This is the accepted version of the journal article:
Villa, Andrea; Wings, Oliver; Rabi, Márton. «A new gecko (Squamata, Gekkota)
from the Eocene of Geiseltal (Germany) implies long-term persistence of Euro-
pean Sphaerodactylidae». Papers in Palaeontology, Vol. 8, Issue 3 (May/June
2022), art. e1434. DOI 10.1002/spp2.1434


This version is available at https://ddd.uab.cat/record/259562

under the terms of the            license
A NEW GECKO (SQUAMATA, GEKKOTA) FROM THE EOCENE OF GEISELTAL
(GERMANY) IMPLIES LONG-TERM PERSISTENCE OF EUROPEAN
SPHAERODACTYLIDAE

by ANDREA VILLA1,*, OLIVER WINGS2, and MÁRTON RABI2,3,*

1
    Institut Català de Paleontologia Miquel Crusafont, Universitat Autònoma de Barcelona, Edifci
ICTA-ICP, c/ Columnes s/n, Campus de la UAB, 08193 Cerdanyola del Vallès, Barcelona, Spain;
e-mail: andrea.villa@icp.cat ORCiD: https://orcid.org/0000-0001-6544-5201

2
    Natural Sciences Collections, Martin Luther University Halle-Wittenberg, Domplatz 4, 06108
Halle (Saale), Germany; e-mail: oliver.wings@zns.uni-halle.de

3
    University of Tübingen, Institute for Geosciences, Hölderlinstraße 12, 72074 Tübingen, Germany;
e-mail: marton.rabi@uni-tuebingen.de ORCiD: https://orcid.org/0000-0002-6482-6683

*corresponding authors

Abstract: Gekkota (geckos and pygopods) constitute a diverse and early diverging clade of
squamates, but their highly fragmentary fossil record allows only limited insights into their
evolutionary history. Even long-known classic Paleogene fossil Lagerstätten, such as the Eocene
Messel and Geiseltal, remained uniformative for gecko evolution. Here we describe an articulated
near-complete crown-gekkotan skull from the Eocene of Geiseltal. The skull represents one of the
most complete and oldest fossil geckos from the Cenozoic and warrants naming the new taxon
Geiseleptes delfinoi gen. et sp. nov. Parsimony analysis of an expanded morphological dataset
comprehensively sampling fossil geckos recovers at least two lineages in the Paleogene of Europe:
Gekkonidae and Sphaerodactylidae. Further clades may have been present but require further
support. The new species, together with several other European extinct taxa, belongs to the extant
clade Sphaerodactylidae. Published molecular phylogenies imply a European dispersal from Africa
for the clade and G. delfinoi constrains the timing of this event to no later than the middle Eocene.
Within Sphaerodactylidae, several Eocene, Oligocene, and Miocene species form the stem-lineage
of the extant European-endemic Euleptes europea of which G. delfinoi is the oldest representative.
We here define the new clade Euleptinae for this group. Euleptine geckos had an unusually long
evolutionary history in Cenozoic Europe and represent one of the very few vertebrate clades with a
persistent presence since the early Paleogene, indicating extreme adaptability. As the oldest known
crown-sphaerodactylid, Geiseleptes delfinoi represents an important fossil calibration for
divergence dating analyses of geckos.

Key words: Paleogene, Europe, phylogeny, Sphaerodactylidae, lizards, geckos
Geckos are an iconic group of squamate reptiles, mostly known for the ability of some of their
representatives to climb using adhesive toepads (Autumn et al. 2002; Russel, 2002; Gamble et al.
2012; also present in skinks and anoles). Their group is rather diverse and widespread, living on all
continents except Antarctica with more than 2000 extant species currently recognised (Vitt &
Caldwell 2009; Uetz et al. 2021). These include mainly nocturnal and diurnal lizards with an
arthropod based diet, even though species that forage on other food resources (such as fruits, nectar
or other vertebrates; Daza et al. 2009; Vitt & Caldwell 2009) exist too. Extant geckos inhabit a
variety of different habitats, such as forests, deserts and even environments shaped by human
presence (Davis 1974; Howard & Parmerlee 2001; Newbery & Jones 2007; Bauer 2013).

The fossil record of geckos consists mostly of isolated skeletal elements and few articulated
specimens. It goes back at least to the late Early Cretaceous (Cretaceogekko burmae; Arnold &
Poinar 2008; Daza et al. 2014), with the majority of taxa coming from the Cenozoic of Europe. A
number of Jurassic taxa have been proposed in the past as being related to or even included in
Gekkota, the crown clade of geckos. These include Ardeosaurus, Bavarisaurus, Eichstaettisaurus,
Palaeolacerta, and Yabeinosaurus. However, these conclusions were later reinterpreted.
Ardeosaurus and Yabeinosaurus were recovered nested within scincoids (and likely closely related
to Scincidae) in a recent phylogenetic analysis by Tałanda (2018). Conrad (2018), on the other
hand, recovered both Bavarisaurus and Eichstaettisaurus as basal ‘scincogekkonomorphs’ and
Palaeolacerta as an ardeosaurid. If the latter is true, Palaeolacerta should also be a scincoid
following Tałanda’s (2018) results. However, given that Palaeolacerta is not included in Tałanda’s
matrix, this still awaits confirmation. Furthermore, Eichstaettisaurus is a sister taxon to Norellius
(which is a basal gekkonomorph; Conrad & Norell 2006; Conrad & Daza 2015) in Tałanda’s
analysis, thus indicating different relationships among the early gekkonomorphs. Therefore,
Eichstaettisaurus might be the only known Jurassic stem-Gekkota, as it seems to be confirmed by
recent combined evidence analyses (e.g., Pyron 2017, and related supplementary files; see also
Simões & Pyron 2021). Nevertheless, the presence of more derived squamates (i.e., putative
“scincomorphs” and anguimorphs; Rage 2013) in the Bathonian of Europe suggests that the
gekkotan lineage must have separated at least in the Middle Jurassic. This is in line with age ranges
retrieved for the split of the total-group gekkotan lineage from other crown squamates in recent
works (Pyron 2017; Simões et al. 2020; Simões & Pyron 2021). Simões et al. (2018) estimated it
even earlier, into the Late Triassic.

Four named taxa referred to either stem or crown gekkotans are known from the Cretaceous,
besides a still unnamed and undescribed Early Cretaceous form from Siberia briefly mentioned by
Skutchas (2006), unnamed specimens preserved in the mid-Cretaceous Burmese amber (Daza et al.
2016; Fontanarrosa et al. 2018), and a possible gekkotan frontal from the Maastrichtian of
Northeastern Spain (Blanco et al. 2016). Eggshells with geckoid morphology are also reported from
the latest Cretaceous of Romania (Csiki-Sava et al. 2016). The oldest material attributed to crown
gekkotans include Cretaceogekko burmae, an amber-preserved specimen coming from the late
Early Cretaceous of Myanmar (Arnold & Poinar 2008), even though the identity of this material as
a member of Gekkota is still debatable (for limitations of amber specimens, see also Fontanarrosa et
al. 2018). It was originally attributed to Gekkonidae, but Daza et al. (2014) treated it as a Gekkota
incertae sedis. Three other amber-embedded specimens attributed to geckos were recently reported
by Daza et al. (2016), also from the Cretaceous of Myanmar. Among these is a well-preserved
hatchling, which was included in a phylogenetic analysis and recovered as sister to all other
gekkotans (Daza et al. 2016). Both Hoburogekko suchanovi (Early Cretaceous) and Gobekko
cretacicus (Late Cretaceous) come from the Gobi Desert in Mongolia. Hoburogekko, originally
described as Gekkonidae s.l. by Alifanov (1989), was recovered as sister taxon to the
carphodactylid Phyllurus by Daza et al. (2012) and later in a basal polytomy within Gekkota by
Daza et al. (2013). This polytomy also included the younger Gobekko (which is therefore not a
gekkonid s.l. as originally inferred by Borsuk-Białynicka, 1990). The same position was recovered
for Gobekko by Conrad & Norell (2006), whereas it is a sister taxon to all gekkotans in the analysis
performed by Conrad (2008). As already mentioned, the Early Cretaceous Norellius nyctisaurops
from the Altai region (Mongolia) is a basal gekkonomorph (Conrad & Norell 2006; Conrad & Daza
2015; Tałanda 2018). Another Mongolian taxon, Myrmecodaptria microphagosa from the Late
Cretaceous locality of Ukhaa Tolgod, was referred to Gekkota by Gao & Norell (2000). However,
Myrmecodaptria is related to more derived squamates according to Conrad & Norell (2006) and
Conrad (2008) and even to Scincoidea according to Gauthier et al. (2012) and Tałanda (2018).

Cenozoic gekkotans are mostly represented by indeterminate and isolated remains (for an overview,
see e.g., Daza et al. 2014; Villa & Delfino 2019a). Paleogene records come almost exclusively from
Europe (Daza et al. 2014). Five extinct species are currently known: Yantarogekko balticus from
the early Eocene of the Amber Coast in the Samland Peninsula (Bauer et al. 2005); Laonogekko
lefevrei from the early Eocene of France (Augé 2003); Rhodanogekko vireti from the middle
Eocene of France (Hoffstetter 1946); Cadurcogekko verus from the late Eocene of France (Augé
2005; Bolet et al. 2015; note that Bolet et al., 2015: fig. 1, incorrectly figured a maxilla of
Cadurcogekko piveteaui instead of the one they selected as the holotype of C. verus); and
Cadurcogekko piveteaui from the late Eocene or early Oligocene of France (Hoffstetter 1946;
Georgalis et al. 2021a). The phylogenetic position of these species is still a matter of debate (Daza
et al. 2014). A peculiar specimen showing primitive features associated to gekkotan
synapomorphies was recently briefly reported by Smith et al. (2018) from the Eocene Messel Pit in
Germany. Outside Europe, Paleogene geckos are reported from the late Paleocene of Brazil
(material currently lost; Estes 1970, 1983; Daza et al. 2014), the late Paleocene of Morocco
(attributed to Gekkonidae; Augé & Rage 2006; Bailon et al. 2017), the latest early-early middle
Eocene of Algeria (Rage et al. 2021), the late Eocene of California (material currently lost;
Schatzinger 1975; Golz & Lillegraven 1977; Estes 1983; Daza et al. 2014), and the late Oligocene
of Tanzania (Müller et al. 2018).

The first undisputed representatives of extant families date back to the Neogene. Again, the
majority of the extinct species currently known from this period were found in Europe. These
European taxa include: Gerandogekko arambourgi and Gerandogekko gaillardi from the early and
late Miocene of France respectively (Hoffstetter 1946; Daza et al. 2014); Euleptes gallica from the
early Miocene of France and Czech Republic (Müller 2001; Čerňanský & Bauer 2010); Euleptes
klembarai from the middle Miocene of Slovakia (Čerňanský et al. 2018); and Palaeogekko
risgoviensis from the middle Miocene of Germany (Schleich 1987). With the only exception of P.
risgoviensis (the phylogenetic relationships of which are still unclear; Daza et al. 2014), all these
species are referred to Sphaerodactylidae. According to Daza et al. (2014), evidence for specific
separation between G. arambourgi and G. gaillardi is weak. They could be conspecific, even
though this cannot be confidently stated based on the material currently available. Euleptes is
further reported from the early Miocene of Germany (Müller & Mödden 2001) and, tentatively, the
late Miocene of Italy (Colombero et al. 2017). A phylogeny of the diverse extinct European
gekkotan species is so far lacking. Extinct sphaerodactylids are also known from the Miocene of the
Dominican Republic, being represented by amber-embedded specimens all attributed to the extant
genus Sphaerodactylus (Sphaerodactylus ciguapa, Sphaerodactylus dommeli, and Sphaerodactylus
spp.; Böhme 1984; Kluge 1995; Grimaldi 1996; Grimaldi et al. 2000; Daza & Bauer 2012; Daza et
al. 2014). Pygopodids are represented in the Neogene by a single early Miocene Australian fossil
attributed to the extinct Pygopus hortulanus (Hutchinson 1997), whereas fossil diplodactylids are
known from the lower Miocene of New Zealand (Diplodactylidae indet.; Lee et al. 2009; Daza et
al. 2014) and the middle Pliocene of Australia (Diplodactylus cf. steindachneri; Hutchinson &
Mackness 2002). Daza et al. (2014) also discussed the possible relationships of Paleogene European
taxa with diplodactylids, but definitive evidence for that was lacking. Neogene fossils referred to
the extant gekkonid genus Alsophylax are reported from the late Miocene of Russia (Vasilyan et al.
2017). The occurrences of Gekko sp. from the Oligocene and Miocene of Europe mentioned by
Böhme & Ilg (2003) were never properly published and need to be confirmed (Villa & Delfino
2019a). Miocene indeterminate gekkonids are also known from few African countries, namely
Kenya (Pickford & Andrews 1981), Morocco (Bailon et al. 2017), and Namibia (Rage 2008).

From Quaternary deposits, only fossils of extant taxa are known. Pleistocene and/or Holocene
geckos are reported from Argentina, Australia, New Zealand, New Caledonia, the Fiji Islands,
Aldabra Atoll in the Seychelles, Africa, the Middle East, and Europe. A number of these
occurrences pertain to indeterminate forms, but remains identified at genus or even species level are
also present. Remains referred to the phyllodactylid Homonota sp. were reported from the late
Quaternary (Late Pleistocene or younger) of Argentina by Albino (2005, 2020) and Agnolin & Jofré
(2011). Dentaries of Pygopus sp. were found in two Australian caves, in Holocene deposits (Mead
et al. 2008). Daza et al. (2015) reported Bavayia crassicollis, Bavayia sp., Gehyra georgpotthasti,
and Nactus pelagicus from Holocene sites on Tiga Island (also called Tokanod) in The Loyalty
Islands. Gehyra oceanica, Gehyra aff. vorax, Gekko sp., and Nactus pelagicus were identified in
various Late Pleistocene to Holocene sites in the Fiji (Worthy & Anderson 2009; Worthy & Clark
2009). Worthy & Holdaway (1996) and Holdaway & Worthy (1997) reported various taxa from
different localities in New Zealand, including representatives of the genera Gekko, Hoplodactylus,
and Naultinus. In the Quaternary of Europe, gekkotan fossils referred to three of the extant genera
inhabiting Europe (namely, Euleptes, Hemidactylus, and Tarentola) are known from the
Mediterranean region, in some cases even showing morphologies that appear closely related with
the extant European species (Villa & Delfino 2019a).

In this paper, we report on gekkotan fossil material coming from the Eocene site of Geiseltal in
Germany, including an articulated, incomplete skull. This specimen represents the worldwide oldest
sphaerodactylid and one of the most complete fossil gekkotan from Europe. It belongs to a new
taxon, which is here included in the first phylogenetic analysis of all European gekkotan fossil
species and warranting the definition of a new clade of European Sphaerodactylidae.

GEOLOGICAL AND STRATIGRAPHICAL SETTING

The former Geiseltal lignite (brown coal) deposits, 20 km SSW of Halle (Saale), Saxony-Anhalt,
Germany (Fig. 1), were exploited in several opencast mines. The strata also yielded a diverse fossil
flora and fauna often characterized by exceptional preservation (e.g., Krumbiegel et al. 1983; Voigt
1988). The Geiseltal fossillagerstätte consisted of three major coal seams intercalated by clastic
deposits. Depending on the influx of carbonate-rich waters which allowed fossil preservation,
dozens of vertebrate fossil localities with variable preservational conditions were distributed
vertically and horizontally within Geiseltal (e.g., Haubold & Thomae 1990). Regional spore-pollen
biostratigraphy suggested a Lutetian age (middle Eocene, Spore/Pollen Zone 14/15–15; Krutzsch
1966, 1970, 1976; Krutzsch et al. 1992) for the fossillagerstätte. Biochronology of mammal faunas
at Geiseltal established the middle Eocene Mammal Paleogene levels (MP) 11–13 (= Geiseltalian);
the whole fossillagerstätte is ranging from MP 11 to MP 14, spanning more than the Lutetian
(Franzen & Haubold 1986, 1987; Haubold 1987, 1989; Schmidt-Kittler et al. 1987; Haubold &
Thomae 1990). The gecko-yielding site Cecilie IV was located in the upper Middle Coal, which
was assigned to MP 13 (Haubold & Thomae 1990). However, a recent study of perissodactyls
indicates a lower diversity and an underestimation of taphonomy and collecting bias in local
mammal biochronology (Ring et al. 2020). We thus refrain from assigning individual coal seams to
MP levels at the moment. Faunal similarities with the Messel fossillagerstätte (e.g., cf. Smith et al.
2018 and Steinheimer & Hastings 2019) roughly indicate an age of the gecko remains from
Geiseltal as around the late early or middle Eocene (Messel is spanning the Ypresian–Lutetian
boundary which is at 47.8 Ma; Cohen et al. 2013; see Lenz et al., 2015, for dating Messel). See
Georgalis et al. (2021b) for further details on the age and duration of Geiseltal.

The Site Cecilie IV is also called “Trichter Nordost” (“Pit NorthEast” in the original descriptions;
Weigelt 1934). It represents a circular sinkhole in the gypsum karst of Upper Buntsandstein strata
underlying the bog (Krumbiegel et al. 1983). The water-filled sinkhole probably served as a
watering point during the dry season (Krumbiegel et al. 1983). The often-steep banks of the
sinkholes acted as trapping pits in which many vertebrates perished (Krumbiegel et al. 1983).
Anaerobic conditions at the bottom of the pit combined with a rapid sedimentation rate enhanced
preservation of the carcasses (Krumbiegel et al. 1983). Cecilie IV was one of the most productive
vertebrate sites in Geiseltal, having produced among other fossils the remains of 14 mammal
skeletons and skulls, five birds, 75 “lizards”, 10 crocodilians, 46 turtles, 16 snakes, and 18 frogs
(Weigelt 1934).

MATERIAL AND METHODS

This paper is based on fossil material stored in the Geiseltal Collection, a part of the Natural
Sciences Collections at the Martin Luther University Halle-Wittenberg in Halle (Saale). The
holotype was collected and prepared by Lothar Nöth in the year 1933. GMH Ce IV-4123-1932 was
excavated by Wiesemann in the year 1932 and subsequently prepared by Ehrhard Voigt. Both
specimens are preserved on nitro cellulose lacquer/resin. Photographs of the holotype of the new
species were taken with a Canon EOS 6D Mark II and a Canon EOS 80D. The holotype was
microCT scanned using a Tomoscope HV 500-Werth at the Fraunhofer-Entwicklungszentrum
Röntgentechnik EZRT of the Fraunhofer-Institut für Integrierte Schaltungen IIS in Deggendorf,
Bavaria, Germany. Taxonomic names above the species level highlighted in italics are formally
defined clade names (de Queiroz et al. 2020) including the new clade name, Euleptinae, defined
below following the PhyloCode. Anatomical terminology follows Villa et al. (2018) and Villa &
Delfino (2019b).

Phylogenetic analysis

As no previous studies performed a phylogeny of European fossil gekkotans, the Geiseltal gecko
was included into the matrix of Conrad (2018). This dataset was originally designed for a global
analysis of squamates, but we here limit the taxon sampling to gekkotans and gekkotan-related taxa.
The outgroups used include Pamelina (as in Conrad’s original analysis), Sphenodon, and Dibamus.
For our analysis, we added the Geiseltal taxon, all published European fossil gekkotans (except the
highly fragmentary Gerandogekko arambourgi, which was removed for our second analysis), and
the four extant gecko species most broadly distributed in Europe (Euleptes europaea, Hemidactylus
turcicus, Mediodactylus kotschyi, and Tarentola mauritanica). We furthermore added ten new
characters (see Appendix S1) and reworded state 9 for C403 (“Dentition, number of teeth in each
premaxilla”) as “10 or more”, to acknowledge the fact that extant European taxa can have 11 teeth.
Of the new characters, one is parsimony uninformative (C845). We implemented a molecular
scaffold for extant taxa using the phylogeny of Gamble et al. (2015) and allowed fossil taxa to float
into this topology.

Conrad’s (2018) matrix is an updated version of previous squamate datasets (e.g., Conrad, 2008,
2015; Conrad et al., 2008, 2011, 2012), some of which already focused on gekkotans and relatives
(Daza et al., 2012, 2013). Although he did not mention it explicitly, Conrad (2018) inverted the
states of C175 (“Mandible, posterior mylohyoid foramen”) in his character definitions (his appendix
II). As a matter of fact, state 0 was "present" and state 1 was "absent" in the matrix used by Conrad
& Norell (2006; the original work from which this character is taken), Conrad (2008), and Daza et
al. (2013), whereas they are reversed in Conrad (2018). The reversal of the states appears to be
intentional and coded consistently: Conrad (2008) figured a mandible of Heloderma clearly
showing a posterior mylohyoid foramen (figure 38E) and Heloderma is consequently scored as 0 in
the matrix of Daza et al. (2013), whereas Conrad (2018) scored H. horridus as 1 (foramen present).
In a similar way, Sphenodon (which lacks the foramen) is scored as 0 by Conrad (2018), whereas
rhynchocephalians were scored as 1 by Daza et al. (2013). For most of the geckos, on the other
hand, C175 was incorrectly scored as 1 in the first version of the matrix (Conrad 2008) and by
inverting the states it was corrected by Conrad (2018; still scored as 1). C176 (related to the position
of the same foramen) is scored as inapplicable for geckos in the old matrix (as correct if the
foramen is not present), but it scored as either 0 or 1 in the later version. Taking all of this into
account (and following a preliminary survey of the morphology in select species), we follow
Conrad (2018) with the exception of Gobekko and Hoburgekko. Adding to this, Conrad (2018)
wrote that he deactivated characters 236, 242 and 364 (he wrote 363 in his appendix II, but it is
actually 364 as written in the text) as in previous analyses, but actually he completely removed the
characters from the final matrix.

We used Daza et al.’s (2013) scorings for Gobekko and Hoburogekko for C1–361 (these updates
were ignored by Conrad, 2018), and for the rest of the characters (not present in Daza et al.’s
matrix), we kept Conrad’s (2018) scorings. For the new characters, we based our scorings on Daza
et al.’s (2012, 2013) descriptions of the two Mongolian taxa. Characters that were scored as "-"
(inapplicable) by Daza et al. (2013) were scored as "?" by Conrad (2018), but are here rescored as
"-". This applies only to taxa that were already present in Daza et al.’s matrix and only to the first
361 characters. We follow Conrad (2018) in treating multistate characters that form a morphocline
ordered.

Extant European gekkotans were scored based on personal observations of material stored in the
Massimo Delfino Herpetological Collection (MDHC) at the Department of Earth Sciences of the
University of Torino (Italy), as well as on data available in the literature (Bauer et al., 1997;
Speybroeck et al., 2016; Villa et al., 2018). Fossil taxa were scored based on published data and
personal observations (type material of C. piveteaui and G. arambourgi stored in the Muséum
national d’Histoire naturelle in Paris). The matrix was constructed in MESQUITE v3.61 (Maddison
& Maddison, 2017) and analyzed with TNT v1.5 (Goloboff et al., 2008). The analysis was run with
the Traditional search with two rounds (second round with starting trees from RAM) of tree
bisection reconnection using 1000 replicates with 10 trees saved per replicate.

Institutional abbreviations. GMH, “Geiseltalmuseum Halle” = Geiseltal Collection, ZNS, Martin
Luther University Halle-Wittenberg, Germany; MDHC, Massimo Delfino Herpetological
Collection, Dipartimento di Scienze della Terra, Università degli Studi di Torino, Italy.

SYSTEMATIC PALAEONTOLOGY



                                      SQUAMATA Oppel, 1811

                                       GEKKOTA Camp, 1923

                            SPHAERODACTYLIDAE Underwood, 1954
                                    EULEPTINAE new clade name

Phylogenetic definition. The largest clade containing Euleptes europaea (Gené, 1839) but not
Teratoscincus scincus (Schlegel, 1858).

Derivation of name. From Euleptes, only extant genus included within the clade.

Reference phylogeny. Gamble et al. (2015) [primary] and this study.

Composition. The following genera currently comprise Euleptinae: Cadurcogekko Hoffstetter,
1946, Euleptes Fitzinger, 1843, Geiseleptes gen. nov., and Gerandogekko Hoffstetter, 1946.

Diagnosis. The clade is unambiguously diagnosed by the following characters: presence of
depressions/grooves on the dorsal surface of the frontal, along the posterolateral margins of the
bone (C837: 0 → 1).

Remarks. As the phylogeny of the present study does not include the external specifier
Teratoscincus scincus (extant sister-species of Euleptes europaea), we add an apomorphy based
diagnosis in order to support the placement of fossil species to Euleptinae. An additional
apomorphy recovered by our phylogenetic analysis include a lateral groove behind the last
ventrolateral foramen of the maxillae (reversed in Gerandogekko, 840: 0 → 1).

                                     Genus GEISELEPTES nov.

LSID. urn:lsid:zoobank.org:act:2DD0FEC3-A7D8-4C92-A496-9E1485301CA3

Type species. Geiseleptes delfinoi gen. et sp. nov.

Derivation of name. From Geisel, the river that gave name to the fossillagerstätte Geiseltal, and -
leptes, in reference to Euleptes, the only extant European sphaerodactylid and close relative to the
herein-described Geiseltal taxon.

Diagnosis. As for type species.



                                     Geiseleptes delfinoi sp. nov.

                                               Figs 2–5

Derivation of name. Specific name honours Massimo Delfino, leading European palaeo- and
neoherpetologist, who greatly contributed to the study of both fossil and leaving amphibians and
reptiles and inspired us (as well as many others) in pursuing the same path.
Holotype. GMH Ce IV-4057-1933, a partially articulated and fragmentary skull.

Locality and horizon. Geiseltal fossillagerstätte, Site Cecilie IV (=”Trichter Nord-Ost”), upper
Middle Coal (oMK), late early or middle Eocene.

Diagnosis. The phylogenetic analysis recovered three unambiguous autapomorphies for the taxon:
frontoparietal suture U-shaped and anteriorly arched in dorsal view (C70: 1 → 0); orbital process of
the prefrontal terminating within the anterior 1/3 of the orbit (C383: 2 → 1); short grooves along
posterolateral margins of frontal (C389: 1 → 0). For a differential diagnosis and comparison, see
the Remarks section below.

Description

The holotype (Fig. 2), and only specimen currently referred to the new taxon, is represented by a
heavily crushed skull. During preparation, it was separated into two parts, herein named a and b
respectively, which are now hosted on the same support. Part a (Fig. 3) includes most of the
cranium and part of the lower jaws, being visible in lateral view. Part b (Fig. 4) includes part of the
two lower jaws and part of the left maxilla, exposed in ventral view. Other bones and fragments of
bones are also represented in part b, but they cannot be confidently identified. The micro-CT scans
revealed that the unexposed portion of the specimen is very poorly preserved and mainly consists of
heavily crushed bone fragments. Because of this, the description is mainly based on what is visible
directly on the exposed surface, even though some additional details were retrieved with the scans.

Cranium. The nasals (Fig. 3A,B) are preserved but rather fragmentary. The internasal suture is still
recognisable. They are anteroposteriorly elongated and dorsally smooth. It seems that they taper
slightly anteriorly, but this might have been enhanced by the crushing. The frontal (Figs. 3, 5A,B) is
unpaired. It was originally hourglass-shaped, with a moderate constriction at midlength. The
anterior end is completely crushed and cannot be described. The posterior end shows long and
slender posterolateral processes. The posterior margin is distinctly concave. The dorsal surface
appears smooth, but a very short and shallow groove is visible extending along the lateral margin of
the distal portion of each posterolateral process (Fig. 5A,B). The ventral surface is partially visible
only with the scans, but just the posterior end of the bone can be described, due to strong
fragmentation of the rest of the bone. In the scan, it is possible to see the low posterior portion of
the left crista cranii running along the lateral margin and the posterolateral process. Medially to the
crista, the ventral surface of the posterior end of the frontal is smooth. A shallow, but rather wide
articulation surface for the postorbitofrontal is visible on the right lateral side of the posterior end of
the bone. The anterior end of this articulation surface is located far from the posterior end of the
dorsal process of the prefrontal. The preserved length of the frontal can be measured as 5 mm, but
the bone was probably slightly more than 6/7 mm (probably even about 8 mm) in origin. The
parietals (Figs. 3, 5A,B) are paired and poorly preserved, but an original rectangular shape is
recognisable based on what remains. The anterior margin is slightly concave. By the anterolateral
corner, the anterolateral process is well developed and has a rounded end. There is no notch
medially to it. The posterior end is not preserved. Judging from the preserved portion of the lateral
margin of the left parietal, the lateral lamina seems poorly developed. The dorsal surface of the
parietal table is smooth, except for few very shallow grooves, possibly representing a light
sculpturing, that are present near the anterolateral corner of the left element (this area is broken off
in the right parietal). The postparietal processes are missing. Only the alveolar portion and a small
part of the facial process of the right maxilla (Fig. 3A,B) are preserved, whereas only the alveolar
portion of the left maxilla (Fig. 4), including the posterior process, is exposed in ventral view. The
lateral surface of the preserved portion of the facial process is smooth. At least six ventrolateral
foramina are recognisable. The posteriormost one is located on the posterior process and is followed
by a long and shallow groove (Fig. 5A,B). Dorsally to these foramina, two other small foramina are
visible: one near the anterior end and one near the posterior end of the facial process. The last
ventrolateral foramen and the related groove are visible on the left maxilla also. The ventral end of
the anterior margin of the facial process displays a gentle curvature. The anterolateral process is
well developed. The posterior process is long and rather slender, tapering posteriorly and ending
with a pointed tip. The teeth of the right maxilla are only visible in the scans, because they are
partially covered by matrix, but they can be easily observed also in the actual specimen in the left
one. They are pleurodont, cylindrical, closely spaced and rather slender. They are small at the
anterior end, slightly increase in size posteriorly, but then become smaller again towards the
posterior end. The tooth row reaches the posterior end of the posterior process. At least 13 tooth
positions can be counted in the right maxilla, but most of the tooth row is hidden. The left element,
on the other hand, shows at least 27 tooth positions. Few teeth in the left maxilla are preserved
enough to show the crowns: these are identical to those of the dentaries (see below). Some of them
clearly display two parallel cutting edges, oriented mesiodistally. The right maxilla is roughly 9.5
mm long. The preserved portion of the left one is 9 mm long, but it should have been longer when
intact, reaching the same size of its counterpart. The right prefrontal (Figs. 3A,B, 5A,B) is crescent-
shaped in lateral view. The orbitonasal flange is large. The exposed dorsal surface and the posterior
surface are smooth. The anterodorsal process is broken and neither the orbitonasal flange projection
nor the distal end of the posteroventral process is visible. The pointed dorsal process is rather robust
and long. It is as long as the orbitonasal flange is wide. The small right postorbitofrontal (Figs.
3A,B, 5A,B) is partially preserved and Y-shaped. The anterior process is long and tapers anteriorly.
Its anterior end is missing. The posterior process is completely missing. Laterally, there is a well-
developed lateral process, which is triangular and shorter than the anterior process. A small foramen
is visible at the base of the posterior process (Fig. 5A,B). The ventral ends of both quadrates (Figs.
3, 5A,B) are preserved. Both have a wide mandibular condyle, the medial side of which is flat and
ventrally less developed than the lateral side. There is no medial lamina. A poorly preserved,
slender and rod-like bone preserved posteriorly to the posterior process of the maxilla and dorsally
to the coronoid might be the right epipterygoid (Fig. 3). Part of the left epipterygoid is also
preserved, being clearly visible only in the scan near the left quadrate (Figs. 3B). A long bone
located ventrally to the mandibular end of the left quadrate might be part of the left pterygoid (Figs.
3A,B, 5A,B), including mainly the quadrate process. This latter process is distinctly curved in
lateral direction, but its posterior end is not visible. A flat area on the medial surface near the
anterior end of the exposed portion of the bone might represent the poorly marked basipterygoid
fossa. The anterior portion of the bone is crushed. A small piece of bone located dorsally to the
posterior process of the right maxilla is most likely the right ectopterygoid (Fig. 3). It has a rather
flat aspect and seems to taper anteriorly. Only a small portion of the bone is visible, but the
articulation surface with the pterygoid flange of the pterygoid is visible posteriorly. Part of the
otooccipital region might be visible posteriorly, but the bones are difficult to interpret. A rather long
and robust bone might be the right paroccipital process seen in dorsal view (Fig. 3A,B), but it seems
rather robust. Anteriorly to it, there is also a void and curved structure that might be one of the
semicircular canals.

Lower jaw. Both dentaries are preserved (Figs. 3, 4). On the medial side, the Meckelian fossa is
closed in a tubular structure by the expansions of the ventral margin and the subdental shelf. The
tubular structure is narrow and tapers anteriorly. The mandibular symphysis is preserved only in the
left dentary, even though in poor condition. It was most probably narrow and slightly dorsally
inclined. The lateral surface of the dentaries is smooth. The complete series of the mental foramina
is not recognisable, given that only three of them are visible on the left dentary and none on the
right one. The ventral margin of these slender bones is straight. The posterior portion of the
dentaries is damaged and so the posterior processes are not clearly recognisable. At least 18 tooth
positions are visible on the right dentary, whereas the left one displays at least 14 of them (four by
the anterior end in part a and the rest in part b). Teeth are pleurodont, cylindrical, slender and
closely spaced. The crown is rather pointed and does not bend significantly in posteromedial
direction. It is also not striated lingually. The preserved part of the right dentary is 8.5 mm long.
Estimating the length of the left one is difficult. The portion preserved on part a (i.e., the anterior
end) is roughly 5.5 mm long, whereas that on part b is about 8 mm. However, it is not clear how
much these two portions are superimposed. Nevertheless, it is possible to estimate the complete
original length of the dentary at around 10 mm. The right coronoid is preserved in part a (Fig. 3). It
is slender and concave in medial direction. The anteromedial process is not visible, but a short labial
process is exposed. The coronoid process is dorsally broken, but it was connected to the
posteromedial process by an osseous lamina. The posteromedial process is long and thin. Its distal
tip is not clearly visible, but it seems rounded in ventral view. Part b preserves the left coronoid
(Fig. 4), which is clearly visible only in the scans and has a morphology that is comparable with that
of the right one. Most of the left compound bone and part of the right one are very poorly preserved
in part b (Fig. 4). In part a (Fig. 3), the right compound bone is visible in lateral view, whereas part
the left one is displaced dorsally and it is visible in ventral view. The angular was most likely fused
to the compound bone, as evidenced by the lack of an evident articular surface for this bone in the
left element. The lateral surface of the bone is smooth, with a large anterior surangular foramen
located anteriorly near the dorsal margin. The posterior surangular foramen is visible only on the
left element and only in the scans. It appears rather large and it seems somehow shifted towards the
dorsal margin of the bone, even though the crushed state of the specimen renders its recognised
position less confident. The right compound bone is broken just by the position of the articular
condyle. The posterior end of the left one shows an expansion, which most likely also represents the
condyle. A narrow posterior projection is probably the retroarticular process, the posterior end of
which is broken.

Remarks

GMH Ce IV-4057-1933 can be clearly referred to gekkotans due to the presence of distinctive
features, such as the unpaired and hourglass-shaped frontal, paired and elongated parietals, presence
of a Y-shaped postorbitofrontal, completely closed Meckelian fossa on the dentary, and pleurodont
and closely-spaced teeth with an unstriated crown (Sumida & Murphy 1987; Evans 2008; Villa et
al. 2018).

However, it displays significant differences compared to all fossil species of geckos previously
reported from the European continent, as well as to the four most common extant European
gekkotans. This allows assignment to its own taxon. Geiseleptes delfinoi differs from
Cadurcogekko piveteaui in having grooves instead of depressions in the posterior portion of the
frontal, a posterior margin of the frontal that is concave instead of straight, and a smooth lateral
surface of the maxilla (Hoffstetter 1946; Augé 2005, Daza et al. 2014). Cadurcogekko piveteaui
probably had also more teeth than Geiseleptes (38 maxillary teeth according to Augé, 2005, but
Daza et al., 2014, estimates between 40 and 44 teeth for the maxilla of the French species).
Geiseleptes delfinoi differs from C. verus in having smooth maxillae and frontal (Augé 2005; Daza
et al. 2014; Bolet et al. 2015). It differs from both Gerandogekko arambourgi and G. gaillardi
because of the shorter grooves on the frontal and at least from G. arambourgi in the presence of the
posterior groove on the maxilla (Hofstetter 1946; Daza et al. 2014). It also differs from Laonogekko
lefevrei in the presence of a groove posteriorly to the last ventrolateral foramen in the maxilla (Augé
2003, 2005; Daza et al. 2014). Similar to C. piveteaui, L. lefevrei may also have more teeth than the
Geiseltal species: Augé (2003, 2005) reported 38 maxillary teeth and 45 dentary teeth, but Daza et
al. (2014) gave a slightly lower (up to 33 tooth positions in the most complete specimen) dentary
teeth count. Geiseleptes delfinoi differs from Palaeogekko risgoviensis in having a posterior groove
on the maxilla and in the distinctly larger size (Schleich 1987; Daza et al. 2014). It also differs from
Rhodanogekko vireti in the frontal that is less constricted at midlength, dorsally smooth and
provided with a concave posterior margin and posterior grooves (Hofstetter 1946; Estes 1983; Daza
et al. 2014). Differences between Geiseleptes delfinoi and Euleptes klembarai include the
significantly larger size (the length of the maxilla is almost double), the posterior groove and the
anterolateral process on the maxilla that are better developed in the German taxon (Čerňanský et al.
2018). Geiseleptes delfinoi differs from E. gallica mainly because of its distinctly larger size
(Müller 2001; Čerňanský & Bauer 2010; Daza et al. 2014); all the diagnostic features of E. gallica
are located either on the premaxilla or on the facial process of the maxilla, which are respectively
completely or partially unknown for Geiseleptes delfinoi. Due to its peculiar preservational nature
(it is preserved in amber), it is difficult to compare Yantarogekko balticus to other fossil geckos.
Nevertheless, Geiseleptes delfinoi is distinctly larger than this early Eocene species (Bauer et al.
2005).

As far as the extant European species are concerned, Geiseleptes delfinoi differs from E. europaea
in (Villa et al. 2018): posterior margin of frontal concave; posterior grooves on the frontal less
marked and shorter; well-marked articular surfaces with postorbitofrontal on the frontal; larger size
(based on both the length of frontal and maxilla); dorsal surface of the parietals largely smooth, but
with few grooves; anterolateral process of the maxilla more developed; posterior process of the
maxilla pointed; dorsal process of the prefrontal robust; presence of a lateral process on the
postorbitofrontal; presence of a foramen on the postorbitofrontal. It differs from Hemidactylus
turcicus in (Villa et al. 2018): frontal and maxilla smooth; presence of posterior grooves on the
frontal; posterior margin of frontal concave; anterolateral process of the parietal narrower; absence
of a notch medially to the anterolateral process of the parietal; lateral lamina of the parietal poorly
developed; groove present posteriorly to last ventrolateral foramen on the maxilla; larger size
(based on maxilla); presence of a lateral process on the postorbitofrontal; presence of a foramen on
the postorbitofrontal; and maybe also posterior surangular foramen shifted dorsally. It differs from
Mediodactylus kotschyi in (Villa et al. 2018): presence of posterior grooves on the frontal; absence
of a notch medially to the anterolateral process of the parietal; lateral lamina of the parietal poorly
developed; dorsal surface of the parietals largely smooth, but with few grooves; anterolateral
process of the maxilla more developed; anterior margin of the facial process not vertical; groove
present posteriorly to last ventrolateral foramen on the maxilla; larger in size (based on maxilla);
dorsal process of the prefrontal robust; presence of a lateral process on the postorbitofrontal;
presence of a foramen on the postorbitofrontal; posterior surangular foramen shifted dorsally. It also
differs from Tarentola mauritanica in (Villa et al. 2018): presence of posterior grooves on the
frontal; posterior margin of the frontal concave; anterolateral process of the parietal well developed;
lateral lamina of the parietal poorly developed; dorsal surface of the parietals largely smooth, but
with few grooves; absence of a notch on the anterior margin of the facial process of the maxilla;
presence of a groove posterior to the last ventrolateral foramen on the maxilla; presence of a
foramen on the postorbitofrontal.

Geiseleptes delfinoi also displays some similarities with both the fossil and the extant European
gekkotans. It shares a moderately constricted frontal with Cadurcogekko, G. arambourgi, and L.
lefevrei, the presence of posterolateral grooves on the frontal with E. europaea and Gerandogekko,
and the concave posterior margin of the frontal with G. arambourgi and M. kotschyi. According to
Bauer et al. (2005), Y. balticus could have an expanded (i.e., not strongly constricted) frontal, which
would be similar to Geiseleptes. The posterolateral grooves on the frontal are shallow in both
Geiseleptes delfinoi and Gerandogekko, whereas they are more marked in E. europaea. The
articular surface with the postorbitofrontal is visible on the frontal also in C. piveteaui, G.
arambourgi, H. turcicus, L. lefevrei, and T. mauritanica. The dorsal surface of the frontal is smooth
in E. europaea, Gerandogekko, L. lefevrei (Augè, 2003, mentions a dermal ornamentation, but Daza
et al., 2014, wrote that it is just the result of erosion of a rather smooth dorsal surface of the bone),
M. kotschyi, T. mauritanica, and probably also Y. balticus (according to Bauer et al., 2005, this
species probably lacks ossified osteoderms on the frontal), whereas the maxilla is laterally smooth
in Euleptes, G. gaillardi, L. lefevrei, M. kotschyi, P. risgoviensis, and T. mauritanica. The posterior
groove on the maxilla is shared by Geiseleptes delfinoi, Cadurcogekko, and Euleptes. Other
maxillary features that are shared by Geiseleptes delfinoi with other species are a similar degree of
development of the anterolateral process (with H. turcicus and T. mauritanica) and the pointed
posterior process (with H. turcicus, M. kotschyi, and T. mauritanica). The parietal of Geiseleptes
delfinoi is similar to the one of E. europaea and M. kotschyi in having a well-developed
anterolateral process, the one of E. europaea in having a poorly-developed lateral lamina and no
notch medially to the anterolateral process, and the one of H. turcicus in the presence of grooves on
the dorsal surface (even though they are fewer in the Geiseltal species). The prefrontal of
Geiseleptes delfinoi recalls those of both H. turcicus and T. mauritanica in the robustness of its
dorsal process. The postorbitofrontal of Geiseleptes delfinoi displays a lateral process as in T.
mauritanica. In the end, the dorsally-shifted posterior surangular foramen (if confirmed) would be
shared by Geiseleptes delfinoi, E. europaea, and T. mauritanica.



                                           ?Gekkota indet.

Material. GMH Ce IV-4123-1932, a slab with a number of disarticulated and very poorly preserved
skeletal elements.

Description

This specimen is a mix of different small-sized bones, mainly postcranial elements. Most of them
are slender limb bones, but there is also a left maxilla. The morphology of the maxilla recalls that of
a gecko, but the bone is too poorly preserved and the teeth are not well visible. The morphology
appears somehow similar to the right maxilla of GMH Ce IV-4057-1933. Two small femoral
condyles are also visible on the slab.

Remarks

The bones in this specimen (at least the maxilla) might indeed pertain to a gecko, maybe even the
same taxon as GMH Ce IV-4057-1933. However, the preservation is not good enough to clearly
confirm that.

PHYLOGENETIC ANALYSIS

A first iteration of the analysis failed to recover the currently recognised topology of gekkotan
phylogenetic relationships based on molecular data (as reported by Gamble et al., 2015), and we
therefore rerun the analysis with the same settings and six constraints to force the monophyly of: a)
Diplodactylidae + (Pygopodidae + Carphodactylidae); b) Pygopodidae + Carphodactylidae; c)
Eublepharidae + (Sphaerodactylidae + (Gekkonidae + Phyllodactylidae)); d) Sphaerodactylidae +
(Gekkonidae + Phyllodactylidae), e) Sphaerodactylidae; and f) Gekkonidae. All fossil gekkotans
were treated as wild-card taxa in this second iteration. The second run resulted in 568 trees with a
minimum length of 783 steps. The strict consensus tree (Fig. 6A) recovered most of the ingroup
taxa, including Geiseleptes, in a wide polytomy, though. Gerandogekko arambourgi, a poorly-
known taxon only scored for a few characters, is a wild-card taxon and since it is operationally
identical to the congeneric G. gaillardi (Daza et al., 2014, even question whether they represent
separate species), we removed it from the analysis. The analysis without G. arambourgi obtained 20
trees with a length of 782 steps. In the strict consensus tree (Fig. 6B), an early-branching clade
including the Cretaceous Norellius and Gobekko is recognised that is sister to Hoburogekko and the
crown clade Gekkota. Geiseleptes is always recovered within Sphaerodactylidae. Pruning Euleptes
gallica, Laonogekko lefevrei, and Palaeogekko risgoviensis resulted in an almost completely-
resolved strict consensus tree (Fig. 6C), in which a polytomy persisted within Sphaerodactylidae
and including Geiseleptes, all remaining species of Euleptes, G. gaillardi, and Cadurcogekko. In the
strict consensus, Sphaerodactylidae is supported by three synapomorphies (C137, 403, 429) and the
less inclusive new clade Euleptinae by five (C242, 837, 840, 843, 846). Alternative positions of the
pruned taxa are shown in Fig. 6C. For the time being, we are not able to further resolve the
relationship within this European sphaerodactylid clade. Regarding other extinct European
gekkotans, Rhodanogekko is nested within the Carphodactylidae in all analyses, whereas
Yantarogekko is sister to Phelsuma inside Gekkonidae. Consistency and Retention indexes of the
analysis without G. arambourgi are 0.579 and 0.397, respectively.

DISCUSSION

Antiquity of European sphaerodactylid gekkotans (Euleptinae)

With few exceptions, the gekkotan fossil record is largely represented by isolated remains, for
which a confident phylogenetic framework is often difficult to assess. Geiseleptes thus presents
itself as an important specimen to shed light on the relationships of fossil geckos, in particular when
it comes to Paleogene representatives of the clade from Europe. This taxon shows unique features
that ally it with a single species within the extant European inhabitants: the Mediterranean endemic
sphaerodactylid Euleptes europaea. In both G. delfinoi and E. europaea a groove is present on the
lateral surface of the posterior process of the maxilla (starting from the last ventrolateral foramen),
as well as on each posterolateral process of the frontal dorsally. The combination of these features is
unknown in any other living European gekkotans (Villa et al. 2018) and is absent in other extant
members of Sphaerodactylidae too (Daza et al. 2008; J.D. Daza, pers. comm. 2020). Short
posterolateral dorsal grooves on the frontal may be present in at least one Asiatic sphaerodactylid
(Pristurus carteri; see Daza et al. 2014: fig. 4D), and some pygopodoids have a maxillary groove
behind the last ventrolateral foramen (Daza et al. 2014: p. 445–446). These features are variably
present also in fossil gekkotans from Europe. Fossil maxillae referred to extinct Euleptes display the
posterior groove on the maxilla. This is unknown for the Czech E. gallica reported by Čerňanský &
Bauer (2010): the authors made no mention of this feature and did not figure the lateral surface of
the maxillae. At the moment, there are no frontals referred to extinct Euleptes species, and thus the
presence and development of grooves on the posterolateral processes cannot be evaluated. When it
comes to extinct European genera, the grooves/depressions on the maxilla and the frontal are
present in Cadurcogekko piveteaui, whereas in Gerandogekko arambourgi the maxillary groove is
absent. Gerandogekko was referred to Sphaerodactylidae (Daza et al. 2014), a consideration that is
supported by, among other things (i.e., wide frontals, amphicoelous vertebrae; Daza et al. 2014), the
frontal groove morphology shared with Euleptes.

Our phylogenetic analysis recovers the presence of both the posterior maxillary grooves and
depressions/grooves on the posterolateral process of the frontals as unambiguous synapomorphies
of a European sphaerodactylid clade composed by Geiseleptes, G. gaillardi, two out of three
Euleptes species, and Cadurcogekko spp. (Fig. 6C; see also Appendix S2). Other synapomorphies
include a broad ascending nasal process of the premaxilla, a rounded dorsal end of the facial
process of the maxilla, and cervical vertebrae devoid of hypapophyseal keel. We here define
Euleptinae for this clade, including sphaerodactylids more closely related to E. europaea than to
Teratoscincus scincus. The maxillary groove is unknown for G. gaillardi and absent in G.
arambourgi (removed from the final analysis). Thus, depending on the position of these taxa, the
ungrooved condition may be plesiomorphic for Euleptinae or, alternatively, a reversal occurred at
least in Gerandogekko arambourgi. Since no other extant sphaerodactylids show the combination of
maxillary/frontal grooves, G. delfinoi is confidently reconstructed as an early representative of the
lineage leading to the extant E. europaea (i.e., a member of Euleptinae), irrespective of a limited
sample size of extant sphaerodactylids in our phylogeny. The origin of euleptine sphaerodactylids
can thus be traced back at least to the middle Eocene (Fig. 7). This is consistent with the molecular
divergence estimate of Gamble et al. (2015), which dated the divergence between E. europaea and
its closest extant relative into the mid-Cretaceous. Based on the temporal distribution of Euleptinae
(Fig. 7), the clade is best interpreted as a permanent resident of Europe since the early Paleogene.
Their survival through major climatic shifts (from greenhouse to icehouse conditions) and faunal
turnovers (Eocene/Oligocene transition, Oligo-Miocene transition, and several others during the
Neogene and Quaternary) implies extreme adaptability (or alternatively geographical shifts and
recolonizations, which are however not supported by a literal interpretation of the current fossil
record). Hardly any other vertebrate groups from the European early Paleogene appear to survive in
the continent till today with all exceptions being reptiles and amphibians. Phylogenies and the
currently known fossil record of lacertid (Čerňanský & Smith 2018), as well as anguid lizards (Rage
2012; Klembara et al. 2019), blanid amphisbaenians (Čerňanský et al. 2015), and discoglossine and
perhaps pelobatid frogs (Rage 2012) are best interpreted as persistent presence of at least some
members of these clades since the Eocene. In the case of lacertids, anguids, and blanids, the fossil
record of these groups in Africa and Asia is incompletely sampled, however, and key Paleogene
species are yet to be included into global phylogenies.

Euleptine gekkotans were more diverse and widespread in the past in Europe (Fig. 7, 8), with
different taxa potentially coexisting (i.e., Gerandogekko and Euleptes during the early Miocene, and
Euleptes and possibly Palaeogekko in the middle Miocene; Fig. 7). Eventually, E. europaea
remained the only survivor of the European euleptine lineage, even though with a large gap
encompassing the whole Pliocene and the Quaternary. Together with the overall scarcity of the
fossil record, this currently prevents any attempt at reconstructing the reasons behind this. A single
mention of a still-unpublished occurrence of a form closely related to E. europaea in Sardinia by
Villa & Delfino (2019a) supports the establishment of this lineage in the Mediterranean already
since the late Early Pleistocene (A.V., pers. obs.) and is consistent with a central Mediterranean
insular center of origin proposed by Delaugerre et al. (2011).

European sphaerodactylid palaeobiogeography

Even considering the extra-European record of sphaerodactylids, Geiseleptes is the oldest definite
occurrence of the clade worldwide (Fig. 7). This cannot be seen as an evidence of a European origin
for the clade, though, given that both splits between Sphaerodactylidae and other closely-related
gekkotan clades, as well as between Old-World and New-World sphaerodactylids were estimated to
occur in the Cretaceous (Gamble et al. 2008). The extant genus Sphaerodactylus is known to have
been present on the island of Hispaniola during the Miocene (Böhme 1984; Kluge 1995; Grimaldi
1996; Grimaldi et al. 2000; Daza & Bauer 2012; Daza et al. 2014), but nothing else is known about
the past histories of African, Asian and American sphaerodactylids (Fig. 7, 8). According to
Gamble et al. (2015), the closest living relative of Euleptes is the Asian Teratoscincus, but these
two taxa are sister to the African Saurodactylus fasciatus. Based on Gamble et al.’s (2015)
phylogeny and the distribution of extant taxa, the most parsimonious scenario is an African origin
of European Sphaerodactylidae with a subsequent dispersal to Europe giving rise to the Euleptes
lineage. Geiseleptes delfinoi implies that this dispersal took place no later than the middle Eocene.
Nevertheless, the biogeographic history of the European sphaerodactylid clade may turn out to be
more complex once the Asian fossil record is better sampled. An Old-World origin for Geiseleptes
is at odds with American affinities suggested for other Geiseltal squamates (e.g., Geiseltaliellus,
glyptosaurine anguids; Smith 2009; Sullivan 2019) and other squamate faunas from the European
early Paleogene (Smith & Gauthier 2013; Rage 2013; Smith 2017; Scanferla & Smith 2020; Smith
& Scanferla 2021) and best explained by the comparably older age of the clade (Euleptes split from
its closest living relative, Teratoscincus, in the Cretaceous; Gamble et al. 2015). It should be noted,
however, that some evidence of the presence in Eocene Europe of African-related taxa are already
known (e.g., putative cordylids/cordyliforms; Augé 2003, 2005; Bolet & Evans 2013).

Relationships of other European extinct gekkotans

Apart from sphaerodactylids, our phylogenetic analysis yielded interesting results regarding other
extinct gekkotans. The Eocene amber-preserved Yantarogekko balticus was recovered as a
gekkonid, in agreement with the study describing this taxon but lacking a phylogeny (Bauer et al.
2005). Rhodanogekko vireti is found nested within the Carphodactylidae, a clade that is endemic of
Australia and has currently no fossil record. This is rather surprising from a palaeobiogeographical
point of view, but should be taken with caution given that Rhodanogekko is known only from an
isolated frontal. Daza et al. (2014) reported similarities shared by Rhodanogekko and the
Afroasiatic sphaerodactylid Pristurus, hinting at either close phylogenetic relationships or
convergence to explain them. Inclusion of at least this latter genus in future analyses would be
needed in order to better clarify the systematics of Rhodanogekko. Our analysis was not focused on
understanding the relationships of Cretaceous stem gekkotans, but it is interesting to note that we
were able to recover a clade formed by Norellius and Gobekko and Hoburogekko as subsequent
sister to all Gekkota.

CONCLUSION

The Eocene squamate assemblage of Geiseltal is characterized by a high diversity of taxa, including
members of anguids, iguanids, eolacertids, palaeovaranids, boids, and a species possibly related to
aniliids (Krumbiegel et al. 1983). To these, a new sphaerodactylid gekkotan is now added,
represented by an incomplete skull. This new gecko taxon, Geiseleptes delfinoi, is part of the stem
lineage of the only extant European sphaerodactylid, the Mediterranean-endemic Euleptes
europaea. A phylogenetic analysis was not able to clearly disentangle the detailed relationships
within European extinct and extant sphaerodactylids, but it resulted in the recovery of almost all
European extinct gekkotan species as members (or possible members in the case of the unstable
taxa) of this clade. The name Euleptinae is here proposed to accommodate all European
sphaerodactylids, both extant and extinct. Euleptinae appears to have a permanent presence in the
European fauna since the early Paleogene, in marked contrast with the Neogene origin of almost all
other vertebrates of the continent, implying exceptional ecological flexibility. Among European
fossil gekkotans, only Yantarogekko and Rhodanogekko are consistently placed outside
Sphaerodactylidae. The first one is confirmed as a gekkonid, whereas the second one is nested
within Carphodactylidae, even though possible affinities with some sphaerodactylids were reported
for this genus as well in the past (Daza et al. 2014) and convergence with the Australian
carphodactylids of the only frontal referred to it should be taken into account before completely
accepting this result. Sphaerodactylids in modern Europe are only represented by E. europaea in a
limited range in the Western Mediterranean, whereas phyllodactylids and gekkonids are more
widely distributed and diverse (Speybroeck et al. 2016, 2020). In contrast to this pattern,
Sphaerodactylidae appears to be the dominant clade of gekkotans in Europe during the Cenozoic. A
literal reading of the fossil record implies that ancestors of extant gekkonids and phyllodactylids
may have entered Europe in more recent times (i.e., late Neogene or Quaternary; Villa & Delfino
2019a).

In any case, Geiseleptes delfinoi is the oldest certain sphaerodactylid known to date, not only for
Europe but worldwide, and it is therefore an important calibration point for future divergence dating
analyses. It will also be important for the reconstruction of character distributions and phylogenetic
relationships within this group of geckos, given its preservation. Euleptes europaea is nested within
a clade of ancestrally-nocturnal gekkotans (Gamble et al. 2015), suggesting that Geiseleptes may
also have been nocturnal. Unfortunately, this is not supported by morphological evidence at the
moment. The tooth shape, on the other hand, recalls the condition typically observed in
insectivorous geckos (Sumida & Murphy 1987; Villa et al. 2018). Thence, Geiseleptes likely had an
arthropod-based diet, a thing that agrees well with the rich record of insects from Geiseltal
(Krumbiegel et al. 1983; Steinheimer & Hastings 2019). Despite our extensive search, we found
only a single specimen referable to G. delfinoi (plus another possible indeterminate gekkotan fossil,
which may represent the same taxon) in the Geiseltal Collection. This is puzzling considering the
large number of often well-preserved lizard remains from the Geiseltal deposits. It is plausible that
this gecko lived further away from the swamps and water bodies of the lignite depositional area and
is therefore poorly represented in the Geiseltal taphocoenosis.

Acknowledgements.

This study was supported by the VolkswagenStiftung Research in Museums grant (Az. 90 978)
awarded to M.R. We are grateful to Juan D. Daza (Sam Houston State University) for help with the
original matrix and discussions on characters. Massimo Delfino, Loredana Macaluso, and Lukardis
C. M. Wencker (all from the Università degli Studi di Torino) are thanked, respectively, for: access
to comparative material of extant geckos; taking and sharing photographs of fossil taxa; and help
with the phylogenetic analysis. Michael Stache and Frank Steinheimer (Martin Luther University,
Halle-Wittenberg) are thanked for hosting M.R. and A.V. during their stay at the Geiseltal Museum.
This study benefited from useful discussions with Aaron M. Bauer (Villanova University), Arnau
Bolet (Institut Català de Paleontologia Miquel Crusafont), and Krister Smith (Senckenberg
Museum). We thank the editors Kenneth Angielczyk and Sally Thomas, as well as the two
reviewers Juan D. Daza and Marc Augé (Muséum national d’Histoire naturelle), for their useful
comments that significantly improved this paper. Original maps used for Fig. 8 are freely available
online at d-maps (https://d-maps.com/carte.php?num_car=2232&lang=it and https://d-
maps.com/carte.php?num_car=231325&lang=it) and from Wikimedia Commons. A.V. was
supported by a Humboldt Research Fellowship from the Alexander von Humboldt Foundation
during part of the development of this work; he is now funded by the Agencia Española de
Investigación, with a Juan de la Cierva-Formación grant (FCI2019-039443-I / AEI /
10.13039/501100011033). This is publication number 359 of the Museum of Geology and
Palaeontology of the Università degli Studi di Torino.

DATA ARCHIVING STATEMENT

Phylogenetic data for this study are available in MorphoBank (O’Leary & Kaufman 2012)
(http://morphobank.org/permalink/? P4069); scan data in MorphoSource
(https://doi.org/10.17602/M2/M409939 (image stack);
https://doi.org/10.17602/M2/M407451(mesh)). This published work and the nomenclatural acts it
contains have been registered with ZooBank: http://zoobank.org/References/D59964BB-2E5E-
492F-BCAE-0D73BF8B5181.

SUPPORTING INFORMATION

Additional Supporting Information can be found in the online version of this article:

Appendix S1. New characters added in the phylogenetic analysis.

Appendix S2. List of apomorphies (character numbers follow TnT: i.e., starting from 0) for selected
       clades, calculated based on the pruned consensus tree (Fig. 6C, with Gerandogekko
       arambourgi inactivated, and without Palaeogekko, Laonogekko, and Euleptes gallica).

AUTHOR CONTRIBUTION

Conceptualization A Villa, M Rabi; Data Curation A Villa, O Wings, M Rabi; Formal Analysis
A Villa, M Rabi; Funding Acquisition M Rabi; Investigation A Villa, M Rabi; Methodology A
Villa, O Wings, M Rabi; Project Administration O Wings, M Rabi; Resources O Wings, M Rabi;
Supervision O Wings, M Rabi; Validation A Villa, O Wings, M Rabi; Visualization A Villa, O
Wings, M Rabi; Writing – Original Draft Preparation A Villa, O Wings, M Rabi; Writing –
Review & Editing A Villa, O Wings, M Rabi.

REFERENCES

AGNOLIN, F. and JOFRÉ, G. 2011. Nuevos registros de Squamata (Reptilia) para el Pleistoceno
Superior del norte de la provincia de Buenos Aires, Argentina. Papéis Avulsos de Zoologia, 51, 49–
58. [In Spanish]

ALBINO, A. M. 2005. A late Quaternary lizard assemblage from the southern Pampean Region of
Argentina. Journal of Vertebrate Paleontology, 25, 185–191.

ALBINO, A. M. 2020. The Cenozoic lizard record of the Pampean Region. Geodiversitas, 42, 175–
183.

ALIFANOV, V. R. 1989. The oldest gecko (Lacertilia, Gekkonidae) from the Lower Cretaceous of
Mongolia. Paleontologicheskii Zhurnal, 1989, 124–126. [In Russian]

ARNOLD, E. N. and POINAR, G. 2008. A 100 million year old gecko with sophisticated adhesive
toe pads, preserved in amber from Myanmar. Zootaxa, 1847, 62–68.

AUGÉ, M. L. 2003. La faune de Lacertilia (Reptilia, Squamata) de l’Éocène inférieur de Prémontré
(Bassin de Paris, France). Geodiversitas, 25, 539–574.

― 2005. Évolution des lézards du Paléogène en Europe. Mémoires du Muséum national d'Histoire
naturelle, 192, 1–369.

― and RAGE, J.-C. 2006. Herpetofaunas from the upper Paleocene and lower Eocene of Morocco.
Annales de Paléontologie, 92, 235–253.

AUTUMN, K., SITTI, M., LIANG, Y. A., PEATTIE, A. M., HANSEN, W. R., SPONBERG, S.,
KENNY, T. W., FEARING, R., ISRAELACHVILI, J. N. and FULL, R.J. 2002. Evidence for van
der Waals adhesion in gecko setae. Proceedings of the National Academy of Sciences of the United
States, 99, 12252–12256.

BAILON, S., HOSSINI, S. and RAGE, J.-C. 2017. Les amphibiens et lépidosauriens du
Cénozoïque du Maroc. 453–484. In ZOUHRI, S. (ed). Paléontologie des vertébrés du Maroc: état
des connaissances. Mémoires de la Société géologique de France, 180, 624 pp.

BAUER, A. M. 2013. Geckos: the animal answer guide. The John Hopkins University Press,
Baltimore, 192 pp.
― BÖHME. W. and WEITSCHAT, W. 2005. An early Eocene gecko from Baltic amber and its
implications for the evolution of gecko adhesion. Journal of Zoology, 265, 327–332.

― GOOD, D. A. and BRANCH, W. R. 1997. The taxonomy of the southern african leaf-toed
geckos (Squamata: Gekkonidae), with a review of old world “Phyllodactylus” and the description of
five new genera. Proceedings of the California Academy of Science, 49, 447–497.

BLANCO, A., BOLET, A., BLAIN, H.-A., FONDEVILLA, V. and MARMI, J. 2016. Late
Cretaceous (Maastrichtian) amphibians and squamates from northeastern Iberia. Cretaceous
Research, 57, 624–638.

BÖHME, M. and ILG, A. 2003. fosFARbase [accessed 2019 April]. www.wahrestaerke.com/

BÖHME, W. 1984. Erstfund eines fossilien Kugelfingergeckos (Sauria: Gekkonidae:
Sphaerodactylinae) aus Dominikanischem Bernstein (Oligozän von Hispaniola, Antillen).
Salamandra, 20, 212–220.

BOLET A., DAZA J. D., AUGÉ, M. and BAUER, A. M. (2015). New genus and species names for
the Eocene lizard Cadurcogekko rugosus Augé, 2005. Zootaxa, 3985, 265–274.

― EVANS, S. E. 2013. Lizards and amphisbaenians (Reptilia, Squamata) from the late Eocene of
Sossís (Catalonia, Spain). Palaeontologia Electronica, 16, 8A.

BORSUK-BIAŁYNICKA, M. 1990. Gobekko cretacicus gen. et. sp. n., a new gekkonid lizard from
the Cretaceous of the Gobi Desert. Acta Palaeontologica Polonica, 35, 67–76.

CAMP, C. L. 1923. Classification of the lizards. Bulletin of the American Museum of Natural
History, 48: 289–481.

ČERŇANSKÝ, A., AUGÉ, M. L. and RAGE, J.-C. 2015. A complete mandible of a new
amphisbaenian reptile (Squamata, Amphisbaenia) from the late middle Eocene (Bartonian, MP 16)
of France. Journal of Vertebrate Paleontology, 35, e902379.

ČERŇANSKÝ, A. and BAUER, A. M. 2010. Euleptes gallica Müller (Squamata: Gekkota:
Sphaerodactylidae) from the Lower Miocene of North-West Bohemia, Czech Republic. Folia
Zoologica, 59, 323–328.

― DAZA J. D. and BAUER, A. M. 2018. Geckos from the middle Miocene of Devínska Nová Ves
(Slovakia): new material and a review of the previous record. Swiss Journal of Geosciences, 111,
183–190.
― SMITH, K. T. 2018. Eolacertidae: a new extinct clade of lizards from the Palaeogene; with
comments on the origin of the dominant European reptile group – Lacertidae. Historical Biology,
30, 994–1014.

COHEN, K. M., FINNEY, S. C., GIBBARD, P. L. and FAN, J.-X. 2013; updated. The ICS
International Chronostratigraphic Chart. Episodes, 36, 199–204.

COLOMBERO, S., ALBA, D. M., D'AMICO, C., DELFINO, M., ESU, D., GIUNTELLI, P.,
HARZHAUSER, M., MAZZA, P. P. A., MOSCA, M., NEUBAUER, T. A., PAVIA, G., PAVIA,
M., VILLA, A. and CARNEVALE, G. 2017. Late Messinian mollusks and vertebrates from
Moncucco Torinese, north-western Italy. Paleoecological and paleoclimatological implications.
Palaeontologia Electronica, 20.1.10A, 1–66.

CONRAD, J. L. 2008. Phylogeny and systematics of Squamata (Reptilia) based on morphology.
Bulletin of the American Museum of Natural History, 310, 1–182.

― 2015. A new Eocene casquehead lizard (Reptilia, Corytophanidae) from North America. PLoS
ONE, 10, e0127900.

― 2018. A new lizard (Squamata) was the last meal of Compsognathus (Theropoda: Dinosauria)
and is a holotype in a holotype. Zoological Journal of the Linnean Society, 183, 584–634.

― AST, J. C., MONTANARI, S. and NORELL, M. A. 2011. A combined evidence phylogenetic
analysis of Anguimorpha (Reptilia: Squamata). Cladistics, 27, 230–277.

― BALCARCEL, A. and MEHLING, C. 2012. Earliest example of a giant monitor lizard
(Varanus, Varanidae, Squamata). PLoS ONE, 7, e41767.

― and DAZA, J. D. 2015. Naming and rediagnosing the Cretaceous gekkonomorph (Reptilia,
Squamata) from Öösh (Övörkhangai, Mongolia). Journal of Vertebrate Paleontology, 35, e980891.

― and NORELL, M. A. 2006. High-resolution X-ray computed tomography of an Early Cretaceous
gekkonomorph (Squamata) from Öösh (Övörkhangai; Mongolia). Historical Biology, 18, 405–431.

― RIEPPEL, O. and GRANDE, L. 2008. Re-assessment of varanid evolution based on new data
from Saniwa ensidens Leidy, 1870 (Squamata, Reptilia). American Museum Novitates, 3630, 1–15.

CSIKI-SAVA, Z., VREMIR, M., VASILE, Ş., BRUSATTE, S. L., DYKE, G., NAISH, D.,
NORELL, M. A. and TOTOIANU, R. 2016. The East Side Story – The Transylvanian latest
Cretaceous continental vertebrate record and its implications for understanding Cretaceouse–
Paleogene boundary events. Cretaceous Research, 57, 662–698.
CUVIER, G. 1817. 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. Déterville libraire,
Imprimerie de A. Belin, Paris, 350 pp.

DAVIS, W. K. 1974. The Mediterranean gecko, Hemidactylus turcicus in Texas. Journal of
Herpetology, 8, 77–80.

DAZA, J. D., ABDALA, V., THOMAS, R. and BAUER, A. M. 2008. Skull anatomy of the
miniaturized gecko Sphaerodactylus roosevelti (Squamata: Gekkota). Journal of Morphology, 269,
1340–1364.

― ALIFANOV, V. R. and BAUER, A. M. 2012. A redescription and phylogenetic reinterpretation
of the fossil lizard Hoburogekko suchanovi Alifanov, 1989 (Squamata, Gekkota), from the Early
Cretaceous of Mongolia. Journal of Vertebrate Paleontology, 32, 1303–1312.

― and BAUER, A. M. 2012. A new amber-embedded sphaerodactyl gecko from Hispaniola, with
comments on the morphological synapomorphies of the Sphaerodactylidae. Breviora, 529, 1–28.

DAZA, J. D., BAUER, A. M., SAND, C., LILLEY, I., WAKE, T. A. and VALENTIN, F. 2015.
Reptile remains from Tiga (Tokanod), Loyalty Islands, New Caledonia. Pacific Science, 69, 531–
557

― ― and SNIVELY, E. D. 2013. Gobekko cretacicus (Reptilia: Squamata) and its bearing on the
interpretation of gekkotan affinities. Zoological Journal of the Linnean Society, 167, 430–448.

― ― ― 2014. On the fossil record of the Gekkota. The Anatomical Record, 97, 433–462.

― HERRERA, A., THOMAS, R. and CLAUDIO, H. J. 2009. Are you what you eat? A geometric
morphometric analysis of gekkotan skull shape. Biological Journal of the Linnean Society, 97, 677–
707.

― STANLEY, E. L., WAGNER, P., BAUER, A. M. and GRIMALDI, D. A. 2016. Mid-Cretaceous
amber fossils illuminate the past diversity of tropical lizards. Science Advances, 2, e1501080.

DE QUEIROZ, K., CANTINO, P. D. and GAUTHIER, J. A. 2020. Phylonyms. A companion to the
PhyloCode. CRC Press, Boca Raton, 1352 pp.

DELAUGERRE, M., OUNI, R. and NOUIRA, S. 2011. Is the European Leaf-toed gecko Euleptes
europaea also an African? Its occurrence on the Western Mediterranean landbrige islets and its
extinction rate. Herpetology Notes, 4, 127–137.
ESTES, R. 1970. Origin of the recent North American lower vertebrate fauna: an inquiry into the
fossil record. Forma et Functio, 3, 139–163.

― 1983. Handbuch der Paläoherpetologie 10A. Sauria terrestria, Amphisbaenia. Friedrich Pfeil,
Munich, 249 pp.

EVANS, S. E. 2008. The skull of lizards and Tuatara. 1–347. In GANS, C., GAUNT, A. S. and
ADLER, K. (eds). Biology of the Reptilia. Volume 20. Morphology H. The Skull of Lepidosauria.
Society for the Study of Amphibians and Reptiles, Ithaca, New York, 758 pp.

FITZINGER, L. 1843. Systema Reptilium (Amblyglossae). Braumüller et Seidel, Vienna, 106 pp.

FONTANARROSA, G., DAZA, J. D. and ABDLA, V. (2018). Cretaceous fossil gecko hand
reveals a strikingly modern scansorial morphology: qualitative and biometric analysis of an amber-
preserved lizard hand. Cretaceous Research, 84, 120–133.

FRANZEN, J. L. and HAUBOLD, H. 1986. The middle Eocene of European mammalian
stratigraphy. Definition of the Geiseltalian. Modern Geology, 10, 159–170.

― ― 1987. The biostratigraphic and palaeoecologic significance of the Middle Eocene locality
Geiseltal near Halle (German Democratic Republic). Münchner Geowissenschaftliche
Abhandlungen A, 10, 93–100.

GAMBLE, T., BAUER, A. M., GREENBAUM, E. and JACKMAN, T. R. 2008. Evidence for
Gondwanan vicariance in an ancient clade of gecko lizards. Journal of Biogeography, 35, 88–104.

― GREENBAUM, E., JACKMAN, T. R. and BAUER, A. M. 2015. Into the light: diurnality has
evolved multiple times in geckos. Biological Journal of the Linnean Society, 115, 896–910.

― ― ― RUSSELL, A. P. and BAUER, A. M. 2012. Repeated origin and loss of adhesive toepads
in geckos. PLoS ONE, 7, e39429.

GAO, K. and NORELL, M. A. 2000. Taxonomic composition and systematics of Late Cretaceous
lizard assemblages from Ukhaa Tolgod and adjacent localities, Mongolian Gobi Desert. Bulletin of
the American Museum of Natural History, 249, 1–118.

GAUTHIER, J., KEARNEY, M., MAISANO, J. A., RIEPPEL, O. and BEHLKE, A. 2012.
Assembling the squamate tree of life: Perspectives from the phenotype and the fossil record.
Bulletin of the Peabody Museum of Natural History, 53, 3–308.
GEORGALIS, G. L., ČERŇANSKÝ, A. and KLEMBARA, J. 2021a. Osteological atlas of new
lizards from the Phosphorites du Quercy (France), based on historical, forgotten, fossil material.
Geodiversitas, 43, 219–293.

― RABI, M. and SMITH, K. T. 2021b. Taxonomic revision of the snakes of the genera
Palaeopython and Paleryx (Serpentes, Constrictores) from the Paleogene of Europe. Swiss Journal
of Palaeontology, 140, 18.

GOLOBOFF, P. A., FARRIS, J. S. and NIXON, K. C. 2008. TNT, a free program for phylogenetic
analysis. Cladistics, 24, 774–786

GOLZ, D. J. and LILLEGRAVEN, J. 1977. Summary of the know occurences of terrestrial
vertebrates from Eocene strata of southern California. Contributions to Geology, 15, 43–65.

GRIMALDI, D. 1996. Amber: Window to the past. Harry N. Abrams, Publishers and American
Museum of Natural History, New York, 216 pp.

― NGUYEN, T. and KETCHAM, R. 2000. Ultra-High-Resolution XRay Computed Tomography
(UHR CT) and the study of fossils in amber. 77–91. In GRIMALDI, D. (ed). Studies on fossils in
amber, with particular reference to the Cretaceous of New Jersey. Backhuys Publishers, Leiden,
498 pp.

HAUBOLD, H. 1987. Geiseltalium: ein neues Landsäugetier-Zeitalter im Paläogen. Hallesches
Jahrbuch für Geowissenschaften, 12, 120–121.

― 1989. Die Referenzfauna des Geiseltalium, MP levels 11 bis 13 (Mitteleozän, Lutetium).
Palaeovertebrata, 19, 81–93.

― THOMAE, M. 1990. Stratigraphische Revision der Wirbeltierfundstellen des Geiseltaleozäns.
Hallesches Jahrbuch für Geowissenschaften, 15, 3–20.

HOFFSTETTER, R. 1946. Sur les Gekkonidae fossiles. Bulletin du Muséum National d'Histoire
Naturelle, 18, 195–203.

HOLDAWAY, R. N. and WORTHY, T. H. 1997. A reappraisal of the late Quaternary fossil
vertebrates of Pyramid Valley Swamp, North Canterbury, New Zealand. New Zealand Journal of
Zoology, 24, 69–121.

HOWARD, J. G. and PARMERLEE, J. S. Jr. 2001. Natural history of the edificarian geckos
Hemidactylus mabouia, Thecadactylus rapicauda, and Sphaerodactylus sputator on Anguilla.
Caribbean Journal of Science, 37, 285–288.
HUTCHINSON, M. N. 1997. The first fossil pygopod (Squamata, Gekkota), and a review of
mandibular variation in living species. Memoirs of the Queensland Museum, 41, 355–366.

― and MACKNESS, B. S. 2002. Fossil lizards from the Pliocene Chinchilla Local Fauna,
Queensland, with a description of a new species. Records of the South Australian Museum, 35, 169–
184.

KLEMBARA, J., HAIN, M. and ČERŇANSKÝ, A. 2019. The first record of anguine lizards
(Anguimorpha, Anguidae) from the early Miocene locality Ulm – Westtangente in Germany.
Historical Biology, 31, 1016–1027.

KLUGE, A. G. 1995. Cladistic relationships of sphaerodactyl lizards. American Museum Novitates,
3139, 1–23.

KRUMBIEGEL, G., RÜFFLE, L. and HAUBOLD, H. 1983. Das eozäne Geiseltal. A. Ziemsen
Verlag, Wittenberg Lutherstadt, 227 pp.

KRUTZSCH, W. 1966. Die sporenstratigraphische Gliederung des Tertiärim nördlichen
Mitteleuropa (Paläozän-Mitteloligozän), methodische Grundlagen und gegenwärtiger Stand der
Untersuchungen. Abhandlungen des Zentralen Geologischen Instituts, 8, 112–149.

― 1970. Die stratigraphisch verwertbaren Sporen-und Pollenformen des mitteleuropäischen
Alttertiärs. Jahrbuch für Geologie, 3, 309–379.

― 1976. Die Mikroflora der Braunkohle des Geiseltales, IV. Die stratigraphische Stellung des
Geiseltalprofils im Eozän und die sporenstratigraphische Untergliederung des mittleren Eozäns.
Abhandlungen des Zentralen Geologischen Instituts, 26, 47–92.

― BLUMENSTENGEL, H., KIESEL, Y. and RÜFFLE, L. 1992. Paläobotanische Klimagliederung
des Alttertiärs (Mitteleozän bis Oberoligozän) in Mitteldeutschland und das Problem der
Verknüpfung mariner und kontinentaler Gliederungen (klassische Biostratigraphien –
paläobotanisch-ökologische Klimastratigraphie – Evolutionsstratigraphie der Vertebraten). Neues
Jahrbuch für Geologie und Paläontologie, Abhandlungen, 186, 137–253.

LEE, M. S. Y., HUTCHINSON, M. N., WORTHY, T. H., ARCHER, M., TENNYSON, A. J. D.,
WORTHY, J. P. and SCOFIELD, R. P. 2009. Miocene skinks and geckos reveal long-term
conservatism of New Zealand’s lizard fauna. Biology Letters, 5, 833–837.
LENZ, O. K., WILDE, V., MERTZ, D. F. and RIEGEL, W. 2015. New palynology-based
astronomical and revised 40Ar/39Ar ages for the Eocene maar lake of Messel (Germany).
International Journal of Earth Sciences, 104, 873–889.

MADDISON, W. P. and MADDISON, D. R. 2017. Mesquite: a modular system for evolutionary
analysis. Available at http://mesquiteproject.org

MEAD, J. I., HOLLENSHEAD, M., SWIFT, S. L., BELL, C. J. and BAYNES, A. 2008. Pygopus
(Squamata: Pygopodidae) from mid-Holocene cave deposits, Western and South Australia. Records
of the Western Australian Museum, 25, 87–93.

MÜLLER, J. 2001. A new fossil species of Euleptes from the early Miocene of Montaigu, France
(Reptilia, Gekkonidae). Amphibia-Reptilia, 22, 341–348.

― MÖDDEN, C. 2001. A fossil leaf-toed gecko from the Oppenheim-Nierstein Quarry (Lower
Miocene, Germany). Journal of Herpetology, 35, 529–532.

― ROBERTS, E., NAYLOR, E. and STEVENS, N. 2018. A fossil gekkotan (Squamata) from the
late Oligocene Nsungwe Formation, Rukwa Rift Basin, Tanzania. Journal of Herpetology, 52, 223–
227.

NEWBERY, B. and JONES, D. N. 2007. Presence of Asian House Gecko Hemidactylus frenatus
across an urban gradient in Brisbane: influence of habitat and potential for impact on native gecko
species. 59–65. In LUNNEY, D., EBY, P., HUTCHINGS, P. and BURGIN, S. (eds). Pest or guest:
the zoology of overabundance. Royal Zoological Society of New South Wales, Mosman, 270 pp.

OPPEL, M. 1811. Die Ordnungen, Familien und Gattungen der Reptilien, als Prodrom einer
Naturgeschichte derselben. Joseph Lindauer, Munich, 86 pp.

PICKFORD, M. and ANDREWS, P. 1981. The Tinderet Miocene sequence in Kenya. Journal of
Human Evolution, 10, 11–33.

PYRON, R. A. 2017. Novel approaches for phylogenetic inference from morphological data and
total-evidence dating in squamate reptiles (lizards, snakes, and amphisbaenians). Systematic
Biology, 66, 38–56.

RAGE, J.-C. 2008. Squamate reptiles from the lower Miocene of the Sperrgebiet, Namibia. Memoir
of the Geological Survey of Namibia, 20, 93–103.

― 2012. Amphibians and squamates in the Eocene of Europe: what do they tell us?
Palaeobiodiversity and Palaeoenvironments, 92, 445–457.
― 2013. Mesozoic and Cenozoic squamates of Europe. Palaeobiodiversity and
Palaeoenvironments, 93, 517–534.

RAGE, J.-C., ADACI, M., BENSALAH, M., MAHBOUBI, M., MARIVAUX, L., MEBROUK, F.
and TABUCE, R. 2021. Latest Early-early Middle Eocene deposits of Algeria (Glib Zegdou,
HGL50), yield the richest and most diverse fauna of amphibians and squamate reptiles from the
Palaeogene of Africa. Palaeovertebrata, 44, 1-32.

RING, S. J., BOCHERENS, H., WINGS, O. and RABI, M. 2020. Divergent mammalian body size
in a stable Eocene greenhouse climate. Scientific Reports, 10, 3987.

RUSSELL, A. P. 2002. Integrative functional morphology of the gekkotan adhesive system
(Reptilia: Gekkota). Integrative and Comparative Biology, 42, 1154–1163.

SCANFERLA, A. and SMITH, K. T. 2020. Exquisitely preserved fossil snakes of Messel: insight
into the evolution, biogeography, habitat preferences and sensory ecology of early boas. Diversity,
12, 100.

SCHATZINGER, R. 1975. Later Eocene (Uintan) lizards from the greater San Diego area,
California. Unpublished MSc thesis, San Diego State University, San Diego, 212 pp.

SCHLEICH, H. H. 1987. Neue Reptilienfunde aus dem Tertiär Deutschlands. 7. Erstnachweis von
Geckos aus dem Mittelmiozän Süddeutschlands: Palaeogekko risgoviensis nov. gen., nov. spec.
(Reptilia, Sauria, Gekkonidae). Mitteilungen der Bayerischen Staatssammlung für Palaeontologie
und Historische Geologie, 27, 67–93.

SCHMIDT-KITTLER, N., BRUNET, M., GODINOT, M., FRANZEN, J. L., HOOKER, J. J., and
LEGENDRE, S. 1987. European reference levels and correlation tables. Münchner
Geowissenschaftliche Abhandlungen A, 10, 13–31.

SKUTSCHAS, P. P. 2006. Mesozoic amphibians from Siberia, Russia. 123–126. In BARRETT, P.
M. and EVANS, S. E. (eds). Ninth International Symposium on Mesozoic Terrestrial Ecosystems
and Biota, Abstracts and Proceedings. The Natural History Museum, London, 188 pp.

SIMÕES, T. R., CALDWELL, M. W., TAŁANDA, M., BERNARDI, M., PALCI, A.,
VERNYGORA, O., BERNARDINI, F., MANCINI, L and NYDAM, R.L. 2018. The origin of
squamates revealed by a Middle Triassic lizard from the Italian Alps. Nature 557, 707–709.

SIMÕES, T. R. and PYRON, R. A. 2021. The squamate tree of life. Bulletin of the Museum of
Comparative Zoology, 163, 47–95.
SIMÕES, T. R., VERNYGORA, O. V., CALDWELL, M. W. and Pierce, S. E. 2020.
Megaevolutionary dynamics and the timing of evolutionary innovation in reptiles. Nature
Communications, 11, 3322.

SMITH, K. T. 2009. Eocene Lizards of the clade Geiseltaliellus from Messel and Geiseltal,
Germany, and the early radiation of Iguanidae (Reptilia: Squamata). Bulletin of the Peabody
Museum of Natural History, 50, 219–306.

― 2017. First crocodile-tailed lizard (Squamata: Pan-Shinisaurus) from the Paleogene of Europe.
Journal of Vertebrate Paleontology, 37, e1313743.

― ČERŇANSKÝ, A., SCANFERLA, A. and SCHAAL, S. F. K. 2018. Lizards and snakes –
Warmth-loving sunbathers. 123–147. In SMITH, K. T., SCHAAL, S. F. K. and HABERSETZER,
J. (eds). Messel. An ancient greenhouse ecosystem. Senckenberg Gesellschaft für Naturforschung,
Frankfurt am Main, 355 pp.

― GAUTHIER, J. A. 2013. Early Eocene lizards of the Wasatch Formation near Bitter Creek,
Wyoming: diversity and paleoenvironment during an interval of global warming. Bulletin of the
Peabody Museum of Natural History, 54, 135–230.

― SCANFERLA, A. 2021. A nearly complete skeleton of the oldest definitive erycine boid
(Messel, Germany). Geodiversitas, 43, 1–24.

SPEYBROECK, J., BEUKEMA, W., BOK, B. and VAN DER VOORT, J. 2016. Field guide to the
amphibians and reptiles of Britain and Europe. Bloomsbury Publishing, London, 432 pp.

― BEUKEMA, W., DUFRESNES, C., FRITZ, U., JABLONSKI, D., LYMBERAKIS, P.,
MARTÍNEZ-SOLANO, I., RAZZETTI, E., VAMBERGER, M., VENCES, M., VÖRÖS, J. and
CROCHET P.-A. 2020. Species list of the European herpetofauna – 2020 update by the Taxonomic
Committee of the Societas Europaea Herpetologica. Amphibia-Reptilia, 41, 139–189.

STEINHEIMER, F. D. and HASTINGS, A. K. 2019. HALLE: The Geiseltal Collection of Martin
Luther University, Halle-Wittenberg. 271–280. In BECK, L. A. and JOGER, U. (eds.).
Paleontological Collections of Germany, Austria and Switzerland. Springer Nature, Cham, 573 pp.

SULLIVAN, R.M. 2019. The taxonomy, chronostratigraphy and paleobiogeography of
glyptosaurine lizards (Glyptosaurinae, Anguidae). Comptes Rendus Palevol, 18, 747–763.

SUMIDA, S. S. and MURPHY, R. W. 1987. Form and function of the tooth crown structure in
gekkonid lizards (Reptilia, Squamata, Gekkonidae). Canadian Journal of Zoology, 65, 2886–2892.
TAŁANDA, M. 2018. An exceptionally preserved Jurassic skink suggests lizard diversification
preceded fragmentation of Pangaea. Palaeontology, 61, 659–677.

UETZ, P., FREED, P., AGUILAR, R. and HOŠEK, J. (eds.), The Reptile Database,
http://www.reptile-database.org, accessed November 2021.

UNDERWOOD, G. 1954. On the classification and evolution of geckos. Proceedings of the
Zoological Society of London, 124, 469–492.

VASILYAN, D., ZAZHIGIN, V. S. and BÖHME, M. 2017. Neogene amphibians and reptiles
(Caudata, Anura, Gekkota, Lacertilia, and Testudines) from the south of Western Siberia, Russia,
and Northeastern Kazakhstan. PeerJ, 5, e3025.

VILLA, A., DAZA, J. D., BAUER, A. M. and DELFINO, M. 2018. Comparative cranial osteology
of European gekkotans (Reptilia, Squamata). Zoological Journal of the Linnean Society, 184, 857–
895.

― and DELFINO, M. 2019a. Fossil lizards and worm lizards (Reptilia, Squamata) from the
Neogene and Quaternary of Europe: an overview. Swiss Journal of Palaeontology, 138, 177–211.

― ― 2019b. A comparative atlas of the skull osteology of European lizards (Reptilia: Squamata).
Zoological Journal of the Linnean Society, 187, 829–928.

VITT, L. J. and CALDWELL, J. P. 2009. Herpetology. An introductory biology of amphibians and
reptiles - third edition. Academic Press, Burlington, 697 pp.

VOIGT, E. 1988. Preservation of soft tissues in the Eocene lignite of the Geiseltal near Halle
(Saale). Courier Forschungsinstitut Senckenberg, 107, 325–343.

WEIGELT, J. 1934. Die Geiseltalgrabungen des Jahres 1933 und die Biostratonomie der
Fundschichten. Nova Acta Leopoldina NF, 1, 552–660.

WORTHY, T. H. and ANDERSON, A. 2009. Results of palaeofaunal research. 41–62. In CLARK,
G. and ANDERSON, A. (eds). The early prehistory of Fiji. Terra Australis, 31, 437 pp.

― and CLARK, G. 2009. Bird, mammal and reptile remains. 231–258. In CLARK, G. and
ANDERSON, A. (eds). The early prehistory of Fiji. Terra Australis, 31, 437 pp.

― and HOLDAWAY, R. N. 1996. Quaternary fossil faunas, overlapping taphonomies, and
palaeofaunal reconstruction in North Canterbury, South Island, New Zealand. Journal of the Royal
Society of New Zealand, 26, 275–361.
FIGURE CAPTIONS




FIG. 1. Map of the Geiseltal fossillagerstätte, with its location in Germany and the position of Site
Cecilie IV.




FIG. 2. Partially articulated skull of Geiseleptes delfinoi Villa et al. gen. et sp. nov. from the early
late to middle Eocene of Geiseltal, Germany (GMH Ce IV-4057-1933) The skull was apparently
split during collection or preparation and both parts were subsequently mounted on the same
lacquer film. Scale bar = 5 mm.




FIG. 3. Geiseleptes delfinoi Villa et al. gen. et sp. nov. from the early late to middle Eocene of
Geiseltal, Germany (GMH Ce IV-4057-1933). Photo (A) and CT-scan images (B, C). C shows the
unexposed surface. Abbreviations: c, coronoid; cb, compound bone; d, dentary; e, epipterygoid; ec,
ectopterygoid; f, frontal; m, maxilla; n, nasal; pa, parietal; pf, prefrontal; po, postorbitofrontal; pt,
pterygoid; p?, paroccipital process?; q, quadrate. Scale bar = 2 mm.
FIG. 4. Geiseleptes delfinoi Villa et al. gen. et sp. nov. from the early late to middle Eocene of
Geiseltal, Germany (GMH Ce IV-4057-1933). Photo (A) and CT-scan images (B, C). C shows
unexposed surface. Abbreviations: c, coronoid; cb, compound bone; d, dentary; m, maxilla. Scale
bar = 2 mm.
FIG. 5. Taxonomically-significant grooves on the skull of Geiseleptes delfinoi Villa et al. gen. et
sp. nov. A, detail of GMH Ce IV-4057-1933. B, CT-scan image of the same area. C, posterior end
of a frontal of Euleptes europaea (MDHC 389), in dorsal view for comparison. D, right maxilla of
Euleptes europaea (MDHC 389), in lateral view for comparison. Arrows mark grooves on the
posterolateral process of the frontal and on the posterior process of the maxilla. Abbreviations: e,
epipterygoid; f, frontal; m, maxilla; pa, parietal; pf, prefrontal; po, postorbitofrontal; pt, pterygoid;
q, quadrate. Scale bars = 1 mm.
FIG. 6. Strict consensus trees resulting from the parsimony analysis using a molecular backbone
constraint. A, Strict consensus tree of the analysis with the complete matrix. B, strict consensus tree
of the analysis without Gerandogekko arambourgi, a poorly known taxon that is operationally
identical to G. gaillardi and may even represent the same species. C, pruned strict consensus tree
after excluding Euleptes gallica, Laonogekko lefevrei, and Palaeogekko risgoviensis. Arrows mark
the possible positions of the pruned taxa in the pruned tree. White circles mark the constraints
implemented in the analysis (not all constraints are mapped in A and B, because some clades are
collapsed in a polytomy due to unstable wild-card taxa). Abbreviations: a, L. lefevrei; b, P.
risgoviensis; c, E. gallica.
FIG. 7. Stratigraphic distribution of fossil Sphaerodactylidae. Age ranges in white represent
unstable taxa that are recovered within sphaerodactylids in some of our trees, whereas age ranges in
grey represent uncertain dates. Data were taken from: Schleich 1987; Müller & Mödden 2001;
Augé 2003, 2005; Čerňanský & Bauer 2010; Daza & Bauer 2012; Daza et al. 2014; Bolet et al.
2015; Colombero et al. 2017; Čerňanský et al. 2018. Abbreviations: Holo., Holocene.
FIG. 8. Geographic distribution of fossil Sphaerodactylidae, in Europe (A) and the island of
Hispaniola in the Caribbean (B). The map includes potential sphaerodactylids according to our
phylogenetic analysis. Symbols represent different genera: Cadurcogekko (star); Euleptes (circle);
Geiseleptes (hexagon); Gerandogekko (rhombus); Laonogekko (inverted triangle); Palaeogekko
(triangle); Sphaerodactylus (quadrate). Grey symbols stand for Paleogene (Eocene) occurrences,
whereas black symbols for Neogene (Miocene) ones. Larger symbols indicate multiple localities.
Data sources are the same as in Fig. 7. Occurrences: 1. Cadurcogekko piveteaui, Le Bretou (MP 16,
France); 2. Cadurcogekko piveteaui, Lavergne (MP 16, France); 3. Cadurcogekko piveteaui, Les
Pradigues (MP 17, France); 4. Cadurcogekko piveteaui, La Bouffie (MP 17, France); 5.
Cadurcogekko piveteaui, Malpérié (MP 17, France); 6. Cadurcogekko piveteaui, Perrière (MP 17,
France); 7. Cadurcogekko piveteaui, Aubrelong 2 (MP 17, France); 8. Cadurcogekko piveteaui,
Gousnat (MP 18, France); 9. Cadurcogekko piveteaui, St. Néboule (MP 18, France); 10.
Cadurcogekko piveteaui, Sindou D (MP 18-19, France); 11. Cadurcogekko piveteaui, Coânac (MP
19, France); 12. Cadurcogekko piveteaui, Rosière 2 (MP 16, France); 13. Cadurcogekko piveteaui,
Escamps (MP 19, France); 14. Cadurcogekko verus, Les Pradigues (MP 17, France); 15.
Cadurcogekko verus, Sindou D (MP 18-19, France); 16. Euleptes gallica, Montaigu-le-Blin (MN 2,
France); 17. Euleptes gallica, Merkur North (MN 3, Czech Republic); 18. Euleptes klembarai,
Zapfe‘s fissure fillings, Devínska Nová Ves (MN 6, Slovakia); 19. Euleptes sp.,
Oppenheim/Nierstein quarry (MN 2, Germany); 20. cf. Euleptes sp., Moncucco Torinese (MN 13,
Italy); 21. Geiseleptes delfinoi, Geiseltal (middle Eocene, Germany); 22. Gerandogekko
arambourgi, Saint-Gérand-le-Puy (MN 2, France); 23. Gerandogekko gaillardi, La Grive-Saint-
Alban (MN 7-8, France); 24. Laonogekko lefevrei, Prémontré (MP 10, France); 25. Palaeogekko
risgoviensis, Steinberg (MN 6, Germany); 26. Sphaerodactylus ciguapa, La Toca mine, Cordillera
Septentrional (late early to early middle Miocene, Dominican Republic); 27. Sphaerodactylus
dommeli, La Toca mine, Cordillera Septentrional (early to middle Miocene, Dominican Republic);
28. Sphaerodactylus spp., Cordillera Septentrional (early to middle Miocene, Dominican Republic).