The oldest known snakes from the Middle Jurassic ...

49 downloads 0 Views 4MB Size Report
snakes (for example, recurved teeth with labial and lingual carinae, long toothed suborbital ramus of .... in recognition of type locality near Devil's Canyon,.
ARTICLE Received 1 Oct 2014 | Accepted 1 Dec 2014 | Published 27 Jan 2015

DOI: 10.1038/ncomms6996

The oldest known snakes from the Middle Jurassic-Lower Cretaceous provide insights on snake evolution Michael W. Caldwell1, Randall L. Nydam2, Alessandro Palci3 & Sebastia´n Apesteguı´a4

The previous oldest known fossil snakes date from B100 million year old sediments (Upper Cretaceous) and are both morphologically and phylogenetically diverse, indicating that snakes underwent a much earlier origin and adaptive radiation. We report here on snake fossils that extend the record backwards in time by an additional B70 million years (Middle Jurassic-Lower Cretaceous). These ancient snakes share features with fossil and modern snakes (for example, recurved teeth with labial and lingual carinae, long toothed suborbital ramus of maxillae) and with lizards (for example, pronounced subdental shelf/gutter). The paleobiogeography of these early snakes is diverse and complex, suggesting that snakes had undergone habitat differentiation and geographic radiation by the mid-Jurassic. Phylogenetic analysis of squamates recovers these early snakes in a basal polytomy with other fossil and modern snakes, where Najash rionegrina is sister to this clade. Ingroup analysis finds them in a basal position to all other snakes including Najash.

1 Department

of Biological Sciences, & Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Alberta T6G2E9, Canada. of Anatomy, Midwestern University, Glendale, Arizona 85308, USA. 3 Earth Sciences Section, South Australian Museum, North Terrace, Adelaide, South Australia 5000, Australia. 4 Fundacio´n Fe´lix de Azara, CEBBAD (CONICET), Universidad Maimo´nides, Buenos Aires 1405, Argentina. Correspondence and requests for materials should be addressed to M.W.C. (email: [email protected]). 2 Department

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

1

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

T

he previous understanding of the fossil record of early snake evolution (Late Mesozoic) relies on isolated vertebrae from Africa1 (100 Myr ago), isolated jaws and vertebrae from North America2,3 (98–65 Myr ago), a number of nearly complete snakes, some with rear limbs4–7, from the circum-Mediterranean region (98–95 Myr ago), and two taxa of relatively complete snakes, one with rear limbs, from Argentina (94–92 Myr ago8,9 and 86–80 Myr ago10,11). This morphologically, ecologically and phylogenetically diverse assemblage of snakes appears in the fossil record around the world almost simultaneously (B100–94 Myr ago). The long standing questions in snake palaeontology have centred on when, where and how snakes evolved and radiated from within Squamata before the early part of the Late Cretaceous. Here we report on four new species of significantly older fossil snakes (167–143 Myr ago) recognized from cranial and postcranial remains found in the United Kingdom, Portugal and the United States. These new data extend the known geological range of snakes by nearly 70 million years into the mid-Mesozoic, indicating that their origin was coincident with the known radiation of most other major groups of squamates in the mid-Jurassic12,13 during the final stages of the break-up of Pangaea into Laurasia and Gondwana. It is also important to note that this new record for early snakes, co-occurring with early anguimorphs such as Dorsetisaurus12,13, fills a major chronological gap predicted by molecular phylogenetics14. In stratigraphic order, the oldest snake recognized here is found in rocks dated as Bathonian (B167 Myr ago), Middle Jurassic, from Southern England15, followed by a North American record dated as Kimmeridgian (B155 Myr ago, Upper Jurassic, Colorado, USA)16, which appears to be a contemporary of another taxon from the Kimmerdigian (Upper Jurassic, Guimarota, Portugal)17. The youngest snake taxon and materials recognized here are found in rocks dated as Tithonian (B150 Myr ago; Upper Jurassic) to Berriasian (B140 Myr ago; Lower Cretaceous) outcropping near Swanage, Dorset, Southern England15. Results Systematic palaeontology. Order Squamata Opell, 1811 Suborder Serpentes Linnaeus, 1758 Parviraptor estesi gen. et sp. (Figs 1a and 2a; Supplementary Figs 1 and 2a–e). Holotype. Left maxilla on block NHMUK R48388, Natural History Museum, London. Locality horizon and age. Durlston Bay, Swanage, Dorset, England; Purbeck Limestone Formation (Upper Jurassic; Tithonian/Lower Cretaceous; Berriasian). Emended Diagnosis. Long low, ascending process of maxilla; premaxillary process turned medially; narrow prefrontal facet on ascending process of maxilla. Differs from maxilla of Coniophis precedens in lacking medial process at anterior end, in having greater degree of recurvature of teeth. Differs from Dinilysia patagonica in having gracile maxilla, relatively smaller teeth, less pronounced medial deviation at anterior tip and relatively smaller palatine process. Differs from Portugalophis lignites in having narrower premaxillary process. Differs from Diablophis gilmorei in having larger palatine process and in lacking medial curvature of anterior end of maxilla. Revised Description. Maxilla exposed in medial and dorsal views; long, 24 tooth positions preserved, bearing low ascending process 2

extending from tooth position 4 to tooth position 16/17; anterior superior alveolar foramen large, positioned at front of ascending process; prominent premaxillary process, narrow and turned medially; supradental shelf narrow and thin with prominent palatine process adjacent to posterior end of ascending process; margin of palatine process damaged; long, tooth-bearing suborbital ramus with 10 tooth positions posterior to prefrontal facet and palatine process; maxillary tooth positions defined by three-sided alveolus with no medial border; preserved teeth attached to rims of alveoli; presence of tooth in alveolus closes small posterolingual notch forming basal nutrient foramen; teeth conical, circular cross sections, recurved, with labial and lingual carinae. aff. Parviraptor estesi (Figs 1b,2s and 3a; Supplementary Figs 3a–d and 4a–i). Referred Material. Right frontal, atlas and associated precloacal vertebrae on multispecimen block NHMUK R8551 (specimen originally assigned to Parviraptor estesi15). Locality, horizon and age. Swanage, Dorset, England; Purbeck Limestone Formation (Upper Jurassic; Tithonian/ Lower Cretaceous; Berriasian). Description. Right frontal (NHMUK R8551) in lateral view with well-developed prefrontal and postfrontal facets and deep, medially curved descensus frontalis with well-developed suboptic shelves on posteroventral margin forming posterior portion of optic nerve foramen (Cranial Nerve II); closely associated vertebrae with tall neural spines, massive, vertical synapophyses with distinct, posteriorly expanded parapophyseal and diapophyseal facets; complete neural arch with short zygosphenes; posterior margin lacks incised median notch; elevated, round condyles with weakly constricted necks; centrum narrow posteriorly and wide anteriorly in ventral view; inferior margin of centrum, no development of haemal keel; possible sacral vertebra with robust transverse processes; left neural arch from an atlas with narrow, posteriorly directed neural spine; no notochordal canal. Remarks. Because of the loss of the provenance of this block with respect to the block NHMUK 48388, the lack of articulation of the elements, and the polytaxonomic composition of the Purbeck blocks, we do not feel that there is sufficient evidence to support referral of the frontal to type maxilla of Parviraptor estesi, although we do recognize that such a referral is a strong possibility. Diablophis gilmorei gen. nov., new combination (Figs 1c–f and 2b,j; Supplementary Figs 10a–k and 11a–n; Supplementary Videos 1–9). Etymology. ‘Diablo’, ‘devil’ (Spanish); ophis, snake (Greek); in recognition of type locality near Devil’s Canyon, Colorado. Holotype. LACM 4684/140572 (Los Angeles County Museum), broken right maxilla, broken right mandible, and broken axis vertebra (revised holotype). Referred Specimens. Several precloacal vertebrae and one possible sacral vertebra LACM 4684/140572; broken right dentary LACM 5572/120732 (all specimens originally referred to Parviraptor gilmorei16); LACM 4684/120472, four precloacal vertebrae, one caudal vertebra with large transverse processes. Locality, horizon and age. From Fruita locality, Morrison Formation, Mesa County, Colorado, USA (Upper Jurassic; Kimmeridgian). Diagnosis. Small-bodied snake that differs from Parviraptor estesi and Portugalophis lignites in having smaller palatine

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

b

m

Palaeocene

Laurasia

Maastrichtian

England

c

Late

Campanian

g

Santonian Coniacian

Cretaceous

Aptian

Eur

NA

Diablophis

Barremian

Eophis Portugalophis Parviraptor aff. Parviraptor

i

Albian

Early

e

Portugal

h Colorado

d

Pouitella Pachyophis Mesophis Simoliophis ‘Coniophis’

Turonian Cenomanian

Gondwana Cerberophis Unnamed ‘alethnophidian’ Najash Australophis Dinylisia Lapparentophis Simoliophis Haasiophis Pachyrachis Eupodophis Menarana Kelyophis Madtsoia Indophis Coniophis ?Scolecophidia Sanajeh

a

SA

Afr

ME

Mad

India

Hauterivian

f

j

Valanginian Berriasian

l England

Oxfordian

Middle

k

Jurassic

Late

Tithonian Kimmeridgian

Callovian Bathonian Bajocian Aalenian Toarcian

n

o

~167mya

Pangaea

p

~155mya

~145mya

Figure 1 | Skull elements assigned to new genera and species of Middle Jurassic to Lower Cretaceous snakes. (a) Parviraptor estesi (NHMUK R48388, generic type), medial view left maxilla. (b) aff. Parviraptor estesi (NHMUK R8551) lateral view right frontal. (c–f) Diablophis gilmorei (c,d) medial and lateral views right maxilla (LACM 140572). (e,f) Lateral and medial views of right dentary (LACM 140572). (g–j) Portugalophis lignites (g,h) medial and lateral view left maxilla (MG-LNG 28091). (i,j) medial and lateral view left dentary (MG-LNG 28094). (k,l) Eophis underwoodi, medial and lateral views holotype (NHMUK R12355) anterior symphyseal section; paratype (NHMUK R12354) mid-dentary portion. (m) Stratigraphic and palaeobiogeographic distributions. (n) Plate tectonic reconstruction for Bathonian (B167 Myr ago), yellow dot indicates locality of Eophis underwoodi. (o) Plate tectonic reconstruction for Kimmerdigian (B155Myr ago), blue dot indicates locality of Portugalophis lignites, green dot indicates locality of Diablophis gilmorei. (p) Plate tectonic reconstruction for Tithonian-Berriasian (B145 Myr ago), red dot indicates locality of Parviraptor estesi and aff. Parviraptor estesi. (note: color of dots on map correspond with bracket color for specimens (a–l). Afr, Africa; Eur, Europe; Mad, Madagascar; ME, Middle East; NA, North America; SA, South America (all scale bars, a–l, 1 mm).

process and having strong medial deviation of anterior end of maxilla. Differs from P. lignites and from Eophis underwoodi in having more fully developed subdental lamina relative to dentary size. Differs from Coniophis precedens in the absence of medial process on anterior tip of maxilla; greater degree of recurvature of teeth; alveoli primarily on lateral parapet of dentary and maxilla; presence of subdental lamina and multiple mental foramina on dentary; and in having tall neural spines and small zygosphenes and zygantra. Description. Right maxilla (LACM 4684/140572) small, preserves 10–11 tooth spaces with broken crowns; ascending process low and long with small notch in posterior margin of apex; supradental shelf broken, thin anteriorly and thickened posteriorly; lateral surface smooth with four nutrient foramina; alveoli/interdental ridges with incised medial walls forming nutrient notch; tooth crowns conical,

recurved, sharp; holotype right dentary missing symphysis and postdentary articulation region; straight in dorsal view with medial turn anteriorly and 11 tooth spaces; well-developed subdental lamina forming medial border of subdental gutter; lateral wall with seven mental foramina; dentary teeth attached to three-sided alveoli; teeth circular in cross-section, conical and strongly recurved; neural spine tall, synapophyses massive and vertical, and condyle elevated above bottom of centrum; neural canals show ‘trefoil’ organization often present in snakes3,8; condyle round, with weakly constricted neck and synapophysis with posteriorly expanded parapophyseal and diapophyseal facets; centrum narrow posteriorly, wide anteriorly; inferior margin of centrum without haemal keel; paralymphatic fossae present; no notochordal canal; small zygosphene-zygantrum articulations. Eophis underwoodi gen. et sp. nov. (Figs 1k,l and 2g–i; Supplementary Figs 5a–i and 6a–g).

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

3

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

ap

prff

asaf

idr pmp 1 mm sdl

ap

prff

sop palp

1 mm ap prff

sop

1 mm

idr

New fossil snakes

palp

1 mm sdl

idr

pmp

1 mm 1 mm

palp

sdl

prf ect

1 cm sop sop

prff

idr

pmp

palp

palp

pmp

Known fossil snakes

jug

1 cm idr

5 mm palp

asaf

5 mm

idr

Modern snake

prff sop

5 mm

5 mm

nf

lin car

0.25 m

idr

0.25 m

idr

suof

psf

prff

psf

0.5 mm

0.5 mm

1 mm

prff

df

Lsos Rsos

df

Rsos

5 mm

Figure 2 | Jaw elements of earliest snakes. (a–c,g–k) compared with Dinilysia patagonica (d,l), Yurlunggur sp. (e,m), and modern Python sp. (f,n). (a) Left maxilla, Parviraptor estesi (NHMUK 48388, part). (b) Right maxilla (image reversed), Diablophis gilmorei (LACM 4684/140572, part). (c) Left maxilla, Portugalophis lignites (MG-LNEG 28091). (d) Left maxilla, prefrontal, jugal, and ectopterygoid, D. patagonica (MACN-RN 1013, part). (e) Right maxilla (image reversed), Yurlunggur sp. (QMF45391, part). (f) Left maxilla, Python sp. (UALVP unnumbered specimen). (g) Eophis underwoodi (Holotype, R12355), anterior symphyseal section. (h) E. underwoodi (Paratype NHMUK R12354: mid-dentary portion). (i) E. underwoodi (referred specimen, NHMUK R12354: posterior dentary portion). (j) Right dentary (image reversed), D. gilmorei (LACM 4684/140572, part), new genus. (k) Left dentary (image reversed), Portugalophis lignites (MG-LNEG 28091), new gen. and sp. (l) Right dentary, splenial, angular, compound, and coronoid, D. patagonica (MACN-RN 1013, part). (m) Right dentary of Yurlunggur sp. (QMF45391, part). (n) Right dentary, Python sp. (UALVP unnumbered specimen). (o) SEM image, posterior tooth sockets, right dentary, E. underwoodi (NHMUK R12370). (p) SEM image, posterior tooth sockets, left maxilla, Xenopeltis unicolor (FMNH 287277). (q) Light photograph, isolated left maxillary tooth with lingual apical carina, Parviraptor estesi (NHMUK 48388, part). (r) SEM image, isolated left maxillary tooth with lingual apical carina, Parviraptor estesi (NHMUK 48388, part). (s) right frontal, lateral view, aff. Parviraptor estesi (NHMUK R8551). (t) Right frontal, Python sp. (UALVP unnumbered specimen). ap, ascending process; asaf, anterior superior alveolar foramen; ect, ectopterygoid; idr, interdental ridges; jug, jugal; lin car, lingual carina; Lsos, left suboptic shelf; nf, nutrient foramina; palp, palatine process; pmp, premaxillary process; prf, prefrontal; prff, prefrontal facet; psf, postfrontal facet; Rsos, right suboptic shelf of frontal; sdl, subdental lamina; sop, suborbital process of maxilla; suof, supraorbital facet.

Etymology. Eos, dawn (Greek); ophis, snake (Greek); underwoodi (Surname); in recognition of being oldest known snake material, and lifelong impact/contributions of Garth Underwood to study of snakes. 4

Syntypes. NHMUK R12355, symphyseal portion of right dentary; NHMUK R12354, midsection right dentary; NHMUK R12370, posterior portion right dentary (all three specimens previously referred to Parviraptor cf. P. estesi15).

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

ns

a

pz

c

ptz

ptz

pz

ns

b

d

cn sy

pz

cn ct

ct

cbr

cbr

sy

ct

sy

cn

sy

pz

ct

ct

cn

cn

zs

pz

cn

zs pz tp

ns ptz

ns

?sr

pz

cbr

pz

?sr

ptz

sy cbr

cn

ptz

cn

sy

ptz

ns

ns

ptz cn

ct

sy

cbr zs

ns pz

cn

zg

ptz

cn

zs

ptz

cbr pz

cbr pz

cn

ns pz

sy

cn

ns

e

ptz

ptz

sy

ptz

ns

zg

ptz zs

zs

pz

pz

zs

ns tp

ptz

ns

ptz

ns

ns

ptz

ptz

ns

zs

zs pz

pz

pz

ptz

ptz pz

cn

cbr 1 mm

sy sy

ct 1 mm

ct

sy

1 mm

ct

1 mm

ct

sy 1 mm

Figure 3 | Vertebral comparisons of earliest snakes (yellow background) to Najash and Coniophis (green background) vertebrae. (a) aff. Parviraptor estesi (NHMUK R8551) in posterior, right lateral, ventral and dorsal views (note: no anterior view is available for aff. Parviraptor estesi). (b–e) Images in posterior, lateral (left for Najash, right for Diablophis and Coniophis), ventral, dorsal and anterior views. (b) Diablophis gilmorei (LACM 4684/140572 [part]). (c) Diablophis gilmorei (LACM 4684/120472 [part]). (d) Najash rionegrina (MPCV 395 (ventral view), 397 (anterior, lateral [reversed for comparative purposes], dorsal views), 386 (posterior view)). (e) Coniophis precedens (UALVP unnumbered specimen). cbr, centrum bony ridge; cn, condyle; ct, cotyle; ns, neural spine; ptz, postzygapophysis; pz, prezygapophysis;?sr, possible sacral rib; sy, synapophysis; zg, zygantrum; tr, transverse process; zs, zygosphene.

Referred specimen. NHMUK R12352, fragment of left maxilla. Locality, horizon and age. Forest Marble, Kirtlington Cement Works Quarry, Oxfordshire, England (Middle Jurassic; Bathonian). Diagnosis. Small-bodied snake with low and shallow subdental lamina of dentary. Differs from both Diablophis gilmorei and Portugalophis lignites in having smaller subdental lamina relative to dentary size. Differs from Coniophis precedens in having a subdental lamina, and multiple mental foramina on dentary. Description. Dentary fragments with three-sided alveoli forming distinct interdental ridges and tall lateral parapet; medial side of alveolus with nutrient notch; short subdental lamina and well-developed Meckelian fossa, narrow anteriorly, wide posteriorly; intramandibular septum fused to floor of Meckelian fossa; mandibular symphysis smooth and broad; lateral surface of dentary convex with 4–5 mental foramina; NHMUK R12370 possesses narrow facet on medial edge of subdental lamina for articulation with anteromedial process of coronoid; ventromedial edge of dentary with narrow facet for articulation with splenial. NHMUK R12352 (Supplementary Fig. 6e–g) is a fragment of the left maxilla preserving three alveoli and an incomplete portion of the dorsal process. The most complete alveolus is similar to those of the dentary in

being three-sided with the interdental ridge canal (irc) piercing the alveolar bone between adjacent alveoli. A small tooth is lodged within the irc. Portugalophis lignites gen. et sp. nov. (Figs 1g–j and 2c,k; Supplementary Figs 7a–j,8a–c and 9a–c; Supplementary Videos 10–14). Etymology. Portugal, Portugal; ophis, snake (Greek); lignites, lignum (Latin); in recognition of Portugal, snake affinities, and coal-lithology of Guimarota Mine. Holotype. Left maxilla (holotype), MG-LNEG 28091 (Museu Geologico, Lisboa, Portugal). Paratype. MG-LNEG 28094, left dentary. Referred Specimen. MG-LNEG 28100, partial left maxilla. Locality. From coal beds, Guimarota mine, Leiria, Portugal (Upper Jurassic; Kimmeridgian). Diagnosis. Differs from Parviraptor estesi, Eophis underwoodi, and Diablophis gilmorei in larger size, dentary tooth bases extend into subdental gutter. Further differs from P. estesi in having wider premaxillary process of maxilla, more deeply concave medial surface of ascending process of maxilla. Differs from Eophis underwoodi in having taller subdental lamina relative to dentary size. Differs from Coniophis in the absence of medial process on anterior tip of maxilla; greater degree of recurvature of teeth; alveoli

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

5

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

primarily on lateral parapet of dentary and maxilla; and the presence of subdental lamina and multiple mental foramina on dentary. Description. MG-LNEG 28091, holotype left maxilla, ascending process long and low with dorsal edge turned medially; ascending process with distinct, broad facet and notch for prefrontal; maxilla with two rows of labial foramina, inferior row large, superior row small; premaxillary process short, broad, mediolaterally expanded with opening of anterior superior alveolar foramen at posterior margin; supradental shelf narrow and rounded with distinct medially expanded palatine process (between tooth positions 11 and 15); 15 preserved alveoli with 9 fragmentary teeth; alveoli with tall lateral walls, well-defined three-sided alveolar margins and interdental ridges, and low medial wall notched by nutrient foramen; paratype dentary, MGLNEG 28094, with 15 tooth spaces and 6 teeth; well-developed subdental lamina thickened posteriorly bordering large and prominent sulcus dentalis; posteriorly, subdental lamina with narrow splenial facet; Meckelian fossa tall posteriorly, narrowing to form shallow channel, open ventrally; anteriormost portion of lateral emargination preserved; 5–6 large, circular mental foramina; teeth conical, sharply pointed, strongly recurved, and bear anteromedial and posterolateral carinae; tooth bases broad, set in margins of tooth sockets/alveoli. Comparative anatomy of early snakes. These new snake taxa are based on specimens that were either previously described and named, or referred to, various species or groups of anguimorph lizards15–17. The problems associated with identifying any of the Parviraptor specimens as the remains of snakes were due to the fact that the original taxon15 was a complex chimaera (see Supplementary Figs 12–23; Supplementary Notes 1,2). The type species, P. estesi15 was assembled from isolated to arguably associated fossils preserved on two different blocks of rock, NHMUK R48388 from slightly older strata at Durlston Bay, Swanage, the United Kingdom15 (Supplementary Fig. 1) and NHMUK R8551 from younger strata at Swanage, the United Kingdom15 (Supplementary Fig. 3). The original characterization of Parviraptor cf. P. estesi15 was based on isolated specimens found more than 150 km away at Kirtlington, Oxfordshire, the United Kingdom (Fig. 1n), in rocks that are B30 million years older than the type and referred block (Fig. 1p). Reinterpretation of these specimens finds that the original type and referred specimens of Parviraptor estesi15 and Parviraptor cf. P. estesi15 include the remains of possibly three separate taxa of snakes, a large number of gekkotan as well as other indeterminate lizard

elements, and also a number of non-squamate elements and one invertebrate fragment (Supplementary Figs 12–23). The identity and affinities of Parviraptor were further confounded by recent phylogenetic investigations that, based on the original descriptions15, found the taxon to have gekkotan affinities18,19. These hypotheses arose from the assignment to Parviraptor cf. P. estesi15 of gekkotan vertebrae from the Kirtlington Quarry (Supplementary Figs 20–23) that bear notochordal canals piercing the centra, as well as amphicoelous condyles/cotyles; it is important to note that no such vertebrae occur on either the original Swanage type block, NHMUK R8551, or the referred block NHMUK R48388. In addition, the isolated parietals preserved on the Swanage blocks (Supplementary Fig. 13) are identified here as gekkotan, which explains the character scorings for Parviraptor as a gekkotan18,19, in contrast to the original description15. The original descriptions of Diablophis gilmorei16 and Portugalophis lignites17 avoided the chimaera problem of Parviraptor estesi/aff. Parviraptor15 as the materials selected from the microvertebrate assemblages were only compared with the maxilla and teeth of P. estesi. Identifying the affinities of disarticulated vertebrate remains is difficult and relies on detailed comparisons of specialized features that characterize one vertebrate over another. However, snakes possess a number of detailed cranial and postcranial anatomies that are definitively characteristic of the group, distinguishing them from other squamates, and that are present in these new Middle Jurassic–Lower Cretaceous forms. The unique features of snake cranial, dental and axial skeletal elements make it possible to identify snakes in the fossil record from isolated or even fragmentary elements2,3,20. Skull, jaw and dental features characterizing fossil and modern snakes21 that are present in the oldest snakes described here (Figs 1a–l and 2a–t; Table 1; Supplementary Figs 1–11; Supplementary Videos 1–6 and 11–14), recognizing of course the limits imposed by incompleteness and preservation, include (1) long, tooth-bearing sub-orbital ramus of maxilla, unique to snakes with exception of some gekkos (Fig. 2a,c–f; Supplementary Figs 2a–c and 7a–f; Supplementary Videos 1, 2, 10–12); (2) distinct and medially projecting palatine process of maxilla lacking supradental shelf as present in lizards (Fig. 2a–f; Supplementary Fig. 2a–c; Supplementary Videos 1, 2, 10–12); (3) prefrontal facets on ascending process of maxilla anterior to palatine process (Fig. 2a–f; Supplementary Video 1, 2, 10–12); (4) low and rounded ascending process of maxilla, with small medial facets for prefrontal (Fig. 2a–f; Supplementary Figs 2a–c and 7a–f; Supplementary Video 1, 2, 10–12); (5) lateral dentary wall emarginated, resulting in exposure of anterior (‘prearticular’)

Table 1 | Distribution in new fossil taxa of diagnostic cranial features present in some or all modern and fossil snakes. Cranial Anatomical Trait

Parviraptor estesi

Portugalophis lignites

Diablophis gilmorei

Eophis underwoodi

aff. Parviraptor

Long, tooth-bearing sub-orbital ramus of maxilla Articulating or abutting palatine process of maxilla Prefrontal facets on maxilla anterior to palatine process Low ascending process of maxilla, small prefrontal facets Dentary laterally emarginated Dorsal exposure, anterior superior alveolar foramen Smooth, rounded anteromedial process maxilla Well-developed descensus frontalis & suboptic shelves Strongly recurved maxillary & dentary teeth Short tooth root Small posterolingual nutrient foramen on alveolus Alveoli three-sided Labial & lingual carinae on apex of tooth crowns

Present Present Present Present — Present Present — Present Present Present Present Present

Present Present Present Present Present Present Present — Present Present Present Present Present

Broken Present Present Present Broken Present Broken — Present Present Present Present Present

— — — — — — — — Teeth missing Teeth missing Present Present Teeth missing

— — — — — — — Present — — — — —

6

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

portion of compound bone (Supplementary Figs 7g and 8c; Supplementary Videos 13, 14); (6) dorsal position and superior exposure of anterior opening of superior alveolar canal (anterior superior alveolar foramen) along anterodorsal margin of maxilla (in lizards, this feature is variably projecting medially, and is covered by nasals, septomaxilla, or overhanging ascending process of maxilla) (Fig. 2a–f; Supplementary Figs 7e and 9a); (7) smooth and rounded anteromedial process of maxilla for ligamentous connection with premaxilla (Fig. 2a–f; Supplementary Videos 10–12); (8) well-developed descensus frontalis and posteriorly, suboptic shelves (Fig. 2s–t; Supplementary Fig. 3c,d); (9) strongly recurved maxillary and dentary teeth where maxillary teeth are more recurved than the dentary teeth; a sigmoidal curvature cannot be observed; however, such curvature is not present in all snakes and can also disappear ontogenetically (see for example, Fig. 2a,c,k with Fig. 2e,f,m,n; Supplementary Figs 2a–c and 8c; Supplementary Videos 1–4 and 10–12); (10) short tooth root, with enamel cap not extending to cementum margin; (11) small posterolingual nutrient foramen at margin of alveolus (in lizards, nutrient foramina, when present, are bounded, at least partially, by base of tooth); (12) alveoli threesided, alveolar bone forming interdental ridges and lining lateral parapet (Figs 1a–l and 2a–p; Supplementary Fig. 6a–d; Supplementary Videos 1–4 and 10–12); (13) labial and lingual carinae on apex of tooth crowns (Fig. 2a,k,e–n,q,r; Supplementary Fig. 2b,d,e). Comparison of the alveolar walls of Eophis underwoodi with those of the modern snake Xenopeltis unicolor (Fig. 2o,p; Supplementary Fig. 6a–d) shows a striking similarity in the construction between the two taxa. In both taxa the interdental ridge is continuous posteriorly with the lateral parapet to form a J-shaped boundary to the alveolus. An obvious difference between the jaws of the oldest known snakes and geologically younger snakes is the presence of a maxillary supradental shelf (Fig. 2a–c) and distinct dentary subdental shelf and sulcus dentalis that runs the length of the tooth-bearing element (Fig. 2g–k). In modern snakes, inclusive of scolecophidians, and geologically younger snakes such as Yurlunggur, the medial bony tissues of the maxilla and dentary grade smoothly20,21 into the tooth row without forming a shelf (Fig. 2d–f,l–m) though a small shelf persists on the dentary immediately superior to the articulation with the posterior portion of the splenials (Fig. 2l–m). Fossil snakes such as Dinilysia are similar to the oldest known snakes, while Coniophis is more similar to Yurlunggur. Associated with the tooth bearing elements of Diablophis gilmorei, is a small elongate element that appears to be the right surangular bone (Supplementary Fig. 10i–k); the element was not included in the parsimony analysis as the association is weak. However, this disarticulated surangular is long, bears a large snake-like adductor fossa and laterally directed coronoid process, but is not fused into a single compound bone (thus, the articular surface for the quadrate is only partly preserved, as the articular bone is missing). It may well be that B140 Myr ago the compound bone characteristic of all other fossil and modern snakes20,21 had not yet appeared in snake evolution. Vertebral features that are present in two of the new fossil snakes (aff. Parviraptor estesi; Diablophis gilmorei) and are shared with fossil snakes such as Najash rionegrina and Coniophis precedens2,3 (Fig. 3a–e; Table 2; Supplementary Figs 4a–g and 11a–n), but not necessarily with all fossil and modern snakes include (1) well-developed synapophyses continuous with prezygopophysis; (2) synapophyses divided into dorsal and ventral surfaces; (3) all four snakes lack accessory processes on the prezygapophyses, and the synapophyses are lateral, not ventral as in more modern snakes; (4) the synapophyses flare laterally from the cotylar margin, while posteriorly the centrum is

Table 2 | Distribution in fossil taxa of diagnostic vertebral features of snakes. Vertebral Anatomical Traits (Cf. Najash & Coniophis) Well-developed synapophyses continuous with prezygopophysis Synapophyses divided in dorsal & ventral surfaces Shallow angle of prezygapophyses Accessory processes absent; rib articulations lateral, not ventral as in modern snakes Synapophyses flare laterally from cotylar margin; posteriorly centrum expanded (except C. precedens) Small zygosphenes & zygantra; zygosphenes not well developed Possession of ‘trefoil’ organization of neural canal Round condyle & cotyle Constriction at base of condyle

Diablophis gilmorei Present

aff. Parviraptor Present

Present

Present

Present Present

Present Present

Present

Present

Present

Present

Present

Present

Present Present

Present Present

expanded mediolaterally just in front of the condyle (except C. precedens); (5) small zygosphenes and zygantra are present in both D. gilmorei and aff. Parviraptor estesi; the zygosphene articulations are present, small, set on a narrow neural arch platform and not as well developed as they are in Coniophis and Najash; (6) the possession of a ‘trefoil’ organization of the neural canal3; (7) round condyle and cotyle; (8) constriction at base of condyle; (9) long, broad, transverse processes on postcloacal vertebrae. The features shared by aff. Parviraptor estesi and Diablophis gilmorei with Najash rionegrina and the much younger (B100 Myr difference) Coniophis precedens are interesting and important, despite the much weaker development of the zygosphenezygantral articulations and width of neural arch platform in the first two snakes (Fig. 3a–e; Supplementary Figs 4a–g and 11a–n). However, in contrast to most known snakes8,9, there is one potential sacral vertebra present on the aff. Parviraptor estesi (NHMUK R8551) slab (Fig. 3a) and another in the vertebral specimens referred to Diablophis gilmorei (4684/140572). Each vertebra is far more robust than the materials described for Najash rionegrina8,9 (Fig. 3d), suggesting the possible presence of robust pelvic girdles and hind limbs. The referred postcloacal vertebra of D. gilmorei (LACM 4684/120472) possesses a large transverse process that is directed laterally, not anteriorly as in modern snakes, and is more similar in size and orientation to postcloacal transverse processes of Najash. As this report extends the fossil record of snakes by B70 million years, it is likely that these early snakes shared with younger fossil snakes4,7,8 the presence of at least rear limbs. However, we remain conservative on this point and have not coded this information for phylogenetic analysis (Supplementary Information Section 3); presence or absence of forelimbs remains mere speculation as it cannot be verified by reference to current specimens. Phylogeny. At the alpha taxonomic level, based on comparative anatomy, Parviraptor, Diablophis, Eophis and Portugalophis are confidently identified as snakes1–3,20,22–24. However, to address questions of phylogeny, we conducted two separate analyses using two vastly different data matrices. To test the possibility that these new snakes might fall outside of the snake clade in an overall analysis of squamate phylogeny, we included them in a recent large matrix of all squamates25. We also examined the

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

7

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

more specific characters related to snakes, by including them in a data matrix originally developed to test snake ingroup relationships3,7,8,26–28 (Supplementary Methods; Supplementary Dataset 1). Both phylogenetic analyses were run using the software programs PAUP29 and TNT30, the latter was also used to calculate bootstrap values of statistical support. The matrix25 analysed in our examination of overall squamate phylogeny included the addition of the maxillae of Parviraptor, Portugalophis and Diablophis as terminal taxa (the maxillae were selected because, of all the isolated elements, they are those that allow for the largest number of character states that can be coded;

Sineoamphisbaena hexatabularis Anniella pulchra Anelytropsis papillosus Dibamus novaeguineae Rhineura floridana Spathorhynchus fossorium Dyticonastis rensbergeri Bipes biporus Bipes canaliculatus Trogonophis wiegmanni Diplometopon zarudnyi Geocalamus acutus Amphisbaena fuliginosa Najash rionegrina Parviraptor maxilla Portugalophis maxilla Diablophis maxilla Anomochilus leonardi Anilius scytale Cylindrophis ruffus Uropeltis melanogaster Haasiophis terrasanctus Eupodophis descouensis Pachyrhachis problematicus Dinilysia patagonica Leptotyphlops dulcis Typhlops jamaicensis Liotyphlops albirostris Typhlophis squamosus Epicrates striatus Boa constrictor Calabaria reinhardtii Eryx colubrinus Lichanura trivirgata Exiliboa placata Ungaliophis continentalis Aspidites melanocephalus Python molurus Loxocemus bicolor Xenopeltis unicolor Xenophidion acanthognathus Casarea dussumieri Trachyboa boulengeri Tropidophis haetianus Xenodermus javanicus Acrochordus granulatus Pareas hamptoni Lycophidion capense Aparallactus werneri Atractaspis irregularis Lampropeltis getula Coluber constrictor Natrix natrix Afronatrix anoscopus Amphiesma stolata Thamnophis marcianus Xenochrophis piscator Notechis scutatus Naja naja Laticauda colubrina Micrurus fulvius Causus rhombeatus Azemiops feae Daboia russelli Agkistrodon contortrix Bothrops asper Lachesis muta

other elements associated with these maxillae were not included to provide the strictest test of phylogenetic relationships); the goal of this analysis was to test whether or not these partial skeletal remains would be recovered as snakes, or fall outside that clade. In the strict consensus tree of 108,981 shortest trees retrieved in PAUP29 (tree length: 5290 steps), all three taxa were reconstructed within a polytomy comprised of Mesozoic snakes (Dinilysia, Pachyrhachis, Haasiophis and so on), scoleocophidians and alethinophidians, with Najash as the most basal snake. The strict consensus of 200 trees retrieved using TNT30 (tree length: 5283) is very similar to the strict consensus of the trees retrieved

Pythoninae

Boinae

Erycniae

Bolyeriidae

Ungaliophiidae

“Basal colubroids”

Acrochordidae

Xenopeltis

Tropidophiidae

Loxocemus

Haasiophis

Pachyrachis

Eupodophis

Anilius

Uropeltidae

Coniophis

Scolecophidia

Najash

Wonambi

Yurlunggur

Sanajeh

Dinilysia

Parviraptor

aff. Parviraptor

Portugolophis

Eophis

Diablophis

Anguimorph root

Ophidia

Figure 4 | Phylogeny of earliest known snakes and other fossil and living snakes. (a) Snake portion of Strict Consensus Tree of 108,981 shortest trees, illustrating overall phylogeny of squamates25 including Parviraptor, Diablophis and Portugalophis (yellow background). All three taxa are nested within a polytomy comprised Mesozoic snakes (Dinilysia, Pachyrhachis, Haasiophis and so on), scoleocophidians and alethinophidians, with Najash9 as the basal most snake. Tree length (TL) ¼ 5290; consistency index (CI) ¼ 0.1841; homoplasy index (HI) ¼ 0.8159; retention index (RI) ¼ 0.7934; Bootstrap support for clade Ophidia: 85. (b) Tree derived from Strict Consensus Tree of 305 most parsimonious trees (MPTs) retrieved by parsimony analysis of matrix composed of 27 terminal taxa of fossil and living snakes, and 237 characters (tree length (TL) ¼ 515; consistency index (CI) ¼ 0.5340; homoplasy index (HI) ¼ 0.4660; retention index (RI) ¼ 0.6935). Terminal taxon ‘Coniophis’ (vertebrae, left maxilla, right dentary only) reconstructed in sistergroup position to all extant snakes with fossil clade Pachyophiidae nested within that extant clade. The earliest snakes (yellow background) form a clade as sistergroup to all other snakes, fossil (green background) and modern (blue background), in 98 of 305 MPTs. In remaining 207 MPTs, the earliest snakes form a paraphyletic assemblage below Dinilysia. 8

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

in PAUP29, and the only differences consist in the repositioning of some basal macrostomatans (for example, Calabaria, Eryx and Lichanura) and in the addition to the basal polytomy of the taxa Casarea, Xenophidion, Loxocemus and Xenopeltis. The clade Ophidia (that is, all extant and fossil snakes) was retrieved with strong bootstrap support (BS ¼ 85) (Fig. 4a; Supplementary Methods). These results are supportive of our identification of these specimens as the remains of snakes, and that derived features of snakes were present among squamates as early as the Middle to Upper Jurassic. In our parsimony analyses of snake ingroup phylogeny, we submitted a taxon-character matrix of 27 terminal taxa and 237 characters, derived in large part from a recent study3 that in turn had adapted characters and taxa from several other works8,11,24,26,27,31 (Supplementary Methods; Supplementary Dataset 1). Our characterization of the Upper Cretaceous snake, Coniophis precedens3 is more conservative than that given in a recent report, and is restricted to the vertebral assemblage, a single maxillary fragment (the other maxillary fragments are identified here as lizards, not snakes) and a fragmentary right dentary3,24. We also followed the more restrictive concept of Najash8 in coding that taxon as was recently suggested in a conservative revision of the materials assigned to the type of that taxon9 (Supplementary Dataset 1). The results of our ingroup phylogenetic analysis suggest that the new Jurassic/Lower Cretaceous fossil snakes either form a clade in the sistergroup position to all other known snakes, or form a paraphyletic assemblage at the base of the radiation of all other snakes (Fig. 4b). The new data, and the results of our analysis of snake phylogeny (see also Supplementary Figs 24 and 25), agree with previous ingroup analyses of fossil snakes3,4,8,26–28,31 that hypothesized that the most basal snakes are not the extant scolecophidians7,32. The new early snakes add to those hypotheses by extending the fossil record of snakes by 70 million years (Fig. 1m–p) coincident with the radiation of other mid-Jurassic squamate groups12,13 and with molecular clock predictions14. Paleobiogeography. An extension of the fossil record of snakes to the Middle Jurassic is not surprising, as it coincides with the radiation of other major groups of squamates12,13, even though these new records of diverse and broadly distributed midMesozoic (B167–145 Myr ago) snake assemblages (Fig. 1m–p) create a significant temporal gap with the previous early records of snakes (B100 Myr ago) from Africa1, North America2, Brazil32 and Europe33. An unexpected result is the paleoecological and paleobiogeographical diversity and distribution of these Middle Jurassic–Early Cretaceous snakes (Fig. 1m–p). These earliest snakes are found preserved in rocks deposited in coal swamps (Portugalophis)17, mixed coastal lake and pond systems (Eophis, Parviraptor, aff. Parviraptor), river systems (Diablophis)16 and on epicontinental islands located up to several hundred kilometers off the coastline of the closest large landmass (Fig. 1n–p). Subsequent snakes appear in rocks deposited in fluvial1–3,8–11 and marine environments4–7, making the paleogeography and paleoecology of the earliest snakes into important reference points for later evolutionary radiations. It is possible that these early snakes were isolated on various islands and continents during the break-up of Pangaea in the mid-Mesozoic (Fig. 1n). Such a scenario finds support based on the similarities of their cranial features to those of Dinilysia10,11 and later madtsoiids22,26–28 (Fig. 2a–p), which certainly appear to be restricted to Gondwana continents during the Mesozoic and Cenozoic. It is also possible that snakes forming these island assemblages arrived as secondarily aquatic invaders. Secondary

invasions of marine environments certainly characterize the subsequent evolutionary histories of numerous clades of fossil and modern snakes4–6,31. In fact, there is no evidence to suggest that these radiation scenarios are mutually exclusive, and so it is possible that both Pangaean-derived vicariance and multiple adaptive radiations (marine and terrestrial) influenced the Jurassic radiation of snakes and the subsequent Gondwanan and Laurasian evolution of these animals through the Mesozoic and Cenozoic4–11. Discussion These earliest snakes provide insights on aspects of snake origins, adaptive radiations, phylogeny and evolution. A recent study on the Late Cretaceous snake Coniophis precedens concluded that the snake body evolved before the snake head3; this conclusion was based on supposed lizard-like features observed in isolated and disassociated skull elements linked to similarly isolated vertebrae from numerous individuals collected at two disparate localities. Observations of living groups of lizards and snakes suggest the opposite transformation of morphology from that proposed in a recent study of material referred to Coniophis3, that is, that the snake skull evolved before the elongate and limb-reduced to limbless postcranial skeleton. For example, scincid lizards, whether fully limbed and short-bodied, or limb-reduced to limbless and elongate, are diagnosable as scincid lizards by their distinctive skull anatomy. This is also true for all limb-reduced to limbless anguids, cordylids, gerrhosaurids, gymnophthalmids and pygopodids34,35. Because snakes are lizards, nested within that monophyletic assemblage31, it is possible to extend this analogy as a test of the conclusions presented in the aforementioned Coniophis study3. The observation that ‘skinkness’ or ‘anguidness’ is diagnosed not by elongation and limb reduction to limblessness, but rather by the shared possession of anatomical features of the head, and the recognition that snakes are lizards, leads to the prediction that the fossil record of snake evolution will likely reveal four legged, short bodied, ‘stem-snakes’ that possess ‘snake’ skull anatomies. Such a prediction is logically consistent with the pattern seen in all groups of living limbless to limb-reduced lizards; the recent claim to the contrary3 violates that prediction and would be extraordinary if supported by fossil evidence. While we cannot ascertain the shape, length, form and so on of any aspect of the body of the earliest snakes (B167 Myr ago) reported herein, these animals are, however, identifiable as snakes by their cranial features. If ‘snakeness’ is recognized in the numerous cranial features of the B167–143 Myr ago snakes described herein, nearly 100 million years before Coniophis precedens3, and if cranial features diagnose squamate groups regardless of postcranial evolution, then the prediction given here is that evolution of the snake skull was the key innovation of the clade, and not elongation and limb reduction, contrary to previous claims3. By the Middle Jurassic–Lower Cretaceous, and amidst the break-up of the old supercontinent Pangaea, snakes were distinguished from their lizard relatives by the possession of a skull with snake cranial and dental features that are certainly present by the Cenomanian4–7 and are retained to the present. Similar to many lineages of squamates34,35, it seems likely that they then subsequently evolved elongate and limb-reduced to limbless bodies, evolving and adapting through more than 167 million years of earth history to become the extremely diverse group they are today. Methods Materials examined. The specimens of fossil (13 taxa) and living snakes (76 taxa) examined in this study belong to the collections of the following institutions: American Museum of Natural History, New York, USA (AMNH); Field Museum

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

9

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

of Natural History, Chicago, Illinois, USA (FMNH); Hebrew University of Jerusalem, Paleontology Collections (HUJ-PAL); Museo Argentino de Ciencias Naturales ‘Bernardino Rivadavia’, Buenos Aires, Argentina (MACN-RN); Museum of Comparative Zoology, Cambridge, Massachusetts, USA (MCZ); Museo de La Plata, La Plata, Argentina (MLP); MPCA, Museo Provincial Carlos Ameghino, Cipolletti, Rı´o Negro, Argentina (MPCA-PV); Museo di Storia Naturale di Milano, Milano, Italy (MSNM); Natural History Museum, London, UK (NHML); Naturhistoriches Museum, Vienna, Austria (NMV); Queensland Museum, Brisbane, Australia (QMF); Natural History Museum of Gannat, Gannat, France (Rh-E.F.); Senkemberg Museum, Frankfurt, Germany (SMF); University of Alberta Zoology Museum (UAZM); University of Florida, Gainesville, Florida, USA (UF); National Museum of Natural History, Washington, DC, USA (USNM); Zoologisches Forschungsmuseum Alexander Koenig, Bonn, Germany (ZFMK). The thirteen fossil snake taxa include Eupodophis descouensi Rh-E.F. 9001, 9002, 9003, MSNM V-3660, 3661, 4014; Haasiophis terrasanctus HUJ-PAL 659; Pachyrhachis problematicus HUJ-PAL 3659, HUJ-PAL 3775; Dinilysia patagonica MLP 26-410, MPCA-PV 527; MACN-RN-976, 1013, 1014; Yurlunggur sp. QMF 45391; Najash rionegrina MCPA-PV 380-388, 389–400; Sanajeh indicus26; Wonambi naracoortensis SAM 30178; Parviraptor estesi NHMUK 48388; Portugalophis lignites MG-LNEG 28091, 28094, 28100; Eophis underwoodi NHMUK R12355, R12354, R12370; Diablophis gilmorei LACM 4684/140572, 4684/140572, 5572/120732; aff. Parviraptor estesi NHMUK R8551; Pachyophis woodwardii NMV A3919. Additional anatomical features were observed by reference to CT scans available through www.digimorph.org (Aspidoscelis tigris; Anomochilus leonardi; Cylindrophis rufus; Uropeltis woodmasoni; Anilius scytale; Xenopeltis unicolor; Loxocemus bicolor; Typhlops jamaicensis; Leptotyphlops dulcis; Python molurus; Boa constrictor; Casarea dussumieri; Ungaliophis continentalis; Lichanura trivirgata; Calabaria reinhardtii; Naja naja; Thamnophis marcianus; Natrix natrix), and from MicroCT scans completed by the authors (Anomalepis flavapices; Atractaspis aterrima). Phylogenetic analysis. The sistergroup and ingroup relationships of fossil and modern snakes were approximated using parsimony methods in both PAUP and TNT. The sistergroup relationships of Parviraptor, Portugalophis and Diablophis, and a large sample of other fossil and modern snake terminal taxa, were tested using the matrix and methods (for example, ordering, outgroup selections and so on) of a recent study testing overall squamate relationships25. Sixteen characters were coded in that matrix25 for the maxillary elements of the three fossil snake taxa described here (see Supplementary Methods for details). The ingroup relationships of snakes, including the fossil taxa described here, were tested using a significantly modified data matrix (see Supplementary Methods; Supplementary Dataset 1) from a recent study reporting on the phylogenetic relationships of the Mesozoic snake taxon, Coniophis3. The same outgroup taxon was used as was employed in that previous study3, although with some modifications to the coding for the ‘anguimorph root’3. Contra the previous study3, all characters were analysed unordered and unweighted in both matrix tests. The data were analysed using the Heuristic Search option in PAUP* 4.0b (ref. 29) and character transformations were optimized using the ACCTRAN assumption. The data set was also tested using two different search routines in TNT30; the Traditional search and the Drift search option from the New Technology searches. The Traditional search was run using 100 Wagner seed trees, 100 replications, but no swapping algorithims were utilized (for example, SPR, TBR) as the data set is not excessively large. The Drift search was chosen because unlike the Ratchet option it does not reweight characters during the analysis. This analysis was run using the default settings for the search. Nomenclatural acts. This published work and the nomenclatural acts it contains have been registered in ZooBank, the proposed online registration system for the International Code of Zoological Nomenclature (ICZN). The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix ‘http://zoobank.org/’. The LSIDs for this publication are: urn:lsid:zoobank.org: act:FDF73A76-1B5C-43C4-957B-479568D7B478; urn:lsid:zoobank.org:act: 5513C88B-AB09-4929-8501-5341EA3DDF46; urn:lsid:zoobank.org:act: 8B66D02D-CC1A-4DAA-B927-7C5D450272C6; urn:lsid:zoobank.org:act: A5F66167-1EF0-4FAE-8723-DF5912D7CF18.

References 1. Cuny, G., Jaeger, J. J., Mahboub, M. & Rage, J.-C. Les plus anciens serpents (Reptilia, Squamata) connus. Mise ua point sur l’age ge´ologique des serpents de la partie moyenne du Cre´tace´. C. R. Acad. Sci. Paris II 311, 1267–1272 (1990). 2. Gardner, J. & Cifelli, R. A primitive snake from the Cretaceous of Utah. Spec. Pap. Palaeont. 60, 87–100 (1999). 3. Longrich, N. R., Bhullar, B. -A. S. & Gauthier, J. A. A transitional snake from the Late Cretaceous period of North America. Nature 488, 205–208 (2012). 4. Caldwell, M. W. & Lee, M. S. Y. A snake with legs from the marine Cretaceous of the Middle East. Nature 386, 705–709 (1997). 10

5. Lee, M. S. Y., Caldwell, M. W. & Scanlon, J. D. A second primitive marine snake: Pachyophis woodwardi from the Cretaceous of Bosnia-Hercegovina. J. Zool. Lond. 248, 509–520 (1999). 6. Rage, J.-C. & Escuillie´, F. Un nouveau serpent bipe`de du Ce´nomanien (Cre´tace´). Implications phyle´tiques. C. R. Acad. Sci. Paris (IIa) 330, 513–520 (2000). 7. Tchernov, E., Rieppel, O., Zaher, H., Polcyn, M. J. & Jacobs, L. L. A fossil snake with limbs. Science 287, 2010–2012 (2000). 8. Apesteguı´a, S. & Zaher, H. A Cretaceous terrestrial snake with robust hindlimbs and a sacrum. Nature 440, 1037–1040 (2006). 9. Palci, A., Caldwell, M. W. & Albino, A. M. Emended diagnosis and phylogenetic relationships of the Upper Cretaceous fossil snake Najash rionegrina Apesteguı´a and Zaher, 2006. J. Vertebr. Paleontol. 33, 131–140 (2013). 10. Caldwell, M. W. & Albino, A. M. Exceptionally preserved skeletons of the Cretaceous snake Dinilysia patagonica Woodward, 1901. J. Vertebr. Paleontol. 22, 861–866 (2002). 11. Zaher, H. & Scanferla, C. A. The skull of the Upper Cretaceous snake Dinilysia patagonica Smith-Woodward, 1901, and its phylogenetic position revisited. Zool. J. Linn. Soc. 164, 194–238 (2012). 12. Estes, R. Sauria Terrestria, Amphisbaenia (Handbuch der Palaoherpetologie 10A) (Gustav Fischer Verlag, 1983). 13. Evans, S. E. At the feet of the dinosaurs: the early history and radiation of lizards. Biol. Rev. 78, 513–551 (2003). 14. Pyron, R. A. A likelihood method for assessing molecular divergence time estimates and the placement of fossil calibrations. Syst. Biol. 59, 185–194 (2010). 15. Evans, S. E. A new anguimorph lizard from the Jurassic and Lower Cretaceous of England. Palaeontology 37, 33–49 (1994). 16. Evans, S. E. in The Continental Jurassic Vol. 60 (ed. Morales, M.) 243–248 (Museum of Northern Arizona Bulletin, 1996). 17. Brochinski, A. in Guimarota: A Jurassic Ecosystem (eds Martin, T. & Krebs, B.) 59–68 (Verlag Dr. Friedrich Pfeil, 2000). 18. Conrad, J. L. Phylogeny and systematics of Squamata (Reptilia) based on morphology. Bull. Amer. Mus. Nat. Hist. 310, 1–182 (2008). 19. Daza, J. D., Bauer, A. E. & Sniveley, E. Gobekko cretacicus (Reptilia:Squamata) and its bearing on the interpretation of gekkotan affinities. Zool. J. Linn. Soc. 167, 430–438 (2013). 20. Rage, J.-C. Serpentes. Handbuch der PalaoherpetologieVol. 11Gustav Fischer Verlag, 1984). 21. Cundall, D. & Irish, F. The snake skull. Biology of the Reptilia 20, 349–692 (2008). 22. Laduke, T. C., Krause, D. W., Scanlon, J. D. & Kley, N. J. A Late Cretaceous (Maastrichtian) snake assemblage from the Maevarano Formation, Mahajanga Basin, Madagascar. J. Vertebr. Paleontol. 30, 109–138 (2010). 23. Estes, R. Fossil vertebrates from the Late Cretaceous Lance Formation, eastern Wyoming. Univ. Calif. Publ. Geol. Sci. 49, 140–141 (1964). 24. Bryant, L. G. Non-dinosaurian lower vertebrates across the CretaceousTertiary boundary in Northeastern Montana. Geol. Sci 134, 1–107 (1989). 25. Gauthier, J. A., Kearney, M., Maisano, J. A., Rieppel, O. & Behlke, A. D. B. Assembling the Squamate tree of life: perspectives from the phenotype and the fossil record. Bull. Peabody Mus. Nat. Hist. 53, 3–308 (2012). 26. Wilson, J. A., Mohabey, D. M., Peters, S. E. & Head, J. J. Predation upon hatchling dinosaurs by a new snake from the Late Cretaceous of India. PLoS Biol. 8, e1000322 (2010). 27. Scanlon, J. D. Skull of the large non-macrostomatan snake Yurlunggur from the Australian Oligo-Miocene. Nature 439, 839–842 (2006). 28. Scanlon, J. D. & Lee, M. S. Y. The Pleistocene serpent Wonambi and the early evolution of snakes. Nature 403, 416–420 (2000). 29. Swofford, D. L. PAUP*: Phylogenetic Analysis Using Parsimony, version 4.0b10 (Sinauer Associates, 2003). 30. Goloboff, P. A., Farris, J. S. & Nixon, K. C. TNT, a free program for phylogenetic analysis. Cladistics 24, 774 (2008). 31. Caldwell, M. W. in Major transitions in vertebrate evolution. (eds Anderson, J. S. & Sues, H.-D.) 253–302 (Indiana University Press, 2007). 32. Rage, J. C. & Escuillie´, F. The Cenomanian: stage of hindlimbed snakes. Carnets de Ge´ologie/Notebooks on Geology 2003, 1–11 (2003). 33. Hsiou, A. S., Albino, A. M., Medeiros, M. A. & Santos, R. A. B. The oldest Brazilian snakes from the Cenomanian (early Late Cretaceous). Acta Palaeo. Polonica 59, 635–642 (2014). 34. Greer, A. E. Limb reduction in squamates: Identification of the lineages and discussion of the trends. J. Herpetol. 25, 166–173 (1991). 35. Wiens, J. J., Brandley, M. C. & Reeder, T. W. Why does a trait evolve multiple times within a clade? Repeated evolution of snakelike body form in squamate reptiles. Evolution 60, 123–141 (2006).

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

ARTICLE

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6996

Acknowledgements

Author contributions

We thank: S. Chapman (NHMUK, London, UK) for assisting during multiple visits to collections (2006-2013); L. Chiappe for access to collections (2008-2009 LACM, Los Angeles, California); M. Ramalho (MG-LNEG, Lisboa, Portugal), for access to collections; P. Currie and Z. Johansen for transport assistance with NHM Purbeck/Kirtlington specimens; A. Richter and students for assisting us in repatriating Guimarota lizard specimens (2008) to MG-LNEG, and O. Mateus who hosted us in his lab (2008). We thank R. Barbieri and C. Mun˜oz (MPCA, Cipoletti, Argentina) for their hospitality while we studied Najash. We thank G.W. Caldwell for hand-carrying MG-LNEG (2012); J.-C. Rage for discussion and opinions, and B. Barr, T. Konishi and A. LeBlanc for discussions. We thank two anonymous referees for their useful and insightful comments. Funding was provided by a Natural Sciences and Engineering Research Council of Canada Discovery Grant (#238458), an Accelerator Award (#412275), and a Faculty of Science Chairs Research Allowance to M.W.C. Funding was provided to R.L.N. by Midwestern University Intramural Research Grants, to A.P. by a Alberta Innovates Ph.D. Student Fellowship, and to S.A. by Maimonides University for collection visits to the Los Angeles County Museum.

M.W.C. and R.L.N. designed the project. M.W.C., R.L.N., A.P. and S.A. wrote the manuscript. M.L.C. and R.L.N. prepared the illustrations. All authors discussed the results and provided input on the manuscript.

Additional information Supplementary Information accompanies this paper at http://www.nature.com/ naturecommunications Competing financial interests: The authors declare no competing financial interests. Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ How to cite this article: Caldwell, M. W. et al. The oldest known snakes from the Middle Jurassic-Lower Cretaceous provide insights on snake evolution. Nat. Commun. 6:5996 doi: 10.1038/ncomms6996 (2015).

NATURE COMMUNICATIONS | 6:5996 | DOI: 10.1038/ncomms6996 | www.nature.com/naturecommunications

& 2015 Macmillan Publishers Limited. All rights reserved.

11

1

Supplem mentary Fiigure 1. Parrviraptor estesi. (top) NHMUK N 488388 - Typee block. (bottom) Color-cooded silhouuettes (bottom m) indicatinng updated ttaxonomic iinterpretatioon of the isoolated elementts. Scale baar equals 1 ccm. The typpe left maxillla and otherr specimenss originally referredd to Parvirapptor are isollated elemennts on a smaall block froom the Purbbeck Limestoone Formatiion, Durlston Bay, Swaanage, Dorseet, England (Upper Juraassic; Tithonnian/Lower Cretaceoous; Berriassian). The block also coontains isolaated elemennts from nonn-squamate taxa t as well aas isolated eelements thaat appear to be squamatte, but so diistinctly nonn-ophidian they t are remooved from P Parviraptorr estesi.

2

Supplementary Figure 2. Holotype left maxilla and details of dissociated maxillary tooth of Parviraptor estesi (NHMUK 48388 - part). (a) photograph of medial view. (b) Illustration of same (box indicates area detailed in d and e. (c) dorsal view. (d) Medial view of isolated tooth (position 3 or 4) of Parviraptor estesi (NHMUK 48388- part). (e) SEM of isolated tooth. (f) Detail of carina along medial edge of tooth. Abbreviations: ap, ascending process; asaf, anterior superior alveolar foramen; idr, interdental ridges; lin car, lingual carina; nf, nutrient foramen; pmxp, premaxillary process; Pl-pr, palatine process.

3

Supplem mentary Fiigure 3. Bloock with frontal and vertebrae v referred to aaff. Parvirap aptor estesi (N NHMUK R R8551). (a) photographh of block, N NHMUK R88551). (b) C Color-coded silhouetttes indicatinng updated taxonomic interpretatio i on of the isoolated elemeents. (c) NH HMUK R8551 ((part). (d) Pen and ink iillustration of o same from m slightly m more dorsal view. Abbreviiations: df, ddecensus froontalis; Lsopp, left subopptic shelf; ppsf, postfronntal facet; prrf, prefronttal facet; Rssop right subboptic shelff.

4

Supplementary Figure 4. Vertebrae from block NHMUK R8551 (Swanage), aff. Parviraptor estesi. In (a) posterior, (b) right lateral, (c) ventral, (d) dorsal, (e) dorsal (neural arch missing), (f) dorsal views, (g) left atlas neural arch in lateral view, (h) detail of block NHMUK R8551, (i) line drawing of block NHMUK R8551 indicating position of specimens. (a–g) Box indicates area of detail in (h). Red indicates specimens, aff. Parviraptor estesi. All scale bars equal 1 mm.

5

Supplem mentary Fiigure 5. Eop ophis underw woodi, com mposite illusstration of rright dentaary specimeens from Foorest Marb ble, Kirtlinggton, Englaand, Late Ju urassic (Baathonian). ((a– c), holottype, NHMU UK R123555: (a) Photograph of meedial view. (b) ( SEM off same (box indicatees area of deetail in i); annd (c) photograph of ventrolateral vview. (d–f) Paratype NHMUK K R12354, mid-portionn: (d) Photoograph of meedial view. (e) SEM off same. (f) Photogrraph of laterral view. (gg–i) NHMUK K R 12370, referred poosterior portiion: (g) Photogrraph of meddial view. (h h) SEM of saame. (i) Phootograph of lateral view w. (j) Detaill of second alveolus a of NHMUK R R12354. (k)) Detail of aalveoli of NH HMUK R122354 showinng light passsing througgh the interddental ridge canals. Abbbreviationss: cf, coronooid articulatiion facet; iddr, interdental lamina; im ms, intermaandibular sepptum; mf, m mental foram mina; Mkf, Meckeliian fossa; sddl, subdentaal lamina; sff, splenial arrticulation fa facet; sym, ssymphysis.

6

Supplementary Figure 6. Comparison of the alveoli of Eophis underwoodi and Xenopeltis unicolor, and details of possible maxillary fragment of cf. E. underwoodi. (a–b) Microphotograph of posterior portin of right dentary of Eophis underwoodi NHMUK R12370 (posterior portion of right dentary). (c–d) SEM image of left maxilla of Xenopeltis unicolor FMNH 287277. (e) Rostral view of NHMUK 12370, posterior portion of right dentary. (f–g) NHMUK R12352, fragment of left maxilla referred to Eophis underwoodi in (f) medial view and (g) detail of tooth imbedded in interdental ridge canal. Abbreviations: ir, interdental ridge; irc, interdental ridge canal; sds, supradental shelf; CNV3, canal for mandibular division of trigeminal nerve and associated vascular bundle. Note: J-shaped boundaries of alveoli are shown in red in (b) and (d). Scale bar equals 1 mm.

7

Supplem mentary Fiigure 7. Poortugalophiss lignites, new taxon, Guimarota, G , Portugal (Kimmeeridgian). (a–f) Holottype left maxxilla (MG-L LNEG 280991): (a) Phottograph in m medial view. (b b) Pen and innk drawing of same. (cc) Photograpph of dorsolateral view.. (d) Pen and ink drawingg of lateral vview. (e) Phhotograph off dorsal view w. (f) Photograph of occlusal view. (g–i) Paratypee left dentarry (MG-LNEG 28094):: (g) Photoggraph of lateeral view. (h h) Pen and innk drawingg of same. (ii) Photograpph of medial view (box indicates arrea of detaill in j). (j) D Detail of teeth annd alveoli. A Abbreviationns: ap, ascennding proceess; asaf, antterior superrior alveolarr foramen; idr, interdental ridgges; le, lateraal emarginaation; mf, meental foramina; Pl-pr, ppalatine proccess; pn, posterioor notch of aascending prrocess; psaff, posterior ssuperior alveolar foram men; Pmx-prr, premaxiillary processs; sdl, subddental laminna.

8

Supplem mentary Fiigure 8. Poortugalophiss lignites, new taxon, Guimarota, G , Portugal (Kimmeeridgian), sstereopairs of paratyp pe left dentaary (MG-LNEG 280944). (a) Meddial view. (b) Supeerior view. ((c) Lateral vview. Abbreeviation: le, lateral emaargination. Scale S bar eqquals 1 mm.

9

Supplem mentary Fiigure 9. Poortugalophiss lignites, new taxon, Guimarota, G , Portugal (Kimmeeridgian), rreferred lefft maxilla (M MG-LNEG G 28100). (aa) Dorsal viiew. (b) Lateral view. (c) Meddial view. A Abbreviationns: ap, ascennding processs; asaf, antterior superiior alveolar foramen; idr, interdental ridgges; Pl-pr, paalatine proccess; psaf, poosterior supperior alveollar foramen;; sds, S bar eqquals 1 mm.. supradental shelf. Scale

10

ementary F Figure 10. Diablolphiss gilmorei L LACM 46844/140572 (ppart), holotyppe Supple specimeens. (a–h) Holotype H rigght maxilla, and (i–k), hholotype rigght surangullar: (a) Phottograph in medial vview. (b) Peen and ink ddrawing of same. s (c) Phhotograph inn lateral view w. (d) Pen and a ink drawingg of lateral. (e–h) Holootype right ddentary: (e) P Photographh in medial view. v (f) Penn and ink drawingg of same. (gg) Photograpph in laterall view. (h) P Pen and inkk drawing off lateral view w. (i) Right suurangular, medial m view.. (j) Right suurangular, lateral l view.. (k) Detail of surangular portion of glenooid surface. Abbreviatioons: adf, addductor fossaa; ap, ascendding processs; gl, glenoiid fossa; idr, interdental ridgges; mf, menntal foraminna; Mkf, Meecklian fossaa; pn, posterrior notch of ascendinng process; sdl, subdenntal lamina; sds, supradeental shelf. Scale bar eequals 5 mm m.

11

Supplementary Figure 11. Diablophis gilmorei, referred vertebrae from LACM 4684/140572. (a–j) and LACM 4684/120472 (k–n). (a–e) precloacal vertebra in dorsal, lateral, ventral, posterior, and anterior views, respectively. (f–j) caudal vertebra in dorsal, lateral, ventral, posterior, and anterior views, respectively. (k–l) possible sacral in dorsal and posterior views, respectively. (m–n) two partial precloacal vertebrae in articulation in anterior and dorsal views, respectively. Abbreviations: cn, condyle; ct, cotyle; ns, neural spine; ptz, postzygapophysis; pz, prezygapophysis; plp, pleuropophysis; sy, synapophysis; tp, transverse process; zs, zygosphene.

12

Supplem mentary Fiigure 12. T Type blocks for Parviraaptor estesi. NHMUK R48388 (toop) and aff. Parviraptoor estesi NH HMUK R85551 (bottom)) from Swannage, Dorset, England, with color-cooded silhoueettes indicatting revised taxonomic identificatioons of isolatted elementts.

13

Supplem mentary Fiigure 13. L Left ‘parietaal’ from NH HMUK R85551 and NH HMUK R488388 and comparrable paired d parietals of Gekko gecko. g (a) Phhotograph oof NHMUK R8551 in dorsal d view. (b b) reconstruction of parrietal table. (c) Gekko ggecko with pparietals higghlighted. (d) Left parietal from NHM MUK R483888 in ventral view. (e) L Line drawingg of same. (ff) Reconstruuction of parietal table. abbrreviations: cccp, crista crranii parietaalis; ?pal, poossible palattine; ppf, poosterior parietal fossa; ppp, posterior paarietal proceess; ppt, possterior parieetal tab; stp, supratempooral process..

14

Supplem mentary Fiigure 14. R Right pteryggoid and rigght ?palatin ne from NH HMUK R488388. (a) Block with w boxes aaround the pterygoid p (uppper left) annd ?palatinee (lower righht). (b) Righht pterygoiid in ventrall view. (c) Right R ?palattine in ventrral view. (d)) Left palataal region of Varanus sp. in veentral view. Abbreviatiions: Bpt-prr; basipteryggoid processs; ect, ectoppterygoid; E Ect-pr, ectopterrygoid proceess; mx, maaxilla; Mx-ppr, maxillaryy process; paal, palatine; Pl-pr, palattine process;; pt, pterygooid; Pt-pr; pterygoid proocess; Q-pr,, quadrate process; v, vvomer; V-pr,, vomerinne process. S Scale bar foor (a) equalss 1 cm, scalee bars for (b b) and (c) eqqual 1 mm.

15

Supplementary Figure 15. NHMUK R12353, partial left parietal reassigned to Scincomorpha indet. (a) SEM in ventral view. (b) SEM in lateral view. (c) Reconstruction of paired parietals. Abbreviations: ccp, crista cranii parietalis; fp, fossa parietalis; pf, parietal fork; pfp, parietal fork process; pfo, pineal foramen; pof, postorbital facet; psf, postfrontal facet; stp, supratemporal process.

16

Supplem mentary Fiigure 16. Squamate frrontals from m Kirtlingtton Quarryy Assemblagge (Bathon nian, Middlle Jurassic)) assigned tto squamata incertae ssedis. (a–c)) NHMUK R R12356, partial left parietal in: (a) Dorssal view. (b)) Lateral vieew. (c) Venttral view. (d d–f) NHMU UK L view w. (f) Ventraal view. (g––j) R123577, partial leftt parietal in:: (d) Dorsal view. (e) Lateral NHMUK K R12359, partial left pparietal in: ((g) Dorsal vview. (h) Laateral view. (i) Ventral view. v (j) Medial vview. Abbrreviations: ddf, descensuus frontalis; ; fs, interfroontal suture; nvc, neuroovascular canal; prf, p prefrontaal articular ffacet; psf, postfrontal faacet.

17

Supplem mentary Fiigure 17. L Left lateral vview compaarison of frrontals (Squ uamata ind det.) from the Old d Cement Works W Quarrry, Kirtlin ngton. (a) N NHMUK R112357. (b) N NHMUK R112356. (c) NHMUK K R12359. Abbreviatiions: df, deccensus frontalis; nvc, neeurovasculaar canal; prf,, prefronttal articular facet; psf, ppostfrontal ffacet. Blackk line indicaates the anteeriormost exxtent of the posttfrontal articcular facet inn all three sppecimens.

18

Supplem mentary Fiigure 18. N NHMUK R112358, reidentified as a mollusk sshell fragm ment (originaally identifiied as a parrtial frontall of a juven nile). (a) Exxternal view w. (b) Internaal view. (c) Detaail of brokenn margin inddicted in box in (b). Abbbreviationss: nl, nacre layer; pl, prrismatic layer.

19

Supplem mentary Fiigure 19. ?P Palatines frrom Purbecck Limestoone. (a) ?Paalatine from m NHMUK R8551. (b) ?Palatinne from NHMUK R483388. Abbrevviation: ?CN NV2, possibble channel for mandibuular branch of trigeminnal nerve. Sccale bare equuals 1 mm.

20

Supplem mentary Fiigure 20. C Comparison n between th he axis (C22) of NHMU UK R123600, cf. Gekkotta taxon inccertae sediss A, and thaat of Gekkoo gecko. (a, b, e–g) NH HMUK R123360 in (a) ventral (SEM). ( (b) Left lateral. (e) Dorsal.. (f) Anterioor. (g) Caudal views (asssuming proocoely). (c, d) C2–C3 of Geekko gecko iin (c) ventraal. (d) Left lateral viewss. Abbreviaations: bp, bbroken pedicle;; ic, intercenntrum; ncc, nnotochord canal; c op, oddontoid proccess.

21

Supplem mentary Fiigure 21. F Fragmentarry vertebraee from Kirttlington Qu uarry, Oxfoordshire, England d, previoussly assigned d to Parvirap aptor cf. esteesi assigned d here to cff. Gekkota ttaxon incertaee sedis A. ((a) NHMUK K R12361, iin (from leftt to right, toop to bottom m) dorsal, veentral, left lateral, caudal, c and cranial view ws (assuminng procoely)). (b) NHMUK R123622, in (from lleft to right) doorsal, obliquue ventral, left l lateral annd caudal vviews (assum ming procoeely). (c) NH HMUK R12363, in (from leeft to right, top to bottoom) dorsal, vventral, left lateral, caudal, cranial views MUK R123664, in (from left to rightt) caudal vieew and cross section (assuming procoelyy). (d) NHM R12366, verrtebra of cf. Gekkota taxxon incertaee sedis A in (from in craniaal view. (e) NHMUK R left to riight) dorsal,, right lateraal, ventral, ccaudal, and ccranial view ws. (f) posteerior of view w of specimeens at the saame scale foor comparisoon of relativve sizes. Abbbreviationss: bp, brokenn pedicle; ncc, nottochord canaal; ncs, neurrocentral suuture; sy, synnapophyses.

22

Supplementary Figure 22. Vertebrae from Kirtlington Quarry, Oxfordshire, England, previously assigned to Parviraptor cf. estesi assigned here to cf. Gekkota taxon incertae sedis B. (a) NHMUK R12367 in (from left to right) left lateral, ventral, posterior, and anterior views. (b) NHMUK R12365, in (from left to right) dorsal, right lateral, ventral, posterior, and anterior views. (c) NHMUK R12369, in (from left to right, top to bottom) dorsal, ventral, left lateral, posterior and anterior cross sectional views (orientation of d and e assuming procoely). Abbreviations: bp, broken pedicle; ncc, notochordal canal; ncs, neurocentral suture (closed and fused); sy, synapophyses.

23

Supplementary Figure 23. NHMUK R12368, partial vertebra of an indeterminate taxon. Views from left to right are ventral, dorsal, and anterior. Abbreviations: bp, broken pedicle; ncc, notochord canal.

24

Supplementary Figure 24. Phylogeny of oldest known snakes and other fossil and living snakes derived from TNT 1.1 analyses of the data set. (a, b), resulting Strict consensus (a) and Frequency Distribution (b) trees from ‘Traditional’ analysis (100 random seed Wagner trees and 100 replications; recovered 18 trees of length 513). (c, d), resulting Strict consensus (c) and Frequency Distribution (d) trees from Drift (New Technology) analysis (default assumptions used; resulted in three trees of length 513).

25

Supplem mentary Figgure 25. Strrict Consen nsus Tree of 108,981 shortestt trees (usin ng TNT 1.1), illustratting overalll phylogenyy of squamaates9 includin ng Parviraptor, Diablopphis, and Portugaalophis. All three taxa aare nested w within a polytom my comprisedd of Mesozooic snakes (D Dinilysia, Pachyrhhachis, Haassiophis, etc..), scoleocopphidians, and alethhinophidianns, with Najaash as the basal most snake.

26

Supplementary Note 1

SYSTEMATIC PALEONTOLOGY OF NON-SNAKE MATERIALS

Identity of non-snake specimens originally referred to Parviraptor

Identity of non-snake remains originally referred to Parviraptor estesi The fossils from the Purbeck Marble (Tithonian–Berriasian), Durlston Bay, Swanage, Dorset, England originally referred to Parviraptor estesi1 comprise a series of disarticulated and disassociated cranial and postcranial remains that are co-preserved on two limestone blocks (Supplementary Fig. 12) collected from the Purbeck Marble (Parviraptor estesi), Swanage, Dorset. Most of the elements on the blocks from Dorset were assigned to Parviraptor estesi1 represents an assembling of these elements. Based on our examination of the blocks from Dorset we have come to a more conservative conclusion regarding which of the exposed elements can be referred to the snake Parviraptor estesi. In addition to the type maxilla of P. estesi there are also crocodilian, fish, and indeterminate vertebrate remains on block NHMUK 48388. Since the type maxilla of P. estesi is clearly that of a snake and with the exception of the nonhypertrophied palatine process, has no discernable residual lacertilian features we expect any other elements of the same taxon to be similarly “snake-like” in morphology. Three other elements, a pterygoid, a parietal, and a possible palatine, on the same block were referred to P. estesi1. Of these the pterygoid is almost certainly that of a squamate (palatine process is missing, but its impression remains on the block), but of lacertilian-, not ophidian-grade. The parietal is also referable to Squamata, but of uncertain affinity. Other specimens include an indeterminate

27

squamate, a crocodylian, and several elements that are too poorly preserved or lacking in distinguishing morphological characteristics for confident referral to any group (Supplementary Fig. 12). We have also established above that the frontal and the vertebrae of NHMUK R8551 belong to a snake with probable affinities to aff. Parviraptor estesi. The other squamate specimens compare more favorably with non-ophidian taxa or are indeterminate. The following is a review of the non-jaw specimens associated with the Parviraptor material from England.

A. Parietal (NHMUK R8551, part) (Supplementary Fig. 13a–b) The isolated left ‘parietal’ from block NHMUK R8551 was the primary specimen among the three parietals identified1 for the original reconstruction of the skull of Parviraptor estesi. Among squamates paired parietals are limited to most xantusiids, pygopodids, and noneublepharid geckos2. The identification of this element as a parietal is questionable since its structure is unlike that of any squamate parietal (particularly any snake) for which we are aware. Notably, the structures identified as supratemporal processes are semicircular and have a deep, subcircular fossa bounded by a prominent ridge. For all squamates that possess supratemporal processes on the parietal the processes are straight, not curved, and typically taper to a point as they project posterolaterally from the parietal table. Additionally, the anterior (or frontal) margin of this ‘parietal’ courses anterolaterally, not transversely. However, due to the ‘palatine’ overlying the ‘parietal’ we are unable to determine if this margin is natural or a broken edge. If it is natural this shape of the anterior margin is generally similar with that of many crown group snakes, but in these taxa the parietal has anterolateral processes that clasp the frontal and not a V-

28

shaped contact with the frontal. While identification as a parietal is problematic, we do agree that the general outline of the specimen suggests no reasonable alternative identification1 and reconstructed the element here as a parietal (Supplementary Fig. 13a–b). In the reconstruction there are some noteworthy anatomical features. There is a small posterior parietal process in line with the main axis of the sagittal margin. It is possible that there was a complementary process on the right parietal, but the sagittal margin deviates to the right and we have reconstructed a single posterior parietal process similar to that in Gekko gecko (Supplementary Fig. 13c). A similar process is also present on the fossil gekkonomorph AMNH 21444 from the Early Cretaceous of Mongolia3 (which also has paired parietals). There is no convincing evidence for or against the presence of a pineal foramen along the visible portion of the sagittal margin of the ‘parietal’ and our reconstruction is commensurately vague in this regard. The supratemporal process is broken distally and its full length and degree of articular relationships are not known.

B. Parietal (NHMUK R48388, part) (Supplementary Fig. 13d–f). A second left parietal from Swanage is on the block NHMUK R48388 (Supplementary Fig. 13d–f). This specimen is preserved in ventral view and has a well-defined, but low crista cranii parietalis. The sagittal margin relatively straight, but is damaged near the anterior end of the element. The anterior margin is also damaged and the nature of the frontal articulation is unknown. Posteriorly there are three processes that extend from the parietal table. Laterally there is the heavily built base of the broken supratemporal process. Just medial to the supratemporal process there is a short posterior parietal tab that projects into the supratemporal fossa. At the medial edge of the parietal is a small posterior parietal process that project

29

posterolaterally from near the end of the sagittal margin. This posterior parietal process diverges sufficiently from the sagittal margin that we have reconstructed the complete parietal complex as having paired posterior parietal projections. These projections are too short to be considered parietal forks as is seen in NHMUK R12353 from Kirtlington (see below). It is clear that this parietal is from a different taxon than that of the ‘parietal’ from NHMUK R8551. The structure of this parietal is also sufficiently different than that of any know snakes that it is not referred to aff. Parviraptor estesi, but rather is considered as of an indeterminate lacertilian-grade squamate.

C. Pterygoid (NHMUK R48388, part) (Supplementary Fig. 14a, b) Another specimen from block NHMUK R48388 that is referable to Squamata is a right pterygoid (Supplementary Fig.14a, b). The specimen is long and gracile. An impression in the limestone preserves the shape of the dorsal surface of the missing palatine process and associated portion of the body. The exact morphology of the palatine articulation (e.g., overlapping vs. interdigitating) cannot be determined from the impression. The ectopterygoid process is short and directed anterolaterally. The quadrate process is long, narrow, and bears a thin crest along the ventromedial surface. The posteriormost portion of the quadrate process was removed when the block was cut and the exact length of the element is unknown. At the anterior end of the quadrate process is a well-developed basipterygoid process with a posteromedially facing articulation facet.

D. ?Palatine (NHMUK R48388, part) (Supplementary Fig. 14c, d)

30

The large, complex element opposite of the parietal on NHMUK R48388 was originally identified1 as a left palatine based on its similarity to the palatine of varanid lizards. We agree that this is the closest comparison and tentatively follow this identification. One difference in this interpretation is that based the relatively flat appearance of the exposed surface we believe that this is more likely a right palatine in ventral view. Based on this tentative identification the element has a long and broad vomerine process, a short and broad pterygoid process, a triangular maxillary process and a semicircular ectopterygoid process. Together these processes form an H-shaped element similar to that of Varanus (Supplementary Fig. 14d). Unlike Varanus the vomer process of the fossil specimen is turned nearly 90° relative to the same process in Varanus (Supplementary Fig. 14c, d). With the dorsal surface embedded in the limestone it is not possible to confirm the presence or absence of the typical choanal and prefronal processes, but there is no apparent arching of the palatine between the medial and lateral sides. The specimen has suffered one obvious fracture across the body of the element that resulted in the posterior displacement of the vomerine process (corrected in Supplementary Fig. 14c).

Review of specimens previously referred to Parviraptor cf. P. estesi The Mammal Bed horizon of the Old Cement Works Quarry, Forest Marble (Bathonian), Kirtlington, Oxfordshire, England, has produced a remarkable and diverse microvertebrate fauna that has been recovered through hand quarrying and screen wash concentration. The almost 50 known taxa4 recovered to date from this fauna include amphibians5–7, turtles8, choristoderes9, 10, mammals4, 11–14, sphenodontians4, and lizards1, 15 and now also a snake. The specimens are all isolated elements (i.e., jaws, cranial elements, vertebrae) making their referral to any single taxon

31

difficult. Under such circumstances any such associations are potential chimaeras and should be regarded carefully prior to utilizing the “taxon” for systematic and taxonomic studies. In the original erection of Parviraptor1 multiple specimens from the Old Cement Works Quarry were assigned to either P. estesi or Parviraptor cf. P. estesi. We have removed the jaw elements (NHMUK R12352, R12354–12356) from Parviraptor cf. P. estesi and reassigned them to the snake Eophis underwoodi. Below is a summary of the other specimens with updated interpretations based on detailed review of each element.

A. Parietal (NHMUK R12353) (Supplementary Fig. 15a–c) This specimen is a partial left parietal with broken processes for the supratemporalparietal fork. Referral to Eophis underwoodi is unlikely. Based on the broken stubs of the processes it is clear that this parietal had a very well developed parietal fork as well as a supratemporal process (Supplementary Fig. 15). The presence of a well-developed parietal fork (= bifid supraoccipital process) also occurs in some scincids (e.g., Amphiglossus, Brachymeles, Chalcides, Eumeces, Sphenomorphus) cordylids (e.g., Angolosaurus, Cordylus), and gerrhosaurids (e.g., Cordylosaurus, Gerrhosaurus, Tracheloptychus, Zonosaurus). Based on the presence of a parietal fork we refer this specimen to Scincoidea2 incertae sedis.

B. Frontals (NHMUK R12356; 12357; 12359) (Supplementary Figs 16a–j; 17a–c) These three specimens (Supplementary Fig. 16a–j) were originally identified as paired frontals of Parviraptor cf. P. estesi from Kirtlington1. This identification was based on the presence of a ventral bony process interpreted to be a decensus frontalis similar to that of the

32

frontal on the Purbeck specimen NHMUK R8551 that articulates with suboptic shelves (see above). Referral of these frontal specimens to a specific taxon is more problematic. As shown herein, the various specimens from Kirtlington originally referred to Parviraptor cf. P. estesi is actually a mix of lepidosauromorphan and non-lepidosauromorphan taxa. While we agree that these specimens are likely partial frontals it is not clear whether they are referable to the early snake Eophis underwoodi, the indeterminate scincomorphan represented by the isolated parietal, the gekkotan represented by the atlas vertebra, or to an hitherto unrecognized squamate in the fauna. Paired frontals are present throughout Squamata ([Gauthier et al., 2012: character 36(1)]), while development of a decensus frontalis is less common ([Gauthier et al., 2012: character 38(1, 2, 3)]). The combination of these two characters is most commonly found in gekkonds and some amphisbaenids, although unlike the Kirtlington frontals, the decensus frontalis in gekkonids and amphisbaenids meet at the midline and enclose the olfactory tracts. In the skink Sphenomorphus the frontals are paired and the decensus frontalis are developed similarly to those in the Kirtlington specimens. Evans1 was of the opinion that the Kirtlington specimens she referred to Parvirpator cf. P. estesi represented juvenile forms. While possible, it seems unlikely that all of the frontals (and vertebrae; see below) recovered would be from a single age class, while no definitive adult forms are present. Pending the recovery of more complete material we feel a referral to Squamata incertae sedis is more appropriate for the isolated frontals from Kirtlington.

C. “Frontal” (NHMUK R12358) (Supplementary Fig. 18a–c) This specimen (Supplementary Fig. 18a–c) has a trapezoidal shape characteristic of the other specimens of “frontals” and was originally identified as a left frontal from a juvenile of

33

Parviraptor cf. P. estesi1. The specimen lacks any of the processes or facets noted for the isolated frontals discussed above. High magnification microscopic examination of the specimen reveals an internal prismatic organization of large, presumably CaCO3 crystals between two thin nacre layers. This specimen is a fragment of a mollusk shell (Supplementary Fig. 18a–c). Mollusk shells are known to be a common component in the Mammal Bed horizon of the Forest Marble16.

D. “Palatine” (NHMUK R8551) (Supplementary Fig. 19a) Lying on the surface of the purported parietal on NHMUK R8551 is a small complex element that was originally identified1 as a left palatine in ventral view. If the specimen is indeed a palatine (see discussion above for specimen from NHMUK R48388), the arched portion exposed indicates that it is more likely to be a right palatine in dorsal view (Supplementary Fig. 19a). The ?palatine from NHMUK R8551 is approximately ¾ of the size of the similar element from NHMUK R48388 (Supplementary Figs 14c, 19b). Aside from size the primary difference between the two specimens is the clear presence of an arching dorsal surface typical of squamate palatines and the presence of a small channel near the anterior end of the ectopterygoid process that, assuming the element is a palatine, would have conducted the neurovascular bundle associated with the maxillary branch of the trigeminal nerve.

E. Vertebrae (Supplementary Figs 20, 21, 22, 23) There are 10 vertebral specimens from Kirtlington that Evans referred to Parviraptor cf. P. estesi. All of these vertebrae share the presence of a notochordal canal in the centrum.

34

Evans1 concluded that the presence of the notochordal canal indicated that all of these specimens were from juveniles of Parviraptor cf. P. estesi. While it is known that an open notochordal canal has been demonstrated to be present in hatchling Anguis and Natrix, these close soon after hatching17. In none of the Kirtlington vertebrae originally referred to Parviraptor cf. P. estesi are there any centra with closed notochordal canals. To assess the relative ontogenetic age of the vertebrae from Kirtlington we used the timing of the terminal fusion of the neurocentral suture18 where it was found that in nearly all lizards (with the exception of xantusiids) the neurocentral suture closes near or at the same time as sexual maturity. In all of the vertebral specimens from Kirtlington for which complete or partial pedicles are present—R12364 (Supplementary Fig. 21d) is only a condyle and does not preserve the rest of the centrum—all of the neurocentral sutures appear to be closed and fused. In most of the vertebrae with missing neural arches the bases of the pedicles are fractured exposing cancellous bone indicating that the neurocentral sutures were fused and remodeled. NHMUK R12363 has a clearly broken pedicle, but the bone is so thin that it lacks a cancellous portion (Supplementary Fig. 21c). We interpret this evidence as indicating that these vertebrae are from adult/nearly adult individuals. Comparison of the condylar portions of the vertebrae (Supplementary Fig. 21f; no condyle is preserved on specimen NHMUK R12368) it appears that there are likely two taxa represented by the vertebral specimens from Kirtlington. In many of the vertebrae the notochord canal at the condyle is displaced dorsally and constricted appreciably (NHMUK R12361, R12362, R12366) or nearly completely closed and present as a narrow slit or foramen (NHMUK R12360, R12363, R12364). The latter morphology does not appear to indicate a stage of closure since these vertebrae include the some of the smallest and as well as largest specimens

35

(Supplementary Fig. 21f). A similar morphology was common in the procoelus vertebrae of eublapharine geckos19, a group considered to be the most primitive of the living gekkotans; recent molecular-based20–23 and morphology-based3 phylogenetic analyses support a sister taxon relationship of eublepharids to gekkonids + sphaerodactylids among the living taxa. The difference in the state of closure of the notochordal canals in these specimens may be related to positional differences along the vertebral column and, pending the recovery of more informative specimens we choose to recognize both as likely grades within a single taxon. A closer comparison of these vertebrae is to the eublepharid gekkos (e.g., procoely, notochord canal present throughout centrum) rather than to juvenile anguimorphans or snakes. We reassign these specimens as follows below.

36

Supplementary Note 2 Systematic Paleontology of Reassigned Vertebral Specimens

Squamata Oppel, 1811 Cf. Gekkota Cuvier, 1807 Taxon incertae sedis A (Supplementary Figs 20a, b e–g; 21a–c)

Materials. NHMUK R12360–R12364, R12366; isolated vertebrae. All specimens recovered from the Mammal Bed horizon of the Old Cement Works Quarry, Forest Marble, Kirtlington, England; Late Jurassic (Bathonian). Description. Notochord canal present throughout ontogeny, notochord canal restricted to small opening at dorsal surface of condyle. Specimen specific descriptions as follows: NHMUK R12360—partial atlas vertebra (Supplementary Fig. 20a, b, e–g). The centrum is short, the odontoid process projects from the dorsal aspect of the centrum and preserves a small foramen for a restricted notochord. The opening for the notochord canal on the condylar end is restricted dorsally such that the overall outline of the opening is triangular. There is a fused intercentrum at the anterior end of the centrum with two posterolateral projections and a short hypapophysis centrally. Dorsally the broken bases of the pedicles expose cancellous bone indicating that the neurocentral suture was closed and fused. This specimen is very similar in morphology to the atlas of a gekkonid (Supplementary Fig. 20c, d). The unusual trifid second intercentrum is also found in gekkonids24. A similar

37

horizontally broad intercentrum (Supplementary Fig. 20a) has been described for some species of the eublepharine genus Coleonyx25. Additonally the presence of a notochord canal in the axis that is closed at the odontoid process, but open posteriorly is a recognized gekkonid trait3, 19 and is illustrated here in Gecko gecko (Supplementary Fig. 20c). Referral to Gekkonidae incertae sedis is more consistent with the morphology of this vertebra. NHMUK R12361—partial vertebra (Supplementary Fig. 21a). Similar to NHMUK R12360, this specimen possesses a notochord canal. Note the cancellous bone exposed in the broken pedicles (Supplementary Fig. 21a). The opening of the notochord canal on the condyle is displaced dorsally, but lacks the restriction present on NHMUK R12363 and R12364. NHMUK R12362—partial vertebra (Supplementary Fig. 21b). Similar to NHMUK R12361 in the condition of the notochordal canal at the condyle. Note again (Supplementary Fig. 21b) the cancellous bone exposed in the broken pedicles. There is a partially preserved synapophysis near the anterior end of the better preserved right side. NHMUK R12363—partial vertebra (Supplementary Fig. 21c). Similar to NHMUK R12360 in the condition of the notochordal canal at the condyle. The opening of the notochord canal at the cotyle is large, circular, and centrally located. The broken pedicle is very thin and preserves only cortical bone (Supplementary Fig. 21c), but the neurocentral suture is clearly closed and fused. NHMUK R12364—partial vertebra preserving only the condylar portion (Supplementary Fig. 21d). Similar to NHMUK R12363 in the position and degree of restriction of the opening of the notochordal canal at the condyle.

38

NHMUK R12366—nearly complete vertebra (Supplementary Fig. 21e). Neural spine moderately tall with posteriorly displaced apex (broken); pre- and postzygapophyses short; transverse processes broken; cotyle and condyle circular, opening for notochordal canal central in cotyle and displaced dorsally in condyle and only slightly smaller than opening at cotyle (similar to NHMUK R12361). Taxon incertae sedis B (Supplementary Fig. 22a–c) Materials. NHMUK R12365, R12367, R12369; isolated vertebrae. All referred specimens recovered from the Mammal Bed horizon of the Old Cement Works Quarry, Forest Marble, Kirtlington, England; Late Jurassic (Bathonian). Description. Vertebrae with complete notochordal canals, condylar opening of notochord canal central and unobstructed. Neurocentral and neural (determined only for NHMUK R12365 and R12367) sutures closed and fused. Neural spines short, condyle ventrally displaced relative to cotyle. Specimen specific descriptions below. Discussion. These vertebrae differ from the gekkonid vertebrae in condylar opening of the notochord canal being circular and located at the center of the condyle. Additionally the condyle is ventral to the cotyle. The closed and fused neurocentral and neural sutures indicates an age at or near sexual maturity17, 18. The specimens are referred to Squamata based on the presence of procoely and referred tentatively to Gekkota based on the presence of a complete notochordal canal late into ontogeny. NHMUK R12369 is tentatively included based on the centrally located and circular condylar opening of the notochordal canal. These specimens are not included with the vertebrate in cf. Gekkota taxon incertae sedis A, based on the differences in

39

the position and shape of the condylar opening of the notochord canal and differences in the relative positions of the condyle and cotyle. NHMUK R12365—nearly complete presacral vertebra (Supplementary Fig. 22a). Neural spine low; prezygapophyses short; postzygapophyses missing; synapophyses well developed; cotyle and condyle circular, opening for notochordal canal central and unobstructed in both cotyle and condyle. NHMUK R12367—nearly complete caudal vertebra (Supplementary Fig. 22b). Neural spine low; prezygapophyses excavated for articulation with postzygapophyses; postzygapophyses missing; transverse processes central, broken; centrum with ventral groove and large subcentral foramina; cotyle and condyle circular, condyle weakly projecting posteriorly; opening for notochordal canal central and unobstructed in both cotyle and condyle. NHMUK R12369—partial ?caudal vertebra (Supplementary Fig. 22c). Condyle circular with centrally placed circular opening for notochordal canal; notochord canal greatly narrowed at mid centrum. Centrum with ventral groove, but lacking subcentral foramina; neural arch and cotyle missing.

Vertebrata Taxon incertae sedis (Supplementary Fig. 23a–c) Materials. NHMUK R12368, partial vertebrae. Recovered from Mammal Bed horizon of Old Cement Works Quarry, Forest Marble, Kirtlington, England; Late Jurassic (Bathonian).

40

Description. Articular surface platy-amphicoelus, circular, and with centrally placed circular opening for notochordal canal; notochord canal greatly narrowed at mid centrum. Discussion. Based on the relatively flat articular surface it is possible that this specimen is of a sphenodontian, but the lack of preserved diagnostic morphology makes referral to Osteichthyes, Lissamphibia, or even Choristodera possible. NHMUK R12368—partial vertebra (Supplementary Fig. 38). Centrum tightly restricted centrally and with ventral groove, but lacking subcentral foramina (on preserved portion); neural arch, cotyle, transverse processes missing.

41

Supplementary Methods

Materials examined The extant snake species examined in collections include: Anomalepis aspinosus MCZ 14785 (cleared and stained); Anomalepis flavapices AMNH R-6966 (CT-scanned specimen); Leptotyphlops dulcis AMNH R-160152; Leptotyphlops humilis AMNH R-73716, USNM 222795; Leptotyphlops scutifrons MCZ 54515 (cleared and stained), MCZ 68781 (cleared and stained); Rhamphotyphlops braminus USNM 509423; Rhamphotyphlops subocularis MCZ 65993, MCZ 65997, MCZ 72084; Rhinotyphlops schlegeli MCZ 29174 (cleared and stained), MCZ 70064 (cleared and stained), MCZ 38551; Typhlops angolensis AMNH R-11633; Typhlops diardi NHML 1930-5-8-3; Typhlops lineolatus MCZ 48063; Typhlops punctatus MCZ 7293, MCZ 2249, NHML 1911-6-9-2, NHML 1975-567, SNHM 320704; Typhlops reticulatus AMNH R-3001; Anilius scytale MCZ 19537, MCZ 2984, MCZ 17645, NHML 58-8-23-48; Cylindrophis maculatus NHML 1930-5-8-50; Cylindrophis ruffus AMNH R-85647, NHML 1930-5-8-47, USNM 297456; Rhinophis planiceps NHML 1930-5-8-69; Rhinophis sanguineus NHML 19305-8-62; Uropeltis ocellatus MCZ 3873; Uropeltis pulneyensis MCZ 3870; Uropeltis rubrolineatus MCZ 47101; Loxocemus bicolor AMNH R-110151, AMNH R-44902, AMNH R19393, NHML 82-8-17-16; Xenopeltis unicolor AMNH R-29969, AMNH R-71531, NHML 1947-1-1-10, NHML 1947-1-1-12, USNM 287277; Tropidophis canus AMNH R-45839, AMNH R-73066; Tropidophis pardalis FMNH 233; Ungaliophis panamensis AMNH R-58845, AMNH R-62639, MCZ 56051; Boa constrictor ZFMK 21661, ZFMK 54844; Calabaria reinhardtii ZFMK 89190, AMNH R-10092, NHML 1911-10-28-17, UAZM R937 (dissected); Charina

42

bottae FMNH 1218, FMNH 22348, FMNH 31300; Corallus caninus AMNH R-57788, AMNH R-63587, AMNH R-73347, AMNH R-155263; Eryx colubrinus ZFMK 50246; Eryx conicus NHML 1930-5-8-14; Eryx jaculus FMNH 19624; Eryx johni NHML 1930-5-8-34; Eunectes murinus AMNH R-54158, AMNH R-29349, AMNH R-29350, AMNH R-57474; Messelophis variatus SMF ME-1828; Messelophis ermannorum SMF ME-759; Liasis albertisi ZFMK 5165, ZFMK 70427; Morelia spilota AMNH R-59880, AMNH R-79043, FMNH 22234, FMNH 22380, ZFMK 84282; Palaeopython fisheri SMF ME-1002; Python breitensteini UAZM R938 (dissected); Python molurus NHML 1972-21-78, ZFMK 5161, ZFMK 83431; Python reticulatus FMNH 15678, FMNH 51631, NHML 1972-2169, ZFMK 5175, ZFMK 70207; Acrochordus javanicus AMNH R-46251, AMNH R-140814, AMNH R-155254; Pareas carinatus NHML 1964-1092, NHML 1964-1094, NHML 1964-1098; Xenodermus javanicus FMNH 67427; Atractaspis aterrima NHML 95-5-3-58, AMNH R-12352 (CT-scanned specimen); Atractaspis bibroni AMNH R-82071; Atractaspis corpulenta MCZ 4826; Atractaspis irregularis FMNH 142994, MCZ 53534, AMNH R-12355; Atractaspis microlepidota FMNH 58397; Homoroselaps lacteus FMNH 187420, FMNH 187421, FMNH 204893, FMNH 206416; Agkistrodon piscivorus ZFMK 21724, AMNH R-81544, AMNH R-57801; Azemiops feae FMNH 218628; Causus rhombeatus FMNH 2268, FMNH 51692, FMNH 51693, FMNH 164744; Bitis gabonica ZFMK 21718, AMNH R-64518, AMNH R-57792, AMNH R-137177; Cerastes cerastes ZFMK 53537, ZFMK 5181; Vipera russelli AMNH R-75739, AMNH R-74818, ZFMK 5187; Bungarus fasciatus AMNH R-56198, AMNH R-76574; Laticauda colubrina FMNH 236242, FMNH 234147, FMNH 234149, FMNH 236242, FMNH 236243; Micrurus fulvius FMNH 34282, FMMNH 229600; Micrurus nigrocintus FMNH 210092; Naja naja AMNH R-57807, AMNH R-

43

74833, ZFMK 21704, ZFMK 21705; Pelamis platurus FMNH 171628, FMNH 171632, FMNH 216510; MCZ 7084, MCZ 131501; Cerberus rhynchops NHML 58-9-21-3, NHML 1964-10-20; Homalopsis buccata NHML 111-18-1-e, NHML 1930-5-8-630, NHML 1930-5-8-631, NHML 1964-11-25; Coluber caspius ZFMK 5221; Coluber viridiflavus AMNH R-67896; Dasypeltis scabra MCZ 30208, MCZ 54894; Heterodon platyrhinos AMNH R-63590, AMNH R-69647, AMNH R-155313; Lampropeltis getulus AMNH R-70097, AMNH R-75539, AMNH R-128202, ZFMK 54259, ZFMK 5205; Malpolon monspessulanus ZFMK 5197; Natrix natrix ZFMK 42502; Oxyrhabdium modestum FMNH 96532; Pseudoxenodon macrops NHML 1930-5-8-271, NHML 1930-5-8-273, NHML 1930-5-8-274; Thamnophis sirtalis AMNH R-74849, AMNH R148084; Thamnophis validus AMNH R-62287. Information concerning the cranial osteology of Scolecophidia was complemented by reference to three previous studies26-28. Phylogenetic analysis The following list details the character codings for the maxillae of Parviraptor, Portugalophis and Diablophis; the original character29 precedes the fossil snake taxon and the state assigned for that character. The remainder of the matrix remained coded as in the original study29 and is not reproduced here. Analysis of this large matrix resulted in 108,981 shortest trees with the followings supporting statistics: Tree Length [TL] 5290; Consistency Index [CI] = 0.1841; Homoplasy Index [HI] = 0.8159; Retention Index [RI] = 0.7934; Bootstrap support for clade Ophidia: 85 (calculated using TNT).

CH. 5: Premaxillary-maxillary fenestra: (0) absent; (1) present. Parviraptor (0), Portugalophis (0), Diablophis (?). CH. 9: Premaxilla-maxilla suture: (0) firm; (1) loose. Parviraptor (1), Portugalophis (?), Diablophis (?).

44

CH. 20: Nasal-maxilla suture: (0) present; (1) absent. Parviraptor (1), Portugalophis (1), Diablophis (?). CH 112: Maxilla premaxillary process dorsal surface grooved (often enclosed) for passage of a deeper and more internally placed ramus of the subnarial artery: (0) absent; (1) present. Parviraptor (0), Portugalophis (0), Diablophis (0). CH. 115: Maxilla facial process height: (0) tall, to skull roof; (1) reduced ; (2) absent; (3) columnar process received in longitudinal concavity on anterior face of prefrontal. Parviraptor (1), Portugalophis (1), Diablophis (1). CH 118: Maxilla narial margin rises at: (0) high angle; (1) low angle. Parviraptor (1), Portugalophis (1), Diablophis (1). CH 119: Maxilla firmly sutured to palatine: (0) present; (1) prominent palatine process of maxilla; (2) loosely ligamentous connection via projecting palatine process of maxilla and distinct maxillary process of palatine, with the former lying anterior to the latter; (3) maxilla free of palatine, suspended from prefrontal; (4) maxilla rotates to erect fang. Parviraptor (?), Portugalophis (1), Diablophis (?). CH 121: Maxilla suborbital process width ventral to ectopterygoid: (0) tapers posteriorly; (1) widens below articulation (i.e., ectopterygoid flange). Parviraptor (0), Portugalophis (0), Diablophis (?). CH 123: Maxilla suborbital process tip shape at jugal articulation: (0) suborbital margin slopes smoothly to tip; (1) with distinct step or V-shaped notch distally at jugal articulation. Parviraptor (0), Portugalophis (0), Diablophis (?). CH 125: Maxilla, intramaxillary joint: (0) absent; (1) present. Parviraptor (0), Portugalophis (0), Diablophis (0). CH 135: Prefrontal-maxilla articulation: (0) prefrontal posteroventromedial corner narrowly (or not at all) in contact with maxilla lateral to palatine; (1) prefrontal broadly contacts maxilla supradental shelf lateral to palatine; (2) prefrontal has mobile contact with maxilla; (3) rodlike prefrontal arched dorsally, bifid at each end, with mobile joints at maxilla and frontal (prefrontal functionally part of upper jaw). Parviraptor (2), Portugalophis (2), Diablophis (2). CH 417: Maxilla, enlarged teeth (fangs) (relative to adjacent teeth): (0) absent; (1) present on anterior maxilla; (2) present on posterior maxilla. Parviraptor (0), Portugalophis (0), Diablophis (?). CH 420: Maxillary tooth count: (0) 0; (1) 2–5; (2) 7–15; (3) 16–27; (4) 31 or more. Parviraptor (3), Portugalophis (3), Diablophis (?). CH 423: Position of marginal teeth relative to tooth-bearing element: (0) on medial side of toothbearing element; (1) near/on apical margin of tooth-bearing element. Parviraptor (1), Portugalophis (1), Diablophis (1). CH 424: Fusion of marginal teeth: (0) unfused to each other; (1) fused to each other. Parviraptor (0), Portugalophis (0), Diablophis (0). CH 430: Tooth replacement: (0) present; (1) absent. Parviraptor (0), Portugalophis (0), Diablophis (0).

45

Character descriptions adapted from recent matrix30 for ingroup analysis of snakes The following characters and character states have been adapted and re-evaluated from a previous data30 that was itself derived from a number of existing data sets31-36. Characters bearing an “*” were recoded for one of more taxa, and most importantly, and noted as such, based on our restudy of the specimens used to create the terminal taxon chimaera for Coniophis precedens given in a recent study30. We took a much more conservative approach to defining our “Coniophis” terminal taxon. For the purposes of this study we used the vertebral assemblage assigned to C. precedens30, and the right dentary (UCMP 50000) and left maxilla (UCMP 53935) as we agree that these elements belong to some kind of snake even if we question their assignment30 to C. precedens (a vertebral form taxon). It remains possible, and beyond the scope of this study, that these unassociated vertebrae and jaw fragments may represent a second or even third snake in the Lance Formation fauna similar to the multiple snake taxa known from the equivalent and nearby Hell Creek Formation fauna30. We also followed the more restrictive concept of Najash31 in coding that taxon as was recently suggested in a conservative revision of the materials assigned to the type of that taxon32. The Traditional Search analysis resulted in 3 trees of length 518 (Supplementary Fig. 24a, b). The main difference between this analysis and PAUP are the positions of Dinilsyia and Najash (sister taxa at base of tree above basal polytomy) and Coniophis (sister taxon to Scolecophidia). The Parviraptoridae were again placed in an unresolved basal polytomy with weak support in the Frequencey Distribution tree (Supplementary Fig. 24b) for a sister taxon relationship between aff. Parviraptor estesi and Diablophis. The result of the Drift search was three trees of length 515. The Strict consensus tree (Supplementary Fig. 24c) of this analysis has

46

the same topology as that of the PAUP analysis. The Frequency Distribution tree (Supplementary Fig. 24d) reconstructs a weakly supported resolution of the relationships of Eophis, aff. Parviraptor estesi, and Diablophis. We converted the language of the character descriptions given below into telegraphic language, and reorganized some sentences for both clarity and brevity. The scores for the five new fossil taxa added to this matrix are also given for each scoreable character. Dentition *1. Tooth implantation on dentary: pleurodont (0); Alethinophidian (1). This character is extremely problematic as “Alethinophidian”38 is not an anatomical feature and thus does not define a homolog statement of any recognizable kind. We interpret the condition of “Alethinophidian” following the recently defined histological characteristics defined39-41 for snake alveoli. We rescored Najash, Scolecophidia, and Coniophis as state “1” as none of these snakes show a pleurodont form of attachment – the original coding30 is contra its own claims that the “interdental” plates are definitive snake morphology and has the effect of forcing Coniophis, Najash and Scolecophidia to the base of the tree. We also note that this character is uninformative but have left it in the analysis for consistency with the earlier study30. Parviraptor, Portugalophis, Diablophis and Eophis were scored as “1” for this state based on the possession of alveoli forming “interdental” plates on three sides of the tooth (similar to Xenopeltis, Dinilysia, Python, etc. [Figs 1a–l; 2a-p; see Supplementary Figs 2, 7]). *2. Plicidentine: present (0); absent (1). This character is uninformative, but is retained in this analysis for consistency with an earlier study30. All snakes, including Parviraptor, Portugalophis, Diablophis and Eophis, are coded as absent following our observations and utilizing recent definitions and arguments on plicidentine39-41. *3. Maxillary and dentary teeth: relatively short conical, upright (0); robust, recurved (1); elongate needle-shaped, distinctly recurved (2). Parviraptor, Portugalophis, and Diablophis are the only parviraptorids that possess teeth and are coded as possessing state “2” (Figs 1a–l; 2a–r; Supplementary Figs 2d, e; 7). Scolecophdia was recoded as polymorphic (1&2) to accurately reflect the recurved teeth in the maxilla of Typhlops, and less elongate teeth in the dentary and maxilla of Liotyphlops. Our Coniophis code reflects the teeth as preserved in the maxilla (UCMP 53935) and dentary (UCMP 50000) that we consider to show snake characteristics. 4. Premaxillary dentition: present (0); absent (1). 5. Alveoli and base of teeth: not expanded transversely (0); wider transversely than anteroposteriorly (1). Parviraptor, Portugalophis, Diablophis and Eophis were scored as “1” for this state based on the assessment of both the alveoli as preserved, and the attached and isolated teeth (Figs 1a–l; 2a–r). 6. Pterygoid teeth: absent (0); present (1).

47

Skull 7. Premaxilla: broadly articulated with maxilla (0); loosely contacting maxilla (1). 8. Transverse processes of premaxilla: curved backwards (0); extending straight laterally or anterolaterally (1). 9. Nasal process of premaxilla: elongate, approaching or contacting frontals (0); short, divides nasals only at anterior margin or not at all (1). 10. Dorsal (horizontal) lamina of nasal: relatively broad anteriorly, with narrow gap between lateral margin and vertical flange of septomaxilla (0); dorsal lamina of nasal distinctly tapering anteriorly, leaving wide gap between lateral margin and vertical flange of septomaxilla (1). 11. Medial flanges of nasal, articulation with median frontal pillars: present (0); absent (1). 12. Anterior margin of nasals: restricted to posteromedial margins of nares (0); extend anteriorly toward tip of rostrum (1). 13. Lateral flanges of nasals: articulate with anterior margin of frontals (0); separated from frontals (1). 14. Posterolateral margin of nasal: contacts posteromedian margin of prefrontal (0); elements in contact along most of their length (1); contact between elements with interfingering of nasal and prefrontal margins (2); nasals do not contact prefrontals (3). 15. Septomaxilla posterior dorsal process of lateral vertical flange: absent (0); short (1); long (2). 16. Septomaxilla articulatioin with median frontal pillars: absent (0); present (1). 17. Ventral portion of posterior edge of lateral flange of septomaxilla and opening of Jacobsen’s organ: located at level of posterior edge or behind (0); distinctly in front (1). 18. Vomeronasal cupola: fenestrated medially (0); closed medially by a sutural contact of septomaxilla and vomer (1). 19. Septomaxilla: forms lateral margin of opening of Jacobson’s organ (0); vomer extends into posterior part of lateral margin, restricting septomaxilla to anterolateral part of lateral margin of opening of Jacobson’s organ (1). 20. Vomeronasal nerve: does not pierce vomer (0); exits vomer through single large foramen (1); through cluster of small foramina (2). 21. Posterior ventral (horizontal) lamina of vomer: long, parallel edged (0); short, tapering to pointed tip (1). 22. Posterior dorsal (vertical) lamina of vomer: well developed (0); reduced or absent (1). 23. Prefrontal: articulates with frontal laterally (0); anterolaterally (1). Only aff. Parviraptor estesi can be scored for this character, and even then, only referring to the preserved prefrontal facet on the frontal (Fig. 1b; Supplementary Fig. 2s, t). 24. Lateral margin of prefrontal: slanting anteroventrally (0); positioned vertically (1). 25. Lacrimal foramen on prefrontal: not completely enclosed (0); enclosed by prefrontal (1). 26. Lateral foot process of prefrontal: absent (0); contacts maxilla only (1); maxilla and palatine (2); palatine only (3). 27. Medial foot process of prefrontal: absent (0); present, low (1); present, high (2). 28. Anterior/lateral flange of prefrontal covering nasal gland and roofing auditus conchae: absent (0); present (1). 29. Ventral margin of lateral surface of prefrontal: articulates with dorsal surface of maxilla (0);

48

retains only posterior contact (1). 30. Dorsal lamina of prefrontal: contacts or forms overlapping contact with nasal posteromedially (0); remains separate from nasal (1). 31. Medial frontal pillars: absent (0), present (1). Though not figured, the type specimen of aff. Parviraptor estesi, NHMUK R8551, can be examined using microscopy and there is no apparent preservation of the median frontal pillars, nor any evidence they were present. The taxon is scored as “0” for this character (Figs 1b; 2s,r; Supplementary Fig. 3). 32. Transverse horizontal shelf of frontal: developed and broadly overlapped by nasals (0); poorly developed and never broadly overlapped by nasals (1); absent (2). 33. Lacrimal: present (0); absent (1). *34. Postfrontal: present (0); absent (1). We refer to a recent study [14) for clarity on identifications of the postfrontal in snakes. Of the new taxa described herein, only aff. Parviraptor estesi can be scored for this character, and even then, only referring to the preserved postfrontal facet on the frontal (Figs 1b; 2s, t). *35. Postorbital (JUGAL): present (0); absent (1). We refer to a recent study39 for clarity on identifications of the postorbital in snakes, and treat the codings for this character as recognizing the presence of the jugal, not the postorbital. This is a non-trivial difference, but is consistent with all tests of similarity for the identification of this element. *36. Ventral tip of postorbital (JUGAL): remains separated by wide gap from ectopterygoid (0); contacts or closely approaches ectopterygoid, forming almost complete posterior margin of orbit (1). Following the clarity of recent argumentation and Tests of Similarity42, we consider the “postorbital” in this character to be the JUGAL, and consider this character to further refine and support previous arguments42. 37. Dorsal head of postorbital: fuses or articulates with posterodorsal surface of postfrontal (0); articulates with parietal (1). 38. Parietal: without lateral wings meeting postorbital bones (0); with lateral wings meeting postorbital bones (1). 39. Distinct lateral ridge of parietal: extending posteriorly from anterior lateral wing up to prootic: absent (0); present (1). 40. Frontoparietal suture: relatively straight (0); frontoparietal suture U-shaped (1). 41. Parietal margin of optic foramen: straight (0); concave (1). 42. Lateral margins of braincase open anterior to prootic (0); descending lateral processes of parietal enclose braincase (1). 43. Supratemporal processes of parietal: distinctly developed (0); not distinctly developed (1). 44. Parietal enters anterior aspect of base of basipterygoid process: absent (0); present (1). 45. Contact between parietal and supraoccipital: V-shaped with apex pointing anteriorly (0); straight transverse line (1); V-shaped with apex pointing posteriorly (2). 46. Ascending process of maxilla: tall, extending to dorsal margin of prefrontal (0); short (1); absent (2). Parviraptor, Portugalophis, and Diablophis were scored as “1” for this character (Figs 1a, c, d, g, h; 2a–c; Supplementary Figs 2; 7; 9–11). 47. Small horizontal shelf on medial surface of anterior end of maxilla: present (0); absent (1). Only Parviraptor and Portugalophis were scored as “0” for this character (Figs 1c, i; 2a, c; Supplementary Figs 2; 7; 9–11), and share this feature with the purported Coniophis maxilla (UCMP 53935).

49

48. Posterior end of maxilla: does not project beyond posterior margin of orbit (0); projects moderately beyond posterior margin of orbit (1); projects distinctly beyond posterior margin of orbit, with broad flat surface (2). Only Parviraptor and Portugalophis can be confidently scored as “1” for this character as the maxilla of Diablophis is broken posterior to the palatine process (Figs 1a,g; 2a,c). 49. Medial (palatine) process of maxilla: located in front of orbit (0); located below orbit (1). Parviraptor, Portugalophis, and Diablophis were scored as “0” for this character (Figs 1a,g; 2a,c; Supplementary Figs 2; 7; 9–11). 50. Medial (palatine) process of maxilla: pierced (0); not pierced (1). Parviraptor, Portugalophis, and Diablophis were scored as “1” for this character as there is no evidence in any of the parviraptorids of a foramen in the palatine process of the maxilla (Figs 1a,g; 2a,c; Supplementary Figs 2; 7; 9–11). 51. Anterior end of supratemporal: located behind or above posterior border of trigeminal foramen (0); anterior to posterior border of trigeminal foramen (1). 52. Supratemporal facet on opisthotic-exoccipital: flat (0); sculptured and delineated with projecting posterior rim that overhangs exoccipital (1). 53. Free-ending posterior process of supratemporal: absent (0); present (1). 54. Supratemporal: present (0); absent (1). 55. Anterior dentigerous process of palatine: absent (0); present (1). 56. Medial (choanal) process of palatine: forms extensive concave surface dorsal to ductus nasopharingeus (0); narrows abruptly to form curved finger-like process (1); forms short horizontal lamina that does not reach vomer (2). 57. Choanal process of palatine: without expanded anterior flange articulating with vomer (0); with anterior flange (1). 58. Pterygoid contacts palatine: complex and finger-like articulations (0); tongue-in-groove joint (1); reduced to flap-overlap (2). 59. Palatine contact with ectopterygoid: present (0); absent (1). 60. Dentigerous process of palatine contact with vomer and/or septomaxilla posterolateral to opening for Jacobson’s organ: present (0); absent (1). 61. Maxillary process of palatine: anterior to posterior end of palatine (0); at posterior end of palatine (1). 62. Lateral (maxillary) process of palatine and maxilla: in well-defined articulation (0); loosely overlapping medial (palatine) process of maxilla, or absent (1). 63. Maxillary branch of trigeminal nerve: pierces lateral (maxillary) process of palatine (0); passes dorsally between palatine and prefrontal (1). 64. Vomerine (choanal) process of palatine: articulates broadly with posterior end of vomer (0); meets vomer in well-defined articular facet (1); touches or abuts vomer without articulation or remains separated from vomer (2). 65. Internal articulation of palatine with pterygoid: short (0); long (1). 66. Pterygoid tooth row: anterior to basipterygoid joint (0); tooth row reaches or passes level of basipterygoid joint (1). 67. Quadrate ramus of pterygoid: robust, rounded or triangular in cross-section, but without groove (0): blade-like and with distinct longitudinal groove for protractor pterygoidei (1). 68. Transverse (lateral) process of pterygoid: forms distinct, well-defined lateral projection (0);

50

gently curved lateral expansion of pterygoid, or absent (1). 69. Lateral edge of ectopterygoid straight (0); angulated at contact with maxilla (1). 70. Anterior end of ectopterygoid: restricted to posteromedial edge of maxilla (0); invades dorsal surface of maxilla (1). 71. Pterygoid attached to basicranium: by strong ligaments at palatobasal articulation (0); pterygoid free from basicranium in dried skulls (1). 72. Quadrate: slender (0); broad (1). 73. Quadrate: slanted clearly anteriorly, posterior tip of pterygoid dislocated anteriorly from mandibular condyle of quadrate (0); positioned slight anteriorly or vertically (cephalic condyle positioned behind or at same level of mandibular condyle) (1); slanted posteriorly (cephalic condyle positioned in front of mandibular condyle) (2). 74. Cephalic condyle of quadrate: elaborated into posteriorly projecting suprastapedial process (0); suprastapedial process absent or vestigial (1). 75. Stapedial footplate: broad and massive (0); narrow and thin (1). 76. Stylohyal: not fused to quadrate (0); fuses to posterior tip of suprastapedial process (1); fuses to ventral aspect of reduced suprastapedial process (2); stylohyal fuses to quadrate shaft (3). 77. Stapedial shaft: straight (0); angulated (1). 78. Stapedial shaft: slender and longer than diameter of stapedial foot-plate (0); thick, and equal to, or shorter than diameter of stapedial foot-plate (1). 79. Paroccipital process of otooccipital: well developed and laterally projected (0); reduced to short projection or absent (1). 80. Juxtastapedial recess defined by crista circumfenestralis: absent (0), present but open posteriorly (1); present and closed posteriorly (2). 81. Crista circumfenestralis: exposes most of stapedial footplate (0); converges upon stapedial footplate (1). 82. Crista interfenestralis: does not form individualized component in ventral rim of crista circumfenestralis (0); does form individualized component in ventral rim of crista circumfenestralis (1). 83. Jugular foramen: exposed in lateral view by crista tuberalis (0); concealed in lateral view by crista tuberalis (1). 84. Otooccipitals: do not contact each other dorsally (0); contact each other dorsally (1). 85. Otooccipital posterolateral processes: short and narrow, do not extend toward posterior margin of occipital condyle (0); wider than condyle and long, combine with crista tuberalis to extend to approximate posterior margin of occipital condyle (1). 86. Supraoccipital contact with prootic: with narrow (0); broad (1). 87. Prootic exclusion of parietal from trigeminal foramen: absent (0); present (1). 88. Laterosphenoid: absent (0): present (1). We note that identification of a “laterosphenoid” for Haasiophis and Eupodophis have recently been contested and revised, and are now recognized as ectopterygoids, not broken flanges underlying the laterosphenoids43. We have therefore rescored these two states as “?”, not present “1”, for these two taxa. 89. Prootic ledge underlap of posterior trigeminal foramen: absent (0); present (1). 90. Prootic: exposed in dorsal view medial to supratemporal or to supratemporal process of parietal (0); fully concealed by supratemporal or parietal in dorsal view (1). 91. Exit hyomandibular branch of facial nerve inside opening for mandibular branch of

51

trigeminal nerve: absent (0); present (1). 92. Vidian canal: does not open intracranially (0); open intracranially (1). 93. Anterior opening of Vidian canal: single (0); divided (1). 94. Sella turcica: bordered posteriorly by well-developed dorsum sellae (0); dorsum sellae low (1); dorsum sellae not developed, sella turcica with shallow posterior margin (2). 95. ‘Lateral wings of basisphenoid’: absent (0); present (1). 96. Ventral surface of basisphenoid: smooth (0); with weakly developed sagittal crest from which protractor pterygoidei originates (1); with strongly projecting sagittal crest (2). 97. Basioccipital: contributes to ventral margin of foramen magnum (0); basioccipital excluded by medial contact of otooccipitals (1). 98. Basisphenoid-basioccipital suture: smooth (0); transversely crested (1). 99. Basipterygoid (= basitrabecular) processes: present (0); absent (1). 100. Crista trabeculares: short and or indistinct (0); elongate and distinct in lateral view (1). 101. Cultriform process of parabasisphenoid: does not extend anteriorly to approach posterior margin of choanae (0); approaches posterior margin of vomer (1). 102. Parabasisphenoidal rostrum behind optic foramen: narrow (0); broad (1). 103. Parabasisphenoid rostroventral surface: flat or broadly convex (0); concave (1). 104. Basioccipital meets parabasisphenoid: suture located at level of fenestra ovalis (0); located at or behind trigeminal foramen (1). 105. Parasphenoid rostrum interchoanal process: absent (0); broad (1); narrow (2). MANDIBLE 106. Anteromedial margin of dentaries: symphyseal articular facet (0); no symphyseal facet (1). Only Eophis is confidently scored for this character, “1”, as the remaining new taxa described herein do not preserve this detail on the identified elements (Figs 1k; 2g; Supplementary Fig. 5a, b). 107. Posterior dentigerous process of dentary: absent (0); present, short (1); present, long (2). 108. Medial margin of adductor fossa: relatively low and smoothly rounded (0); forms distinct dorsally projecting crest (1). 109. Mental foramina on lateral surface of dentary: two or more (0); one (1). Only Portugalophis, Diablophis and Eophis can be scored for this character, “0” and they all large, and two or more mental formina (Figs 1e, j, l; Supplementary Figs 5c, f, i; 7g, h; 8c; 10g, h). 110. Coronoid process of coronoid bone: high, tapering distally (0); high, with rectangular shape (1); low, not exceeding significantly coronoid process of compound bone (2). 111. Coronoid bone: present (0); absent (1). 112. Posteroventral process of coronoid: present (0); absent (1). 113. Coronoid process on lower jaw: formed by coronoid bone only (0); or by coronoid and compound bone (1); or by compound bone only (2) (i.e. coronoid absent). 114. Posdentary elements: presence of separate elements (0); fusion of surangular /articular into compound bone (1). VERTEBRAE *115. Chevrons: present (0); absent (1). We have recoded Wonambi, Eupodophis and Haasiophis

52

as possessing distinct chevron bones43. *116. Hemapophyses: absent (0); present (1). We have recoded Wonambi, Eupodophis and Haasiophis as possessing distinct hemapophyses43. 117. Hypapophyses: restricted to anterior-most precloacal vertebrae (0); present throughout precloacal skeleton (1). 118. Para-diapophysis: confluent (0); separated into dorsal and ventral facet (1). Diablophis and aff. Parviraptor estesi are coded as “1” for this character (Fig. 3a,b; Supplementary Fig. 4a–g). 119. Prezygapophyseal accessory processes: absent (0); present (1). Diablophis and aff. Parviraptor estesi are coded as “0” for this character (Fig. 3a,b; Supplementary Fig. 4a– g). 120. Subcentral paralymphatic fossae on posterior precloacal vertebrae: absent (0); present (1). Diablophis and aff. Parviraptor estesi are coded as “0” for this character (Fig. 3a,b; Supplementary Fig. 4a–g). 121. Subcentral foramina: absent (0); present, consistently small (1); present, of variable size (2). Diablophis and aff. Parviraptor estesi are coded as “0” for this character (Fig. 3a,b; Supplementary Fig. 4a–g). 122. Well-developed, consistently distributed paracotylar foramina: absent (0); present (1). 123. Ventral margin of centra: smooth (0); median prominence from cotyle to condyle (1). Diablophis and aff. Parviraptor estesi are coded as “0” for this character (Fig. 3a,b; Supplementary Fig. 4a–g). 124. Axis intercentrum: not fused to anterior region of axis centrum (0); fused (1). 125. Neural spine height: well-developed process (0); low ridge or absent (1). Diablophis and aff. Parviraptor estesi are coded as “0” for this character (Fig. 3a,b; Supplementary Fig. 4a–g). 126. Posterior margin of neural arch: shallowly concave in dorsal view (0); with deep V-shaped embayment in dorsal view (1). Diablophis and aff. Parviraptor estesi are coded as “0” for this character (Fig. 3a,b; Supplementary Fig. 4a–g). 127. Cotyle shape of precloacal vertebrae: oval (0); circular (1). Diablophis and aff. Parviraptor estesi are coded as “1” for this character (Fig. 3a,b; Supplementary Fig. 4a–g). 128. Parazygantral foramen: absent (0); present (1). 129. Lymphapophyses: absent (0); present (1). 130. Lymphapophyses: three or fewer (0); three lymphapophyses and one forked rib (1); more than three lymphapophyses and one forked rib (2). 131. Sacral vertebrae: present (0); absent (1). 132. Position of synapophyses in relation to lateral edge of prezygapophyses: at same level or slightly more projected laterally (0); clearly medial to edge of prezygapophyses (1). Diablophis and aff. Parviraptor estesi are coded as “1” for this character Fig. 3a,b; Supplementary Fig. 4a–g). 133. Pachyostotic vertebrae: absent (0); present (1). Diablophis and aff. Parviraptor estesi are coded as “0” for this character (Fig. 3a,b; Supplementary Fig. 4a–g). 134. Precloacal vertebrae number: fewer than 100 (0); more than 100 (1). 135. Caudal vertebrae number: greater than 50% of precloacal number (0); approximately 10% or less than precloacal number (1).

53

136. Tuber costae absent from ribs (0), tuber costae present (1). HINDLIMBS 137. Pectoral girdle and forelimbs: present (0); absent (1). 138. Tibia, fibula, and hind foot: present (0); absent (1). 139. Trocanter externus: present (0); absent (1). 140. Pelvis: external to sacral-cloacal ribs (0); internal to sacral-cloacal ribs (1). 141. Ilium and pubis length: ilium longer than pubis (0); ilium and pubis of same size (1); pubis much longer than ilium (2). 142. Pelvic elements: with strongly sutured contact (0); with weak (cartilaginous) contact (1); fused together (2). 143. Pelvic elements: present (0); absent (1). NEW CHARACTERS (1) 144. Medial vertical flanges of nasals: absent (0); present (1). 145. Preorbital ridge: dorsally exposed (0); overlapped by prefrontal (1). 146. Lateral foot process of prefrontal: articulates with lateral edge of maxilla via thin anteroposteriorly directed lamina (0); articulates with maxilla via large contact that runs from lateral to medial dorsal surface of maxilla (1). 147. Medial finger-like process of ectopterygoid articulating with medial surface of maxilla: present (0); absent (1). 148. Posterolateral corners of basisphenoid: strongly ventrolaterally projected (0); not projected (1). 149. Basioccipital: expanded laterally to form floor of recessus scalae tympani (0); excluded from floor of recessus scalae tympani by otooccipital (1). 150. Frontal subolfactory process: absent or present as simple horizontal lamina (0); present and closing tractus olfactorius medially (1). In anterior view, the type specimen of aff. Parviraptor estesi, NHMUK R8551, can be observed as possessing state “1” as the subolfactory processes angle sharply to the midline of the element. The taxon is scored as “1” for this character (Fig. 1b; Supplementary Figs 2s-r; 3). 151. Ectopterygoid contact with pterygoid: restricted to transverse (lateral) process of pterygoid (0); contact expanded significantly on dorsal surface of pterygoid body (1). 152. Maxillary process of palatine: main element bridging contact with maxilla and palatine in ventral view (0); covered ventrally by expanded palatine process of maxilla (1). 153. Coronoid bone contributes to anterior margin of adductor fossa: present (0); absent (1). 154. Coronoid bone: sits mostly on dorsal and dorsomedial surfaces of compound bone, being exposed in both lateral and medial views of mandible (0); applied to medial surface of compound bone (1). TEETH 155. Teeth, implantation: interdental ridges absent (0): interdental ridges present (1). This character is clearly redundant with Character 1 but is retained here for consistency with the study from which this data set was derived and is being compared to30 Parviraptor, Portugalophis, Diablophis and Eophis were scored as “1” for this state based on the

54

possession of alveoli forming “interdental” plates on three sides of the tooth (similar to Xenopeltis, Dinilysia, Python, etc. [Figs 1a–l; 2a-p; Supplementary Figs 2, S7]). 156. Teeth, replacement: replacement teeth lie vertically (0); lie horizontally in jaws (1). 157. Teeth, replacement: single replacement tooth per tooth position (0); two or more replacement teeth per tooth position (1). 158. Teeth, attachment: ankylosed to jaws (0) teeth loosely attached by connective tissue (1). This character is clearly redundant with Character 1 but is retained here for consistency with the study from which this data set was derived and is being compared to30. Many extant snakes possess hinged teeth, i.e., the teeth are attached by uncalcified periodontal ligaments to the margins of the alveoli, and as such this state overlaps with and is not independent of, Character 1, the “Alethinophidian” attachment condition. Parviraptor, Portugalophis, and Diablophis were scored as “0” for this state based on the possession of alveoli forming “interdental” plates on three sides of the tooth (similar Dinilysia, Python, etc. [Figs 1a–l; 2a–p; Supplementary Figs 2, 7]). *159. Teeth, size: crowns isodont or enlarged at middle of tooth row (0) crowns large anteriorly, and decrease in size posteriorly (1); anterior teeth conspicuously elongate, length of crown significantly exceeds height of dentary at midlength (2). We recoded our combination for “Coniophis precedens” for this character as “?” as the anterior tip of the dentary is missing and the character cannot be scored. Among the parviraptorids, only Portugalophis, was scored for this character with a state “0” [Figs 1i, j; 2k; Supplementary Figs 7, 8). SKULL 160. Premaxilla: teeth borne medially on premaxilla (0); teeth absent from midline of premaxilla (1). 161. Premaxilla: ascending process transversely expanded, partly roofing external nares (0); ascending process mediolaterally compressed, blade-like or spine-like (1). *162. Premaxilla: premaxilla medial to maxillae (0); located anterior to maxillae (1). We recoded our combination for “Coniophis precedens” for this character as “?” as the anterior tip of the maxilla is missing and the character cannot be scored. 163. Prefrontal: prefrontal socket for dorsal peg of maxilla absent (0); present (1). 164. Prefrontal extends medially across frontal for more than 75% of width of frontal: absent (0); present (1). 166. Frontal: nasal processes of frontal project between nasals (0); nasal processes absent (1). 167. Frontals: frontals taper anteriorly, distinct interorbital constriction (0); frontals broad anteriorly, interorbital region broad (1). 168. Frontal: subolfactory process abuts prefrontal in immobile articulation (0); subolfactory process articulates with prefrontal in mobile joint (1); subolfactory process with distinct lateral peg or process that clasped dorsally and ventrally by prefrontal (2). 169. Frontals and parietals: do not contact ventrally (0); descending wings of frontals and parietals contact ventrally to enclose optic foramen (1). 170. Parietal, sagittal crest: absent (0); present posteriorly but not anteriorly, and extending for no more than 50% of parietal midline length (1); present anteriorly and posteriorly, and extending more than 50% of parietal midline length (2).

55

171. Parietal: narrow (0); inflated (1). 173. Skull, postorbital region relative length: short, less than half (0); elongate, half or more (1). 174. Supraoccipital region of skull: nuchal crests absent (0); present (1). 175. Supratemporal: supratemporal short, does not extend posterior to paroccipital process (0); elongate, extending well beyond paroccipital process (1). 176. Maxilla: palatine process short, weakly developed (1); palatine process long, strongly projecting medially (1). Parviraptor, Portugalophis, and Diablophis were scored as “1” for this character (Figs 1a,c,d,g,h; 2a-c; Supplementary Figs 2; 7; 9–11). 177. Maxilla: facial process (ascending process as described in this study) projects up strongly, caudal margin inclined steeply relative to maxilla (0); facial process weakly projecting, caudal margin of facial process lies at angle of 30º to horizontal or less (1). Parviraptor and Portugalophis were scored as “1” for this character (Figs 1a,c,d,g,h; 2a-c; Supplementary Figs 2; 7; 9–11). NOTE: this character strongly overlaps on Character 46, but is retained here for consistency with that previous study30. 178. Maxilla, premaxillary process: medial projection articulating with vomers present (0); premaxillary process does not contact vomers (1). Parviraptor and Portugalophis show a similar morphology to the condition as observed in the maxilla (UCMP 53935) assigned to Coniophis30 and scored in that data matrix. While we agree that the morphology of these three elements is similar, none of them are preserved as intact skulls demonstrating empirically observable articulations. Many extant snakes (e.g., Xenopeltis) possess medial processes of the maxilla in the same position as those observed in Parviraptor, Portugalophis and Coniophis, which articulate with no other elements, though they approach the septomaxilla or premaxilla. Only in lizards, does this process contact the vomer. We have elected to score this character as “?” for these three taxa as we cannot confirm or discount a vomer contact (Figs 1a,c,d,g,h; 2a-c; Supplementary Figs 2; 7; 9–11). 179. Maxilla, number of mental foramina: 5 or more (0); 4 or fewer (1). Portugalophis shows state “0”, while Diablophis shows at least state “1” with 4 preserved foramina. However, we have elected code Diablophis as “?” because the posterior portion of the maxilla is missing (Figs 1d, h; Supplementary Figs 7c,d; 10c,d). *180. Maxilla, supradental shelf development: extending full length of maxilla (0); reduced anterior to palatine process (1). Parviraptor, Portugalophis, and Diablophis were scored as “0” for this character (Figs 1a,c,d,g,h; 2a-c; Supplementary Figs 2; 7; 9–11). Coniophis was rescored as “?” because UCMP 53935 preserves only the anterior portion of the maxilla. *181. Maxilla: medial surface of facial process with distinct naso-lacrimal recess demarcated dorsally by anteroventrally trending ridge: (0) present; (1) absent. Parviraptor, Portugalophis, and Diablophis were scored as “1” for this character (Figs 1a,c,d,g,h; 2a-c; Supplementary Figs 2; 7; 9–11). As UCMP 53935 does not preserve the ascending/facial process, this character was scored as “?” for Coniophis. *182. Maxilla: medial surface of facial process with well-defined fossa for lateral recess of nasal capsule: present (0); reduced and present as small fossa on back of facial process (1); absent, fossa for lateral recess developed entirely on prefrontal (2). Parviraptor, Portugalophis, and Diablophis were scored as “2” for this character (Figs 1a,c,d,g,h; 2a-c; Supplementary Figs 2; 7; 9–11). As UCMP 53935 does not preserve the ascending/facial process, this

56

character was scored as “?” for Coniophis. *183. Maxilla: extensive contact of dorsal margin of maxilla with nasal (0); nasal-maxilla contact lost (1). As UCMP 53935 does not preserve the ascending/facial process, this character was scored as “?” for Coniophis. *184. Maxilla: maxilla overlaps prefrontal laterally in tight sutural connection (0); overlap reduced, mobile articulation (1). As UCMP 53935 does not preserve the ascending/facial process, this character was scored as “?” for Coniophis. *185. Maxilla: excluded from anteroventral margin of orbit by jugal (0); maxilla forms anteroventral margin of orbit (1). Coniophis was rescored as “?” because UCMP 53935 preserves only the anterior portion of the maxilla. *186. Maxilla: palatine process of maxilla projects medially (0); palatine process of maxilla downturned (1). Parviraptor, Portugalophis, and Diablophis were scored as “0” for this character (Figs 1a,c,d,g,h; 2a-c; Supplementary Figs 2; 7; 9–11). Coniophis was rescored as “?” because UCMP 53935 preserves only the anteriormost portion of the maxilla, exclusive of the palatine process. *187. Maxilla: superior alveolar foramen: positioned near middle of palatine process, opening posterodorsally (0); positioned near anterior margin of palatine process, opening medially (1). Parviraptor, Portugalophis, and Diablophis were scored as “-” for this character (Figs 1a,c,d,g,h; 2a-c; Supplementary Figs 2; 7; 9–11). Coniophis was rescored as “?” because UCMP 53935 preserves only the anteriormost portion of the maxilla, exclusive of the palatine process. *188. Maxilla, accessory foramen posterior to palatine process: absent (0); present (1). Only Portugalophis was scored for this character, with the state assignment being “0”. Coniophis was rescored as “?” because UCMP 53935 preserves only the anteriormost portion of the maxilla, exclusive of the palatine process. *189. Maxilla, ectopterygoid process: absent (0); present (1). Coniophis was rescored as “?” because UCMP 53935 preserves only the anterior portion of the maxilla. *190. Maxilla: articulates with distally expanded postorbital element to form complete postorbital bar: present (0); absent (1). Coniophis was rescored as “?” because UCMP 53935 preserves only the anterior portion of the maxilla. *191. Maxilla: 15 or more maxillary teeth (0); fewer than 15 maxillary teeth (1). Coniophis was rescored as “?” because UCMP 53935 is incomplete. 192. Postfrontal: anterior and posterior processes clasping frontals and parietals (0); anterior and posterior processes present, but postfrontal abuts frontals and parietals (1); anterior and posterior processes absent (2). 193. Supratemporal: free caudal end of supratemporal projects posteroventrally (0); posteriorly or posterodorsally (1). 194. Quadrate, lateral conch: present (0); absent (1). 195. Quadrate, maximum length relative to proximal width: quadrate elongate, maximum length at least 125% of maximum width of quadrate head (0); quadrate short, length less than 125% of width of quadrate head (1). 196. Quadrate, proximal end platelike: absent (0); present (1). 197. Palatine, palatine teeth small relative to lateral teeth (0); or enlarged, palatine teeth at least half diameter of posterior maxillary teeth (1).

57

198. Palatine, elongate lateral process projecting to lateral edge of orbit to articulate with caudal margin of prefrontal: absent (0); present (1). 199. Epipterygoid: present (0); absent (1). 200. Ectopterygoid: clasps pterygoid anteromedially (0); ectopteryoid overlaps pterygoid (1); ectopterygoid abuts pterygoid medially (2). 201. Vidian canals: enclosed in sphenoid (0); open intracranially (1). 202. Vidian canals: posterior openings symmetrical (0); asymmetrical (1). 203. Exoccipitals: separated ventral to foramen magnum (0); contact below foramen magnum (1). 204. Exoccipital-opisthotic: horizontal, winglike crista tuberalis absent (0); present (1). 205. Otoccipitals: do not project posteriorly to level of occipital condyle (0); project posteriorly to conceal occipital condyle in dorsal view (1). 206. Sclerotic ring: present (0); absent (1). MANDIBLE *207. Dentary, enlarged mental foramen: absent (0); present (1). The original coding (1) had this character coded as polymorphic. Our observations of scolecophidian snakes find that all mental foramina present are in fact very large in relationship to the size of the dentary. We recoded the character to only state “1”. Portugalophis, Diablophis and Eophis were scored as “1” for this character (Figs 1e, j, l; Supplementary Figs 5c,f,i; 7g,h; 8c; 10g,h). *208. Dentary, depth of Meckelian groove anteriorly: deep slot (0); shallow sulcus (1). We recoded our combination for “Coniophis precedens” for this character as “1” as the Meckelian groove is not deep. Portugalophis, Diablophis and Eophis were scored as “0” for this character (Figs 1f, i, k; 2g–k; Supplementary Figs 5a–i; 7i, j; 10e, f). *209. Dentary, angular process shape: posteroventral margin of dentary angular process weakly wrapped around underside of jaw (0); dentary angular process projects more nearly horizontally to wrap beneath jaw (1). We recoded our combination for “Coniophis precedens” for this character as “?” as the posterior portion of the dentary, key to the character description, is missing30 and is not comparable to Dinilysia or any other snake. Portugalophis, Diablophis and Eophis were scored as “1” for this character (Figs 1e, j, l; 2g–k; Supplementary Figs 5a–i; 7i, j; 10e, f). 210. Dentary, angular process length relative to coronoid process: angular process disinctly shorter than coronoid process, former terminating well anterior to latter (0); subequal in length posteriorly (1). 211. Dentary, symphysis: weakly projecting medially (0); hooked inward and strongly projecting medially (1). This character is redundant with Character 106, however for consistencies sake it is retained in this analysis. Only Eophis is scored for this character with state “0”. *212. Dentary, ventral margin: unexpanded, medial margin of dentary straight in ventral view (0); expanded, medial margin crescentic in ventral view (1). We recoded our combination for “Coniophis precedens” for this character as “?” as the posterior portion of the dentary, key to the character description, is missing. Portugalophis, Diablophis and Eophis were scored as “0” for this character (Figs 1f, i, k; 2g–k; Supplementary Figs 5a–i; 7i, j; 10e, f). 213. Dentary, coronoid process: wraps around surangular laterally and medially (0); broad and sits atop surangular (1).

58

*214. Dentary, coronoid process with slot for medial tab of surangular: absent (0) or present (1). We recoded our combination for “Coniophis precedens” for this character as “?” as the posterior portion of the dentary, key to the character description, is missing. *215. Dentary, subdental shelf: present along entire tooth row (0); present only along posterior portion of tooth row (1); absent (2). We completely rewrote the original characters 215 and 216 as recently presented29, condensing them into our Character 215. Portugalophis, Diablophis and Eophis were scored as “0” for this character (Figs 1f, i, k; 2g–k; Supplementary Figs 5a–i; 7i, j; 10e, f). 216. Dentary, enlarged mental foramen position: near tip of dentary (0); displaced from tip of jaw (1); displaced further to lie halfway between symphysis and surangular notch (2). 217. Surangular, dentary process with distinct triradiate cross-section: absent (0); present (1). 218. Surangular, adductor fossa: small (0); extended caudally towards jaw articulation (1). 219. Surangular: ventrolateral surface of surangular bearing distinct crest for attachment of adductor muscles: absent (0); present (1). 220. Coronoid, lateral overlap of coronoid onto dentary: absent (0); present (1). 221. Splenial attachment to dentary above Meckel’s canal: close throughout length (0); loose, with dorsal dentary suture confined to posterodorsal corner of splenial (1); (2) contact with subdental shelf reduced to small spur of bone or contact lost entirely. 222. Splenial – angular articulation: splenial overlaps angular (0); splenial abuts against angular to form hinge joint (1). 223. Splenial, size: splenial elongate, extends more than half distance from angular to dentary symphysis (0); splenial short, extends less than half distance from angular to symphysis (1). 224. Splenial, anterior mylohyoid foramen: present (0); absent (1). 225. Angular, lateral exposure (with coronoid region pointing dorsally): angular broadly exposed laterally along length (0); angular narrowly exposed laterally (1). 226. Angular, length posteriorly relative to glenoid (quadrate articulation): relatively long, extends more than half distance from anterior end of angular to glenoid; (0) relatively short, half or less of distance to glenoid (1); very short, one third or less of distance to glenoid (2). 227. Surangular, enlarged anterior surangular foramen: absent (0); or present (1). 228. Coronoid eminence: (0) well-developed; (1) weakly developed or absent. 229. Glenoid, shape: quadrate cotyle shallow (0), anteroposteriorly concave and transversely arched, ‘saddle shaped’ (1). While we recognize a probable surangular bone amongst the Diablophis materials, we are conservative in scoring that element here for its contribution to the glenoid shape as the element does not appear to be a compound bone, but an isolated surangular. As such it either represents an early stage in snake compound bone evolution, or, it is not part of the Diablophis taxon and belongs to a contemporary lizard (Supplementary Fig. 10i–k). 230. Retroarticular process: retroarticular process elongate (0) or shortened (1). *231. Ventral projections (pedicles) of anterior precloacals: short, about 50% length of centrum (0); long, subequal to or longer than centrum (1). We note that the “pedicles” [4] are not homologs of intercentra, but rather are the hypapophyses and potentially the fused intercentra (pedicles) as represented by snakes such as Pachyrhachis, Dinilysia, Eupodophis, and Haasiophis43 where the intercentra are not fused to the hypapophyses. 232. Vertebrae, ridgelike or bladelike ventral keels developed posterior to pedicles: (0) absent;

59

(1) present. Diablophis and aff. Parviraptor estesi were scored as “0” for this character (Fig. 3a, b; Supplementary Fig. 4a–g). 233. Vertebrae, dorsolateral ridges of neural arch: (0) absent; (1) present. Diablophis and aff. Parviraptor estesi were scored as “0” for this character (Fig. 3a, b; Supplementary Fig. 4a–g). *234. Vertebrae, vertebral centrum: narrow in ventral view (0); broad and subtriangular in shape (1); broad and square (2). We added a character state “2” to this character in order to recognize a primary homology between pachyophiid/simoliophiid snakes concerning centrum shape and pachyostosis. Diablophis and aff. Parviraptor estesi were scored as “0” for this character Fig. 3a, b; Supplementary Fig. 4a–g). 235. Vertebrae, arterial grooves: absent in neural arch (0); present (1). The vertebrae of both aff. Parviraptor estesi and Diablophis appear to show this snake synapomorphy, but only Diablophis was assigned state “1” as the aff. Parviraptor estesi material is not well enough prepared. The “trefoil”29 condition of the neural canal is easily observed in Diablophis. 236. Vertebrae, posterior condyle: confluent with centrum ventrally (0); distinctly separated from centrum by groove/constriction between centrum and condyle (1). Diablophis and aff. Parviraptor estesi were scored as “1” for this character (Fig. 3a, b; Supplementary Fig. 4a–g). 237. Vertebrae: narrow, width across zygapophyses not significantly greater than distance from prezygapophyses to postzygapophyses (0); vertebrae wide, width across zygapophyses 150% of length or more (1). Diablophis and aff. Parviraptor estesi were scored as “0” for this character (Fig. 3a, b; Supplementary Fig. 4a–g).

60

Supplementary References 1. 2.

3. 4.

5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Evans S. E. A new anguimorph lizard from the Jurassic and Lower Cretaceous of England. Palaeontology 37, 33–49 (1994). Estes R., de Quieroz K., Gauthier J. Phylogenetic relationships within Squamata. In Phylogenetic Relationships of the Lizard Families: Essays Commemorating Charles L. Camp (eds R Estes, G Pregill), pp. 119–281. Stanford, U.S.A.: Stanford University Press (1988). Conrad J., Norell M. High-resolution X-ray computed tomography of an Early Cretaceous gekkonomorph (Squamata) from Öösh (Övörkhangai; Mongolia). Hist. Biol. 18, 405–431 (2006). Evans S. E., Milner A. Middle Jurassic microvertebrate assemblages from the British Isles. In In the Shadow of the Dinosaurs: Early Mesozoic Tetrapods (eds NC Fraser, H-D Sues), pp. 303–321. Cambridge, UK: Cambridge University Press (1994). Evans S. E., Milner A. R., Mussett F. The earliest known Salamanders (Amphibia, Caudata): A record from the Middle Jurassic of England. Geobios 21, 539–552 (1988). Evans S. E., Milner A. R., Mussett F. A Discoglossid frog from the Middle Jurassic of England. Paleontology 33, 299–311 (1990). Gardner J. D., Evans S. E., Sigogneau-Russell D. New albanerpetontid amphibians from the Early Cretaceous of Morocco and Middle Jurassic of England. Acta Palaeontol. Polonica 48, 301–319 (2003). Scheyer T. M., Anquetin J. Bone histology of the Middle Jurassic turtle shell remains from Kirtlington, Oxfordshire, England. Lethaia 41, 85–96 (2008). Evans S. E. The skull of Cteniogenys, a choristodere (Reptilia: Archosauromorpha) from the Middle Jurassic of Oxfordshire. Zoo. J. Linn. Soc. 99, 205–237 (1990). Evans S. E. The postcranial skeleton of the choristodere Cteniogenys (Reptilia: Diapsida) from the Middle Jurassic of England. Geobios 24, 187–199 (1991). E. F. Freeman, Mammal Teeth from the Forest Marble (Middle Jurassic) of Oxfordshire, England. Science 194, 1053–1055 (1976). K. A. Kermack, D. M. Kermack, P. M. Lees, J. R. E. Mills, New multituberculate-like teeth from the Middle Jurassic of England. Acta Palaeontologica Polonica 43, 581–606 (1995). K. A. Kermack, A. J. Lee, P. M. Lees, F. Mussett, A new docodont from the Forest Marble. Zoological Journal of the Linnean Society 89, 1–39 (1987). D. Sigogneau-Russell, Holotherian mammals from the Forest Marble (Middle Jurassic of England). Geodiversitas 25, 501–537 (2003). S. E. Evans, A new lizard-like reptile (Diapsida: Lepidosauromorpha) from the Middle Jurassic of England. Zoo. J. Linn. Soc. 103, 391–412 (1991). E. F. Freeman, Fossil bone recovery from sediment residues by the 'interfacial method'. Palaeontology 25, 471–484 (1981). L. Winchester, A. d. A. Bellairs, Aspects of vertebral development in lizards and snakes. Journal of Zoology 181, 495–525 (1977). J. A. Maisano, Terminal fusions of skeletal elements as indicators of maturity in squamates. Journal of Vertebrate Paleontology 22, 268–275 (2002). A. G. Kluge, Cladistic relationhips in the Gekkonoidea (Squamata, Sauria). Miscellaneous Publications of the Museum of Zoology, University of Michigan 173, 1–54 (1987). N. Vidal, S. B. Hedges, The molecular evolutionary tree of lizards, snakes, and amphisbaenians. Comptes Rendus Biologies 332, 129–139 (2009). J. J. Wiens et al., Resolving the phylogeny of lizards and snakes (Squamata) with extensive sampling of genes and species. Biology Letters 8, 1043–1046 (2012). J. J. Wiens et al., Combining Phylogenomics and Fossils in Higher-Level Squamate Reptile Phylogeny: Molecular Data Change the Placement of Fossil Taxa. Systematic Biology 59, 684–688 (2010). R. A. Pyron, F. T. Burbrink, J. H. Wiens, A phylogeny and revised classification of Squamata, including 4161 species of lizards and snakes. BMC Evolutionary Biology 13, 1–53 (2013). R. Hoffstetter, J. P. Gasc, in Biology of the Reptilia, 1, C. Gans, T. S. Parsons, Eds., vol. 1, pp. 201–310 (1969).

61

25. 26. 27. 28. 29.

30. 31. 32.

33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43.

A. G. Kluge, Comparative osteology of the eublepharid lizard genus Coleonyx Gray. Journal of Morphology 110, 299–332 (1962). Cundall D., Irish F. The snake skull. Biology of the Reptilia 20, 349–692 (2008). List J. C. Comparative osteology of the snake families Typhlopidae and Leptotyphlopidae. Urbana: University of Illinois Press 36, 1–122 (1966). Rieppel O., Kley N. J., Maisano J. A. Morphology of the skull of the white-nosed blindsnake, Liotyphlops albirostris (Scolecophidia: Anomalepididae). J. Morph. 270, 536–557 (2009). Gauthier J. A., Kearney M., Maisano J. A., Rieppel O., Behlke A. D. B. Assembling the Squamate Tree of Life: Perspectives from the Phenotype and the Fossil Record. Bull. Peabody Mus. Nat. Hist. 53, 3–308 (2012). Longrich N. R., Bhullar B. –A. S., Gauthier J. A. A transitional snake from the Late Cretaceous period of North America. Nature 488, 205–208 (2012). Apesteguía S., Zaher H. A Cretaceous terrestrial snake with robust hindlimbs and a sacrum. Nature 440, 1037–1040 (2006). Palci, A., Caldwell, M. W., & Albino, A. M. Emended diagnosis and phylogenetic relationships of the Upper Cretaceous fossil snake Najash rionegrina Apesteguía and Zaher, 2006. J. Vertebr. Paleontol. 33, 131–140 (2013). Lee M. S. Y., Scanlon J. D. Snake phylogeny based on osteology, soft anatomy and ecology. Biological Reviews 77, 333–401 (2002). Scanlon J. D. Skull of the large non-macrostomatan snake Yurlunggur from the Australian Oligo-Miocene. Nature 439, 839–842 (2006). Wilson J. A., Mohabey D. M., Peters S. E., Head J. J. Predation upon hatchling dinosaurs by a new snake from the Late Cretaceous of India. PLoS Biol 8, e1000322 (2010). Swofford D. L. PAUP*: phylogenetic analysis using parsimony, version 4.0b10 Sunderland, Massachusetts: Sinauer Associates (2003). Goloboff P. A., Farris J. S., Nixon K. C. TNT, a free program for phylogenetic analysis. Cladistics 24, 774 (2008). Zaher H., Rieppel O. Tooth implantation and replacement in squamates, with special reference to mosasaur lizards and snakes. Amer. Mus. Nov. 3271, 1–19 (1999). Budney L. A., Caldwell M. W., Albino A. Tooth socket histology in the Cretaceous snake Dinilysia, with a review of amniote dental attachment tissues. Journal of Vertebrate Paleontology 26, 138–145 (2006). Maxwell E. E., Caldwell M. W., Lamoureux D. O. The structure and phylogenetic distribution of amniote plicidentine. Journal of Vertebrate Paleontology 31, 553–561 (2011). Kearney M., Rieppel O. An investigation into the occurrence of plicidentine in the teeth of squamate reptiles. Copeia 2006, 337-350 (2006). Palci A., Caldwell M. W. Primary homologies of the circumorbital bones of snakes. J. Morph. 274, 973– 986 (2013). Palci A., Caldwell M. W., Nydam R. L. Reevaluation of the anatomy of the Cenomanian (Upper Cretaceous) hind-limbed marine fossil snakes Pachyrachis, Haasiophis, and Eupodophis. J Vert. Paleontol. 33, 1328–1342 (2013).

Supplementary Data File: Data matrix for ingroup analysis of snakes. #NEXUS BEGIN DATA; DIMENSIONS NTAX=28 NCHAR=237; FORMAT DATATYPE = STANDARD GAP = - MISSING = ? SYMBOLS = " 0 1 2 3"; MATRIX Anguimorph_root 00001(0 1)0001?00(0 3)0?0000010000000(0 1)0000010000?00010000-000000000?00000000100010100000?100?0000?0000000000000000000000000000(0 1)(0 1)0000000?00(0 1)0000000000000000(0 1)0000?0000000000000000000000000000(0 1)0000000(0 1)00000000000000000100000-00000000000000000(0 1)000 Najash 1???1?????????????????????????????????0??1000?????00?0????????????0?????10 0???010?00?000000???00?000?110?0??0?????11010?1?101011100001?1?000110????? ??????1??????????????{1 2}00?0??????????????????110??????????100?10??20????1?????????0111100 Scolecophidia 11(1 2)1101110?1010??000??00???0000211(0 1)??000?11?11?(0 1)1?000(0 1)0(0 1)020---1(0 1)0?01000000111112100100000?0?02000010111(0 1)010000000110001120(0 1)(0 1)10(0 1)01(1 2)11011111(0 1)11101??1110(0 1)0(0 1)(0 1)10001100000111101010?0011112111000011--1(0 1)0--1-10000111?0(0 1)0102(0 2)000121110(0 2)10(0 1)11100100 Dinilysia 112?111??1?100??????0?0102100000100100100100110100000000?01?00000000000010 001101010100000001(0 1)100000?11100120?00001??011?(0 1)01?00(0 1)0???101?1???????001000000001???????0001120200100111111111101000011100010000-011010101011000111001110110111101 Sanajeh 111??????????????????????2??????1?????0??10?01020?00?0?0??1???1??????1???? ????01?1111100??0???12?10????1?12?1????1??010???1?0111???001?1???????????1 1?????????1???????????01?0???????????????????????????0?1??1??0?????1???1?? ????????1111??1 Wonambi 112?111?????0?????????0????????1?????010110011120011?010101?011101?0?1???? ????0101111000000011121101?111?1201?0??101010?111?01111??001?1???????????1 1?0???1??01???00?11202011??11101111110100?0?????10?00110?111011111110011???1110101111111 Yurlunggur 112111101??103??0?01000001?00?1110010011010??1020001?01010100111000011?010 ????01011?1?000?00111211011111?12?0????1??010111100111???001(0 1)1???????10?011001??1??01110000112020110?11101??1110??0001?11010?000110?? ?10??1?1?????1?00??????1111111 Anilius 11101110000100200001001011100011111?1001010?110100000010011001111100000010 11110100010011000102100001101011101200111101111010101011110111111111011101 11011111000110100112120010011111121111110111110001110000111111000201101??10??10111100110 Uropeltidae 11111(0 1)(0 1)0(0 1)(0 1)010(0 1)100(0 1)0100100110001111(0 1)01001010?110(1 2)00000(0 1)(0 1)00110010111000(0 1)00101111(0 1)1000100010001021(0 1)(0 1)0(0 1)(0 1)111011{1 2}01200111101111010101011110111111111011101110111110(0 1)0110(0 1)001121(0 2)(0 1)01001(0 1)(0 1)1112111(0 1)11(0 1)11--1101011101001111(0 1)00(0 1)(0 1)20(0 1)001111(0 1)(0 1)2101111001(0 1)0

Pachyrhachis 112??11?1??10?????????00???00???11011??0?11??112???0101??11??????10111?111 ?20?1??????????1??????????10???120110001??0101111?001?121111?11011110????? ??0???1????110?001???2??1???1?1???1?1????1-?1011????????111?0?1???1?111?????1?01111002?1 Haasiophis 112?111?1??10???????0?00???00???110110?1?11?01121?00101??11?11??11?111?011 ????1????0????11?????1?0??10???120110001010101111?10??121?11011011110????? ??01111???11?0?0011??{1 2}101?0??1????1??????10?1011????????111?0????20???111101?101111??2?1 Eupodophis 112?011?1??10????????????1???????1011????1???1101???1011?11????2?101???1?1 ????1????????????????1????1????120110001010001111?00??0?1111111011?10????? ??0???1???11??????1??????????1?1??1??????1????011????????1?1??????1??????? 1????0?1111?2?1 Xenopeltis 11200110110102200101001103100011111??0010111020200001011011001111111110010 0011110101000101010111000110012120120111110111101110101211010111---1111111101111101111101111211101110?110?2111-11110011111110000011110101020111121111211111100110 Loxocemus 11200110110102100101001102100011100010010111020200101011011001111111110010 121011010100(0 1)10101011101111001212012011111011110111010121101011111?201111111011111011 111001112111011111111?2111111010201011111010001111111112011011111111011110 1111 Erycinae 1121011111011(0 1)1011011011012(0 1)10111(0 1)0(0 1)1100111(0 1)(1 2)211101010(0 1)1011111121111(0 1)10011120012011100(0 1)1011(0 1)?11(1 2)0(0 1)011(0 1)01012112011111011110100110121101011111220111111101111110211100011(1 2)11101(0 1)11?111?211111101(0 1)-11011012110011111110112111(0 1)121101(1 2)10111001111 Ungaliophiidae 11210110110110201101101(0 1)0120101111001100111?021(0 2)1000101101111112(0 1)111010011120012010100010111011000111001?1211?12111011110100110121101011111220111111101??11(0 1)0111100011211111111111112111111011010110121100111111111121111121111210111100110 Boinae 1121011(0 1)11011(0 1)2011011011(0 1)1211011110111(0 1)0111122121011101(1 2)011111120111110011120012111(0 1)00(0 1)1111011120101100121211201111101111(0 1)100110121101011111220111111101111110211101011212111111111112111111110110110121100111111101121111111101210111101111 Pythoninae 112(0 1)011(0 1)11010020010110111121101110011100111?0212101110120111(0 1)1021111110011120012111100(0 1)101(0 1)00112(0 1)101100121211201111101111010011012110101111122011111110111111021110101121 21111111111121111110102110110121100111111101121111111111110111111111 Tropidophiidae 1121011111011320010100110110111111011100111?021011101010011101121111010011 120012011101110111011100111001?1211201211111111110011011110101111122(0 1)111111111111110111100011211111111?111?2111111110010110111000111110111122111121101211111100110

Bolyeriidae 1121011011011320010100110110101111011100111?021200101011011101121111010011 12011201110001011??110001110012121120111110111111001101211010111---1111111101111110111100011211111111?111?2111110110010110111000111111111120(0 1)1112110121(0 1)111100110 Acrochordidae 1121011110111310011211111200111111011110?11?021211101010011101120111111021 12001???010111011101110011100101201?1-21111111111(0 1)01101211010(0 1)11---1111111111??1110111100011110111001?111?2111110110110110110000011010?1111210?121101211111101110 Basal_Colubroides 1121011(0 1)1(0 1)(0 1)113110112111(0 1)0(0 2)(0 1)(0 1)111111(0 1)01100111?021(0 1 2)(0 1)1101010011101020111(0 1)1102113001211110111011(0 1)(0 1)1(0 1)0001110(0 1)101211-1211111110110(0 1)1101211010111---1111111111??1110111100011211111111?111?21110-1110110110110000111110111022110121101211101100111 Coniophis 111?1?????????????????????????????????????????0??????????????????????????? ??????????????????????????????????????????0001201?10(0 1)01??10?????????????????????1??0????????????????????0???????????????????? ???????11????1?10??????????????0100100 Parviraptor_estesi 112?1?1??????????????????????????????????????10101???????????????????????? ?????????????????????????????????????????????????????????????????????????? ??????1??0?????????????????11??012???0?????????????????????????????????????????????????? Portugalophis_lignites 112?1?1??????????????????????????????????????10101???????????????????????? ??????????????????????????????????0??????????????????????????????????????? ??????1??00????????????????11?0012???01??????????????????101??0??0?????????????????????? Diablophis_gilmorei 112?1????????????????????????????????????????1??01???????????????????????? ??????????????????????????????????0????????1000?0?001????10??????????????? ??????1??0?????????????????1???012???0???????????????????101??0??0????????????????000110 Eophis_woodwardi 1???1????????????????????????????????????????????????????????????????????? ???????????????????????????????1??0??????????????????????????????????????? ??????1???????????????????????????????????????????????????101?00??0??????? ??????????????? aff. Parviraptor estesi ??????????????????????0???????0??0???????????????????????????????????????? ???????????????????????????????????????????1000?0?001????10??????????????? ?1???????????????????????????????????????????????????????????????????????? ?????????000?10 ; END;