

infrafenestral type may have arisen independently in Synapsida

and Parareptilia, or may represent a synapomorphy of Amniota

(see Section III.1e). In caimans, this morphotype emerged by

secondary closure of the supratemporal fenestra.

(4) Infrafossal

From Latin infra =‘below’and fossa =‘cavity’, infrafossal

describes skulls with a large ventral excavation in the ‘cheek’

as the only temporal opening, with this excavation occupying

more than 30% of the temporal height (Fig. 1D). Smaller

excavations are not considered for classiﬁcation. Skulls of this

type have been sometimes referred to as ‘anapsid’, but the

best historical analogues may be ‘hemi-stegokrotaph’

(Gaupp, 1895a), ‘second series’(Fuchs, 1909), and ‘pleuro-

keiroid’(Smith, Chiszar & Frey, 1983). The morphotype

occurs most notably in Pleurodira, but also in Eunotosaurus

africanus, Scincoidea, some Parareptilia, and some

Lepospondyli.

(5) Suprafenestral

From Latin supra =‘above’and fenestra =‘window’, suprafe-

nestral describes skulls that possess a supratemporal fenestra

or lateromedial widened infratemporal fenestra as the only

temporal opening (Fig. 1E). This morphotype corresponds

mostly to ‘euryapsid’(after Colbert, 1945), but also ‘para-

psid’(after Williston, 1917), and ‘metapsid’(after Boettger,

1952). It occurs especially in various Sauropterygia and

Ichthyosauromorpha, as well as in non-mammalian Euther-

iodontia. It is also present in the early diapsid Araeoscelis graci-

lis, a few Archosauriformes, and some Choristodera. In all

reptiles with a suprafenestral skull, this morphotype likely

evolved due to closure of the infratemporal fenestra. In

Eutheriodontia, however, the suprafenestral type evolved

by a lateromedial widening of the infratemporal fenestra.

(6) Suprafossal

From Latin supra =‘above’and fossa =‘cavity’, suprafossal

describes skulls with a large posterodorsal excavation in the

skull roof as their only temporal opening, with this excavation

occupying more than 30% of the temporal length (Fig. 1F).

Skulls of this type have been sometimes referred to as ‘ana-

psid’. The best historic analogues may be ‘ﬁrst series’

(Fuchs, 1909) and ‘opisthokeiroid’(Smith et al., 1983). This

morphotype occurs most notably in Cryptodira, but also in

several Gymnophiona. Arguably may also be referred to var-

ious Mammalia.

(7) Bifenestral

From Latin bis =‘two’and fenestra =‘window’, bifenestral

describes skulls that possess an infratemporal and a supra-

temporal fenestra (Fig. 1G). This morphotype corresponds

mainly to ‘diapsid’(after Osborn, 1903). Occurs mostly in

non-avialan Archosauriformes, but also in some early

Diapsida, several Lepidosauria, and late Rhynchosauria.

Arguably also present in some Anura, but overall restricted

to reptiles. May be the ancestral condition in Diapsida, how-

ever, the bifenestral morphotype in many Sauria is likely a

secondary condition that emerged through the development

(Figure legend continued from previous page.)

Broom, 1913); (XVII) Llistrofus pricei (Lepospondyli; after Bolt & Rieppel, 2009); (XVIII) Brachydectes newberryi (Lepospondyli; after

Pardo & Anderson, 2016); (XIX) Milleropsis pricei (Parareptilia, after Gow, 1972); (XX) Cricosaura typica (Squamata; after

Maisano, 2003); (XXI) Emydura maquarii (Testudines; after Gaffney, 1979); (XXII) Gallus gallus (Avialae; after Jollie, 1957); (XXIII)

Cynognathus platyceps (early Eutheriodontia; after Broom, 1911); (XXIV) Temnodontosaurus trigonodon (Ichthyosauria; after Maisch &

Hungerbühler, 2001); (XXV) Placodus gigas (Placodontia; after Sues, 1987); (XXVI) Pliosaurus kevani (Plesiosauria; after Benson

et al., 2013); (XXVII) Scolecomorphus sp. (Gymnophiona; after DigiMorph Staff, 2002); (XXVIII) Zalambdalestes lechei (Mammalia;

after Wible et al., 2004); (XXIX) Pelodiscus sinensis (Testudines; after Ogushi, 1911); (XXX) Petrolacosaurus kansensis (early Diapsida;

after Reisz, 1977); (XXXI) Clevosaurus hudsoni (Rhynchocephalia; after Fraser, 1988); (XXXII) Erythrosuchus africanus (early

Archosauriformes; after Nesbitt, 2011); (XXXIII) Thalattosuchus superciliosus (Thalattosuchia; after Andrews, 1913); (XXXIV) Testudo

graeca (Testudines; after Gaffney, 1979); (XXXV) Delorhynchus cifellii (Parareptilia; after Haridy et al., 2016); (XXXVI) Postosuchus

kirkpatricki (early Pseudosuchia; after Weinbaum, 2011); (XXXVII) Pleurodeles walti (Caudata; after AmphibiaTree & Gosselin-

Ildari, 2008); (XXXVIII) Lystrosaurus murrayi (Anomodontia; after Ray, 2005); (XXXIX) Claudiosaurus germaini (early Diapsida; after

Carroll, 1981); (XL) Iguana iguana (Squamata; after Conrad & Norell, 2010).

Society.

Morphology of the tetrapod temporal skull region 2235

of a lower temporal bar in a fossafenestral ancestor. The

shape and proportions of both fenestrae can vary drastically.

(8) Bifossal

From Latin bis =‘two’and fossa =‘cavity’, bifossal describes

skulls that possess a large posterodorsal and ventral excava-

tion as temporal openings, occupying more than 30% of

the temporal height or temporal length, respectively

(Fig. 1H). There is no perfectly ﬁtting historical analogue.

Overall restricted to Testudines. Arguably, most mammal

skulls fall under this type, however, in their case the postero-

dorsal excavation is a drastically widened infratemporal

fenestra, and both excavations are separated by the former

lower temporal bar (i.e. zygomatic arch), whereas in turtles

the excavations are secondary reductions of a former scutal

skull and separated by a temporal bridge.

(9) Additofenestral

From Latin additus =‘additional’and fenestra =‘window’,

additofenestral describes skulls that possess two pairs of infra-

temporal fenestrae often in addition to a supratemporal

fenestra (Fig. 1I). There is no perfectly ﬁtting historical ana-

logue, but skulls of this type are often referred to as ‘diapsid’.

The second pair of infratemporal fenestrae can emerge by

the subdivision of a large infratemporal fenestra by a bony

process or within another suture of the temporal dermatocra-

nium. Occurs in Tyrannosauridae, several early Loricata,

and some Parareptilia. Infrafenestral types with an ‘auxil-

iary’fenestra on one side of the skull (as in some early synap-

sids) may, arguably, be referred to this type.

(10) Fossafenestral

From Latin fossa =‘cavity’and fenestra =‘window’, fossafe-

nestral describes skulls that possess a supratemporal fenestra

together with a large ventral excavation in the ‘cheek’that

occupies at least 30% of the temporal height (Fig. 1J). Corre-

sponds mainly to ‘parapsid’(after Williston, 1917) and

‘katapsid’(after Boettger, 1952). Widespread in Diapsida,

especially within Lepidosauria, in most Anomodontia and

some Salamandridae. However, in Anomodontia the supra-

temporal fenestra is evolutionary derived from a widened

infratemporal fenestra. The size of the suprafenestral fenes-

tra can differ markedly. This morphotype is often associated

with a streptostylic jaw.

III. RESEARCH HISTORY OF THE TEMPORAL

SKULL REGION

(1) Temporal skull diversity as a classiﬁcation tool

Comparative studies of the vertebrate skull, including differ-

ences in the temporal region, pre-date the Darwin revolution

with a notable contribution by Cuvier (1829). The ﬁrst

detailed monograph on the temporal skull region was by

Hallmann (1837). He described the osteology and myology

of the temporal region as well as the associated neurobiology,

and the auditory system mainly of extant vertebrates. Hall-

mann (1837) did not examine the skull under a phylogenetic

perspective, but focussed particularly on embryological dif-

ferences among vertebrates and how the various morpho-

types formed during prehatching/prenatal ontogeny.

(a)Early phylogenetic inferences

Scientiﬁc interest in the morphology of the temporal skull

region increased by the end of the 19th century (Fig. 3).

Due to increasing attempts to understand evolutionary

changes metaphorically as a ‘tree’(Tassy, 2011), together

with the growing acceptance of Darwinian theories of evolu-

tion (Darwin, 1859) and better availability of palaeontologi-

cal data, morphological differences were increasingly

viewed in terms of phylogenetic relationships. The early phy-

logenetic discussions were mostly focussed on the homology

of particular temporal bones (mainly the squamosal and

quadratojugal). To our knowledge, the ﬁrst attempt explicitly

to discuss phylogenetic aspects of temporal anatomy was

Günther (1867), who highlighted similarities in the temporal

region between tuatara (Sphenodon punctatus) and crocodiles,

contrasting these with the morphology of squamates and tur-

tles. Baur (1889) was probably the ﬁrst to describe the evolu-

tion of the ‘temporal arches’(i.e. temporal bars) in different

tetrapods from early stegocephalians and their assumed

gar-like ancestors with a complete dermal covering of the

temporal region.

Cope (1892), who argued for homology of the temporal

arches among reptiles, further distinguished between a

‘Series I’,‘Series II’, and ‘Series III’in the evolution of rep-

tiles with respect to the formation of temporal bars (Fig. 3).

Like Baur (1889), Cope (1892) considered a closed dermato-

cranium as the ancestral condition from which the reduced

dermatocranium of turtles (Series I), the supratemporal

fenestra [‘supramastoid foramen’in Cope (1892)] of

Ichthyopterygia (Series II), as well as different appearances

and losses of temporal bars in various other amniotes

(Series III) evolved.

Baur (1894) discussed the works of Cope (1892) and

Gaupp (1895a,b, see Section III.1b) with regard to homolo-

gies of the temporal bones among different taxa. Case (1898)

argued that the non-fenestrated Pareiasauria were the ances-

tral ‘reptilian’group, of which, ﬁrst, the ‘Proganosauria’

with an infratemporal and supratemporal fenestra arose, fol-

lowed by branches leading to extant reptiles (‘saurocepha-

lous group’) and mammals (‘mastocephalous group’; Fig. 3).

Gegenbaur (1898) postulated that taxa with several ‘Span-

gen’(‘clasps’, meaning temporal bars and bridges) like rhyn-

chocephalians or crocodiles represent the condition that

arose ﬁrst from taxa with a fully roofed dermatocranium,

from which taxa with only one ‘clasp’(turtles, ‘enaliosaurs’,

and ‘theromeres’) and also the morphotypes of squamates

evolved.

Society.

2236 Pascal Abel and Ingmar Werneburg

Fig. 3. Overview of previous descriptive classiﬁcation schemes based on the morphology of the temporal skull region. Skulls are as

listed in legend to Fig. 2 with the addition of ‘Holo-stegocrotaph’:Limnoscelis paludis after Romer (1946); ‘Anasynapsid’:

Tachyglossus aculeatus after Macrini (2004); ‘Type 3’,‘Type 8’:Edaphosaurus boanerges after Modesto (1995); ‘Type 6’,‘Type 9’:

Candelaria barbouri after Cisneros et al. (2004).

Society.

Morphology of the tetrapod temporal skull region 2237

(b)Handling morphological diversity

Gaupp (1895a,b) was the ﬁrst to introduce anatomical terms

for different temporal morphotypes, coining the terms ‘stego-

crotaph’,‘zygocrotaph’,and‘gymnocrotaph’(Fig. 3). With

reference to the taxon Stegocephali, he adopted the term ‘ste-

gocrotaph’[from Ancient Greek στέγος (stégos)=‘roof’and

κρόταφος (krotaphos)=‘temple’] for species with a fully closed

dermatocranium in their temporal region, including early ste-

gocephalians, marine turtles, and ‘Cotylosauria’. In fact, Ste-

gocephali [from Ancient Greek στέγος (see above) and

κεφαλή(kephalé)=‘skull’], which was introduced by

Cope (1868) for a set of Paleozoic tetrapods, may be the ﬁrst

time a taxon was named based on its temporal morphotype.

Gaupp (1895a) proposed a further subdivision of ‘stegocro-

taph’skulls into ‘holo-stegocrotaph’and ‘hemi-stegocrotaph’

types. He used ‘holo-stegocrotaph’[from Ancient Greek ὅλος

(holos)=‘complete’] for skulls that possess a fully enclosed tem-

poral region, whereas ‘hemi-stegocrotaph’skulls [from

Ancient Greek ἥμισυς (hḗmisys)=‘half’] have a completely

ossiﬁed skull roof but a ventrally emarginated ‘cheek’region

like that found in scincoids. ‘Zygocrotaph’[from Ancient

Greek ζύγωμα (zýgoma)=‘bar’], which Gaupp (1895b)subdi-

vided into ‘mono-zygocrotaph’and ‘di-zygocrotaph’,was

used for skulls with one or two fully formed temporal bars.

Gaupp (1895b)deﬁned a ‘mono-zygocrotaph’condition

[from Ancient Greek μόνος (monos)=‘single’] to be present

in anurans, mammals, birds, the majority of turtles, and non-

ophidian squamates, whereas a ‘di-zygocrotaph’cranium

[from Ancient Greek δίς (dís)=‘double’)] is found in rhyncho-

cephalians, crocodiles, and some fossil taxa like non-avian

dinosaurs. ‘Gymnocrotaph’[from Ancient Greek γυμνός

(gymnos)=‘naked’] referred to crania missing most of the der-

mal bones in the temporal region, which Gaupp (1895b)used

for most urodeles, snakes, gekkotans, and chelid turtles [‘Che-

lydae’in Gaupp (1895a,b)]. Gaupp (1895b)alsopostulatedthe

‘stegocrotaph’skull to be the ancestral condition from which

ﬁrst the ‘zygocrotaph’morphology, and from this the ‘gymno-

crotaph’morphology evolved. Fürbringer (1900) suggested an

extension of Gaupp’s (1895a,b) scheme with the terms ‘anazy-

gocrotaph’[from Ancient Greek ἀνα (ana)=‘up’]and‘kata-

zygocrotaph’[from Ancient Greek κατά(kata)=

‘downward’], depending on a ventral or dorsal position of

the temporal bar, respectively (Fig. 3).

Jaekel (1909a) criticized the terms introduced by

Gaupp (1895b), not only because he considered them to be

‘almost unpronounceable’(‘fast unaussprechlich’), but also

because he emphasized the distinct osteological differences

present, for example between early Stegocephali and turtles.

Jaekel (1909a) suggested as an alternative the similarly

deﬁned terms ‘stegal’and ‘zygal’, with the latter also subdi-

vided into ‘monozygal’and ‘dizygal’(Fig. 3). The ‘stegal’-

like type in extant turtles, he considered to be a specialised

condition derived from a ‘zygal’skull, hence he referred to

them as ‘tegal’[from Ancient Greek τέγος (tégos), also mean-

ing ‘roof’]. Jaekel (1909a) also highlighted differences among

the ‘stegal’skulls of early stegocephalians, ‘placoderms’and

other ‘ﬁshes’with fully roofed dermal armour in their skull,

arguing for a need for similar rigorous comparison as already

performed for fenestrated skulls in reptiles.

Fuchs (1909) also described the development and evolu-

tion of the temporal region in tetrapods (‘Quadrupeda’).

He demonstrated a trend for simpliﬁcation of the temporal

dermatocranium during tetrapod evolution [see also ‘Willis-

ton’s law’,sensu Williston (1914)] and highlighted its unique

position covering the large jaw adductor musculature. He

ordered skulls with temporal openings using a new morpho-

logical classiﬁcation system that partly considered phylogeny.

He erected two ‘main groups’(‘Hauptgruppen’), subdivided

into ‘subgroups’(‘Untergruppen’)or‘series’(‘Reihen’;

Fig. 3). Fuchs’(1909) ﬁrst main group incorporated all skulls

with temporal fenestrae (‘zentral gelegene Reduktion’), his

second main group included skulls with temporal excavations

(‘randständige Reduktion’). Fuchs (1909) adopted Jaekel’s

(1909a) terms ‘stegal’and ‘zygal’, but ordered his ﬁrst main

group mostly in the sense of Osborn (1903; see

Section III.1c). Accordingly, his ‘subgroup a’of the ‘ﬁrst

main group’contained all skulls with two temporal fenestrae,

which he identiﬁed with Osborn’s (1903) ‘Diapsida’, and all

skulls that Fuchs (1909) thought were derived from them

(i.e. squamates, birds, ‘pelycosaurs’). ‘Subgroup b’of this

‘ﬁrst main group’contained all skulls with only a dorsally

open temporal fenestra, which the author described as

present in most early Synapsida, Sauropterygia, and

Ichthyopterygia. Fuchs (1909) restricted his ‘second main

group’to Testudinata: the ‘ﬁrst series’of this group contained

turtle skulls predominantly with a large posterior excavation

in the skull roof; his ‘second series’incorporated skulls with

a large ventral excavation in the ‘cheek’region (Fig. 3).

Discussing the evolutionary origins of procolophonian

reptiles, von Huene (1912; Fig. 3) described amniote skulls

using a scheme derived from Gaupp (1895a,b) and Fürbrin-

ger (1900). He hypothesized that from the ‘stegocrotaph’

‘cotylosaurs’,ﬁve amniote radiations with convergent tem-

poral openings emerged: ‘dizygocrotaph’forms

(he presumably meant forms like tuatara, although not

named explicitly), ‘katazygocrotaph’therapsids, ‘pseudo-

monozygocrotaph’turtles [from Ancient Greek ψ ευδής

(pseudes)=‘false’], ‘hypozygocrotaph’‘deuterosaurs’[from

Ancient Greek ὑπό(hỳpo)=‘under’], as well as ichthyosaurs

together with mesosaurids, which von Huene (1912) appar-

ently also counted as ‘zygocrotaph’. von Huene (1912) inter-

preted procolophonians as late-surviving ‘cotylosaurs’and

the only such taxon to evolve temporal openings. However,

he distinguished them from ‘normal zygocrotaph’taxa by

their evolution of an orbitotemporal opening, suggesting

the term ‘pseudostegocrotaph’. Note that von Huene (1912)

did not include the ‘stegocrotaph’Pareiasauromorpha inside

Procolophonia, in contrast to more recent analyses

(e.g. MacDougall & Reisz, 2014). The ‘deuterosaurs’men-

tioned by von Huene (1912) as examples of the ‘hypozygo-

crotaph’type are now nested within Dinocephalia

(Kammerer, 2011). As they had a large infratemporal fenes-

tra in their ‘cheek’, it could be argued that von Huene (1912)

Society.

2238 Pascal Abel and Ingmar Werneburg

understood hypozygocrotaph as similar to Fürbrin-

ger’s (1900) term ‘katazygocrotaph’(Fig. 3), although von

Huene (1912) used katazygocrotaph to refer to taxa that he

considered to be therapsids.

Williston (1912) modiﬁed the terms of Gaupp (1895b),

referring to them as ‘stegocrotaphic’and ‘zygocrotaphic’.

Additionally, for the type found in ‘Theromorpha’he coined

the term ‘therocrotaphic’[from Ancient Greek θηρίον

(theríon)=‘beast’] and for the condition in Diapsida with

two temporal bars he used ‘saurocrotaphic’[from Ancient

Greek σα~

υρος (saûros)=‘lizard’; Fig. 3].

Versluys (1919) summarized the temporal region in terms

of the then-understood phylogeny of Reptilia. Like previous

authors, he interpreted the fully roofed condition as the

ancestral state. From this he derived four ‘types’(Fig. 3):

the ﬁrst type (‘1. Typus’) referred to a skull with infra- and

supratemporal fenestrae, as found in Diapsida; a second type

(which he also called ‘type A’when it appears in synapsid

skulls) was used for skulls with only the infratemporal fenestra

present; a third type (or ‘type B’in the synapsid skull) had

only the supratemporal fenestra present, as found in Saurop-

terygia and Ichthyosauria; and, lastly, a fourth type applied

to skulls with marginal reductions as found in turtles.

Versluys (1919) argued that phylogenetic relevance (‘genetische

Bedeutung’) was present only in the second and fourth

‘Typus’, whereas the third and probably the ﬁrst could

have developed more than once or have been derived from

each other (in case of Ichthyosauria).

Ford (2018) introduced a new scheme to classify the tem-

poral morphology of early amniotes based on the suturing

of the dermal bones and their contribution to the temporal

openings (Fig. 3). In addition to the ‘anapsid’condition

[see Williston (1917) in Section III.c], Ford (2018) distin-

guished nine different morphotypes, taking both temporal

fenestrae and temporal excavations, as well as co-occurences

of both opening types into account.

Gaupp’s (1895b) original terms (rephrased as ‘stegokro-

taphic’,‘zygokrotaphic’,and‘gymnokrotaphic’)have

remained relevant in the comparative anatomy of non-amniote

tetrapods, most notably in the study of caecilians

(e.g. Nussbaum, 1983; Anderson, 2008; Kleinteich et al.,

2012; Schoch, 2014b; Bardua et al., 2019). Similarly,

Jaekel’s (1909a) terms have remained in use (e.g. Jaekel, 1913,

1922, 1927; Zdansky, 1923–1925; Dombrowski, 1924; von

Huene, 1954, 1956; Kilias, 1957; Iordansky, 1990;

Vorobyeva, 2007), although far less frequently than terms

introduced by Osborn (1903; see Section III.1c).

(c)Formal taxa based on temporal morphology –the beginning of a

taxonomic and nomenclatural tradition

Simultaneously with the introduction of morphological ter-

minology, other researchers used differences in temporal

morphology formally to erect new taxa. This led to the start

of a taxonomic and nomenclatural tradition, which resulted

in the introduction of at least nine etymologically similarly

named taxa during the ﬁrst half of the 20th century

(i.e. ‘Synapsida’,‘Diapsida’,‘Heterapsida’,‘Anapsida’,

‘Parapsida’,‘Anomapsida’,‘Euryapsida’,‘Metapsida’,

‘Katapsida’). Most of these taxa have since been revised fol-

lowing the rise of large-scale phylogenetic analyses (see

Section III.1d) or in some cases never received widespread

acceptance in the contemporary scientiﬁc literature.

The initiator of this tradition was Osborn (1903), who sub-

divided the Reptilia, as then understood, into the taxa

‘Synapsida’[from Ancient Greek σύν (sýn)=‘with’and

άψίϛ(apsís) =‘arch’] and ‘Diapsida’(from Ancient Greek

δύο (dýo)=‘two’], mainly based on the osteology of their tem-

poral region (Fig. 4). Osborn (1903) summarised under his

‘Synapsida’species with skulls having “primarily […] single,

or united temporal arches”, whereas the ‘Diapsida’possess

“primarily […] double, or separated temporal arches”

(Osborn, 1903, p. 276). While still regarded as a monophy-

letic taxon (with a different delimitation) (e.g. Ford &

Benson, 2020), Osborn’s (1903) original ‘Synapsida’also

incorporated Placodontia, Plesiosauria, Testudinata, and

‘Cotylosauria’.

Instead of using his own terms from Jaekel (1909a),

Jaekel (1909b) referred to temporal morphotypes using the

terms of Osborn (1903): ‘diapsid’and ‘synapsid’. Due to

his opinion that Sauropterygia should be classiﬁed as Dia-

psida instead of Synapsida (as done by Osborn, 1903), he

coined the term ‘pseudosynapsid’with regard to their

morphology.

Williston (1917) elaborated Osborn’s (1903) classiﬁcation

by coining the names ‘Anapsida’[from Ancient Greek ἄνευ

(aneu)=‘without’] and ‘Parapsida’(from Ancient Greek

παρά(para)=‘near’]. Containing ‘cotylosaurs’and turtles,

Williston’s (1917) ‘Anapsida’corresponded basically to ‘ste-

gocrotaph/stegal’reptiles. ‘Parapsida’incorporated, among

others, Ichthyosauria and Squamata, with Williston (1917)

arguing that their fenestrated morphotype evolved separately

from this morphology in Synapsida and Diapsida (Fig. 4).

Contrary to Osborn (1903) but like Williston (1917),

Fuchs (1909) considered turtles not to be assignable to Synap-

sida with conﬁdence but also neither to Diapsida (Fig. 4).

Hence, he erected the new taxon ‘Heterapsida’[from

Ancient Greek ἔτερος (héteros)=‘different’].

Broom (1922) raised doubts regarding homology of the

temporal arches in Ichthyosauria, Plesiosauria, and Placo-

dontia with those of ‘mammal-like reptiles’. Broom (1922)

also highlighted similarities of turtles with plesiosaurs and

placodonts, arguing that the turtle skull likely derived from

an ancestor with a supratemporal fenestra. Subsequently,

Broom (1922) distinguished between four taxa of Reptilia,

partly deﬁned by their temporal region: Anapsida, Diapsida,

Synapsida, and a new taxon he named ‘Anomapsida’[from

Ancient Greek ἀνώμαλος (anomalos)=‘irregular’], which

comprised ichthyosaurs, plesiosaurs, placodonts, and tur-

tles (Fig. 4).

Colbert (1945) classiﬁed Sauroptergyia, Placodontia, and

Protorosauria into ‘Euryapsida’[from Ancient Greek ε~

υρύς

Society.

Morphology of the tetrapod temporal skull region 2239

(eurús)=‘broad’], identifying them by the presence of a

supratemporal fenestra only, laterally bounded by the post-

orbital and squamosal bones (Fig. 4). In fact, the term

‘Euryapsida’was originally suggested by Romer in a per-

sonal correspondence as indicated by a footnote in

Colbert (1945).

Fig. 4. Overview of amniote systematics as proposed by different authors based on the temporal region. Skulls are as listed in legend

to Fig. 2 with the addition of Pareiasauria: Scutosaurus karpinskii after Lee (2000); ‘Protorosauria’:Prolacerta broomi after Nesbitt (2011);

Mesosauridae: Mesosaurus tenuidens after Modesto (2006).

Society.

2240 Pascal Abel and Ingmar Werneburg

Kilias (1957) argued for phylogenetic relevance of the tem-

poral excavations (‘Hiatus’) in the turtle skull, at least regard-

ing differentiation between Pleurodira and Cryptodira.

Additionally, and contrasting with some previous studies

(e.g. Jaekel, 1909a; Versluys, 1919; Zdansky, 1923–1925),

he interpreted the fully roofed temporal region in sea turtles

as the retention of an ancestral condition due to their highly

aquatic lifestyle.

The ﬁnal author to name new amniote taxa based on their

temporal morphology was Boettger (1952), who coined the

terms ‘Metapsida’[from Ancient Greek μέτα

(méta)=‘between’] and ‘Katapsida’[from Ancient Greek

κατά(kata)=‘downward’]. He chose the name Katapsida

to reﬂect his belief that there was a loss of (or at least a ten-

dency to lose) the lower temporal bar in these taxa. ‘Katap-

sida’comprises Rhynchocephalia (with choristoderes),

Squamata, and Thalattosauria. ‘Metapsida’incorporates

only Ichthyosauria, justiﬁed by Boettger (1952) on the

grounds that ichthyosaurs only possess a temporal fenestra

on the top of their skull (Fig. 4).

Criticism of phylogenies based on temporal morphology

was present long before large-scale cladistic analyses became

possible. Fuchs (1909) had already highlighted that differ-

ences in temporal morphology may not necessarily be of high

phylogenetic value as similar arrangements could be seen in

distantly related taxa. Goodrich (1916), while accepting the

existence of a ‘reptile’branch leading to mammals (‘therop-

sidan’) and one leading to birds (‘sauropsidan’), cautioned

against the use of temporal morphology as a phylogenetic

trait, as an appearance of a temporal opening could not be

distinguished properly from disappearance of the same in

extinct taxa.

In the second half of the 20th century, Kuhn-Schny-

der (1954, 1963, 1967, 1980) worked extensively on the phy-

logenetic implications of temporal openings, especially with

regard to sauropterygians and squamates. Kuhn-Schny-

der (1954) discussed the ancestry of lizards (‘Lacertilia’)

mostly with respect to their temporal morphology, although

he highlighted that similar structures may be the result of

convergent evolution. Kuhn-Schnyder (1963) provided a lit-

erature review on reptile systematics, especially with regard

to previous interpretations of temporal morphotypes.

Regarding the ancestry of lizards, Kuhn-Schnyder (1963)

reinterated his earlier arguments and highlighted the many

routes for the formation of a supratemporal fenestra. He also

explicitly used temporal anatomy as an argument against a

close relationship between placodonts and other sauroptery-

gians. He argued that the ancestors of placodonts never pos-

sessed an infratemporal fenestra, in contrast to the assumed

‘diapsid’ancestors of Nothosauria and Plesiosauria. Conse-

quently, Kuhn-Schnyder (1963, 1967) considered placodonts

to be part of Synapsida and the only large taxon representing

Colbert’s (1945) ‘Euryapsida’, with other sauropterygians

classiﬁed as Diapsida. Kuhn-Schnyder (1980) again reviewed

the association between temporal fenestration and reptilian

phylogeny. While accepting there may be temporal morpho-

types unique to the respective reptilian clades, he cautioned

that the assumed biomechanical factors leading to temporal

fenestration could argue against the use of temporal mor-

phology as a reliable phylogenetic trait.

Nevertheless, new terms in the tradition of Osborn (1903)

have been coined occasionally in recent decades. These were

explicitly introduced as descriptive terms to order the vast

diversity of temporal morphotypes, as originally done by

Gaupp (1895a,b) and Jaekel (1909a), and not to describe

novel taxa. Smith et al. (1983), for example (Fig. 3), intro-

duced a large number of morphological terms applying to

the amniote temporal region, partly based on what was

interpreted to be a secondary or derived condition, such as

‘eusynapsid’[from Ancient Greek εὖ(eû)=‘good’]formost

mammal skulls and ‘hemidiapsid’for the skulls of most squa-

mates. Smith et al. (1983) introduced a novel ‘keiroid’type

[from Ancient Greek κεὶρειν (keírein)=‘to cut’]basedon

the position of excavations in the turtle skull [‘opisthokeir-

oid’, from Ancient Greek όπισθεν (opisthen)=‘backwards’;

‘pleurokeiroid’, from Ancient Greek πλευρά(pleura)=

‘ﬂank’] and how much space these occupy in the temporal

region [e.g. ‘meiopisthokeiroid’, from Ancient Greek μείων

(meíos)=‘less’;or‘metapleurokeiroid’from Ancient Greek

μετά(meta)=‘in-between’].

Being aware of cranial plasticity during evolution, Werne-

burg (2019) took a simpliﬁed approach, subdividing the mor-

phology of the amniote temporal region into ‘anapsid’

(no temporal openings), ‘monapsid’[infratemporal fenestra

present; from Ancient Greek μόνος (monos)=‘single’],

‘diplapsid’[infratemporal and supratemporal fenestrae pre-

sent; from Ancient Greek διπλο~

υς (diploûs)=‘double’], and

‘excavation’types [all skulls with any kind of temporal emar-

gination or embayment (Fig. 3)] and illustrated the diverse

phylogenetic shifts among morphotypes.

The descriptive terminology introduced by Gaupp (1895b)

occasionally inspired researchers to allocate names to new

formal taxa. Based on their often ‘stegokrotaphic’skulls,

Cannatella & Hillis (1993) erected the clade ‘Stegokrotaphia’

for all non-rhinatrematid caecilians.

There are additional descriptive terms in the tradition of

Osborn (1903), mostly used in educational material, that

sometimes include the morphology of the preorbital skull.

As we were unable to ﬁnd scientiﬁc literature that ofﬁcially

introduced this terminology, we do not include a discussion

of these terms herein.

(d)Modern phylogeny and morphological patterns

With the rise of cladistics and large-scale morphological and

molecular phylogenetic analyses in the second half of the

20th century, as well as many new fossil ﬁnds, especially of

parareptiles (e.g. DeBraga & Reisz, 1996; Tsuji, 2006;

Modesto et al., 2009; Tsuji & Müller, 2009; MacDougall &

Reisz, 2014) and potential stem-turtles (Bever et al., 2015;

Schoch & Sues, 2015, 2018), phylogenies based on temporal

fenestration fell out of favour. It can be demonstrated that

various temporal morphotypes evolved convergently, repeat-

edly, and can vary intraspeciﬁcally (e.g. DeBraga & Rieppel,

Society.

Morphology of the tetrapod temporal skull region 2241

1997; Cisneros, 2008; MacDougall & Reisz, 2014; Ford,

2018; Ford & Benson, 2020). Many taxa erected in the tradi-

tion of Osborn (1903) have been shown to be para- or poly-

phyletic and are only used now in a descriptive

sense. These include ‘Anapsida’of Williston (1917) and

‘Euryapsida’of Colbert (1945), and to a lesser extent ‘Para-

psida’(Williston, 1917), ‘Katapsida’(Boettger, 1952), and

‘Metapsida’(Boettger, 1952). ‘Heterapsida’(Fuchs, 1909)

and ‘Anomapsida’(Broom, 1922) are now mostly forgotten.

Only Osborn’s (1903) original Synapsida and Diapsida are

still used as monophyletic taxa, although in a more delimited

way. Nevertheless, ﬁnding phylogenetic trends may be feasi-

ble when the actual conﬁguration of the temporal dermato-

cranium is considered (MacDougall & Reisz, 2014; Ford,

2018), instead of the sole presence or absence of temporal

openings. In fact, even under the recent perspective, some

morphological patterns can be observed in the temporal skull

region.

The ancestral condition in Tetrapodomorpha and Tetra-

poda is scutal (1 and 2 in Fig. 5) with a temporal region roofed

completely by dermal bone (e.g. Sawin, 1941; Clack, 1997,

2002; Blom, 2005; Daeschler et al., 2006; Sigurdsen & Bolt,

2010). Some extinct crownward groups had evolved small

temporal fenestrae (3 in Fig. 5; Panchen, 1977; Clack,

1987), or distinct excavations (4 in Fig. 5), sometimes accom-

panied by drastic reduction of their dermal armour (Bolt &

Rieppel, 2009; Pardo & Anderson, 2016).

The scutal type (Fig. 1) is likely also the ancestral state in

Gymnophiona (5 in Fig. 5; Maddin, Jenkins & Anderson,

2012) and is present in the caecilian clade Stegokrotaphia

(6 in Fig. 5; Cannatella & Hillis, 1993). Yet, even within ste-

gokrotaphians a suprafossal skull evolved several times inde-

pendently (7 in Fig. 5; Kleinteich et al., 2012). In Batrachia,

the temporal dermatocranium is distinctly reduced. Conse-

quently, the great majority of batrachians possesses a nudital

skull (8 and 10 in Fig. 5). Usually, only the frontal, parietal

(or frontoparietal), squamosal, and quadratojugal are present

(Schoch, 2014b). However, especially within Salamandridae

(9 in Fig. 5), a supratemporal fenestra can be present between

the squamosal, parietal, and frontal, leading to a fossafenestral

skull (AmphibiaTree & Gosselin-Ildari, 2008), whereas in

hyperossiﬁed Anura scutal (12 in Fig. 5), infrafenestral (11 in

Fig. 5), or even bifenestral forms can appear (Evans, Jones &

Krause, 2008; Paluh et al., 2020). Due to the loss of the jugal

and postorbital in anurans, the infratemporal fenestra in such

hyperossiﬁed taxa is anteriorly bordered by the maxilla

(e.g. Wild, 1997).

The ancestral morphotype of Amniota is ambiguous

(Piñeiro et al., 2012) and highly dependent on the nesting of

certain key clades (see Section III.1e). The ancestral amniote

could have possessed a scutal skull (13 in Fig. 5), as tradition-

ally assumed, or an infrafenestral as seen in early Synapsida and

Parareptilia (e.g. Romer & Price, 1940; Cisneros et al., 2004,

2021). Synapsida share ancestrally an infratemporal fenestra

(infrafenestral morphotype; 14 in Fig. 5) that enlarged during

early therapsid evolution (e.g. Kemp, 1984). In most non-

therapsidan synapsids, the infratemporal fenestra is usually

bordered by the jugal, postorbital, quadratojugal, and squa-

mosal (e.g. Romer & Price, 1940). The contribution of the

quadratojugal declines in Metopophora (e.g. Ophiacodonti-

dae, Edaphosauridae, Sphenacodontidae; Romer & Price,

1940). In many Therapsida, but also in some metopophor-

ans, the parietal contributes to the infratemporal fenestra

(e.g. Boonstra, 1936; Modesto, 1995; Kammerer, 2011).

In most Anomodontia (15 in Fig. 5), the infratemporal

fenestra lateromedially expanded and a distinct ventral exca-

vation formed in the cheek region, overall resulting in a fossa-

fenestral morphotype (e.g. Ray, 2005; Sullivan & Reisz, 2005).

Comparable to a ‘true’supratemporal fenestra, the jugal no

longer contributes to the temporal fenestra. Instead, it is only

bordered by the postorbital, squamosal, and parietal

(Sullivan & Reisz, 2005). The infratemporal fenestra in early

Eutheriodontia (16 in Fig. 5) expanded in a similar manner

(e.g. Kemp, 1984), leading to a suprafenestral morphotype.

However, the non-expanded fenestra in early anomodon-

tians (Angielczyk, 2004; Cisneros et al., 2015), as well as in

the possible Eutheriodontia sister-clade Gorgonopsia

(Gebauer, 2007), argue for an independent evolution of the

lateromedially expanded fenestra. Crownward, the temporal

fenestra often became conﬂuent with the orbit by loss of the

postorbital (e.g. Kemp, 1984). Consequently, this orbitotem-

poral opening in most mammals and their nearest relatives is

bordered by the jugal, squamosal, parietal, and frontal.

Arguably, this orbitotemporal opening may be referred to

as a temporal excavation, making this a suprafossal morpho-

type (17 in Fig. 5), or even bifossal if the zygomatic arch is dis-

tinctly dorsally bended to form a ventral excavation that

occupies more than 30% of the temporal height (see

Section II.H). In some mammals, the zygomatic arch became

conﬂuent with the braincase (18 in Fig. 5), effectively forming

ascutal morphology (Macrini, 2004), although functionally

with little similarity to other scutal tetrapods (Murray, 1981).

The ancestral condition in Reptilia may have been scutal

(e.g. Müller & Reisz, 2006; Ford & Benson, 2020), however,

this is highly dependent on the nesting and intrarelationships

of Parareptilia (Piñeiro et al., 2012; Cisneros et al., 2021). The

earliest parareptiles were likely infrafenestral (20 in Fig. 5; Cis-

neros et al., 2021; but see the controversy surrounding Meso-

sauridae in Section III.1e). Later, also infrafossal (19 in Fig. 5;

Gow, 1972; Tsuji, Müller & Reisz, 2010), scutal (21 in Fig. 5; -

Lee, 1997), and additofenestral (Haridy et al., 2016) forms

appeared. In contrast to synapsids and later reptiles, the

infratemporal fenestra in infrafenestral parareptiles often forms

between the jugal, squamosal, and quadratojugal (Broom,

1913; DeBraga & Reisz, 1996; Tsuji, 2006), probably indi-

cating independent evolution of this temporal opening. Pro-

colophonoids have large orbitotemporal openings (Colbert,

1946; Zaher, Coram & Benton, 2019). In some procolopho-

noids, the infratemporal fenestra can co-occur with a ventral

excavation in the ‘cheek’(Cisneros et al., 2004).

Independent of the current hypotheses on parareptile

interrelationships, Diapsida likely emerged from a scutal

ancestor (22 in Fig. 5; Müller & Reisz, 2006; Ford &

Benson, 2020). The ancestral condition of diapsids may have

Society.

2242 Pascal Abel and Ingmar Werneburg

been bifenestral (23 in Fig. 5; Reisz, 1977; Ford & Benson,

2020). However, the fossafenestral condition in several other

early diapsids (Carroll, 1981; Bickelmann, Müller & Reisz,

2009), as well as the different bone conﬁguration in later bife-

nestral taxa (Müller, 2003), could also argue for an ancestrally

fossafenestral condition in Diapsida (Evans, 2008) or would at

Fig. 5. Composed and simpliﬁed phylogenetic tree of Tetrapoda, depicting the distribution of the temporal morphotypes proposed in

this review. Relationships that are currently controversial (early Tetrapoda, early Amniota, Parareptilia, late Diapsida) are depicted as

polytomies. Note that the colours depict the presence of one representative morphotype within a clade but do not indicate its relative

abundance within the respective clade. Skulls are listed as in legend to Fig. 2, unless otherwise indicated below. (1) Acanthostega gunnari;

(2) Eryops megacephalus (after Sawin, 1941); (3) Anthracosaurus russelli;(4)Llistrofus pricei;(5)Eocaecilia macropodia (after Jenkins et al., 2007);

(6) Microcaecilia iwokramae;(7)Scolecomorphus sp.; (8) Cryptobranchus alleganiensis;(9)Pleurodeles walti; (10) Bombina orientalis;

(11) Calyptocephalella gayi; (12) Gastrotheca galeata (after Paluh et al., 2020); (13) Limnoscelis paludis;(14)Cotylorhynchus romeri;(15)Lystrosaurus

murrayi; (16) Cynognathus platyceps;(17)Zalambdalestes lechei; (18) Tachyglossus aculeatus; (19) Milleropsis pricei; (20) Bolosaurus striatus;

(21) Scutosaurus karpinskii;(22)Captorhinus aguti;(23)Petrolacosaurus kansensis;(24)Claudiosaurus germaini; (25) Temnodontosaurus trigonodon;

(26) Clevosaurus hudsoni;(27)Iguana iguana; (28) Argyrogena fasciolata;(29)Henodus chelyops (after Rieppel, 2001); (30) Keichousaurus hui (after

Holmes et al., 2008); (31) Pliosaurus kevani;(32)Proganochelys quenstedti;(33)Emydura maquarii; (34) Pelodiscus sinensis;(35)Testudo graeca;

(36) Erythrosuchus africanus;(37)Postosuchus kirkpatricki;(38)Thalattosuchus superciliosus; (39) Gallus gallus.

Society.

Morphology of the tetrapod temporal skull region 2243

least imply secondary evolution of the bifenestral morphotype

in these forms (26, 36, 38 in Fig. 5). In fact, Ichthyosauromor-

pha, Lepidosauromorpha, Sauropterygia, Archosauromor-

pha, and Pantestudines likely were all ancestrally

fossafenestral (Waldman & Evans, 1994; Evans & Borsuk-Bia-

łynicka, 2009; Nesbitt, 2011; Neenan, Klein & Scheyer,

2013; Motani et al., 2015; Zhou et al., 2017) or at least infrafos-

sal in the case of Pantestudines (Li et al., 2008; Bever et al.,

2015). Nevertheless, the supratemporal fenestra would have

been ancestrally present in either scenario (Müller, 2003;

Evans, 2008; Ford & Benson, 2020), originally likely bor-

dered by the postorbital, squamosal, parietal, and supratem-

poral (Reisz, 1977). Its size (e.g. Nicholls, 1999; Benson et al.,

2013), as well as the contributions of the supratemporal, fron-

tal, and postfrontal varied in later representatives

(e.g. Müller, 2003). Additionally, the temporal opening in

the ‘cheek’is closed in several extinct diapsids (Kuhn-Schny-

der, 1967; Tarsitano, 1983; Reisz, Berman & Scott, 1984),

most notably in Ichthyosauria (25 in Fig. 5) and various Saur-

opterygia (31 in Fig. 5), leading to a suprafenestral morphotype.

The fossafenestral state was retained in most Lepidosauria

(27 in Fig. 5; Evans, 2008), although in Rhynchocephalia

(26 in Fig. 5; Whiteside, 1986; Jones, 2004), as well as a few

Squamata (Mo, Xu & Evans, 2010), the bifenestral state reap-

peared. In Scincoidea, the supratemporal fenestra is closed

(infrafossal; Gaupp, 1895a), whereas in Ophidia (28 in

Fig. 5) and some other squamate clades, large portions of

the temporal dermatocranium have been reduced, forming

anudital morphotype (e.g. Das et al., 2019).

Testudinata are ancestrally scutal (32 in Fig. 5), however

with morphological differences from the scutal ancestors of

diapsids (e.g. Jaekel, 1909a; Rieppel & deBraga, 1996;

Müller, 2003; Werneburg, 2012). This condition formed

likely by closure of the supratemporal fenestra, followed by

closure of the ventral excavation (Werneburg, 2015;

Schoch & Sues, 2018). In Testudines, distinct ventral and

posterodorsal excavations formed in the temporal dermato-

cranium (Gaffney, 1979; Werneburg, 2012). Pleurodira usu-

ally possess a large ventral opening (infrafossal; 33 in Fig. 5),

whereas in Cryptodira the posterodorsal opening is often

dominant (suprafossal; 34 in Fig. 5). Sometimes both the ven-

tral and posterodorsal excavations can be enlarged, leaving

only a narrow temporal bridge between them (bifossal;

Gaffney, 1979) or even become entirely conﬂuent (nudital;

Gaffney, 1979). The set of dermal bones that contribute to

the testudine temporal bridges varies considerably

(Werneburg, 2012). Notably, the scutal morphotype reap-

peared in Testudines, especially in Chelonioidea (Jones

et al., 2012).

The ancestral fossafenestral state in Archosauromorpha

evolved twice independently into a bifenestral condition by clo-

sure of the lower temporal bar: in late Rhynchosauria

(Benton, 1983), as well as in Archosauriformes (36 and

38 in Fig. 5), where it became the dominant morphotype in

all non-avialan taxa (Nesbitt, 2011). In some early archo-

sauriforms, the infratemporal fenestra was closed again

(suprafenestral; Heckert et al., 2012), whereas in some

Caimaninae the supratemporal fenestra is closed instead

(infrafenestral; Mook, 1921). In some early Pseudosuchia

(37 in Fig. 5), an additional pair of infratemporal fenestrae

even appeared (additofenestral; Sulej, 2005; Weinbaum,

2011). A similar arrangement formed in Tyrannosauridae

by subdivision of the previous infratemporal fenestra by a

bony process (e.g. Carr, 1999). Finally, within

Ornithothoraces, the supratemporal fenestra was closed

and the lower temporal bar disappeared (e.g. Jollie, 1957;

O’Connor & Chiappe, 2011; Field et al., 2018), leading to

the typical infrafossal morphotype of birds (39 in Fig. 5).

Within Neognathae, the upper temporal bar reformed sev-

eral times independently (fossafenestral; Elzanowski &

Mayr, 2017).

(e)Unresolved controversies

In spite of moving away from using temporal morphology as

a major trait in phylogenies, ongoing debates remain regard-

ing the temporal morphology of phylogenetically unstable

taxa and on the ancestral condition for some radiations. As

indicated in Section III.1d, such debates surround the ances-

tral condition in early amniotes, as well as the evolution of the

turtle skull.

While there is mostly a consensus on the content of early

Synapsida and early Reptilia (e.g. Müller & Reisz, 2006;

Benson, 2012; MacDougall et al., 2018; Spindler et al.,

2018; but see also Laurin & Piñeiro, 2017, 2018; Ford &

Benson, 2020), the relationships of some Paleozoic tetra-

pod groups relative to Amniota are subject to debate. For

example, the Diadectomorpha, traditionally seen as the

sister taxon to Amniota (e.g. Laurin & Reisz, 1995, 1997;

Lee & Spencer, 1997; Reisz, 1997; Laurin & Piñeiro,

2017; Ford & Benson, 2020), also have been repeatedly

argued to nest within the amniote crown-group as sister

to synapsids (e.g. Berman, Sumida & Lombard, 1992;

Sumida, Lombard & Berman, 1992; Berman, 2000,

2013; Marjanovic & Laurin, 2019; Klembara et al.,

2019). Similarly, Recumbirostra, a potential clade of

‘microsaurs’and some other lepospondyls has been pro-

posed actually to represent one of the earliest reptilian

radiations (Pardo et al., 2017; Mann, Pardo & Maddin,

2019). In fact, the monophyly and phylogenetic position

of lepospondyls within Tetrapoda is still controversial

(e.g. Marjanovic & Laurin, 2019). The potential nesting

of diadectomorphs and recumbirostrans within Amniota

would have implications for the ancestral condition and

early evolution of the temporal region in the amniote

crown-group. Like many other Paleozoic tetrapods, dia-

dectomorphs typically had a fully roofed dermatocranium

(Kissel, 2010). If they are indeed the sister-clade to synapsids,

the infratemporal fenestra in synapsids could have evolved

later, after the split of the synapsid-diadectomorph clade from

Reptilia, or the fully roofed (scutal) morphotype in diadecto-

morphs could have been a derived condition from a fenestrated

ancestor. Piñeiro et al. (2012) argued that the possession of an

infratemporal fenestra could be the ancestral condition in

Society.

2244 Pascal Abel and Ingmar Werneburg

amniotes. However, this would be dependent on the phyloge-

netic position and temporal morphology in Mesosauridae (see

below). Lastly, nesting recumbirostrans within reptiles would

add another set of temporal morphotypes to the early morpho-

logical diversity of Reptilia, including excavations and distinct

reductions of the dermatocranium (e.g. Pardo & Anderson,

2016; Mann et al., 2019).

A second major discussion relates to the presence of an

infratemporal fenestra in the Permian reptile taxon Meso-

sauridae. Despite an uncertain phylogenetic history [see

Modesto (2006) and references therein], mesosaurids are

usually seen as an early-diverging branch of parareptiles

(e.g. Tsuji & Müller 2009; Ford & Benson 2020), or taking

an even more stemward position within reptiles

(e.g. Laurin & Piñeiro, 2017). Thus, they could be a key to

reconstructing the ancestral temporal morphotype for Para-

reptilia, and potentially also for Reptilia or Amniota as a

whole (Piñeiro et al., 2012). Previously interpreted to possess

both an infratemporal and supratemporal fenestra

(MacGregor, 1908), later work (von Huene, 1941) led to a

new consensus that only the infratemporal fenestra was pre-

sent. This changed with Modesto (2006), who redescribed

the cranial anatomy of mesosaurids, interpreting them to

possess a fully roofed temporal region. By contrast, Piñeiro

et al. (2012) reported the infratemporal fenestra to be present

in the specimens they described and even to be demonstrable

in disarticulated remains due to the shape of the jugal and

postorbital. Their interpretation has since been disputed

[MacDougall & Reisz, 2014; MacDougall et al., 2018; but

see Laurin & Piñeiro (2018) for a response]. It has been also

suggested that the infratemporal fenestra may have been

ontogenetically (MacDougall et al., 2018) or intraspeciﬁcally

variable, given that both scenarios have been demonstrated

for other parareptiles (Gow, 1972; Cisneros, 2008; Haridy

et al., 2016). Despite being able to examine the mesosaurid

material of the Senckenberg collections in Tübingen and

Frankfurt am Main, Germany, we remain unable to agree

conﬁdently with either hypothesis. At present, there seems

to be no consensus on mesosaurid phylogeny and temporal

morphology. Their highly aquatic lifestyle (Silva et al.,

2017) may also mean that their morphology is too derived

to be able to draw conclusions regarding the ancestral cranial

morphotype in amniotes.

Another controversial topic in terms of the evolution of the

temporal region remains the ancestry and phylogenetic posi-

tion of turtles. Traditionally depicted as possessing an ances-

tral fully roofed temporal region (e.g. Cope, 1892; Williston,

1917; Boettger, 1952; Kilias, 1957), many early researchers

highlighted the derived morphology of the turtle skull rela-

tive to the condition in early tetrapods (e.g. Baur, 1889;

Fuchs, 1909; Jaekel, 1909a; Zdansky, 1923–1925). Different

phylogenetic analyses have nested turtles deep within Dia-

psida and, hence, implied their descent from ancestors with

a fenestrated temporal region (e.g. Rieppel & deBraga,

1996; DeBraga & Rieppel, 1997; Bhullar & Bever, 2009;

Schoch & Sues, 2015). An example of a fenestrated stem-

turtle possibly has been found: the description of the Triassic

diapsid Pappochelys rosinae suggests that it had a supratemporal

fenestra and distinct ventral excavation in the ‘cheek’region

(Schoch & Sues, 2015, 2018). An even more stemward stem-

turtle may be the Permian Eunotosaurus africanus (Lyson et al.,

2010; Bever et al., 2015; Schoch & Sues, 2015). Bever

et al. (2015) argued that in juvenile specimens of E. africanus,

in addition to the distinct ventral excavation observed for

assumed adults, a supratemporal fenestra was present. The

supratemporal fenestra would have been closed during

ontogeny by the extreme anterior, autapomorphic expansion

of the supratemporal and hence may perhaps represent an

early stage of the condition in turtles. However, we consider

this scenario as unlikely, because such an extraordinary bone

growth in post-hatching ontogeny is extremely rare among

amniotes, and when present usually concerns multiple bones

aligning to each other (e.g. Hall, 2014). For closure, even sep-

arate ossiﬁcation inside the upper temporal opening may

occur (Klembara et al., 2017). However, we argue that the

juvenile presented by Bever et al. (2015) rather shows a taph-

onomic disruption than a preservation of an early ontoge-

netic state of fenestral development. Compared to the adult

condition, the jugal is broken off the postorbital and unnatu-

rally reaches into the ventral temporal excavation in the juve-

nile suggesting that some pressure on the snout travelled

along the maxilla to break the jugal. Before breakage, the

force was likely further transmitted via the postorbital

towards the dorsal temporal region and the scute-like supra-

temporal (Bever et al., 2015) was likely spalled off. A precise

description and reconstruction of the juvenile bones, by

which a premature and unbroken supratemporal could be

discovered would, however, support the hypothesis of Bever

et al. (2015). It is worth noting that E. africanus could also rep-

resent a species of parareptile or synapsid, not closely related

to turtles (Lee, 1995, 2013; Tsuji & Müller, 2009; MacDou-

gall & Reisz, 2014; Lichtig & Lucas, 2021); correspondingly,

P. rosinae has been also argued by some to nest outside of the

turtle stem-group (Lichtig & Lucas, 2021).

Ford & Benson (2020) provided the most recent assess-

ment of the phylogenetic implications of temporal fenestra-

tions on the relationships of early amniotes. In contrast to

most previous workers (although see Laurin & Piñeiro,

2017), they found parareptiles as well as the traditional syn-

apsid group Varanopidae to nest within Diapsida. This

would have major implications for the ancestral condition

in diapsids as, if varanopids and parareptiles are early-

branching diapsids, the infrafenestral morphotype could

either have been the ancestral state of all diapsids (similar

to synapsids) or evolved secondarily by closure of the supra-

temporal fenestra. This would mean there could have been

diverse temporal morphotypes around the base of Diapsida

with taxa possessing both temporal fenestrae (Reisz, 1977;

Müller, 2003; Ford & Benson, 2019), with a secondarily

closed infratemporal fenestra (Reisz et al., 1984), with a sec-

ondarily closed or never evolved supratemporal fenestra

(Tsuji & Müller, 2009; Piñeiro et al., 2012; MacDougall &

Reisz, 2014), with a supratemporal fenestra and distinct ven-

tral excavation (Carroll, 1981; Modesto & Reisz, 2003), or

Society.

Morphology of the tetrapod temporal skull region 2245

with a ventral excavation alone (Tsuji et al., 2010; MacDou-

gall & Reisz, 2014).

(2) Understanding the functional origins of temporal

openings

Early in the research history of the temporal region, investi-

gators not only described the morphological patterns they

observed, but also attempted to explain how the vast struc-

tural diversity in lissamphibian and especially in amniote

skulls had arisen. For example, Hallmann (1837) already

emphasized the functional background to the observed tem-

poral anatomy.

(a)Comprehending diversity (1880s–1900s)

Dollo (1884) compared the temporal anatomy of non-avian

dinosaurs and other extinct taxa with extant groups to pre-

dict the size of different jaw adductor muscles. He argued

that dominance of either the internal or external jaw muscu-

lature would affect the feeding mode of a taxon. Dollo (1884)

also argued that dominance of internal jaw adductors in cro-

codylians and sauropods was related to the posterior move-

ment of the choanae or external nares, respectively,

enabling more space for these muscles in the anterior palate.

Dollo (1884) saw this as the reason for the small size of the

supratemporal fenestra in crocodylians compared with

the enlarged fenestra in extinct marine crocodiles, with the

latter having their choanae positioned more anteriorly.

Gaupp (1895b) related the transition of a closed, ‘stegocro-

taph’dermatocranium to a fenestrated, ‘zygocrotaph’

appearance to the presence of hydrostatic forces acting on

the skull. According to Gaupp (1895b), the lack of external

hydrostatic forces in terrestrial vertebrates as well as the

increased need to counter gravitational forces, favoured a

reduction of skull mass. The remaining temporal bars in

the ‘zygocrotaph’skull would then be aligned and strength-

ened or weakened depending on the forces applied by the

jaw musculature. Gaupp (1895b) interpreted the ‘gymnocro-

taph’morphology (i.e. a temporal region, in which the tem-

poral bars are reduced) as the result of limited forces acting

on the temporal bones.

Gegenbaur (1898) also provided an in-depth discussion of

the relationships between jaw musculature and temporal mor-

phology. He was probably the ﬁrst explicitly to interpret the

evolution of temporal openings as an adaptation to provide

attachment sites for the jaw muscles. Particularly with refer-

ence to mammals, Gegenbaur (1898) made a connection

between a more ‘massive’dentition and stronger jaw muscula-

ture, which would inﬂuence the morphology of the zygomatic

arch and parietal crests relative to the feeding strategy of any

taxon. By contrast, a reduction of the dentition in insectivorous

taxa like Tachyglossus would allow a corresponding reduction of

the zygomatic arch and parietal crest.

Fuchs (1909) argued that simpliﬁcation of the dermal cov-

ering of the temporal region would not inevitably lead to the

evolution of temporal openings. Fuchs (1909) postulated that

the relative proportions of the braincase and otic capsules as

well as the evolution of a streptostylic jaw (i.e. with a mobile

quadrate) were major factors explaining the observed varia-

tions in the temporal region. Fuchs (1909) described how

the dermal bones of the temporal region can extend inter-

nally to the jaw adductor musculature (internal lamella) with

the potential to replace the primary cartilaginous braincase

of the embryonic skull (as seen in mammals and turtles; Wer-

neburg & Maier, 2019). With changes in the size of the brain-

case, the size of the internal lamellae and hence temporal

anatomy would also have been affected. Also affecting the

size of the internal lamellae would be the size of the otic cap-

sules, which Fuchs (1909) argued would ancestrally have

been large but occupied less space in the interior cranium

in later taxa. Lastly, Fuchs (1909) correlated the presence of

‘monimostyly’versus ‘streptostyly’with the extent of the tem-

poral dermatocranium. A fully closed temporal region or rig-

idly sutured temporal bars would create a ﬁxed quadrate

and, hence, monimostylic jaw, whereas a reduction of a rigid

temporal coverage would have enabled the evolution of a

streptostylic jaw.

(b)The interplay between muscle and bone (1910s–1950s)

Gregory & Adams (1915) considered the relationship

between temporal morphology and jaw musculature. They

observed that in extant taxa like Sphenodon and turtles the

jaw adductor musculature is attached to the sagittal crest

and/or to temporal bars. Gregory & Adams (1915) argued

that the temporal openings provide more space for action

of the temporal musculature and concluded that in extinct

taxa with a closed dermatocranium the temporal muscula-

ture had to be positioned medially to the dermal bones. Stres-

ses induced by the jaw and neck musculature were thus partly

responsible for modiﬁcations of temporal osteology in later

taxa. Adams (1919) extended these ideas to the relationships

between osteology, diet, and myology, using evidence from

numerous extant taxa and embryology.

Versluys (1919) discussed in detail how different actions of

the internal and external jaw adductor musculature could be

related to the evolution of different temporal morphotypes.

Versluys (1919) argued that in early aquatic vertebrates with

a fully closed temporal region, greater muscle movement

could be achieved by simply enlarging the jaw adductor

chamber. Further evolution of the neck and increased mobil-

ity of the tetrapod skull would have favoured the evolution of

a lighter and narrower skull. This would have restricted the

possible action of the jaw adductor musculature, which could

have been offset by opening the temporal dermatocranium.

Versluys (1919) argued that the best way to create temporal

fenestrae would have been the loss of the intertemporal bone

in amniotes or degeneration between the sutures of the jugal,

squamosal, and quadratojugal. Depending on the taxon, the

temporal fenestrae also could have developed at other sites in

the temporal region. Versluys (1919) interpreted the selection

pressure involved to be differences in feeding mode, which

would require different proportions of internal and external

Society.

2246 Pascal Abel and Ingmar Werneburg

adductor musculature, depending on whether a high biting

force or rapid closure of the jaw was required.

Zdansky (1923–1925) was interested in the temporal mor-

phology of turtles, which he considered to be more diverse

than in other vertebrates. Similar to Jaekel (1909a, 1916),

Zdansky (1923–1925) interpreted the fully roofed temporal

dermatocranium of modern turtles to have evolved second-

arily in forms that could not, or could only partly, retract

their head under the shell (e.g. in the sea turtles, Platysternon,

Macrochelys, and Podocnemis). A similar conclusion had been

drawn by Hay (1908). Zdansky (1923–1925) interpreted the

return to a fully roofed dermatocranium in these taxa as a

defensive adaptation to compensate for the lack of

a retractable neck.

Inspired by observations on postcranial bones, Case (1924)

investigated the role of muscle action on the formation of

temporal openings. In the postcranial bones of humans, such

as the scapula or ilium, the associated muscles are attached to

thickened margins of the respective bones, whereas the cen-

tres of the bones, which experience little direct loading, are

distinctly thinned. According to Case (1924), this pattern

represented an adaptation to greater muscle force acting on

the marginal regions of the bones, triggering the develop-

ment of additional trabecular bone in these areas. Applying

this logic to the temporal region, Case (1924) argued that in

forms with a closed dermatocranium, the load would have

been concentrated on marginal sections of the temporal

bones, which would thicken in response, whereas sections

subject to less loading could be thinner and subsequently

reduced, forming a skull with temporal fenestrae and thick-

ened temporal bars.

Lakjer (1926) considered in depth the functional differ-

ences in the jaw muscles, the associated bones, and skull kine-

sis of reptiles (‘Sauropsida’) and other vertebrates.

Lakjer (1926) particularly highlighted the relationship

between muscle size and bite force for non-ophidian squa-

mates [‘Sauria’in Lakjer (1926)], including taxa that were

herbivores, omnivores, or predators of large prey. He also

observed a relationship between muscle size and temporal

dermal armour in lissamphibians, as demonstrated in the dif-

ference in size of the jaw muscles between fully roofed caeci-

lians and batrachians with heavily reduced temporal dermal

armour. He hypothesized that the fully roofed condition of

some anurans had evolved secondarily.

To explain the different temporal morphotypes present in

turtles, Kilias (1957) also argued that the load acting on the

cranial bones could explain the formation of temporal exca-

vations. He suggested that turtles lost the functional require-

ment for a roofed dermatocranium following the rigid

attachment of the quadrate to the palate and braincase.

Kilias (1957) interpreted the temporal bridges as remnants

of the dermatocranium that function as attachment sites for

the external jaw musculature. In agreement with

Zdansky (1923–1925), Kilias (1957) argued that the evolu-

tion of neck retraction was another important factor in turtle

skull evolution. He described a trend of decreasing size of the

anterior aperture of the shell as an anti-predation adaptation,

which simultaneously would have required a decrease in size

of the skull which he suggested was achieved by reduction of

the parts of the skull with the least functional importance.

The hypotheses of Zdansky (1923–1925) and Kilias (1957)

have recently inspired a number of modern analyses

(Werneburg, 2012, 2015; Ferreira et al., 2020).

(c)Biomechanical studies (1960s–1980s)

Olson (1961) analysed the morphofunctional differences

within the sarcopterygian jaw system and how these relate

to skull morphology. Olson (1961) proposed that the different

jaw systems can be broadly divided into a ‘kinetic inertial sys-

tem’(K–I) and a ‘static pressure system’(S–P). In the K–I

system, force is mostly applied by the kinetic energy gener-

ated by rapid and vertical jaw closing. In the S–P system,

the highest force is applied when the jaws are almost closed

and involves a multidimensional array of jaw movements.

According to Olson (1961), non-tetrapod Sarcopterygia as

well as early Tetrapoda ancestrally possessed a simple K–I

system that only enabled vertical movement of the mandible.

The jaw adductor musculature would have been positioned

medially to the dermatocranium. While adaptations to the

neck, torso, and gill apparatus in early Stegocephali could

have led to rearrangement of the posterior skull region,

Olson (1961) did not ﬁnd evidence for major changes in the

jaw musculature in these forms; however, Olson (1961) pos-

tulated that with the rise of Tetrapoda, the evolution of sev-

eral modiﬁcations of the K–I and S–P systems took place.

While early Amphibia and predominantly aquatic Amniota

evolved only a derived K–I system, Olson (1961) associated

the evolution of the S–P system in amniotes with doming of

the skull and a tendency to develop temporal openings as well

as other osteological adaptations for muscle attachments.

Comparable to the conclusions of Versluys (1919) and

Case (1924), Olson (1961) argued that the terrestrial lifestyle

of early amniotes would have been accompanied by higher

mobility of the head and changes in feeding habits, which

inﬂuenced the origin sites of the jaw musculature: dermato-

cranium strengthening was concentrated on the few func-

tional attachment sites for muscles, whereas the regions of

the dermal roof without a function in muscle attachment

were reduced. This morphology, Olson (1961) described as

a‘network of lines of stress’.

The notion of ‘lines of stress’to explain the evolution of

temporal openings was developed further by Fox (1964),

who observed that in the early Permian reptile Captorhinus,

the centre of the fully covered ‘cheek’region was thinned.

Similar observations had been made by Jaekel (1902) for

the possible reptiliomorph Gephyrostegus bohemicus, who pro-

posed that they represented an early stage in the evolution

of fenestrated morphotypes. Fox (1964) also interpreted this

morphology to result from lower levels of stress in the cheek

centre. As these areas apparently did not function as muscle

attachment sites, and additionally were not needed as sources

of calcium for attached muscles, there would be no selection

against reduction of the bone. Fox (1964) considered all

Society.

Morphology of the tetrapod temporal skull region 2247

possible stresses affecting the ‘cheek’region, including forces

arising from the weight of the anterior part of the cranium,

jaw articulation, and muscle action. Fox (1964) also

highlighted that selective pressures leading to the develop-

ment and enlargement of temporal openings were likely to

be multidimensional and to differ among various clades.

For the further enlargement of temporal openings in later

amniotes, Fox (1964) proposed that additional space for the

jaw adductor muscles and a reduction in weight of the skull

were likely to be the main factors involved.

Frazzetta (1968) also argued that selection for the initial

development of a temporal opening could differ from selection

for retaining or enlarging it in later forms. Nevertheless, Fraz-

zetta (1968) considered it likely that there was a common

adaptive reason for the development of temporal openings in

all clades. Frazzetta (1968) suggested that the lack of temporal

openings in early tetrapods and their ancestors could be

explained by their plesiomorphically ﬂat skull. The main jaw

action in these skulls would involve the pterygoid musculature,

which would create a predominantly tangential force affecting

the temporal region [i.e. the K–I system Olson (1961)]. In

most amniotes and other tetrapods with a more domed skull,

the main jaw action instead would arise from the jaw adductor

musculature attached tothe skull roof, creating predominantly

perpendicular forces. Frazzetta (1968) emphasized the

restricted attachment site of these muscles, which correlated

with the development of medial bony ridges as described by

Fox (1964) for Captorhinus. These ridges, representing the pre-

viously described ‘network of lines of stress’, thus functioned to

strengthen the skull at the muscle attachment sites. Another

factor affecting the restricted attachment area of the jaw mus-

culature within the adductor chamber might be the pennate

structure of the jaw adductors, which is related to more efﬁ-

cient adduction of the mandible. Frazzetta (1968) saw the

restricted attachment of the jaw adductors as a reason for

the development of fenestrae in less-loaded areas of the skull.

Frazzetta (1968) supported this argument with skulls of the

early synapsid Ophiacodon retroversus, in which some specimens

have a second pair of infratemporal fenestrae between the

jugal, squamosal, and quadratojugal (i.e. additofenestral). When

this ‘accessory temporal opening’was not present, or was pre-

sent on only one site of the cranium, the area between the

three bones was still ‘paper thin’. Romer & Price (1940) had

argued there would have been no disadvantage to leaving this

suture open, indeed jaw adductor action may even have

beneﬁted from the additional free space, and that the develop-

ment of large fenestrae with thickened margins (i.e. bony

ridges) would provide sites for a more concentrated attach-

ment of the jaw adductor musculature. Frazzetta (1968) fur-

ther suggested that the meeting points of three dermal bones

would favour the development of fenestrae where they were

not subjected to loading. Finally, Frazzetta (1968) argued that,

in ﬂat-skulled tetrapods like temnospondyls, the low vertical

resistance and muscle attachment without the

development of bony ridges would oppose the formation of

temporal fenestrae. Even in more domed tetrapods without

temporal fenestration, the skull roof was only connected to

the cheek region by the remnant of an intracranial joint.

Development of fenestrae in these taxa, according to Fraz-

zetta (1968), would have signiﬁcantly weakened the skull.

Yet, Frazzetta (1968) did acknowledge that this condition

may be present in some fenestrated amniotes such as the early

synapsid Varanosaurus.IncontrastwithbothGregory&

Adams (1915) and Fox (1964), Frazzetta (1968) disagreed that

bulging of the jaw musculature or conservation of structural

materials could explain the formation of temporal openings,

as the earliest temporal fenestrae would have been too small

to function in this way. [Correction added on 8 June 2021

after ﬁrst publication: in the fourth sentence of the preceding

paragraph, the citation to Fox (1964) has been corrected to

Olson (1961)]

Like previous authors (Fuchs, 1909; Kuhn-Schnyder,

1954, 1963), Gow (1972) noted that similar traits may have

evolved independently in distantly related taxa, even though

for similar reasons. He disagreed with Fox (1964) that the

weak zone in the ‘cheek’of Captorhinus could be the predeces-

sor of a temporal fenestra. Instead, he argued that the Perm-

ian parareptilian group Millerettidae may be key to

understanding the evolution of temporal fenestration in rep-

tiles. Gow (1972) demonstrated that in juveniles of Milleretta

rubidgei a small infratemporal fenestra was present that disap-

peared completely in adults, mostly due to an anterior exten-

sion of the squamosal. He explained this by contraction/

expansion in the ‘cheek’induced by expansion of the palate

at the basicranial articulation (i.e. the articulation between

the pterygoid and braincase). According to Gow (1972), this

could represent a condition derived from the intracranial

kinetics of anthracosaurs, which could likely move their

braincase and skull roof relative to their palate and ‘cheek’

(Thomson, 1967). In contrast to the typical sarcopterygian

intracranial joint between ‘cheek’and skull roof found in

the putatively reptiliomorph anthracosaurs, the squamosal

of M. rubidgei was ﬁrmly attached to the skull roof. This would

result in a ‘line of weakness’between the jugal, squamosal,

and quadratojugal on whose dorsal termination the temporal

fenestra was formed. Gow (1972) noted that widening of the

temporal region would have been a signiﬁcant precondition

to allow such a development in M. rubidgei. Further reduc-

tions of the temporal dermatocranium (i.e. the formation of

a ventral excavation) in other milleretids, especially Millerop-

sis, would have resulted in even greater intracranial mobility.

Notably, Gow (1972) highlighted that temporal fenestrae in

the Permian Youngina and later diapsids were likely to have

different causes.

Over the course of the 20th century, a growing consensus

on the relationship between temporal openings and jaw

adductor musculature inspired several studies on jaw biome-

chanics in extinct and extant taxa (e.g. Crompton, 1963;

Barghusen, 1973; Kemp, 1969, 1980, 1984; Rieppel &

Gronowski, 1981). The notion that temporal fenestrae func-

tioned as an adaptation to allow bulging of the jaw muscula-

ture (see Gregory & Adams, 1915) fell out of favour.

Kemp (1980) proposed that the infratemporal fenestra in

early synapsids did not evolve from a skull similar to that of

Society.

2248 Pascal Abel and Ingmar Werneburg

early Eureptilia (‘Romeriidae’), but instead from a type more

similar to the diadectomorph Limnoscelis. Kemp (1980)

highlighted that the appearance of supratemporal and tabu-

lar bones in synapsids is comparable to the condition in Lim-

noscelidae. Like other early tetrapods, Limnoscelis retained a

loose connection between ‘cheek’and skull roof as a remnant

of the sarcopterygian skull with its intracranial joint. Accord-

ing to Kemp (1980), the synapsid infratemporal fenestra

could have formed by the adductor musculature attaching

to the bony margins of the intracranial joint, which was then

strengthened by the processes described above. The fenestra

would subsequently form in the connecting area between the

‘cheek’bones and skull roof.

Kuhn-Schnyder (1980), agreeing with Frazzetta (1968)

that temporal fenestrae predominantly form in the sutures

between three bones, identiﬁed the jugal–squamosal–

quadratojugal and postorbital–squamosal–parietal contacts

as key regions. While apparently considering muscle action

as the main selective force underlying fenestration, Kuhn-

Schnyder (1980) also argued for a role of the atlanto-occipital

joint (see also Versluys, 1919). Elongation of the neck and

higher mobility of the head would require a lighter skull

and hence favour the formation of fenestrae. He cautioned

that such biomechanical causes would make taxonomic clas-

siﬁcations based on temporal morphology less reliable, but he

considered temporal morphotypes to be mostly constant

within single reptilian groups.

Carroll (1982) interpreted early synapsids predominantly as

macropredators with the relatively short jaw adductor chamber

enabling a wider jaw gap, whereas early diapsids were insecti-

vores in which the presence of temporal fenestrae would have

created a lighter skull and the temporal bars formed as areas

of maximal resistance against forces generated during feeding

(see also Evans, 2008). Similar ideas had been proposed by Ver-

sluys (1919), who interpreted temporal differences between

early Synapsidaand Diapsida in termsof their different feeding

ecology requiring a different arrangement of the jaw muscula-

ture. Despite the presence of two pairs of temporal fenestrae in

early diapsids, Carroll (1982) argued that their jaw mechanics

and muscle distribution would have been similar to their non-

fenestrated reptilian ancestors.

The loss of the lower temporal bar in Lepidosauria and a

variety of other fossil taxa was discussed by several

researchers (Robinson, 1973; Evans, 1980; Rieppel &

Gronowski, 1981; Whiteside, 1986). They generally dis-

agreed that this loss had a functional relationship with the

evolution of streptostyly as proposed by previous researchers

(Fuchs, 1909; Romer, 1956). Based on an incomplete lower

temporal bar and ﬁxated quadrate in the early rhynchoce-

phalian Clevosaurus hudsoni, Robinson (1973) argued that this

may be an auditory adaptation, separating the quadrate

from the lower temporal bar to reduce interference with

the ears during feeding. Evans (1980) disagreed with this

interpretation, because action of the dentulous palatine and

stretching of the post-quadrate tympanic membrane by the

mandible would still affect hearing ability during feeding,

making any effect of a jugal–quadrate separation negligible.

She also suggested that this and similar excavations of the

‘cheek’bones were not related to quadrate movement and

that skulls with a mobile quadrate represented a derived con-

dition relative to skulls lacking the lower temporal bar

(in squamates) or in which the lower temporal bar was pre-

sent (in birds). Evans (1980) favoured a relationship with

the size and attachment sites of the jaw adductor muscula-

ture, probably making the lower temporal bar unnecessary

for muscle attachment. Rieppel & Gronowski (1981) also

did not support a relationship with streptostyly, instead pos-

tulating that loss of the lower temporal bar was a conse-

quence of differentiation of the external jaw adductor

musculature: a newly developed muscle unit extended pos-

teroventrally from the posterior temporal region to the lat-

eral mandible. Whiteside (1986) drew connections between

the weak but rapid bite that would have accompanied the

punctuating dentition in these taxa, and interpreted the reap-

pearance of the lower temporal bar in the extant tuatara as

an adaptation to ensure precise occlusion of the jaws in taxa

with a more powerful bite.

In a study on turtles, Lakjer (1926) interpreted a ligament

between the quadrate and jugal as the retention of a lower tem-

poral arch from diapsid ancestors. Werneburg (2013b), how-

ever, demonstrated that this ligament was derived from a

superﬁcial craniocervical aponeurosis and argued that adjacent

bones could not develop histologically into one consistent liga-

ment. The quadratojugal ligament would serve in turtles as ten-

sion cord (sensu Klenner et al., 2015) to buffer bite forces

(Werneburg, 2013b;seealsoJoneset al., 2012). In other amni-

otes, this ligament is either detached from the quadrate in mam-

mals (i.e. as the external masseter fascia) or from the jugal in

non-turtle reptiles (i.e. as the quadrate ligament; cf. Iordansky,

1996) and serves as a morphogenetic element to differentiate

the external jaw musculature in these taxa (Werneburg, 2013b).

(d)Quantitative modelling and other recent approaches (1990s–present)

The recent increase in computational power has enabled

researchers to perform quantitative analyses on the functional

morphology of tetrapod skulls. In addition to large-scale statis-

tical analyses of diversity patterns (e.g. Jones, 2008; Young

et al., 2010; Ferreira et al., 2020; Paluh et al., 2020), these

include muscle reconstructions and ﬁnite element analyses of

the skull, and modelling of strain distributions along the tem-

poral bones (e.g. Hylander & Johnson, 1997; Holliday &

Witmer, 2007; Lautenschlager, 2015; Lautenschlager et al.,

2017; Ferreira et al., 2020; Nabavizadeh, 2020). An in-depth

discussion of this literature is outside the scope of this review,

but these studies provide a valuable basis for future large-scale

studies on the comparative functional morphology of the tetra-

pod temporal region.

Of particular interest are the functional causes underlying

the temporal morphology of caecilians (Gymnophiona).

Extant caecilians either possess a distinct posterodorsal exca-

vation between the parietal and squamosal (‘zygokrotaphic’)

or a fully roofed dermatocranium (‘stegokrotaphic’;

e.g. Kleinteich et al., 2012; Sherratt et al., 2014). In contrast

Society.

Morphology of the tetrapod temporal skull region 2249

to earlier interpretations (Nussbaum, 1983), the ancestral

morphology of caecilians was likely ‘stegokrotaphic’

(i.e. scutal) and probably symplesiomorphic with the condi-

tion in dissorophoid temnospondyls, the putative sister-clade

to lissamphibians (Maddin et al., 2012; see Section III.1d)

with ‘zygokrotaphic’(i.e. suprafossal in caecilians) forms hav-

ing evolved several times independently in the crown-group

(Kleinteich et al., 2012). ‘Stegokrotaphy’, together with eyes

covered by bone and a subterminal mouth, has been inter-

preted as an adaptation to a fossorial lifestyle including

head-ﬁrst burrowing (Sherratt et al., 2014; Bardua et al.,

2019). ‘Stegokrotaphic’forms differ in their fossorial behav-

iour from ‘zygokrotaphic’taxa, as they are apparently better

adapted to burrow in more compact soil (Gower et al., 2004).

However, Kleinteich et al. (2012) demonstrated that the tem-

poral bones are relatively unaffected by forces assumed to act

during head-ﬁrst burrowing. They postulated instead that

‘zygokrotaphy’evolved in caecilians to provide more space

for the jaw adductor musculature. Nevertheless, even among

‘zygokrotaphic’caecilians, the alignment of the jaw adduc-

tors can vary (Nussbaum, 1983). We propose that an inﬂu-

ence of the neck musculature might be relevant if head

mobility varies among taxa (sensu Werneburg, 2015).

In another group of lissamphibians, namely anurans,

Paluh et al. (2020) inferred a relationship between a hyperos-

siﬁed cranium and dietary or defence adaptations. In

anurans, the temporal region is usually reduced to the fronto-

parietal, squamosal, and quadratojugal (e.g. Schoch, 2014b).

In several lineages, hyperossiﬁcation of the frontoparietal,

squamosal, as well as the maxilla led independently to sec-

ondary closure of the temporal region, sometimes with for-

mation of an infratemporal fenestra in the ‘cheek’(e.g.

Ceratophrys,Calyptocephalella) or a casque-like cranium

(e.g. Hemisus,Myobatrachus) (Paluh et al., 2020). In a large-scale

quantitative analysis, the authors reported that such hyperos-

siﬁcation occurs in hypercarnivorous taxa adapted to feed on

other large vertebrates and is correlated with high and wide

skulls that are also anteroposteriorly short. Paluh et al.

(2020) further argued that this may represent an adaptation

to the high forces acting on the skull during feeding. The

casque-like cranium was found in taxa capable of phragmotic

behaviour, i.e. the ability to retract the eyes within the orbital

cavity, which is likely a defensive strategy, protects the eyes

during feeding, or prevents water loss through evaporation.

Holliday et al. (2020) showed that, at least in archosaurs,

only part of the supratemporal fenestra serves as an attach-

ment site for the jaw adductor musculature. Small excava-

tions adjacent to the supratemporal fenestra, which they

referred to as the ‘frontoparietal fossa’, may instead have

contained vascular and adipose structures. Holliday

et al. (2020) argued that these structures could have func-

tioned in thermoregulation, especially of the eyes and/or

brain, or they may have supplied integumentary structures

used in display [see Carr (2020) for a discussion on Holliday

et al.’s (2020) interpretation of the frontoparietal fossa].

Werneburg (2012, 2013a, 2015, 2019) and Werneburg

et al. (2015a,b) reported a series of studies focussing on the

temporal region of turtles. Werneburg (2012) reviewed

the morphological, evolutionary, and ecological factors poten-

tially inﬂuencing temporal morphology in turtles and other

amniotes. Werneburg (2015) identiﬁed a correlation between

neck retraction in cryptodiran turtles and the formation of pos-

terior excavations in their temporalregion. He postulated that

the tensile forces generated during neck retraction would

require the skull to form deep posterodorsal emarginations

for better stress distribution. If the ancestral state in the turtle

stem-group was indeed a fenestrated skull, this would have also

resulted in closing of the temporal fenestrae to produce the

non-fenestrated condition seen in Testudinata. Indeed, the

turtle skull may have evolved in response to a complex set of

adaptations initiated by an increase in neck mobility and the

evolution of the turtle shell (Ferreira et al., 2020; Werneburg,

2020). The loss, reduction, or absence of the ability to retract

the neck could also explain the distinct reduction in temporal

excavations in many extant taxa like sea turtles, Platysternon,

or the recently extinct meiolaniids, a conclusion also drawn

in some previous publications (Zdansky, 1923–1925;

Gaffney, 1983; Jones et al., 2012).

(e)Developmental studies

The ﬁeld of evo-devo has grown considerably in recent

decades (Olsson, Hoßfeld & Breidbach, 2006), offering a per-

spective on macroevolutionary patterns often not feasible

using anatomy alone. Considering embryology, ontogenetic

changes, and the genetic mechanisms behind skull develop-

ment may be key to understanding the evolution of temporal

openings in tetrapods.

Tarsitano et al. (2001) considered mineralization of the jaw

muscle tendons as a main driver behind the evolution of tem-

poral fenestrae in amniotes. They proposed that the evolu-

tion of temporal fenestrae allowed a size increase in jaw

musculature while ensuring a low angle of bone attachment.

For muscles originally attaching to the sutures of temporal

bones, incomplete closure of these sutures would enlarge

the available attachment area for the jaw muscles

(as already argued by previous authors) while maintaining a

low angle of attachment. Over time, the developmental rela-

tionship between bone and jaw muscles would select for a

more circular shape of the fenestrae as this would maximize

the available surface for muscle attachment. A low angle of

attachment between muscle and bone is not only optimal

for biomechanical reasons but also would be beneﬁcial dur-

ing prenatal ontogeny due to the physiological need to main-

tain a large attachment area for the transport of biomolecules

from bone to tendon.

Tokita, Chaeychomsri & Siruntawineti (2013) identiﬁed a

possible relationship between temporal morphology in rep-

tiles and the distribution of mesenchymal cells expressing

Runx2 and Msx2 genes. They reported a higher distribution

of Runx2 and Msx2 expression in the lateral skull of turtle

embryos relative to a focal distribution on regions forming

the temporal bars in a crocodile embryo, and a distribution

restricted to the precursor of the braincase in a snake

Society.

2250 Pascal Abel and Ingmar Werneburg

embryo. Tokita et al. (2013) interpreted these distribution

patterns of Runx2 and Msx2 expression as a key element in

the diversity of the temporal region of amniotes.

Schoch (2014b) provided a hypothesis for the origin of

emarginated ‘gymnokrotaphic’(i.e. nudital) skulls in batra-

chians (frogs and salamanders). He argued that the cranial

osteology of adult batrachians can be compared to the larvae

of branchiosaurid temnospondyls but with a trend for succes-

sive ﬂattening of the skull that led to the ‘gymnokrotaphic’

morphotype. According to Schoch (2014b), this skull ﬂatten-

ing made it necessary for the jaw musculature to increase in

length [see also Frazzetta (1968); Section III.2c], which

would require a reduction of the ancestrally fully roofed der-

mal armour. This was achieved both by a lack of ossiﬁcation

in some of the dermal bones present in their presumed tem-

nospondylian ancestors and by early fusion of primordial

bones. The latter enabled the jaw adductors to attach dor-

sally to the frontal and parietal primordia before other der-

mal bones were formed. A dorsal attachment of the jaw

adductors has been also reported for suprafossal caecilians

(Nussbaum, 1983; Kleinteich et al., 2012). However, note

that the content of the lissamphibian stem-group, as well as

their temnospondylian afﬁnities are still debated [Schoch

(2014a) and references therein; Laurin, Lapauze &

Marjanovic (2019; Marjanovic & Laurin (2019)], and the

ossiﬁcation sequences of branchiosaurids may be different

to those of extant amphibians (Laurin et al., 2019).

Werneburg (2019) postulated ontogenetic plasticity as an

underlying cause for the high diversity in amniote temporal

anatomy. He argued that the absence of a larval stage in amni-

otes enabled the jaw musculature to attach directly to the devel-

oping temporal dermatocranium. Whereas in larvae the dermal

bones are not fully differentiated and jaw musculature has to

attach to the primordial skull to be functional (Ziermann et al.,

2018), in the direct development of amniotes, the jaw muscula-

ture is not restricted to an insertion onto the chondro- respec-

tively neurocranium, and has more freedom to respond to the

functional requirements of the hatchling/newborn.

The scenarios proposed by Schoch (2014b) and Werne-

burg (2019) may explain how both tetrapod crown-groups

could reduce their temporal dermatocranium despite differ-

ent developmental strategies. Yet, these developmental strat-

egies are derived relative to those of extinct tetrapod groups

(e.g. Packard & Seymour, 1997; Schoch, 2014a; Laurin

et al., 2019). The lack of temporal openings in most other

Paleozoic tetrapods must be explained by selective forces that

prevented the development of temporal openings in these

taxa. While the developmental strategies of several Paleozoic

tetrapod groups were strikingly diverse [Schoch (2014a) and

references therein], especially in groups that possessed tem-

poral openings such as the embolomere Anthracosaurus russelli

(Panchen, 1977; Clack, 1987) or some lepospondyls

(e.g. Bolt & Rieppel, 2009; Pardo & Anderson, 2016), their

ontogenetic trajectories are not understood in detail

(Schoch, 2014a; Laurin et al., 2019). Nevertheless, many

lepospondyls apparently had an uniphasic life cycle, possibly

comparable to amniotes or terrestrial salamanders (Schoch,

2014a). Hence, the development of temporal openings in

some lepospondyls may be explained in the same way

(Werneburg, 2019). In fact, several lepospondyls with dis-

tinctly reduced regions of the dermatocranium recently have

been proposed to nest within Amniota (Mann et al., 2019; see

Section III.1e). However, the proposed absence of larvae in

lepospondyls could be also explained by a preservation bias

(Michel Laurin, personal communication) and the absence

of a consensus on early tetrapod interrelationships compli-

cates any attempt to make developmental inferences regard-

ing this essential period of skull evolution.

(3) Summary on the origins of temporal openings

The majority of researchers have made a connection between

jaw muscular arrangement and the presence and morphology

of temporal openings. Yet, the evolution of the temporal

region is likely to be multidimensional (e.g. Fox, 1964;

Kuhn-Schnyder, 1980; Ferreira et al., 2020). Temporal open-

ings seem predominantly to form in weak areas of the derma-

tocranium, characterized by relatively thin bone (Jaekel, 1902;

Romer & Price, 1940; Fox, 1964) and along the sutures of

three or more bones (Frazzetta, 1968; Kuhn-Schnyder,

1980) or at intracranial joints (Kemp, 1980). These ‘weak’

areas may be more likely to form temporal openings due to

a lack of ossiﬁcation (Romer & Price, 1940; Cisneros, 2008),

as an adaptation to reduce skull weight (Gaupp, 1895b; Fox,

1964), or to reduce bone volume in functionally less important

areas (Case,1924; Olson, 1961; Fox, 1964;Frazzetta, 1968) to

provide more space for the jaw adductors (Dollo, 1884; Greg-

ory & Adams, 1915; Versluys, 1919; Lakjer, 1926; Kleinteich

et al., 2012). The arrangement of the latter determines the

force distribution on the temporal dermatocranium, thereby

favouring the development of bony ridges along ‘networks of

lines of stress’for attachment of the musculature. This could

subsequently facilitate reduction of areas subject to lower

applied forces and the formation of strong temporal bars in

high-stress regions (Gaupp, 1895b; Gegenbaur, 1898; Case,

1924; Kilias, 1957; Olson, 1961; Fox, 1964; Frazzetta, 1968;

Carroll, 1982; Tarsitano et al., 2001).

The arrangement of the jaw adductors is then dependent on

the shape of the skull (Olson, 1961; Frazzetta, 1968; Tarsitano

et al., 2001; Schoch, 2014b),therelativedimensionsandposi-

tions of different skull regions like the choanae, otic capsules,

braincase, and orbits (Dollo, 1884; Fuchs, 1909; Lakjer, 1926),

and on the ontogenetic strategy (Schoch, 2014b; Werneburg,

2019). Temporal morphology will be further related to aspects

of feeding mechanics like jaw movement (Dollo, 1884;

Versluys, 1919; Olson, 1961; Whiteside, 1986), jaw articulation,

and cranial kinesis (Fuchs, 1909; Romer, 1956; Kilias, 1957).

Temporal morphology may be also dependent on neck anat-

omy, including the mobility of the head–neck joint (Versluys,

1919; Olson, 1961; Kuhn-Schnyder, 1980; Werneburg 2015)

and the ability to retract the head (Zdansky, 1923–1925;

Kilias, 1957; Werneburg, 2012, 2015; Ferreira et al., 2020).

Finally, external mechanical stresses during an aquatic or fosso-

rial lifestyle may also be relatedtoareductioninthetemporal

Society.

Morphology of the tetrapod temporal skull region 2251

region (Gaupp, 1895b; Olson, 1961; Sherratt et al., 2014; Bar-

dua et al., 2019).

There is obviously no universal answer explaining the for-

mation of temporal openings in Tetrapoda, with selective

regimes varying according to taxon (Fig. 6). For example, the

ability to retract the head arguably plays an important role

for the turtle skull, but is unlikely to be ableto explain the pres-

ence of temporal openings in lepidosaurs or lissamphibians.

Temporal morphology will be a compromise of the various

factors that act on the skull, including potential phylogenetic

constraints. Nevertheless, examples of independent evolution

of similar temporal morphotypes (e.g. MacDougall & Reisz,

2014; Ford & Benson, 2020) suggest that in some cases we will

be able to uncover general patterns of selection leading to a

particular temporal morphology.

IV. CONCLUSIONS

(1) We introduced a novel morphological classiﬁcation

scheme that subdivides tetrapod temporal

morphology into 10 morphotypes: scutal, infrafenestral,

suprafenestral, bifenestral, additofenestral, fossafenestral, infra-

fossal, suprafossal, bifossal, and nudital. This scheme rep-

resents an alternative to traditional classiﬁcation

schemes in being independent of phylogeny, homology

criteria, and functional interpretations. Plotting these

10 morphotypes onto a phylogenetic tree illustrates

the broad range of character distributions among taxa.

The ancestral condition of major clades remains uncer-

tain –including those of Amniota, Reptilia, and Synap-

sida –because many morphotypes evolved in parallel in

the early members of these groups. Future research

should focus on obtaining a detailed and homology-

based character deﬁnition of the temporal region,

including bone-to-bone contacts, suture anatomy, and

suitable metrics to enable us to understand the detailed

pathways of evolution of the temporal region. This will

allow us to clarify how similar morphotypes evolved in

different taxa in response to (potentially different) selec-

tive forces.

(2) Research on the tetrapod temporal region has a long

history, extending back to the early 19th century.

Fig. 6. Summary of the different factors proposed to affect the morphology of the temporal skull region in tetrapods. The relative

importance of these factors differs among taxa. For some taxa, speciﬁc factors may be not applicable; additionally, factors not

mentioned in this ﬁgure may also play a relevant role in some cases.

Society.

2252 Pascal Abel and Ingmar Werneburg

The morphological differences observed compara-

tively early in this ﬁeld of research were placed into

an evolutionary context by the late 19th and early

20th century. Within this period, several authors inde-

pendently devised naming conventions for the differ-

ent temporal conﬁgurations, and for taxa based on

the latter, with several of these terms still in com-

mon use.

(3) A relationship between jaw musculature and the for-

mation of temporal openings was proposed early on

and was generally accepted during the last century.

Temporal openings appear to form predominantly in

relatively weak areas of the dermal armour, such as

in the contact zone of three or more bones or at the

intracranial joint between the parietal and ‘cheek’.

Many researchers interpreted the formation of tem-

poral openings, or more precisely the associated tem-

poral bars and thickened bony margins, as an

adaptation to force distribution and muscle size.

The development of temporal openings has been con-

sidered in the context of tetrapod terrestrialisation

and postulated to be favoured by the absence of

external hydrostatic pressure, weight reduction, evo-

lution of the head–neck joint, doming of the skull,

but also skull ﬂattening (in batrachians), changes in

thejawhingeandcranialkinesis,evolutionoftheotic

capsules and auditory apparatus, or by the absence of

a larval stage in early ontogeny. Postulated selective

pressures include changes in feeding mechanics,

enlarging of the braincase, evolution of neck retrac-

tion, fossoriality, and differentiation of the jaw adduc-

tor musculature.

V. ACKNOWLEDGEMENTS

We thank Agnes Fatz (Tübingen) for the specimen photo-

graph in Fig. 6 and its graphical editing, Patrick Hänsel

(Erlangen) for help with translating Greek terminology, and

Gerardo Antonio Cordero (Tübingen) for helpful

discussions. We also thank John Welch and Alison Cooper

for editing, Michel Laurin and an anonymous reviewer for

their valuable comments. This study was funded by DFG

grant WE 5440/6-1. Open Access funding enabled and

organized by Projekt DEAL.

VI. REFERENCES

[Correction added on 8 June 2021 after ﬁrst publication: publisher name and location

have been added to the reference Hay (1908)]

Adams, L. A. (1919). A memoir on the phylogeny of the jaw muscles in recen t and fossil

vertebrates. Annals of the New York Academy of Sciences 28(1), 51–166.

AmphibiaTree (2004). Bombina orientalis.Digital Morphology.

AmphibiaTree &Gosselin-Ildari, A. (2008). Pleurodeles walti.Digital

Morphology.

Anderson, J. S. (2008). Focal review: the origin(s) of modern amphibians. Evolutionary

Biology 35(4), 231–247.

Andrews, C. W. (1913). A Descriptive Catalogue of the Marine Reptiles of the Oxford Clay: Based

on the Leeds Collection in the British Museum (Natural History) Library, London, Edition

(Volume 2). Order of the Trustees of the British Museum, London.

Angielczyk, K. D. (2004). Phylogenetic evidence for and implications of a dual

origin of propaliny in anomodont therapsids (Synapsida). Paleobiology 30(2),

268–296.

Bardua, C.,Wilkinson, M.,Gower, D. J.,Sherratt, E. &Goswami, A. (2019).

Morphological evolution and modularity of the caecilian skull. BMC Evolutionary

Biology 19(1), 1–23.

Barghusen, H. R. (1973). The adductor jaw musculature of Dimetrodon (Reptilia,

Pelycosauria). Journal of Paleontology 47(5), 823–834.

Baur, G. (1889). On the morphology of the vertebrate-skull. Journal of Morphology 3(3),

467–474.

Baur, G. (1894). Bemerkungen über die Osteologie der Schläfengegend der höheren

Wirbeltiere. Anatomischer Anzeiger 10(10), 315–330.

Benson, R. B. J. (2012). Interrelationships of basal synapsids: cranial and postcranial

morphological partitions suggest different topologies. Journal of Systematic

Palaeontology 10(4), 601–624.

Benson, R. B. J.,Evans, M.,Smith, A. S.,Sassoon, J.,Moore-Faye, S.,

Ketchum, H. F. &Forrest, R. (2013). A giant pliosaurid skull from the late

Jurassic of England. PLoS One 8(5), e65989.

Benton, M. J. (1983). The Triassic reptile Hyperodapedon from Elgin: functional

morphology and relationships. Philosophical Transactions of the Royal Society of London.

B: Biological Sciences 302(1112), 605–718.

Benton, M. J. (1985). Classiﬁcation and phylogeny of the diapsid reptiles. Zoological

Journal of the Linnean Society 84(2), 97–164.

Berman, D. S. (2000). Origin and early evolution of the amniote occiput. Journal of

Paleontology 74(5), 938–956.

Berman, D. S. (2013). Diadectomorphs, amniotes or not? New Mexico Museum of Natural

History and Science Bulletin 60,22–35.

Berman, D. S.,Sumida, S. S. &Lombard, R. E. (1992). Reinterpretation of the

temporal and occipital regions in Diadectes and the relationships of

diadectomorphs. Journal of Paleontology 66(3), 481–499.

Bever, G. S.,Lyson, T. R.,Field, D. J. &Bhullar, B. A. (2015). Evolutionary

origin of the turtle skull. Nature 525(7568), 239–242.

Bhullar, B. A. &Bever, G. S. (2009). An archosaur-like laterosphenoid in early

turtles (Reptilia: Pantestudines). Breviora 518(1), 1–11.

Bickelmann, C.,Mu

¨ller, J. &Reisz, R. R. (2009). The enigmatic diapsid

Acerosodontosaurus piveteaui (Reptilia: Neodiapsida) from the upper Permian of

Madagascar and the paraphyly of ‘younginiform’reptiles. Canadian Journal of Earth

Sciences 46(9), 651–661.

Blom, H. (2005). Taxonomic revision of the late Devonian tetrapod Ichthyostega from

East Greenland. Palaeontology 48(1), 111–134.

Boas, J. E. V. (1915). Die Schläfenüberdachung und das Palatoquadratum in ihrem

Verhältnis zum übrigen Schädel bei den Dipnoern und den terrestren

Wirbeltieren. Gegenbaurs Morphologisches Jahrbuch 49(2), 229–307.

Boettger, C. R. (1952). Die Stämme des Tierreichs in ihrer systematischen

Gliederung. Abhandlungen der Braunschweigischen Wissenschaftlichen Gesellschaft 4,

238–300.

Bolt, J. R. &Rieppel, O. (2009). The holotype skull of Llistrofus pricei Carroll and

Gaskill, 1978 (Microsauria: Hapsidopareiontidae). Journal of Paleontology 83(3),

471–483.

Boonstra, L. D. (1936). Some features of the cranial morphology of the

tapinocephalid deinocephalians. Bulletin of the American Museum of Natural History 72

(2), 75–98.

Broom, R. (1911). On the Structure of the Skull in Cynodont Reptiles. Paper

presented at the Proceedings of the Zoological Society of London, Oxford.

Broom, R. (1913). On the structure and afﬁnities of Bolosaurus.Bulletin American Museum

of Natural History 32(33), 509–516.

Broom, R. (1922). On the Temporal Arches of the Reptilia. Paper presented at the

Proceedings of the Zoological Society of London, Oxford.

Cannatella, D. C. &Hillis, D. M. (1993). Amphibian relationships: phylogenetic

analysis of morphology and molecules. Herpetological Monographs 7,1–7.

Carr, T. D. (1999). Craniofacial ontogeny in Tyrannosauridae (Dinosauria,

Coelurosauria). Journal of Vertebrate Paleontology 19(3), 497–520.

Carr, T. D. (2020). A high-resolution growth series of Tyrannosaurus rex obtained from

multiple lines of evidence. PeerJ 8, e9192.

Carroll, R. L. (1981). Plesiosaur ancestors from the upper Permian of Madagascar.

Philosophical Transactions of the Royal Society of London. B: Biological Sciences 293(1066),

315–383.

Carroll, R. L. (1982). Early evolution of reptiles. Annual Review of Ecology and Systematics

13(1), 87–109.

Carroll, R. L. (1988). Vertebrate Paleontology and Evolution. W.H. Freeman and

Company, New York.

Carroll, R. L. &Holmes, R. (1980). The skull and jaw musculature as guides to the

ancestry of salamanders. Zoological Journal of the Linnean Society 68(1), 1–40.

Society.

Morphology of the tetrapod temporal skull region 2253

Case, E. C. (1898). The signiﬁcance of certain changes in the temporal region of the

primitive Reptilia. The American Naturalist 32(374), 69–74.

Case, E. C. (1924). A possible explanation of fenestration in the primitive reptilian

skull, with notes on the temporal region of the genus Dimetrodon.Contributions from

the Museum of Geology University of Michigan 2(1), 1–12.

Cisneros, J. C. (2008). Taxonomic status of the reptile genus Procolophon from the

Gondwanan Triassic. Palaeontologia Africana 43,7–17.

Cisneros, J. C.,Damiani, R.,Schultz, C.,da Rosa, A.,Schwanke, C.,

Neto, L. W. &Aure

´lio, P. L. P. (2004). A procolophonoid reptile with temporal

fenestration from the middle Triassic of Brazil. Proceedings of the Royal Society of

London. Series B: Biological Sciences 271(1547), 1541–1546.

Cisneros, J. C.,Abdala, F.,Jashashvili, T.,de Oliveira Bueno, A. &

Dentzien-Dias, P. (2015). Tiarajudens eccentricus and Anomocephalus africanus, two

bizarre anomodonts (Synapsida, Therapsida) with dental occlusion from the

Permian of Gondwana. Royal Society Open Science 2(7), 1–27.

Cisneros, J. C.,Kammerer, C. F.,Angielczyk, K. D.,Fro¨ bisch, J.,

Marsicano, C.,Smith, R. M. H. &Richter, M. (2021). A new reptile from the

lower Permian of Brazil (Karutia fortunata gen. et sp. nov.) and the interrelationships

of Parareptilia. Journal of Systematic Palaeontology 18(23), 1–12.

Clack, J. A. (1987). Two new specimens of Anthracosaurus (Amphibia: Anthracosauria)

from the Northumberland coal measures. Palaeontology 30(1), 15–26.

Clack, J. A. (1997). The Scottish carboniferous tetrapod Crassigyrinus scoticus

(Lydekker)—cranial anatomy and relationships. Transactions of the Royal Society of

Edinburgh: Earth Sciences 88(3), 127–142.

Clack, J. A. (2002). The dermal skull roof of Acanthostega gunnari, an early tetrapod from

the late Devonian. Transactions of the Royal Society of Edinburgh: Earth Sciences 93(1),

17–33.

Colbert, E. H. (1945). The Dinosaur Book: The Ruling Reptiles and their Relatives. The

American Museum of Natural History, New York.

Colbert, E. H. (1946). Hypsognathus, a Triassic reptile from New Jersey. Bulletin of the

American Museum of Natural History 86(5), 231–274.

Conrad, J. L. &Norell, M. A. (2010). Cranial autapomorphies in two species of

Iguana (Iguanidae: Squamata). Journal of Herpetology 44(2), 307–312.

Cope, E. D. (1868). Synopsis of the extinct Batrachia of North America. Proceedings of the

Academy of Natural Sciences of Philadelphia 20, 208–221.

Cope, E. D. (1892). On the homologies of the posterior cranial arches in the Reptilia.

Transactions of the American Philosophical Society 17(1), 11–26.

Crompton, A. W. (1963). The evolution of the mammalian jaw. Evolution 17(4),

431–439.

Cuvier, G. L. C. F. D. (1829). Le Règne Animal Distribué d’Après son Organisation,

Pour Servir de Base à l’Histoire Naturelle des Animaux et d’Introduction à

l’Anatomie Comparée (Vol. 2): Chez Déterville, Libraire, rue Hautefeuille, No

8 et Chez Crochard, Libraire, Cloitar Saint-Ehnoit, No 16.

Daeschler, E. B.,Shubin, N. H. &Jenkins, F. A. Jr. (2006). A Devonian tetrapod-

like ﬁsh and the evolution of the tetrapod body plan. Nature 440(7085), 757–763.

Darwin, C. (1859). The Origin of Species by Means of Natural Selection, or the Preservation of

Favoured Races in the Struggle for Life, 1st Edition. John Murray, London.

Das, S.,Campbell, P.,Roy, S.,Mukherjee, S.,Pramanick, K.,Biswas, A. &

Raha, S. (2019). Cranial osteology and molecular phylogeny of Argyrogena fasciolata

(Shaw, 1802) (Colubridae: Serpentes). Vertebrate Zoology 69(3), 311–325.

DeBraga, M. &Reisz, R. R. (1996). The early Permian reptile Acleistorhinus pteroticus

and its phylogenetic position. Journal of Vertebrate Paleontology 16(3), 384–395.

DeBraga, M. &Rieppel, O. (1997). Reptile phylogeny and the interrelationships of

turtles. Zoological Journal of the Linnean Society 120, 281–354.

DigiMorph Staff (2002). Scolecomorphus sp. Digital Morphology.

Dollo, L. (1884). Cinquième note sur les dinosauriens de Bernissart. Bulletin du Musée

Royal d’Histoire Naturelle de Belgique - Mededeelingen van het Koninklijk Natuurhistoris ch

Museum van Belgi 3(8), 129–146.

Dombrowski, B. (1924). Über einige Gesetzmäßigkeiten im Aufbau des

Schalleitenden und Kieferapparates der Tetrapoden. Zeitschrift für Anatomie und

Entwicklungsgeschichte 75(1–2), 69–81.

Elzanowski, A. &Mayr, G. (2017). Multiple origins of secondary temporal fenestrae

and orbitozygomatic junctions in birds. Journal of Zoological Systematics and Evolutionary

Research 56(2), 248–269.

Evans, S. E. (1980). The skull of a new eosuchian reptile from the lower Jurassic of

South Wales. Zoological Journal of the Linnean Society 70(3), 203–264.

Evans, S. E. (2008). The skull of lizards and tuatara. In Biology of the Reptilia (Volume .

20, eds C. GANS,A.S.GAUNT and K. ADLER), pp. 1–347. , Ithaca: Society for the

Study of Amphibians and Reptiles.

Evans, S. E.,Jones, M. E. H. &Krause, D. W. (2008). A giant frog with South

American afﬁnities from the Late Cretaceous of Madagascar. Proceedings of the

National Academy of Sciences of the United States of America 105(8), 2951–2956.

Evans, S. E. &Borsuk-Białynicka, M. (2009). A small lepidosauromorph reptile

from the Early Triassic of Poland. Palaeontologia Polonica 45, 179–202.

Ezcurra, M. D.,Butler, R. J. &Benson, R. B. J. (2015). Taxonomy of the

proterosuchid archosauriforms (Diapsida: Archosauromorpha) from the earliest

Triassic of South Africa, and implications for the early archosauriform radiation.

Palaeontology 58(1), 141–170.

Ferreira, G. S.,Lautenschlager, S.,Evers, S. W.,Pfaff, C.,Kriwet, J.,

Raselli, I. &Werneburg, I. (2020). Feeding biomechanics suggests progressive

correlation of skull architecture and neck evolution in turtles. ScientiﬁcReports10(1), 1–11.

Field, D. J.,Hanson, M.,Burnham, D.,Wilson, L. E.,Super, K.,Ehret, D.,

Ebersole, J. A. &Bhullar, B.-A. S. (2018). Complete Ichthyornis skull

illuminates mosaic assembly of the avian head. Nature 557(7703), 96–100.

Ford, D. P. (2018). The evolution and phylogeny of early amniotes. Dissertation,

University of Oxford.

Ford, D. P. &Benson, R. B. J. (2019). A redescription of Orovenator mayorum

(Sauropsida, Diapsida) using high-resolution μCT, and the consequences for early

amniote phylogeny. Papers in Palaeontology 5(2), 197–239.

Ford, D. P. &Benson, R. B. J. (2020). The phylogeny of early amniotes and the

afﬁnities of Parareptilia and Varanopidae. Nature Ecology & Evolution 4(1), 57–65.

Fox, R. C. (1964). The adductor muscles of the jaw in some primitive reptiles. University

of Kansas Publications 12(15), 657–680.

Fox, R. C. &Bowman, M. C. (1966). Osteology and relationships of Captorhinus aguti

(Cope) (Reptilia: Captorhinomorpha). The University of Kansas Paleontological

Contributions Vertebrata 11,1–79.

Fraser, N. C. (1988). The osteology and relationships of Clevosaurus (Reptilia:

Sphenodontida). Philosophical Transactions of the Royal Society of London. B: Biological

Sciences 321(1204), 125–178.

Frazzetta, T. H. (1968). Adaptive problems and possibilities in the temporal

fenestration of tetrapod skulls. Journal of Morphology 125(2), 145–157.

Fuchs, H. (1909). Betrachtungen über die Schläfengegend am Schädel der

Quadrupeden. Anatomischer Anzeiger 35(5–7), 113–167.

¨rbringer, M. (1900). Zur vergleichenden Anatomie des Brustschulterapparates

und der Schultermuskeln. Jenaische Zeitschrift für Naturwissenschaft 34(4), 215–718.

Gaffney, E. S. (1979). Comparative cranial morphology of recent and fossil turtles.

Bulletin of the American Museum of Natural History 164(2), 65–376.

Gaffney, E. S. (1983). The cranial morphology of the extinct horned turtle, Meiolania

platyceps, from the Pleistocene of Lord Howe Island, Australia. Bulletin of the American

Museum of Natural History 175(4), 361–480.

Gaffney, E. S. (1990). The comparative osteology of the Triassic turtle Proganochelys.

Bulletin of the American Museum of Natural History 194,65–376.

Gaupp, E. (1895a). Beiträge zur Morphologie des Schädels. III. Zur vergleichenden

Anatomie der Schläfengegend am knöchernen Wirlbelthier-Schädel. Morphologische

Arbeiten 4(1), 77–129.

Gaupp, E. (1895b). Ueber die Jochbogen-Bildungen am Schädel der Wirbelthiere.

Jahres-Bericht der Schlesischen Gesellschaft für vaterländische Cultur 72(2), 56–63.

Gauthier, J. A. &de Queiroz, K. (2020). Diapsida. In Phylonyms, a Companion to the

PhyloCode, 1st Edition (ed. J. A. GAUTHIER), pp. 1033–1040. Taylor & Francis

Group, Ltd, London.

Gebauer, E. V. I. (2007). Phylogeny and evolution of the Gorgonopsia with a special

reference to the skull and skeleton of GPIT/RE/7113 (‘Aelurognathus?’parringtoni).

Dissertation, Eberhard-Karls Universität Tübingen.

Gegenbaur, C. (1898). Vergleichende Anatomie der Wirbelthiere mit Berücksichtigung der

Wirbellosen, Edition (Volume 1). Verlag von Wilhelm Engelmann, Leipzig.

Goodrich, E. S. (1916). On the classiﬁcation of the Reptilia. Proceedings of the Royal

Society of London. Series B, Containing Papers of a Biological Character 89(615), 261–276.

Gow, C. E. (1972). The osteology and relationships of the Millerettidae (Reptilia:

Cotylosauria). Journal of Zoology 167(2), 219–264.

Gower, D. J.,Loader, S. P.,Moncrieff, C. B. &Wilkinson, M. (2004). Niche

separation and comparative abundance of Boulengerula boulengeri and Scolecomorphus

vittatus (Amphibia: Gymnophiona) in an east Usambara forest, Tanzania. African

Journal of Herpetology 53(2), 183–190.

Gregory, W. K. &Adams, L. A. (1915). The temporal fossae of vertebrates in relation

to the jaw muscles. Science 41(1064), 763–765.

¨nther, A. C. L. G. G. (1867). Contribution to the anatomy of Hatteria

(Rhynchocephalus, Owen). Proceedings of the Royal Society of London 15(19), 595–629.

Hall,B.K.(2014). Bones and Cartilage.AcademicPress,London,SanDiego,Waltham,Oxford.

Hallmann, E. (1837). Die Vergleichende Osteologie des Schläfenbeins. Verlag der Hahn’schen

Hofbuchhandlung, Hannover.

Haridy, Y.,Macdougall, M. J.,Scott, D. &Reisz, R. R. (2016). Ontogenetic change

in the temporal region of the early Permian parareptile Delorhynchus cifellii and the

implicationsfor closure ofthe temporalfenestra in amniotes.PLoS One 11(12),e0166819.

Hay, O. P. (1908). The Fossil Turtles of North America. Carnegie Institution of

Washington, Washington D.C.

Heckert, A. B.,Lucas, S. G. &Spielmann, J. A. (2012). A new species of the

enigmatic archosauromorph Doswellia from the upper Triassic Bluewater Creek

formation, New Mexico, USA. Palaeontology 55(6), 1333–1348.

Holliday, C. M.,Porter, W. R.,Vliet, K. A. &Witmer, L. M. (2020). The

frontoparietal fossa and dorsotemporal fenestra of archosaurs and their

signiﬁcance for interpretations of vascular and muscular anatomy in dinosaurs. The

Anatomical Record 303(4), 1060–1074.

Society.

2254 Pascal Abel and Ingmar Werneburg

Holliday, C. M. &Witmer, L. M. (2007). Archosaur adductor chamber evolution:

integration of musculoskeletal and topological criteria in jaw muscle homology.

Journal of Morphology 268(6), 457–484.

Holmes, R.,Cheng, Y. N. &Wu, X.-C. (2008). New information on the skull of

Keichousaurus hui (Reptilia: Sauropterygia) with comments on sauropterygian

interrelationships. Journal of Vertebrate Paleontology 28(1), 76–84.

Hylander, W. L. &Johnson, K. R. (1997). In vivo bone strain patterns in the

zygomatic arch of macaques and the signiﬁcance of these patterns for functional

interpretations of craniofacial form. American Journal of Physical Anthropology 102(2),

203–232.

Iordansky, N. N. (1990). Evolution of cranial kinesis in lower tetrapods. Netherlands

Journal of Zoology 40(1–2), 32–54.

Iordansky, N. N. (1996). The temporal ligaments and their bearing on cranial kinesis

in lizards. Journal of Zoology 239(1), 167–175.

Jaekel, O. (1902). Ueber Gephyrostegus bohemicus nov. gen. nov. sp. Zeitschrift der Deutschen

Geologischen Gesellschaft 54, 127–132.

Jaekel, O. (1909a). Über die Klassen der Tetrapoden. Zoologischer Anzeiger 34(7–8),

193–212.

Jaekel, O. (1909b). Über das System der Reptilien. Zoologischer Anzeiger 35(11),

324–341.

Jaekel, O. (1913). Über den Bau des Schädels. Paper presented at the Verhandlungen

der Anatomischen Gesellschaft, Greifswald.

Jaekel, O. (1916). Die Wirbeltierfunde aus dem Keuper von Halberstadt.

Paläontologische Zeitschrift 2,88–214.

Jaekel, O. (1922). Neues über Hemispondyla. Paläontologische Zeitschrift 5(1), 1–25.

Jaekel, O. (1927). Neue Forschungen über das Primordial-Cranium und Gehirn

paläozoischer Fische. Paläontologische Zeitschrift 8(1), 161–184.

Jenkins, F. A.,Walsh, D. M. &Carroll, R. L. (2007). Anatomy of Eocaecilia

micropodia, a limbed caecilian of the early Jurassic. Bulletin of the Museum of

Comparative Zoology 158(6), 285–365.

Jollie, M. T. (1957). The head skeleton of the chicken and remarks on the anatomy of

this region in other birds. Journal of Morphology 100(3), 389–436.

Jones, M. E. H. (2004). Skull evolution in the Rhynchocephalia (Diapsida: Lepidosauria). Paper

presented at the Sixty-Fourth Annual Meeting Society of Vertebrate Paleontology,

Denver.

Jones, M. E. H. (2008). Skull shape and feeding strategy in Sphenodon and other

Rhynchocephalia (Diapsida: Lepidosauria). Journal of Morphology 269(8), 945–966.

Jones, M. E. H.,Werneburg, I.,Curtis, N.,Penrose, R.,O’Higgins, P.,

Fagan, M. J. &Evans, S. E. (2012). The head and neck anatomy of sea turtles

(Cryptodira: Chelonioidea) and skull shape in Testudines. PLoS One 7(11), e47852.

Kammerer, C. F. (2011). Systematics of the Anteosauria (Therapsida: Dinocephalia).

Journal of Systematic Palaeontology 9(2), 261–304.

Kemp, T. S. (1969). On the functional morphology of the gorgonopsid skull. Philosophical

Transactions of the Royal Society of London. B: Biological Sciences 256(801), 1–83.

Kemp, T. S. (1980). Origin of the mammal-like reptiles. Nature 283(5745), 378–380.

Kemp, T. S. (1984). Mammal-Like Reptiles and the Origin of Mammals, 2nd Edition.

Academic Press Inc. (London) LTD, London.

Kilias, R. (1957). Die funktionell-anatomische und systematische Bedeutung der

Schläfenreduktion bei Schildkröten. Mitteilungen aus dem Museum für Naturkunde in

Berlin 33(2), 307–354.

Kissel, R. (2010). Morphology, Phylogeny, and Evolution of Diadectidae

(Cotylosauria: Diadectomorpha). Dissertation, University of Toronto.

Kleinteich, T.,Maddin, H. C.,Herzen, J.,Beckmann, F. &Summers, A. P.

(2012). Is solid always best? Cranial performance in solid and fenestrated caecilian

skulls. The Journal of Experimental Biology 215(Pt 5), 833–844.

Klembara, J.,Dobias

ˇov

a, K.,Hain, M. &Yaryhin, O. (2017). Skull anatomy and

ontogeny of legless lizard Pseudopus apodus (Pallas, 1775): heterochronic inﬂuences on

form. The Anatomical Record 300(3), 460–502.

Klembara, J.,Hain, M.,Berman, D. S.,Pierce, S. E. &Henrici, A. C. (2019).

Inner ear morphology of diadectomorphs and seymouriamorphs (Tetrapoda)

uncovered by high-resolution x-ray microcomputed tomography, and the origin of

the amniote crown group. Paléo 63(1), 131–154.

Klenner, S.,Witzel, U.,Paris, F. &Distler, C. (2015). Structure and function of

the septum nasi and the underlying tension chord in crocodylians. Journal of Anatomy

228(1), 113–124.

Kuhn, O. (1961). Die Familien der rezenten und fossilen Amphibien und Reptilien. Verlagshaus

Meisenbach KG, Bamberg.

Kuhn-Schnyder, E. (1954). The origin of lizards. Endeavour 13(52), 213–219.

Kuhn-Schnyder, E. (1963). Wege der Reptiliensystematik. Paläontologische Zeitschrift

37(1–2), 61–87.

Kuhn-Schnyder, E. (1967). Das Problem der Euryapsida. Mitteilungen aus dem

Paläontologischen Institut der Universität Zürich 49, 335–348.

Kuhn-Schnyder, E. (1980). Observations on temporal openings of reptilian skulls

and the classiﬁcation of reptiles. In Aspects of Vertebrate History, Essays in Honor of

Edwin Harris Colbert (ed. L. L. JACOBS), pp. 153–175. Museum of Northern Arizona

Press, Flagstaff.

Lakjer, T. (1926). Studien über die Trigeminus-versorgte Kaumuskulatur der

Sauropsiden (A. Luther & C. Wesenburg-Lund Eds.). Kopenhagen.

Laurin, M. (1991). The osteology of a lower Permian eosuchian from Texas and a

review of diapsid phylogeny. Zoological Journal of the Linnean Society 101,59–95.

Laurin, M. (2020a). Stegocephali. In Phylonyms, a Companion to the PhyloCode, 1st Edition

(eds P. CANTINO,J.A.GAUTHIER and K. DE QUEIROZ), pp. 741–745. Taylor & Francis

Group, Ltd, London.

Laurin, M. (2020b). Tetrapoda. In Phylonyms, a Companion to the PhyloCode, 1st Edition

(eds P. CANTINO,J.A.GAUTHIER and K. DE QUEIROZ), pp. 759–764. Taylor &

Francis Group, Ltd, London.

Laurin, M. &Anderson, J. S. (2004). Meaning of the name Tetrapoda in the

scientiﬁc literature: an exchange. Systematic Biology 53(1), 68–80.

Laurin, M.,Lapauze, O. &Marjanovi

c, D. (2019). What do ossiﬁcation sequences

tell us about the ofthe origin of extant amphibians?Peer Community in Paleontology,1–36.

Laurin, M.,Arntzen, J. W.,B

aez, A. M.,Bauer, A. M.,Damiani, R.,

Evans, S. E.,Kupfer, A.,Larson, A.,Marjanovi

c, D.,Mu

¨ller, H.,

Olsson, L.,Rage, J.-C. &Walsh, D. (2020a). Amphibia. In Phylonyms, a

Companion to the PhyloCode, 1st Edition (eds P. CANTINO,J.A.GAUTH IER and K. DE

QUEIROZ), pp. 765–770. Taylor & Francis Group, Ltd, London.

Laurin, M.,Arntzen, J. W.,B

aez, A. M.,Bauer, A. M.,Damiani, R.,

Evans, S. E.,Kupfer, A.,Larson, A.,Marjanovi

c, D.,Mu

¨ller, H.,

Olsson, L.,Rage, J.-C. &Walsh, D. (2020b). Lissamphibia. In Phylonyms, a

Companion to the PhyloCode, 1st Edition (eds P. CANTINO,J.A.GAUTH IER and K. DE

QUEIROZ), pp. 773–778. Taylor & Francis Group, Ltd, London.

Laurin, M. &Pin

˜eiro, G. H. (2017). A reassessment of the taxonomic position of

mesosaurs, and a surprising phylogeny of early amniotes. Fro ntiers in Earth Science 5

(88), 1–13.

Laurin, M. &Pin

˜eiro, G. H. (2018). Response: commentary: a reassessment of the

taxonomic position of mesosaurs, and a surprising phylogeny of early amniotes.

Frontiers in Earth Science 6(220), 1–9.

Laurin, M. &Reisz, R. R. (1995). A reevaluation of early amniote phylogeny.

Zoological Journal of the Linnean Society 113(2), 165–223.

Laurin, M. &Reisz, R. R. (1997). A new perspective on tetrapod phylogeny. In

Amniote Origins, Completing the Transition to Land (eds S. S. SUMIDA and K. L. M.

MARTIN), pp. 9–59. Academic Press, San Diego.

Laurin, M. &Reisz, R. R. (2020). Synapsida. In Phylonyms, a Companion to the PhyloCode,

1st Edition (ed. J. A. GAUTHIER),pp. 811–814. Taylor & Francis Group, Ltd, London.

Lautenschlager, S. (2015). Estimating cranial musculoskeletal constraints in

theropod dinosaurs. Royal Society Open Science 2(11), 150495.

Lautenschlager, S.,Gill, P.,Luo, Z. X.,Fagan, M. J. &Rayﬁeld, E. J. (2017).

Morphological evolution of the mammalian jaw adductor complex. Biological Reviews

92(4), 1910–1940.

Lee, M. S. (1995). Historical burden in systematics and the interrelationships of

‘parareptiles’.Biological Reviews 70, 459–547.

Lee, M. S. (1997). Pareiasaur phylogeny and the origin of turtles. Zoological Journal of the

Linnean Society 120(3), 197–280.

Lee, M. S. (2000). The Russian Pareiasaurs. In The Age of Dinosaurs in Russia and Mongolia

(eds M. J. BENTON,M.A.SHISHKIN,D.M.UNWIN and E. N. KUR OCHKIN), pp. 71–85.

Cambridge University Press, Cambridge.

Lee, M. S. (2013). Turtle origins: insights from phylogenetic retroﬁtting and molecular

scaffolds. Journal of Evolutionary Biology 26(12), 2729–2738.

Lee, M. S. &Spencer, P. S. (1997). Crown-clades, key characters and taxonomic

stability: when is an amniote not an amniote. In Amniote Origins, Completing the

Transition to Land (eds S. S. SUMIDA and K. L. M. MARTIN), pp. 61–84. Academic

Press, San Diego.

Li, C.,Wu, X.-C.,Rieppel, O.,Wang, L.-T. &Zhao, L.-J. (2008). An ancestral

turtle from the late Triassic of southwestern China. Nature 456(7221), 497–501.

Lichtig, A. J. &Lucas, S. G. (2021). Chinlechelys from the upper Triassic of New

Mexico, USA, and the origin of turtles. Palaeontologia Electronica 24(1), a:13.

Lyson, T. R.,Bever, G. S.,Bhullar, B. A.,Joyce, W. G. &Gauthier, J. A.

(2010). Transitional fossils and the origin of turtles. Biology Letters 6(6), 830–833.

MacDougall, M. J.,Modesto, S. P.,Brocklehurst, N.,Verrie

`re, A.,

Reisz, R. R. &Fro¨ bisch, J. (2018). Commentary: a reassessment of the

taxonomic position of mesosaurs, and a surprising phylogeny of early amniotes.

Frontiers in Earth Science 6(99), 1–5.

MacDougall, M. J. &Reisz, R. R. (2014). The ﬁrst record of a nyctiphruretid

parareptile from the early Permian of North America, with a discussion of

parareptilian temporal fenestration. Zoological Journal of the Linnean Society 172(3),

616–630.

MacGregor, J. H. (1908). On Mesosaurus brasiliensis nov. sp. from the Permian of

Brasil. In Commissao dos Estudos das Minas de Carvao de Pedra do Brasil, Parte II

(ed. I. D. WHITE), pp. 301–336. National Press, Rio de Janeiro.

Macrini, T. (2004). Tachyglossus aculeatus.Digital Morphology.

Maddin, H. C.,Jenkins, F. A. Jr. &Anderson, J. S. (2012). The braincase of

Eocaecilia micropodia (Lissamphibia, Gymnophiona) and the origin of caecilians.

PLoS One 7(12), e50743.

Society.

Morphology of the tetrapod temporal skull region 2255

Maisano, J. (2003). Cricosaura typica.Digital Morphology.

Maisch, M. W. &Hungerbu

¨hler, A. (2001). New evidence for a discrete

supratemporal bone in the Jurassic ichthyosaur Temnodontosaurus.Historical Biology: A

Journal of Paleobiology 15(4), 335–345.

Mann, A.,Pardo, J. D. &Maddin, H. C. (2019). Infernovenator steenae, a new

serpentine recumbirostran from the ‘Mazon Creek’Lagerstätte further clariﬁes

lysorophian origins. Zoological Journal of the Linnean Society 187, 506–517.

Marjanovi

c, D. &Laurin, M. (2019). Phylogeny of Paleozoic limbed vertebrates

reassessed through revision and expansion of the largest published relevant data

matrix. PeerJ 6, e5565.

Mo, J.-Y.,Xu, X. &Evans, S. E. (2010). The evolution of the lepidosaurian lower

temporal bar: new perspectives from the Late Cretaceous of South China. Proceedings

of the Royal Society B: Biological Sciences 277,331–336.

Modesto, S. P. (1995). The skull of the herbivorous synapsid Edaphosaurus boanerges

from the lower Permian of Texas. Paléo 38(1), 213–239.

Modesto, S. P. (2006). The cranial skeleton of the early Permi

Morphology of the temporal skull region in tetrapods: research history, functional explanations, and a new comprehensive classification scheme

Abstract