Access to this full-text is provided by Springer Nature.
Content available from Scientific Reports
This content is subject to copyright. Terms and conditions apply.
1
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports
Largest known madtsoiid snake
from warm Eocene period
of India suggests intercontinental
Gondwana dispersal
Debajit Datta * & Sunil Bajpai *
Here we report the discovery of fossils representing partial vertebral column of a giant madtsoiid
snake from an early Middle Eocene (Lutetian, ~ 47 Ma) lignite-bearing succession in Kutch, western
India. The estimated body length of ~ 11–15 m makes this new taxon (Vasuki indicus gen et sp. nov.)
the largest known madtsoiid snake, which thrived during a warm geological interval with average
temperatures estimated at ~ 28 °C. Phylogenetically, Vasuki forms a distinct clade with the Indian
Late Cretaceous taxon Madtsoia pisdurensis and the North African Late Eocene Gigantophis garstini.
Biogeographic considerations, seen in conjunction with its inter-relationship with other Indian and
North African madtsoiids, suggest that Vasuki represents a relic lineage that originated in India.
Subsequent India-Asia collision at ~ 50 Ma led to intercontinental dispersal of this lineage from the
subcontinent into North Africa through southern Eurasia.
Abbreviations
AMNH American Museum of Natural History, New York
CPAG Centre of Pure and Applied Geology, University of Sindh, Pakistan
DGM Departamento Nacional de Produção Mineral, Rio de Janeiro, Brazil
FMNH e Field Museum, Chicago, USA
IITR/VPL/SB Vertebrate Paleontology Laboratory, Indian Institute of TechnologyRoorkee, Roorkee, India
LPB FGGUB Laboratory of Paleontology, Faculty of Geology and Geophysics, University of Bucharest,
Bucharest, Romania
MPEF-PV Vertebrate Paleontological collection, MuseoPaleontológico Egidio Feruglio, Trelew, Chubut
Province, Argentina
NTM Northern Territory Museum, Australia
NHMUK e Natural History Museum, London, U.K.
QMF Queensland Museum, Brisbane, Australia
PVL Vertebrate Paleontological Collection of the Instituto Miguel Lillo, San Miguel de Tucumán,
Argentina
WIF/A Wadia Institute of Himalayan Geology, Dehradun, India
Madtsoiidae are an extinct clade of primarily Gondwanan terrestrial snakes with a temporal range spanning
about 100 Myr from the Late Cretaceous–Late Pleistocene1–3. eir geographic range during the Late Cretaceous
encompassed Madagascar, South America, India, Africa and the European archipelago1,4–9. e Cenozoic forms
are restricted to North Africa, South America, the Indian subcontinent and Australia2,10–17. Madtsoiids display
a broad spectrum of body-sizes and include some of the largest known terrestrial snakes that ever lived2,7,9.
Although a speciose clade, most taxa are known exclusively from vertebrae, resulting in poorly constrained in-
group relationships2,8,16. Additionally, the phylogenetic position of Madtsoiidae within Ophidia has remained
contentious, as some studies recover it within Serpentes whereas others place it outside the crown group3,9,17–20.
ese phylogenetic uncertainties have hampered our understanding of madtsoiid biogeography and radiation
events2,8.
In the Indian subcontinent, Late Cretaceous (Maastrichtian) madtsoiids are known from the Deccan vol-
canic province, including the large-sized Madtsoia pisdurensis from the Lameta Formation6,8. Among Tertiary
OPEN
Department of Earth Sciences, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand 247667, India. *email:
debajitdatta.pd@es.iitr.ac.in; debajitdatta9@gmail.com; sunil.bajpai@es.iitr.ac.in
Content courtesy of Springer Nature, terms of use apply. Rights reserved
2
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
madtsoiids, indeterminate forms are known from the early Paleocene Khadro Formation (Pakistan16) and the
early Eocene Cambay Shale (India15). e latter also yielded the large madtsoiid Platyspondylophis21. e Eocene
and Late Oligocene records include indeterminate taxa from Kutch and Ladakh, respectively14,22. Here we report
the discovery of a giant madtsoiid snake, one of largest snakes ever reported, from an interval corresponding to
a warm Middle Eocene period (~ 47Ma) of India. Fossils were collected from an early Lutetian grey shale unit
from Panandhro Lignite Mine, Kutch, Gujrat State, western India (Supplementary Note 1,Fig.1), and includes
an excellently preserved, partial vertebral column. e discovery of a giant Eocene snake has important implica-
tions for madtsoiid biogeography in the context of Gondwanan inter-continental dispersal, and the evolution of
large body-sizes possibly driven by high temperatures in the Middle Eocene tropical zones.
Results
Systematic paleontology
Squamata Oppel, 1811
Ophidia Brongniart, 1800
Madtsoiidae (Hostetter 1961) McDowell, 1987
Vasuki indicus gen. et sp. nov.
Etymology
Generic name aer the well-known Hindu mythical serpent ‘Vāsuki’ around the neck of Lord Shiva; specic
name is for the country of origin i.e., India.
Holotype
IITR/VPL/SB 3102-1-21; a partial vertebral column representing the precloacal region (Figs.2, 3; Supplementary
Table1).
Horizon and locality
Naredi Formation; Panandhro Lignite Mine,district Kutch, Gujarat state, western India.
Diagnosis
Vasuki exhibits a unique combination of the following characters: presence of prominent paracotylar foramina
(shared with Madtsoiidae); middle-sized cotyle (shared with Madtsoiidae); median prominence on ventral mar-
gin of centrum (shared with Madtsoiidae); prezygapophyseal process absent; high angle of synapophysis with
horizontal in anterior view (avg. 71.5°); MTV diapophysis level with dorsoventral midpoint of neural canal
(shared with Madtsoia madagascariensis, Madtsoia camposi, Wonambi barriei and Adinophis); prezygapophyseal
buttress succeeded posteriorly by elliptical fossa (shared with Madtsoia pisdurensis); deep V-shaped embayment
(shared with Gigantophis garstini and Madtsoia pisdurensis); oval precloacal cotyle (shared with Gigantophis
garstini and Madtsoia pisdurensis); transversely wide vertebrae (shared with Gigantophis garstini and Madtsoia
pisdurensis); neural spine posteriorly canted (shared with Gigantophis garstini and Madtsoia pisdurensis); broad
hemal keel with posterior process (shared with Gigantophis garstini and Madtsoia pisdurensis); strongly notched
anterior zygosphenal margin; endozygantral foramen present (shared with Madtsoia madagascariensis, Powello-
phis and Gigantophis garstini).
Autapomorphies: exceptionally large vertebrae [centrum length (cL): 37.5–62.7mm and prezygapophyseal
width (prW): 62.4–111.4mm]; neural spine cross-section spade-shaped; poorly developed hemal keel which
remains dorsal to the parapophyses; chisel-shaped posterior process of the hemal keel.
Figure1. Geological map of Kutch Basin showing fossil locality (a); stratigraphic column at Panandhro
Lignite Mine showing the position of madtsoiid snake-yielding horizon with age diagnostic dinoagellate cyst
assemblage and δ13C curve marking hyperthermal event ETM2 (modied aer Agrawal etal.23) (b); panoramic
view of the fossil site (c). Map and stratigraphic column were drawn by D.D. using CorelDRAW 2019 (Version
number: 21.0.0.593, URL link: http:// www. corel. com/ en/). ETM2 age estimate aer Westerhold etal.24.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
3
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
Description
e collection comprises 27 associated vertebrae which are mostly well-preserved and include a few in articula-
tion (Figs.2A–T, 3A–W). 22 out of the 27 specimens can be condently assigned to the precloacal region based
on the absence of hemapophyses, pleurapophyses and lymphapophyses, and are further constrained to a position
anterior to the posterior trunk region as suggested by a greater mediolateral width of the neural arch compared
to centrum length (sensu LaDuke1; Rio and Mannion2; Supplementary Tables1, 2; Supplementary Fig.2). Such
vertebral dimensions are usually found in large-bodied madtsoiids such as, Gigantophis2; Yurlunggur11, Madt-
soia1,10,13, and Wonambi25. Moreover, the closure of vertebral sutures suggests these specimens likely reached
skeletal maturity, similar for instance to Madtsoia pisdurensis8.
Vasu ki is characterized by exceptionally large vertebrae where centrum length (cL) and prezygapophyseal
width (prW) range between 37.5–62.7 and 62.4–111.4mm, respectively (Supplementary Table2). We recognize
this as an autapomorphy since these proportions eclipse all large-sized madtsoiids [Madtsoia (cL = 18–25mm;
prW = 35–65mm; LaDuke etal.1), Gigantophis (cL = 28–41mm; prW = 44–66mm; Rio and Mannion2), Platyspon-
dylophis (cL = 18–21mm; prW = 26–43mm; Smith etal.21) and Yurlunggur (cL = 15–22mm; prW = 19–41mm)].
Some caution, however, is warranted here because of uncertainties as to whether the largest size of these large-
bodied madtsoiids has been captured, although, the same is true for Vasuki.
In overall form, the vertebrae of the new Indian taxon are massive (prW >> cL) and comprise a procoelous
centrum. Anteriorly, the centrum preserves an anteroventrally inclined cotyle, whereas the posterior condyle
is deected posterodorsally resulting in considerable visibility of the condyle and cotyle in dorsal and ventral
views, respectively (Fig.2C,E). In anterior view, the cotyle is strongly concave with its ventral margin recessed
relative to the dorsal. e cotyle is mediolaterally wider than dorsoventrally high (Figs.2P, 3A,F,K; IITR/VPL/SB
3102-4, coW/coH = 1.2; Supplementary Table2) as in all madtsoiids [e.g., Gigantophis garstini2 (NHMUK R8344,
coW/coH = 1.2), Madtsoia madagascariensis (FMNH PR 2551, coW/coH = 1.24) Yurlunggur (NTM P8695-243,
coW/coH = 1.22), and Wonambi (QMF23038, coW/coH = 1.4]. Laterally, the cotyle is bordered on each side by
a well-developed and moderately deep paracotylar fossa (Figs.2K,P, 3A,K). e dorsal and ventral margins of
the fossa are prominent and dened by bony struts emanating from the dorsolateral and lateral cotylar margins,
respectively. e lateral margin of the fossa, however, is ush with the surface. Furthermore, in some specimens
the paracotylar fossa is divided into a shallower dorsal and deeper ventral sub-fossa by a weak secondary strut
extending laterally from the dorsolateral margin of the cotyle. A tiny paracotylar foramen is present on the dorsal-
most part of one or both paracotylar fossae, immediately lateral to the neural canal (Figs.2F,K,P, 3A,K). While
the presence of paracotylar fossae and foramina is a synapomorphy of Madtsoiidae16,26, the exact morphology
of these features is variable across the clade. “Gigantophis sp.” (CPAG-RANKT-V-1), Menarana nosymena and
Adinophis saka (FMNH PR 2572) dier from Vasuki in the presence of paired paracotylar foramina on each
side1,16,27. In Madtsoia and Eomadtsoia (MPEF-PV 2378) the foramina are deep and comparatively large, whereas
in Yurlunggur these occur in clusters7,8,10,11,13. Eomadtsoia, however, shares with Vasuki the presence of prominent
ventral rim of the paracotylar fossa7. In Gigantophis garstini the paracotylar fossa lacks a ventral margin and in
Platyspondylophis the paracotylar foramen is absent altogether2,21.
e posterior condyle is transversely wider than high (IITR/VPL/SB 3102–4, cnW/cnH = 1.2; Supplementary
Table2) with the width progressively increasing from ATV (Fig.2B,G; cnW/cnH = 1.1) to MTV (Fig.3G, Q;
cnW/cnH = 1.2–1.3). Similar proportions of the posterior condyle characterize most madtsoiids [e.g., Nido-
phis (LPB FGGUB v.547/3, ATV, cnW/cnH = 1.1; LPB FGGUB v.547/1, MTV, cnW/cnH = 1.2); Gigantophis
garstini (NHMUK R8344, MTV, cnW/cnH = 1.2 Rio and Mannio2); Madtsoia camposi (DGM 1310b, MTV,
cnW/cnH = 1.3] (Fig.3G,I,Q). Furthermore, in posterior view, two small, distinct fossae are discernible on the
lateral surface of the centrum immediately posterior to the le diapophysis (Fig.3G,I,Q). e fossae are vertically
arranged, on top of each other, and separated by a prominent ridge. Whether these unilateral fossae represent an
individual condition or a general feature cannot be currently ascertained and will require additional specimens
of Vasuki.
e synapophysis is dorsoventrally high and comprises a distinct diapophysis and parapophysis (Figs.2M,R,
3C,R) unlike in Gigantophis garstini, Madtsoia madagascariensis, and Madtsoia pisdurensis1,2,8. In anterior
view, the orientation of the synapophysis changes from ventrolateral (Fig.2F,K) to somewhat laterally facing
(Fig.3K,P,U) across the precloacal series. is change is marked by an increase in the synapophyseal angle (α),
with the horizontal, from ATV (α = avg. 56.6°) to MTV (α = avg. 71.5°). A narrower synapophyseal angle was
observed in most of the comparative madtsoiid taxa including Eomadtsoia [MPEF-PV 2378 (MTV), α = 45°],
Gigantophis garstini [NHMUK R8344 (MTV) α = 48], Madtsoia madagascariensis [FMNH PR 2549 (ATV),
α = 47°; FMNH PR 2551 (MTV), α = 56°], “Gigantophis sp.” [CPAG-RANKT-V-1 (MTV), α = 56°], Madtsoia
camposi [DGM 1310c (MTV), α = 57°], Wonambi [QMF23038 (MTV) α = 58°] and Madtsoia bai [AMNH 3155
(MTV), α = 62°]. In lateral view, the synapophysis is inclined at (β) 20°–27° from the vertical in Vasuki. is
is similar to Wonambi [QMF23038, β = 25°], Nanowana [QMF19741, β = ~ 25°], Madtsoia camposi [DGM
1310c, β = 26°] and Yurlunggur [P8695, β = 22°–26°]. In contrast, wider angles characterize Gigantophis garstini
[NHMUK R8344, β = 30°2], Platyspondylophis [β = 30°–35°], Madtsoia madagascariensis [FMNH PR 2549, β = 33°]
and “Gigantophis sp.” [CPAG-RANKT-V-1, β = ~ 90°], whereas in Patagniophis [β = 7°–9°], Powellophis [PVL
4714–4, β = 18°] and Madtsoia pisdurensis [225/GSI/PAL/CR/10, β = 12°] the angles are narrower.
An arcuate paracotylar notch (sensu LaDuke etal.1), between the ventral cotylar rim and the parapophysis, is
consistently present in all specimens (Fig.2A,F). e parapophysis comprises a sub-rectangular facet, in lateral
view, and extends below the ventral cotylar rim in ATV (Fig.2F,H,P,R). In MTV it lies dorsal to the ventral cotylar
rim (Fig.3F,P) unlike Madtsoia pisdurensis and Gigantophis garstini where the parapophyseal base is ventral and
in level with the ventral cotylar rim, respectively2,8. e diapophysis is bulbous and extends laterally beyond the
prezygapophysis (Figs.2F,H, 3P,R), contrary to Powellophis3, Patagoniophis australiensis28, Madtsoia pisdurensis8,
Madtsoia madagascariensis1 and Nidophis9. e dorsal margin of the diapophysis remains ventral to the dorsal
Content courtesy of Springer Nature, terms of use apply. Rights reserved
4
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
cotylar margin in ATV (Fig.2A,F), but becomes level with the dorsoventral midpoint of the neural canal in MTV
(Fig.3K,P). A similar disposition of the MTV diapophysis is observed in Madtsoia madagascariensis, Madtsoia
camposi, Wonambi barriei and Adinophis1,2,13,27. e dorsal diapophyseal margin lies between the ventral margin
of the neural canal and the dorsoventral midpoint of the cotyle in “Gigantophis sp .”16, Gigantophis garstini2, Nido-
phis9, Yurlunggur11 and Powellophis3. In Platyspondylophis the diapophysis extends beyond the ventral margin of
the neural canal in all preserved precloacal vertebrae21.
e prezygapophyseal buttress is massive, lacks a prezygapophyseal process and bears an oblique, blunt ridge
anteriorly (Fig.2F,K). In lateral view, the buttress is succeeded posteriorly by an elliptical fossa (Fig.2C,H,R).
e fossa occurs immediately ventral to the interzygapophyseal ridge and medial to the diapophysis, similar to
Madtsoia pisdurensis (Mohabey etal.8). e prezygapophyseal facets are elliptical (5022–4, przL/przW = 1.3) and
inclined ventromedially (prα = 20°–28°; Fig.2A,D,F,I). In dorsal view, these facets diverge at 45° from the sagittal
plane, contrary to the transversely oriented facets in Madtsoia bai10, Madtsoia madagascariensis1, Platyspondylo-
phis21, and Yurlunggur11. Strongly divergent prezygapophyses are also observed in Gigantophis garstini2 (~ 70°)
and Eomadtsoia7 (60°–80°). e postzygapophyseal facets in Vasuki are also elliptical (IITR/VPL/SB 3102-8II,
pozL/pozW = 1.2; Supplementary Table2) and medioventrally oriented (poα = 12°–26°; Figs.2G,J, 3B,E). e
interzygapophyseal ridge is thick and posterodorsally directed, acting as a bridge between the pre- and postzyga-
pophyses. A small lateral foramen is present ventral to the ridge (Fig.3L,Q) as in Powellophis3. In dorsal view the
Figure2. Anterior trunk vertebrae of Vasuki indicus. IITR/VPL/SB 3102-3, partial vertebra in anterior view (a);
posterior view (b); le lateral view (c); dorsal view (d); ventral view (e). IITR/VPL/SB 3102-5, complete vertebra
in anterior view (f); posterior view (g); le lateral view (h); dorsal view (i); ventral view (j). IITR/VPL/SB 3102-
7I-II, partial vertebra in anterior view (k); posterior view; (l); le lateral view (m); dorsal view (n); ventral view
(o). IITR/VPL/SB 3102-6, complete posterior anterior trunk vertebra in anterior view (p); posterior view (q);
le lateral view (r); dorsal view (s); ventral view (t). Grey arrows indicate anterior direction. Red arrowheads
and arrows indicate fossae on neural spinal base and endozygantral foramina, respectively. Roman numerals
on gures (m–o) refer to individual vertebrae in articulated specimens where ‘I” is towards the anterior. White
arrowhead and arrow indicate fossa medial to diapophysis and foramen on dorsal surface of neural arch. co
cotyle, cn condyle, da diapophysis, hyp hypapophysis, izr interzygapophyseal ridge, msf median sha, nc neural
canal, nrl neural arch lamina, ns neural spine, pa parapophysis, pcof paracotylar foramen, pcofo paracotylar
fossa, pcon paracotylar notch, po postzygapophysis, pr prezygapophysis, psl prespinal lamina, pzgf parazygantral
foramen, pzgfo parazygantral fossa, scf subcentral foramen, scfo subcentral fossa, zg zygantrum, zs zygosphene.
Scale bar represents 50mm.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
5
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
interzygapophyseal ridges are straight and dier from the arcuate ridges seen in most madtsoiids [e.g., Madtsoia,
Gigantophis garstini, Wonambi, Yurlunggur and Platyspondylophis]2,8,10,11,13,18,21,28.
e neural canal is reniform (Figs.2P,Q, 3F,G) in cross-section and signicantly wider than high (ncW/
ncH = 3–3.6). It diers from the comparatively narrower and trilobate neural canal in Gigantophis garstini2
(NHMUK R8344, ncW/ncH = 2.3), Platyspondylophis (WIF/A 2271, ncW/ncH = 2.1), Madtsoia (ncW/
ncH = 1.3–2.3), Yurlunggur (NTM P8695-243, ncW/ncH = 2.3), “Gigantophis sp.” (CPAG-RANKT-V-1, ncW/
Figure3. Precloacal vertebrae of Vasuki indicus. IITR/VPL/SB 3102-10I-II, complete posterior anterior trunk/
mid-trunk vertebrae in anterior view (a); posterior view (b); right lateral view (c); dorsal view (d); ventral view
(e). IITR/VPL/SB 3102-9I-II, partial mid-trunk vertebrae in anterior view (f); posterior view (g); le lateral
(reversed) view (h); dorsal view (i); ventral view (j). IITR/VPL/SB 3102-4, nearly-complete mid-trunk vertebra
in anterior view (k); posterior view; (l); le lateral (reversed) view (m); dorsal view (n); ventral view (o). IITR/
VPL/SB 3102-8I-II, partial mid-trunk vertebrae in anterior view (p); posterior view (q); right lateral view (r);
dorsal view (s); ventral view (t). IITR/VPL/SB 3102-11I-III, partial mid-trunk vertebrae in posterior view (u);
right lateral view (v); dorsal view (w); ventral view (x). Grey arrows indicate anterior direction. Roman numerals
on gures (c–e,h–j,r–t,v–w) refer to individual vertebrae in articulated specimens where ‘I” is towards the
anterior. Pink and white arrows indicate fossae and foramen on lateral surface of centrum, respectively. Red
arrow indicates endozygantral foramen. White arrowheads indicate paired protuberance on ventral median
sha. co cotyle, cn condyle, da diapophysis, hyp hypapophysis, izr interzygapophyseal ridge, msf median sha,
nc neural canal, nrl neural arch lamina, ns neural spine, pa parapophysis, pcof paracotylar foramen, pcofo
paracotylar fossa, po post-zygapophysis, pr prezygapophysis, psl prespinal lamina, scf subcentral foramen, scfo
subcentral fossa, zg zygantrum, zs zygosphene. Scale bar represents 50mm.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
6
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
ncH = 1.8) and Powellophis (PVL 4714–4, ncW/ncH = 1.6), and the sub-elliptical canal in Wonambi (QMF23038,
ncW/ncH = 1.3).
e zygosphene is trapezoidal and mediolaterally wider than high (zsW/zsH = 1.4–1.8; Fig.2A,K), as in
Gigantophis garstini (NHMUK R8344, zsW/zsH = 22), Madtsoia bai (AMNH 3155, zsW/zsH = 1.8) and Madtsoia
madagascariensis (FMNH PR 2551, zsW/zsH = 1.9). Transversely much wider zygosphenes characterize Nidophis
(LPB FGGUB v.547/1, zsW/zsH = 5), Madtsoia camposi (DGM 1310a, zsW/zsH = 2.8), Eomadtsoia (MPEF-PV
2378, zsW/zsH = 2.6), Platyspondylophis (WIF/A 2269, zsW/zsH = 2.2) and Patagoniophis (QMF 19717, zsW/
zsH = 5). In Vasuki, the zygosphene is wider than the cotyle, contrary to Gigantophis garstini, “Gigantophis sp.”,
Platyspondylophis and Madtsoia1,8,10,13,16,21. In anterior view, dorsal margin of the zygosphene is straight and
the articular facets are steeply inclined (~ 40° form the vertical; Figs.2F,P, 3A). ese facets are oval in lateral
view (IITR/VPL/SB 3102–6, zsfL/zsfW = 1.1). e anterior zygosphenal margin is markedly notched in dorsal
view (zsα = 118°–128°; Figs.2I,N, 3N), and diers from the non-notched zygosphene in Madtsoia pisdurensis8,
Madtsoia camposi13, Eomadtsoia7 and Platyspondylophis21. In “Gigantophis sp.” (zsα = 145°) and Madtsoia mada-
gascariensis (zsα = 145°–147°) the zygosphene is weakly notched.
e zygantrum is mediolaterally wider than high, with steeply inclined facets (50°–60° from the horizontal;
Fig.2B,G,Q). e facets are elliptical in posterior view, but devoid of a median wall present in Gigantophis
garstini2. An anteroventrally directed fossa is present at the base of each facet, and accommodates an endozygan-
tral foramen (Figs.2G, 3B). e latter is also present in Madtsoia madagascariensis1, Powellophis3 and Gigantophis
garstini2. In Vasuki, the zygantral roof above each facet is medio-dorsally convex and descends as sub-vertical
ridges into the zygantrum (Fig.2Q) as in Madtsoia madagascariensis1. e roof is ventrally convex in Eomadt-
soia and Madtsoia pisdurensis, and straight in Powellophis, Platyspondylophis, Yurlunggur and Gigantophis garst
ini3,7,8,11,21. A large, dorsolaterally oriented, elliptical parazygantral fossa anks the zygantrum laterally on either
side and bears a small parazygantral foramina (Fig.2B,G,Q).
e neural spine is dorsoventrally high (MTV, nsH/tvH = 0.21–0.29, Supplementary Table2) and buttressed
posteriorly by the neural arch laminae (Fig.3B–D,V,W). e latter extend anterodorsally from the dorsolateral
margin of the postzygapophyses up to the dorsal spinal margin, resulting in a deep median embayment. In lat-
eral view, the spine is steeply inclined posterodorsally (12°–19° from the vertical) with a concave anterior and
a straight posterior margin. While a high neural spine characterizes most large madtsoiids [Madtsoia camposi
(DGM, 1310b, MTV, nsH/tvH = 0.22), Madtsoia madagascariensis (FMNH PR 2551, MTV, nsH/tvH = 0.33),
Madtsoia bai (AMNH 3154, MTV, nsH/tvH = 0.22), Wonambi (QMF23038, MTV, nsH/tvH = 0.27)], it is more
gently inclined in these large-sized taxa [e.g., Madtsoia madagascariensis (27°–33°), Wonambi (30°), Gigantophis
garstini (30°)]. A convex anterior margin in Madtsoia madagascariensis as well as Powellophis and Nanowana fur-
ther distinguishes them from Vasuki. Furthermore, the presence of a sharp postspinal lamina (sensu Tschopp29)
on the posterior spinal surface and a spade-shaped cross-section of the spine dierentiates Vasuki from other
madtsoiids (Figs.2D,S, 3D). In dorsal view, the neural spine base is anked on either side by a prominent fossa
(Fig.2I,S), as in Madtsoia pisdurensis8 and Madtsoia madagascariensis1. e fossae occur immediately posterior
to the zygosphene and are bordered ventrally by weak, rounded bony struts emanating from the posterolateral
zygosphenal margin. Ventral to these struts, a prominent foramen is present on the dorsal surface of the neural
arch posterior to the zygosphene (Fig.2I), similar to Madtsoia madagascariensis1.
In ventral view, the centrum is triangular and widest across the parapophyses. Large paired subcentral fos-
sae, more prominent in the anterior trunk vertebrae (ATV), occupy most of the ventral surface of the centrum
(Figs.2J,O, 3E,T). e fossae are bordered laterally by robust subcentral ridges that extend posteromedially
from the parapophyses to the dorsoventral midpoint of the condyle. ese ridges are straight to weakly convex
in ventral view and dier from the concave ridges in Patagoniophis28 and Madtsoia madagascariensis1. e
subcentral fossae are separated by a transversely convex low hemal keel (Figs.2T, 3E,O,T). e latter is broad,
weakly raised and terminates anterior to the precondylar constriction. e hemal keel is not prominent, unlike
the narrow/sharp keel in “Gigantophis sp.”, Eomadtsoia, Nidophis, Nanowana and Powellophis2,3,7,9,16. In Vasuki,
this keel remains dorsal to the ventral parapophyseal margin (Figs.2M,R, 3M,R,V) unlike the hemal keel of other
madtsoiids which descends below the parapophysis. Consequently, we identify the disposition of the hemal keel
as an autapomorphy of Vasuki.
A small subcentral foramen is present on either side of the ventral sha in Vasuki (Fig.2E,O,T), as in Madt-
soia madagascariensis1, Madtsoia camposi13, Nidophis9, and Patagoniophis28. e hypapophysis is paddle-like
with sharp lateral margins and extends up to the level of the ventral condylar rim in ATV (Fig.2G,H,J,L,M,O).
e hypapophysis is directed posteroventrally unlike the ventrally directed hypapophysis in Madtsoia madagas-
cariensis1 and Patagoniophis28. Across the precloacals, the hypapophysis progressively reduces in prominence
and is replaced by a chisel shaped structure with paired protuberances separated from the ventral condylar rim
by a short, sharp ridge in the mid-trunk vertebrae (MTV; Fig.3J,O,T,X). is chisel shaped structure appears
autapomorphic for Vasuki as it diers from the condition in other madtsoiids.
Phylogenetic analysis
e position of Vasuki within Madtsoiidae was tested in a modied version of the character-taxon matrix of Zaher
etal.30 (Analysis 1; see “Methods” section and Supplementary Note 2). 50 most parsimonious trees were recovered
with a tree length of 1610, consistency index (CI) of 0.386 and retention index (RI) of 0.73. e resultant tree
topologies are largely consistent with Zaher etal.30 as Madtsoiidae was recovered as a distinct clade within crown
Serpentes (Fig.4, Supplementary Fig.3). Madtsoiidae, however, was poorly resolved and did not provide insights
into the inter-relationship of Vasuki with the other members of the clade. e poor resolution is likely a reection
of the absence of cranial material in majority of madtsoiids and a function of the large matrix where very few
vertebral characters could be scored for most madtsoiid taxa. We, therefore, ran a second analysis (Analysis 2)
Content courtesy of Springer Nature, terms of use apply. Rights reserved
7
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
by removing all non-madtsoiid Serpentes and combining the cranial and vertebral characters of Zaher etal.30
and Garberoglio etal.3, respectively (see “Methods” section and Supplementary Note 3). e latter dataset was
used because as the study focused on madtsoiid ingroup relationships. Our analysis recovered only two most
parsimonious trees with a tree length of 191, CI of 0.634 and RI of 0.62. Both trees (Fig.5, Supplementary Fig.4)
were mostly well resolved and the resultant topologies largely consistent with recent studies2,3,7 on madtsoiid
inter-relationships. Madtsoiidae shows size-based clustering with the small (< 2m) and medium–large bodied
(> 3m) taxa recovered as separate clades (Fig.5). Vasuki is nested within a distinct clade (Bremer support = 3) as
a sister taxon to Indian Late Cretaceous Madtsoia pisdurensis + North African Late Eocene Gigantophis garstini.
Figure4. Phylogenetic position of Vasuki indicus gen. et sp. nov. IITR/VPL/SB 3102 in 50% majority-rule tree
of Analysis 1. Clade comprising Vasuki indicus highlighted in pink. Numbers above and below nodes indicate
the frequency a clade is represented in the most parsimonious trees and Bremer support values, respectively.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
8
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
Estimation of body length
Quantitative estimates of total body length (TBL) of Vasuki were made based on two separate methods which
have been used in recent years for size estimation of extinct large-bodied snakes (see “Methods” section and
Supplementary Tables3–5). In these methods TBL was regressed on the postzygapophyseal width (following
Head etal.31; Rio and Mannion2) and the prezygapophyseal width (= trans-prezygapophyseal width; following
McCartney etal.32, Garberoglio etal.3), respectively. In the present study estimates were made from MTV (IITR/
VPL/SB 3102-4, 3102-8I–II, 3102-11II–III), the largest specimens in the collection, following Rio and Mannion2,
McCartney etal.32 and Garberoglio etal.3. Both regression models were statistically signicant (p < 0.05) and
had a high explanatory power (r2 = 0.83–0.96) which asserts their validity. e TBL estimates following Head
etal.31 ranges between 10.9 and 12.2m (Fig.6A,B), whereas those following McCartney etal.32 is between 14.5
and 15.2m (Fig.7A). ese estimates, however, should be treated with caution as the collection lacks posterior
precloacal and cloacal vertebrae, and an understanding of the intracolumnar variation in madtsoiids is currently
non-existent.
It is worth noting that the largest body-length estimates of Vasuki appear to exceed that of Titanoboa, even
though the vertebral dimensions of the Indian taxon are slightly smaller than those of Titanoboa. We acknowl-
edge that this observation may be a reection of the dierent datasets used to formulate the predictive equa-
tions. However, we do not disregard the results based on the dataset of MacCartney etal.32, since the equations
derived from the dataset of Head etal.31 involve measurements of extant boine taxa that are taken from vertebrae
60–65% posteriorly along the column. Caution is warranted here because of the uncertainties surrounding the
phylogenetic position of Madtsoiidae relative to crown snakes which make estimations based on a model depict-
ing intracolumnar variation in vertebral morphology of a particular extant family/taxa tentative. Consequently,
Figure5. Phylogenetic position of Vasuki indicus gen. et sp. nov. IITR/VPL/SB 3102 in 50% majority-rule tree
of Analysis 2. Clade comprising Vasuki indicus highlighted in pink. Numbers above and below nodes indicate
the frequency a clade is represented in the most parsimonious trees and Bremer support values, respectively.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
9
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
Figure6. Regressions of vertebral metrics on total body length in extant boine taxa. Regression of
postygapophyseal width on total body length in extant boine taxa from vertebrae 60% posteriorly along the
vertebral column; p = 0.00000003, standard error = ± 0.3m (a). Regression of postygapophyseal width on total
body length in extant boine taxa from vertebrae 65% posteriorly along the vertebral column; p = 0.00000001,
standard error = ± 0.2m (b). Measurements of extant boine snakes taken from Head etal.31 and plotted as black
circles. Estimated body lengths of Vasuki indicus shown in red.
Figure7. Regression of total body length on prezygapophyseal width in extant snakes. Measurements of extant
snakes taken from McCartney etal.32 and plotted as black circles. Estimated body lengths of Vasuki indicus
shown in red. p = 0.000000000000003; standard error = ± 0.09m.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
10
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
predictive regression equations following McCartney etal.32, which comprise vertebral data from an array of
extant snakes, are also considered in our study.
Discussion
Phylogenetic implications
e analyses presented here recovered a monophyletic Madtsoiidae with the clade placed within crown Ser-
pentes in Analysis 1 (Fig.4, Supplementary Fig.3). is is in accordance with most phylogenetic studies which
assessed the relationship of snake total group within Squamata30,33–35. Furthermore, similar to Zaher etal.30, the
tree topology in Analysis 1 recovered Sanajeh, Diniliysia, Najash stemward of crown Serpentes. Although the
clade Madtsoiidae remains poorly resolved in Analysis 1, we found a combination of ve synapomorphies sup-
porting the placement of Vasuki within Madtsoiidae [centrum broad and subtriangular (ch 613); deep V-shaped
embayment along posterior margin of neural arch (ch 614); presence of well-developed paracotylar foramina
(ch 615); absence of prezygapophyseal accessory process (ch 616); presence of parazygantral foramina (ch 617)].
On the other hand, Analysis 2 gave insights into the ingroup relationships of Madtsoiidae (Fig.5, Sup-
plementary Fig.4). e resultant topologies are largely comparable with previous phylogenetic results2,3,7,9, as
the taxa were found to resolve into two size-based clades (large vs small). While the possibility of size-related
features driving such groupings cannot be ruled out, the recovery of small–medium sized taxa (e.g., Adinophis,
Menarana, Powellophis) within the large bodied clade suggests the presence of size-independent characters
supporting these clades. A similar argument was also put forward by Garberoglio etal.3 while discussing the
occurrence of size-based clades within Madtsoiidae. However, none of the speciose genera (e.g., Madtsoia,
Nanowana, Menarana) included in this study formed monophyletic clades. e Bremer support for most internal
nodes within Madtsoiidae remains low, although a few have comparatively higher support (Fig.5). ese results
highlight the need for more rigorous sampling involving a better anatomical coverage of madtsoiids, leading to
more robust phylogenetic relationships.
A unique combination of 7 synapomorphies nest Vasuki within Madtsoiidae [well-developed paracotylar
foramina (ch 610); median prominence on ventral margin of centrum (ch 611); coW:dW between 0.5 and 0.3
(ch 634); lateral ridge on precloacal vertebrae below lateral foramen (ch 635); thick zygosphene (ch 645); mod-
erately high neural spine (ch 648); lateral foramina present dorsal to subcentral ridges (ch 650)]. Furthermore, a
combination of 6 unambiguous synapomorphies [posterior neural arch margin with deep V-shaped embayment
(ch 614); oval precloacal cotyle (ch 615); transversely wide vertebrae (ch 629); hemal keel not sharp and narrow
(ch 633); neural spine posteriorly canted (ch 652); presence of posterior process of hemal keel (ch 653)] support
the placement of Vasuki with Gigantophis garstini and Madtsoia pisdurensis. Moreover, a single autapomorphy
characterises Vasuki—chisel-shaped process of hemal keel (ch 654).
It is noteworthy that some of the synapomorphies mentioned above may be individually plesiomorphic
characters, it is the unique combination of characters that justies the recovery of Vasuki within Madtsoiidae.
Previous studies (e.g., Head etal.31; Mohabey etal.8) have used character combinations to diagnose Madtsoiidae
and other snake taxa.
Body length estimation and paleoecology
Our TBL estimations show that Vasuki was not only the largest madtsoiid (Table1) but one of the largest snakes
ever reported. Its vertebral dimensions are second only to the Paleocene Boinae Titanoboa (Head etal.31). We
attempted to infer the paleoecology of this large Indian madtsoiid from vertebral morphology since several
previous studies on other extinct snakes (e.g., Palaeophis colossaeus, Powellophis and Madtsoia madagascaren-
sis) have highlighted the importance of vertebrae in paleoecological reconstructions1,3,32. e transversely wide
vertebrae of Vasuki bear mainly laterally-directed synapophyses which would have been associated with laterally
directed ribs, suggesting a broad and cylindrical body (see McCartney etal.32). ese features suggest a non-
aquatic lifestyle for Vasuki as opposed to aquatic snakes which may possess high pterapophyses and have laterally
Table 1. Comparison of body length estimates of Vasuki indicus and other large-bodied matdsoiids. All
estimates are in meters. Superscripts indicate source of estimates where a–c refer to predictive equations used
in the present study (ay = 100.72x + 436.24, by = 105.98x + 390, cy = 1.0739x + 1.9842); dLaDuke etal.1; eRio and
Mannion2; fSmith etal.21; gScanlon26. See Figs.6 and 7 for p-values and standard errors of predictive equations;
see Abbreviations section.
Taxa Specimen number Estimated body length
Vasuki indicus IITR/VPL/SB 3102-8II 11.6a; 12.1b; 15.2c
Madtsoia pisduriensis 225/GSI/PAL/CR/10 4.7a; 4.9b; 5.2c
Madtsoia camposi DGM1311 3.6a; 3.7b; 4.1c
Madtsoia bai AMHN 3154 3.8a; 3.9b; 4c
“Gigantophis s p.” (CPAG-RANKT-V-1) 6a;6.3b; 6.8c
Madtsoia madagascarensis – 5–8d
Gigantophis garstini NHMUK R8344A ~ 7e
Platyspondylophis tadkeshwarensis WIF/A 2269 ~ 5f
Yurlunggur sp. – 4.7–5.7g
Content courtesy of Springer Nature, terms of use apply. Rights reserved
11
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
compressed vertebrae with ventrally facing synapophyses, thereby placing the ribs beneath the vertebrae3,32,36.
A high pterapophysis, however, is absent in many aquatic snakes and changes in the orientation of synapophy-
ses from ventral to lateral across the vertebral column have been previously noted in aquatic snakes such as
Simoliophis37. In hydrophiine sea snakes the vertebrae show true lateral compression only in the caudal region.
erefore, the possibility of an aquatic lifestyle for this giant Indian madtsoiid cannot be completely ruled out. An
arboreal lifestyle is unlikely, judging from the large size of Vasuki and the fact that arboreal snakes tend to have
elongated vertebrae with short zygapophyses38. A non-fossorial habitat is inferred here for Vasuki based on large
body-size and non-depressed neural arch-spine complexes which would have placed the dorsal muscles (e.g., M.
Semispinalis et spinalis, M. Interarticularis superior) away from the sagittal plane (sensu Auenburg39)1,3. is
is further supported by the inferred presence of dorsoventrally thick M. multidus, which originates from the
anterodorsal neural spinal surface and inserts anteriorly onto the posterior margin of the neural arch laminae
of the preceding vertebra. Gross similarity in vertebral morphology with extant large-bodied pythonids (e.g.,
Python and Malayopython)40 suggests a terrestrial/semi-aquatic paleohabitat for Vasuki. Corroborative evidence
comes from the depositional environment of the Vasuki-yielding horizon, which was reconstructed as a back
swamp marsh23,41–45, similar to the habitat of modern large pythonids.
Vasuki is envisaged as a slow-moving snake that possibly adopted a rectilinear locomotory mechanism as
indicated by its large size, anteroposteriorly short and transversely wide vertebrae and absence of accessory
prezygapophyseal processes1,38,46. A similar, anatomy-based inference was also drawn for the large Malagasy
Madtsoia madagascariensis1, although rectilinear locomotion has also been documented in extant snakes with
well-developed prezygapophyseal processes, such as vipers47. In spite of the uncertainties associated with the
locomotory mechanism of Vasuki, it was perhaps too large to be an active forager and was more likely an
ambush predator that would subdue its prey through constriction, similar to modern anacondas and large-
bodied pythonids1,42,48.
e new Indian madtsoiid suggests a relatively warm climate (~ 28°C) for the Middle Eocene (early Lutetian)
paleogeographic position of India within the tropical zone49,50. is inference stems mainly from empirically
derived dependence of poikilotherm body temperature on the ambient environmental temperature, which in
turn controls the maximum body size31,32,51. Following Head etal.31, the mean annual paleotemperature (MAPT)
for the Middle Eocene was estimated based on a relationship between the present mean annual temperature
(MAT), TBL dierence between Vasuki and reticulated python (Malayopython reticulatus, the longest known
extant snake)42 and the mass-specic metabolic rate of pythons (see “Methods” section). e predicted MAPT
falls between 27.2 and 28.6°C, corresponding to the temperature range necessary for the survival of an 11–15m
snake, and suggests the Middle Eocene tropics were 0.7–2.1°C (ΔT) warmer than at present (MAT = 26.5 °C52).
ese estimates are largely comparable to those for the Palaeocene and Late Cretaceous based on the extinct
Titanoboa (ΔT = 1.9–3.7°C) and the frog Beelzebufo ampinga (ΔT = 2.1°C), respectively52. Studies based on δ18O
isotopic ratios from foraminifera and TEX86 index53–56 have predicted high tropical sea surface temperatures
(≥ 30°C) during the Middle Eocene at ~ 47Ma, whereas some estimates suggest tropical cooling for the early
Middle and Late Eocene, but particularly during 45–34Ma57,58. e paleotemperature inferred here (< 30°C) are
lower than the afore-mentioned estimates (≥ 30°C), but suggests that the Middle Eocene (early Lutetian, ~ 47Ma)
climate was warmer than at present.
A possible limitation of this study could be the use of a pythonid (Malayopython reticulatus) as the modern
analog, especially since Pythonoidae and Madtsoiidae are phylogenetically distant. However, our choice of a
modern analog is based on the inferred foraging mode and terrestrial/semi-aquatic paleohabitat of Vasuki, using
anatomical data and the depositional environment of the fossiliferous horizon. e latter are similar to those of
modern large pythonids which known to inhabit swamps, marshes and lowland forests41–43,45.
In India, Paleogene hyperthermal events, such as PETM and ETM2 are well documented from the Kutch and
Cambay basins of western India based on δ13C negative excursions23,59–61. In comparison, studies on Paleogene
paleotemperatures are scarce. Based on oxygen isotopic ratios, temperatures in excess of 30°C were determined
for the late Paleocene and early Eocene, whereas lower temperatures, ranging between 22 and 28°C, were
reported for the Middle–Late Eocene (~ 45–37Ma)62,63. Our new estimates show that while the paleoclimate
during the Middle Eocene (~ 47Ma) became cooler compared to the Late Paleocene and early Eocene, it was still
higher than at present. Further studies on Paleogene climates in the context of squamate speciation and extinction
pattern are necessary in view of their suggested correlation with temperature patterns64–66.
Paleobiogeography
Madtsoiids were a major group of terrestrial snakes whose temporal range straddles the Cretaceous–Paleogene
boundary. Fossil occurrences depict a skewed distribution of these snakes as most taxa are known from the
Gondwanan landmasses, except Antarctica1,2 (Figs.8, 9). e Laurasian record is extremely poor with madtsoiids
known only from the Late Cretaceous (upper Campanian–Maastrichtian) of southern Europe9. e distributional
pattern also shows the appearance of taxa on landmasses which were separated during the Late Cretaceous and
Cenozoic but which share close phylogenetic relations indicating biogeographic links (sensu LaDuke etal.1). is
conundrum is aptly illustrated by the presence of Madtsoia in the Late Cretaceous (Maastrichtian) of Madagascar
and India and the Early Paleogene of South America, and Menarana in the Maastrichtian of Madagascar and
Spain1,8 (Fig.8). Previous studies put forward multiple scenarios for madtsoiid paleobiogeography including—a
pan-Gondwanan distribution, albeit unsampled, during the Early Cretaceous followed by regional extinctions
and/vicariance; presence of land bridges allowing dispersal between dierent Gondwanan landmasses and to
Europe; sweepstakes dispersal between continents separated by oceanic barriers1,8,9. However, Rio and Mannion2
argued in favour of an early pan-Gondwanan distribution and trans-Tethyan dispersals between Africa and
Europe in the Late Cretaceous. e new Middle Eocene Indian madtsoiid further adds to the complexity of
Content courtesy of Springer Nature, terms of use apply. Rights reserved
12
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
madtsoiid biogeography owing to its close phylogenetic ties with the Late Cretaceous Madtsoia pisdurensis from
India and the Late Eocene North African Gigantophis garstini (Figs.5, 8).
To assess the biogeographic signicance of Vasuki we constructed a time-calibrated phylogenetic tree since
this approach has been widely used in several previous studies for evaluating the paleobiogeographic signicance
of dierent vertebrate groups, including snakes and dinosaurs2,67–69. e rationale behind this approach is that
phylogenetic relations are widely considered to be suggestive of biogeographic ties1,70–72. e phylogenetic tree
used here is based on anatomically sparse data because most madtsoiid taxa are known exclusively from vertebrae
and lack cranial material, resulting in weak support (Bremer support) for a majority of internal nodes within
Madtsoiidae. For this reason, we restricted our biogeographic interpretations only to those nodes which had
comparatively higher support (Bremer support ≥ 2; Fig.8). Overall, the paleobiogeographic scenarios presented
here should be treated with caution as future fossil discoveries may alter the phylogenetic position of some
madtsoiid taxa and, in turn, the present biogeographic inferences.
Notwithstanding the above-mentioned limitations, the resultant tree in our study is consistent with the
current consensus on madtsoiid origins as it suggests a Gondwanan origin reecting the fact that all known
early-diverging taxa are from erstwhile Gondwanan landmasses (Fig.8). e tree topology argues for biotic
Figure8. Time-calibrated phylogenetic tree, based on the 50% majority-rule tree of Fig.5. Red star indicates
position of Vasuki indicus. Clade for which biogeographic scenarios have been discussed are marked with
colored nodes.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
13
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
exchanges between South America, Madagascar and Australia since the Malagasy Madtsoia madagascariensis
(Late Cretaceous) and the South American Madtsoia bai (Eocene) are successive outgroups to the clade com-
prising the Neogene Yurlunggur and Wonambi from Australia. Paleogeographic reconstructions depict frag-
mentation of most major Gondwana landmasses by the early Cenomanian, with Indo-Madagascar separating
from Australia–Antarctica by ~ 110 Ma73–75. However, previous studies suggested that land connections between
South America and Australia facilitating faunal dispersal through Antarctica persisted till the early Eocene1,75
(Fig.9). On the other hand, the Malagasy–South American–Australian biotic link can likely be explained by
the presence of madtsoiids or their most recent common ancestors in these continental blocks prior to their
break-up. Recent studies on madtsoiid biogeography envisage an Early Cretaceous pan-Gondwana dispersal of
these snakes, with ghost lineages from time-calibrated trees predicting an Aptian origin of Madtsoiidae1,2,8,9,19.
e fossil record, however, is inconsistent with the hypothesized Early Cretaceous madtsoiid origins since their
currently known earliest representatives are from the Coniacian–Santonian of Niger1,2,9,76,77. Future sampling
from the pre-Maastrichtian horizons of Africa and Indo-Madagascar may help resolve this conundrum.
e Indian madtsoiids, namely Vasuki indicus, Madtsoia pisdurensis, and Platyspondylophis tadkeshwarensis,
are resolved into two distinct sub-clades (Fig.8). Platyspondylophis (Ypresian) and the Malagasy Adinophis saka
(Maastrichtian) are recovered as sister-taxa, whereas Vasuki (early Lutetian) is the earliest-diverging member of a
clade comprising Madtsoia pisdurensis (Maastrichtian) and the North African Gigantophis garstini (Priabonian).
ese phylogenetic relations suggest Late Cretaceous–Paleogene biotic exchanges between the Indian subcon-
tinent, Madagascar and North Africa. Among the various competing hypotheses explaining such faunal links,
Krause etal.74 hypothesized connections (stepping stones) between the Indian subcontinent, Madagascar and
Africa during the Late Cretaceous, which were possibly destroyed in subsequent tectonic events (e.g., subduction,
hotspot related volcanism). e Oman-Kohistan-Ladakh arc (OKL) is another biogeographic pathway which is
considered to have facilitated biotic interchanges between North Africa and India following the subcontinent’s
collision with OKL at ~ 80 Ma78. While there is some support from paleomagnetic and radiometric data for the
80Ma Indo–OKL collision78, subsequent studies based on detrital zircon ages and dating of post-collisional olas-
ses have provided alternate explanations bearing on the sequence of accretion of the OKL with India/Asia79,80.
ese studies support OKL–Eurasia collision by ~ 100–80Ma, with India colliding with Asia + OKL only during
the Paleogene. is makes the possibility of Late Cretaceous Indo–African faunal exchange less likely2. More
recent studies based on paleomagnetic data propose an initial collision between India and Kohistan-Ladakh
arc at ~ 60–50Ma followed by their nal collision with Asia at ~ 45–50Ma, with the arc being positioned at
8.3 ± 5.6°N at ~ 66–62 Ma81,82.
Among the scenarios discussed above we consider the following to be the most plausible explanation for the
Indo-Madagascar-North African biotic links suggested by phylogenetic disposition of the Indian madtsoiids:
Figure9. Palaeogeographic distribution of madtsoiids with taxa of dierent ages plotted together in a
simplied Middle Eocene (50Ma) map to show their global spatio-temporal occurences. Dashed-lines indicate
possible dispersal routes between South America and Australia and the Indian subcontinent and North Africa.
Palaeogeographic map aer Scotese43 and sourced from https:// www. earth byte. org/ paleo map- paleo atlas- for-
gplat es/ [is work is licensed under the Creative Commons Attribution 4.0 International License. http:// creat
iveco mmons. org/ licen ses/ by/4. 0/]. Source of information on madtsoiid distribution from the Paleobiology
database (https:// www. paleo- biodb. org/).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
14
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
(i) A sister taxa relationship between the Maastrichtian Malagasy Adinophis saka and Indian Platyspon-
dylophis (Ypresian) suggests a dispersal event at or before Indo-Madagascar separation at ~ 88 Ma83. e
direction of dispersal, however, remains uncertain as the available fossil evidence does not allow a critical
evaluation of this hypothesis due to the poor sampling record of pre-Maastrichtian Malagasy and Indian
deposits. However, recovery of Madtsoia from the Maastrichtian of both India and Madagascar8 (Fig.8)
supports the prevalence of their biotic links, as also suggested by other groups including cordyliform
lizards and the nigerophid Indophis75,84.
(ii) Post Indo-Madagascar separation at ~ 88Ma, there was extended periods of isolation which ended with
collision of the Indian subcontinent + Kohistan-Ladakh arc with Asia in the early Paleogene50,81,82 result-
ing in biogeographic pathways with North Africa through southern Eurasia (Fig.9).
(iii) Vasuki, Madtsoia pisdurensis and Gigantophis garstini form a distinct clade to the exclusion of others, with
the earliest-diverging taxa from India (Fig.8). is clade also shows close phylogenetic links between
Late Cretaceous and Middle Eocene Indian taxa, suggesting a possible Indian origin for this clade. e
placement of Gigantophis garstini within this clade indicates possible dispersal events from India to
North Africa following India-Asia collision, consistent with the Late Eocene (Priabonian, 37–35 Ma2)
age of Gigantophis and recent paleobiogeographic reconstructions showing dispersal routes between
India and North Africa via southern Eurasia following the collision43 (Fig.9). Whereas an African
origin of Gigantophis garstini cannot be ruled out considering the recovery of madtsoiids from the Late
Cretaceous deposits of that continent, the taxonomic and phylogenetic uncertainties oer little support
for this hypothesis. However, Rio and Mannion’s2 alternative explanation that an Early Cretaceous pan-
Gondwanan dispersal and long ghost lineages may have led to close phylogenetic relations between
Gigantophis garstini and the Indian madtsoiids, though potentially valid, is currently weakly supported
because of poor sampling.
To summarize, we identify a lineage of exceptionally large-bodied madtsoiids (represented by the largest
known madtsoiids, Vasuki and Gigantophis garstini) which originated in the Indian subcontinent and subse-
quently spread to Africa via southern Eurasia during the Eocene. e discovery of Vasuki, and the sparse ana-
tomical coverage of known madtsoiids highlight the need for rigorous sampling of Late Cretaceous and Paleogene
Gondwanan deposits. Recovery of additional material and new taxa (including large-sized forms) may provide
further insights into madtsoiid systematics and biogeography.
Methods
Osteological description
e osteological description of the skeletal specimens was carried out following the nomenclature of LaDuke
etal.1, Rio and Mannion2 and Mohabey etal.8. Dierent parameters of the fossil specimens were measured
(Supplementary Fig.2) using Mitutoyo digital callipers with a precision of 0.01mm. Explanatory line drawings
are used wherever necessary. e terminology for vertebral laminae and fossae follows Rio and Mannion2 and
Tschopp29.
Phylogenetic analysis
e phylogenetic anity of Vasuki was assessed in two separate analyses (Analysis 1 and 2). In Analysis 1 (Sup-
plementary Dataset 1) the character-taxon matrix of Zaher etal.30 was used. All non-Pan-Serpentes toxicoferans
were removed except for Varanus exanthematicus which was used as the outgroup. 15 madtsoiid taxa, including
Vasuki, were added. e character-taxon matrix included 72 taxa and 785 characters. e phylogenetic analysis
was performed using TNT version 1.685 where the soware memory was set to retain 10,000 trees and a display
buer of 10Mb. e Traditional Search option was used to analyse the dataset. e constraints for the analysis
included 50 replications of Wagner trees, in which the swapping algorithm was bisection reconnection with 10
trees saved per replication. To determine the robustness of the nodes, Bremer support values were calculated
using the script bremer.run in which only trees suboptimal by 20 steps were retained.
In Analysis 2 (Supplementary Dataset 2) all non-madtsoiid Serpentes were removed except for the basal
ophidians Najash and Sanajeh. e latter taxon was used as the outgroup. e dataset combined the cranial
and vertebral characters of Zaher etal.30 and Garberoglio etal.3, respectively. 3 additional madtsoiid taxa were
included. e character-taxon matrix included 22 taxa and 656 characters. e analysis was performed using
TNT version 1.685 following the soware settings and search parameters of Analysis 1. e script bremer.run was
used to calculate Bremer support values in which only trees suboptimal by 20 steps were retained.
Time-calibrated tree
is was constructed by plotting the temporal ranges of the snake taxa onto the majority rule tree of Analysis
2 against a numerically calibrated geological time-scale. e temporal ranges of the taxa used in this study
have been obtained from the Paleobiology Database (https:// www. paleo biodb. org/), Rio and Mannion2, and
Garberoglio etal.3.
Body length estimation
e body-length estimates of Vasuki were based on the datasets of Head etal.31 and McCartney etal.32. e
dataset of Head etal.31 comprises measurements of trans-postzygapophyseal width (poW) and TBL of 21 extant
boine taxa, whereas that of McCartney etal.32 include measurements of trans-prezygapophyseal width and total
body length of 21 extant snakes.
e following predictive regression equations were formulated aer
Content courtesy of Springer Nature, terms of use apply. Rights reserved
15
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
1. Head etal.31:
where postzygapophyseal width (x) is equated with the total body length (y). e dataset was from vertebrae
60% posteriorly along the vertebral column, and was not log transformed as the measured parameters were
approximately normally distributed (sensu Head etal.31).
where postzygapophyseal width (x) is equated with the total body length (y). e dataset was from vertebrae
65% posteriorly along the vertebral column, and was not log transformed as the measured parameters were
approximately normally distributed (sensu Head etal.31).
2. McCartney etal.32 and Garberoglio etal.3:
where trans-prezygapophyseal width (x) is equated with the total body length (y). Log transformed values
of the measured parameters were used to normalize the dataset.
In previous studies, body lengths have been estimated for extinct snakes, which are part of extant clades,
using maximum likelihood methods31,32. Head etal.31 developed a model depicting intracolumnar variation of
vertebral morphology in extant boines to assign vertebral specimens of the giant extinct boid Titanoboa to their
most likely position in the vertebral column. Based on vertebral landmarks, the specimens of Titanoboa were
matched to a position 60–65% posteriorly along the column (MTV, sensu Rio and Mannion2), and size estimates
were obtained by regressing TBL on poW based on vertebrae of extant boines from those positions. However,
such models showing intracolumnar variation in madtsoiids are currently non-existent as very few of these
snakes are known from complete/nearly complete vertebral column2. Size estimates of Vasuki were calculated
in this study using MTV following Rio and Mannion2, although, these estimates should be considered tentative
as the specimens of Vasuki cannot be assigned to the same position as the boine vertebrae used to formulate the
equations. Also, there may be dierences in the relationship between poW and TBL between extant boines and
Vasuki. Furthermore, uncertainties associated with the phylogenetic position of Madtsoiidae relative to crown
snakes, preclude formulation of models showing intracolumnar variation in vertebral morphology based on any
extant clade. Consequently, predictive regression equations, based on data from an array of extant snakes from
McCartney etal.32, were used to determine the body length of the new Indian taxon and therefore, the estimated
lengths, though reasonable, should also be treated with caution.
Estimation of paleotemperature
Paleotemperature estimates were obtained using the following equation provided in Head etal.31:
where MAPT is the mean annual paleotemperature; MAT is the present mean annual temperature (26.5 °C52);
TBLM = 10.05m is the maximum total body length of Malayopython reticulatus41; TBLV is the maximum estimated
body length of Vasu ki (15.2m); Q10 (mass specic metabolic rate of pythonids) = 2.686; α (metabolic scaling
component) = 0.1752,87.
Since Madtsoiidae are an extinct clade, the body length of Malayopython reticulatus (Serpentes, Pythonidae)
was used in the study as it is the longest known extant snake42. e choice of Malayopython as the modern analog
is based on the similarity in gross vertebral morphology and, inferred mode of life and habitat between Vasuki
and extant large-bodied pythonids40–43,45. However, in the absence of extant representatives of madtsoiids or their
close relatives, the estimated paleotemperature values should be treated with caution.
Data availability
All data associated with the manuscript are provided in the Supplementary File.
Code availability
Nomenclatural acts. is published work and the nomenclatural acts it contains have been registered in Zoo-
Bank, the proposed online registration system for the International Code of Zoological Nomenclature (ICZN).
e LSIDs for this publication are urn:lsid:zoobank.org:act: 2F44E9BE-AE99-45E8-A132-D36A935D3B36
(Vasuki) and urn:lsid:zoobank.org:act: 0DD3FB9F-A500-4FFE-842C-EFE51EC76E4D (V. indicus).
Received: 18 October 2023; Accepted: 28 March 2024
y
=
100.72x
+
436.24,
y
=
105.98x
+
390,
y
=
1.0739x
+
1.9842,
MAPT
=MAT +3α10 ◦C
log10(TBLv/TBLM)
log10Q10 ,
MAPT
=MAT +5.1 ◦C
log10(TBLv/TBLM)
0.41 ,
Content courtesy of Springer Nature, terms of use apply. Rights reserved
16
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
References
1. 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).
2. Rio, J. P. & Mannion, P. D. e osteology of the giant snake Gigantophis garstini from the upper Eocene of North Africa and its
bearing on the phylogenetic relationships and biogeography of Madtsoiidae. J. Vertebr. Paleontol. 37, e1347179 (2017).
3. Garberoglio, F. F., Triviño, L. N. & Albino, A. A new madtsoiid snake from the Paleogene of South America (northwestern Argen-
tina), based on an articulated postcranial skeleton. J. Vertebr. Paleontol. 42, e2128687 (2022).
4. Albino, A. M. Simposio Evolución de los Ver tebrados Mesozóicos. In Actas IV Congreso Argentino de Paleontología y Bioestratigrafía
(ed. Bonaparte, J. F.) (Formación Los Alamitos, 1986).
5. Albino, A. M. Una nueva serpiente (Reptilia) en el Cretácico Superior de Patagonia, Argentina. Pesquisas 21, 58–63 (1994).
6. Rage, J. C., Prasad, G. V. R. & Bajpai, S. Additional snakes from the uppermost Cretaceous (Maastrichtian) of India. Cretac. Res.
25, 425–434 (2004).
7. Gómez, R. O., Garberoglio, F. F. & Rougier, G. W. A new Late Cretaceous snake from Patagonia: Phylogeny and trends in body
size evolution of madtsoiid snakes. C. R. Palevol. 18, 771–781 (2019).
8. Mohabey, D. M., Head, J. J. & Wilson, J. A. A new species of the snake Madtsoia from the Upper Cretaceous of India and its paleo-
biogeographic implications. J. Vertebr. Paleontol. 31, 588–595 (2011).
9. Vasile, Ş, Csiki-Sava, Z. & Venczel, M. A new madtsoiid snake from the Upper Cretaceous of the Haţeg Basin, western Romania.
J. Vertebr. Paleontol. 33, 1100–1119 (2013).
10. Simpson, G. G. A new fossil snake from the Notostylops beds of Patagonia. Bull. Am. Mus. Nat. Hist. 67, 1–22 (1933).
11. Scanlon, J. D. A new large madtsoiid snake from the Miocene of the Northern Territory. e Beagle 9, 49–60 (1992).
12. Scanlon, J. D. Nanowana gen. nov., small madtsoiid snakes from the Miocene of Riversleigh: Sympatric species with divergently
specialised dentition. Mem. Qd Mus. 41, 393–412 (1997).
13. Rage, J.-C. Fossil snakes from the Palaeocene of São José de Itaboraí, Brazil. Part I. Madtsoiidae, Aniliidae. Palaeovertebrata 27,
109–144 (1998).
14. Rage, J.-C. et al. Early Eocene snakes from Kutch, Western India, with a review of the Palaeophiidae. Geodiversitas 25, 695–716
(2003).
15. Rage, J.-C. et al. A diverse snake fauna from the early Eocene of Vastan Lignite Mine, Gujarat, India. Acta Palaeontol. Pol. 53,
391–403 (2008).
16. Rage, J.-C. et al. First report of the giant snake Gigantophis (Madtsoiidae) from the Paleocene of Pakistan: Paleobiogeographic
implications. Geobios 47, 147–153 (2014).
17. Scanlon, J. D. & Lee, M. S. Y. e Pleistocene serpent Wonambi and the early evolution of snakes. Nature 403, 416–420 (2000).
18. Wilson, J. A., Mohabey, D. M., Peters, S. E. & Head, J. J. Predation upon hatchling dinosaurs by a new snake from the Late Creta-
ceous of India. PLoS Biol. 8, e1000322 (2010).
19. 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).
20. Martill, D. M., Tischlinger, H. & Longrich, N. R. A four-legged snake from the Early Cretaceous of Gondwana. Science 349, 416–419
(2015).
21. Smith, T. et al. New early Eocene vertebrate assemblage from western India reveals a mixed fauna of European and Gondwana
anities. Geosci. Front. 7, 969–1001 (2016).
22. Wazir, W. A. et al. A nd from the Ladakh Himalaya reveals a survival of madtsoiid snakes (Serpentes, Madtsoiidae) in India
through the late Oligocene. J. Vertebr. Paleontol. 41, e20584012021 (2021).
23. Agrawal, S. et al. Lignite deposits of the Kutch Basin, western India: Carbon isotopic and palynological signatures of the early
Eocene hyperthermal event ETM2. J. Asian Earth Sci. 146, 296–303 (2017).
24. Westerhold, T., Röhl, U., Donner, B. & Zachos, J. C. Global extent of early Eocene hyperthermal events: A new Pacic benthic
foraminiferal isotope record from Shatsky Rise (ODP Site 1209). Paleoceanogr. Paleoclimatol. 33, 626–642 (2018).
25. Smith, M. J. Small fossil vertebrates from Victoria Cave, Naracoorte, South Australia IV. Reptiles. Trans. R. Soc. S. Aust. 100, 39–51
(1976).
26. Scanlon, J. D. Skull of the large non-macrostomatan snake Yurlunggur from the Australian Oligo-Miocene. Nature 439, 839–842
(2006).
27. Pritchard, A. C., McCartney, J. A., Krause, D. W. & Kley, N. J. New snakes from the Upper Cretaceous (Maastrichtian) Maevarano
Formation, Mahajanga Basin, Madagascar. J. Vertebr. Paleontol. 34, 1080–1093 (2014).
28. Scanlon, J. D. Australia’s oldest known snakes: Patagoniophis, Alamitophis, and cf. Madtsoia (Squamata: Madtsoiidae) from the
Eocene of Queensland. Mem. Queensl. Mus. 51, 215–235 (2005).
29. Tschopp, E. Nomenclature of vertebral laminae in lizards, with comments on ontogenetic and serial variation in Lacertini (Squa-
mata, Lacertidae). PLoS ONE 11, e0149445 (2016).
30. Zaher, H., Mohabey, D. M., Grazziotin, F. G. & Wilson Mantilla, J. A. e skull of Sanajeh indicus, a Cretaceous snake with an
upper temporal bar, and the origin of ophidian wide-gaped feeding. Zool. J. Linn. Soc. 197, 656–697 (2023).
31. Head, J. J. et al. Giant boid snake from the Palaeocene neotropics reveals hotter past equatorial temperatures. Nature 457, 715–717
(2009).
32. McCartney, J. A., Roberts, E. M., Tapanila, L. & Oleary, M. A. Large palaeophiid and nigerophiid snakes from Paleogene Trans-
Saharan Seaway deposits of Mali. Acta Palaeontol. Pol. 63, 207–220 (2018).
33. Zaher, H. & Scanferla, C. A. e skull of the Upper Cretaceous snake Dinilysia patagonica Smith-Woodward, 1901, and its phylo-
genetic position revisited. Zool. J. Linn. Soc. 164, 194–238 (2012).
34. Hsiang, A. Y. et al. e origin of snakes: Revealing the ecology, behavior, and evolutionary history of early snakes using genomics,
phenomics, and the fossil record. BMC Evol. Biol. 15, 1–22 (2015).
35. Harrington, S. M. & Reeder, T. W. Phylogenetic inference and divergence dating of snakes using molecules, morphology and fossils:
New insights into convergent evolution of feeding morphology and limb reduction. Biol. J. Linn. Soc. 121, 379–394 (2017).
36. Georgalis, G. L. First potential occurrence of the large aquatic snake Pterosphenus (Serpentes, Palaeophiidae) from Nigeria, with
further documentation of Pterosphenus schweinfurthi from Egypt. Alcheringa 1, 1–9 (2023).
37. Rage, J. C., Vullo, R. & Néraudeau, D. e mid-Cretaceous snake Simoliophis rochebrunei Sauvage, 1880 (Squamata: Ophidia) from
its type area (Charentes, southwestern France): Redescription, distribution, and palaeoecology. Cretac. Res. 58, 234–253 (2016).
38. Johnson, R. G. e adaptive and phylogenetic signicance of vertebral form in snakes. Evolution 9, 367–388 (1955).
39. Auenberg, W. Additional remarks on the evolution of trunk musculature in snakes. Am. Midl. Nat. 65, 1–16 (1961).
40. Szyndlar, Z. & Georgalis, G. L. An illustrated atlas of the vertebral morphology of extant non-caenophidian snakes, with special
emphasis on the cloacal and caudal portions of the column. Vertebr. Zool. 73, 717–886 (2023).
41. Murphy, J. C. & Henderson, R. W. Tales of Giant Snakes: A Natural History of Anacondas and Pythons (Krieger, 1997).
42. Auliya, M., Mausfeld, P., Schmitz, A. & Böhme, W. Review of the reticulated python (Python reticulatus Schneider, 1801) with the
description of new subspecies from Indonesia. Naturwissenschaen 89, 201–213 (2002).
43. Reed, R. N. & Rodda, G. H. Giant Constrictors: Biological and Management Proles and an Establishment Risk Assessment for Nine
Large Species of Pythons, Anacondas, and the Boa Constrictor (US Department of the Interior, US Geological Survey, 2009).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
17
Vol.:(0123456789)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
44. Babar, M. E. et al. Biology, habitat and conservation of Indian Rock Python—A brief review. J. Anim. Plant Sci. 29, 349–352 (2019).
45. Burger, R. Ecology of the Reticulated Python (Malayopython reticulatus): Life in an Altered Landscape 212. Doctoral dissertation,
Cardi University (2022).
46. Mosauer, W. On the locomotion of snakes. Science 76, 583–585 (1932).
47. Marvi, H., Bridges, J. & Hu, D. L. Snakes mimic earthworms: Propulsion using rectilinear travelling waves. J. R. Soc. Interface 10,
20130188 (2013).
48. Rivas, J. e Life History of the Green Anaconda (Eunectes murinus), with Emphasis on Its Reproductive Biology. Dissertation, Univ.
Tennessee (2000).
49. Shukla, A. et al. Cool equatorial terrestrial temperatures and the South Asian monsoon in the Early Eocene: Evidence from the
Gurha Mine, Rajasthan, India. Palaeogeogr. Palaeoclimatol. Palaeoecol. 412, 187–198 (2014).
50. Scotese, C. R. PALEOMAP PaleoAtlas for GPlates and the PaleoData Plotter Program, PALEOMAP Project. http:// www. earth byte.
org/ paleo map-- ‐paleo atlas-- ‐for-- ‐gplat es/, https:// doi. org/ 10. 13140/ RG2.2. 34367. 00166 (2016).
51. Makarieva, A. M., Gorshkov, V. G. & Li, B.-L. Gigantism, temperature and metabolic rate in terrestrial poikilotherms. Proc. R. Soc.
Lond. B 272, 2325–2328 (2005).
52. Makarieva, A. M., Gorshkov, V. G. & Li, B. L. Re-calibrating the snake palaeothermometer. Nature 460, E2–E3 (2009).
53. Pearson, P. N. et al. Warm tropical sea surface temperatures in the Late Cretaceous and Eocene epochs. Nature 413, 481–487 (2001).
54. Pearson, P. N. et al. Stable warm tropical climate through the Eocene Epoch. Geology 35, 211–214 (2007).
55. Cramwinckel, M. J. et al. Synchronous tropical and polar temperature evolution in the Eocene. Nature 559, 382–386 (2018).
56. Evans, D. et al. Eocene greenhouse climate revealed by coupled clumped isotope-Mg/Ca thermometry. Proc. Natl. Acad. Sci. U.S.A.
115, 1174–1179 (2018).
57. Inglis, G. N. et al. Descent toward the Icehouse: Eocene sea surface cooling inferred from GDGT distributions. Paleoceanography
30, 1000–1020 (2015).
58. Zachos, J. et al. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science 292, 686–693 (2001).
59. Clementz, M. et al. Early Eocene warming events and the timing of terrestrial faunal exchange between India and Asia. Geology
39, 15–18 (2011).
60. Samanta, A. et al. Do the large carbon isotopic excursions in terrestrial organic matter across Paleocene–Eocene boundary in India
indicate intensication of tropical precipitation? Palaeogeogr. Palaeoclimatol. Palaeoecol. 387, 91–103 (2013).
61. Mitra, A., Dutta, R. & Halder, K. A study on benthic molluscs and stable isotopes from Kutch, western India reveals early Eocene
hyperthermals and pronounced transgression during ETM2 and H2 events. Swiss J. Palaeontol. 141, 1–22 (2022).
62. Saraswati, P. K., Ramesh, R. & Navada, S. V. Palaeogene isotopic temperatures of western India. Lethaia 26, 89–98 (1993).
63. Khanolkar, S. & Saraswati, P. K. Eocene foraminiferal biofacies in Kutch Basin (India) in context of palaeoclimate and palaeoecol-
ogy. J. Palaeogeogr. 8, 1–16 (2019).
64. Prothero, D. R. e late Eocene–oligocene extinctions. Annu. Rev. Earth Planet. Sci. 22, 145–165 (1994).
65. Rage, J. C. Mesozoic and Cenozoic squamates of Europe. Paleobiodivers. Paleoenviron. 93, 517–534 (2013).
66. Cyriac, V. P. & Kodandaramaiah, U. Paleoclimate determines diversication patterns in the fossorial snake family Uropeltidae
Cuvier, 1829. Mol. Phylogenet. Evol. 116, 97–107 (2017).
67. Delcourt, R. & Grillo, O. N. Tyrannosauroids from the Southern Hemisphere: Implications for biogeography, evolution, and
taxonomy. Palaeogeogr. Palaeoclimatol. Palaeoecol. 511, 379–387 (2018).
68. Whitlock, J. A. & Mantilla, J. A. W. e Late Jurassic sauropod dinosaur ‘Morosaurus’ agilis Marsh, 1889 reexamined and reinter-
preted as a dicraeosaurid. J. Vertebr. Paleontol. 40, e1780600 (2020).
69. Bajpai, S. et al. Fossils of the oldest diplodocoid dinosaur suggest India was a major centre for neosauropod radiation. Sci. Rep. 13,
12680 (2023).
70. Krause, D. W., Sampson, S. D., Carrano, M. T. & O’Connor, P. M. Overview of the history of discovery, taxonomy, phylogeny, and
biogeography of Majungasaurus crenatissimus (eropoda: Abelisauridae) from the Late Cretaceous of Madagascar. J. Vertebr.
Paleontol. 27, 1–20 (2007).
71. Gallina, P. A., Apesteguia, S., Haluza, A. & Canale, J. I. A diplodocid sauropod survivor from the Early Cretaceous of South America.
PLoS ONE 9, e97128 (2014).
72. Mohabey, D. M., Samant, B., Vélez-Rosado, K. I. & Wilson Mantilla, J. A. A review of small-bodied theropod dinosaurs from the
Upper Cretaceous of India, with description of new cranial remains of a noasaurid (eropoda: Abelisauria). J. Vertebr. Paleontol.
43, e2288088 (2023).
73. Ali, J. R. & Krause, D. W. L ate Cretaceous bioconnections between Indo-Madagascar and Antarctica: Refutation of the Gunnerus
Ridge causeway hypothesis. J. Biogeogr. 38, 1855–1872 (2011).
74. Krause, D. W. et al. e Mesozoic biogeographic history of Gondwanan terrestrial vertebrates: Insights from Madagascar’s fossil
record. Annu. Rev. Earth Planet. Sci. 47, 519–553 (2019).
75. Wilf, P. et al. Splendid and seldom isolated: e paleobiogeography of Patagonia. Annu. Rev. Earth Planet. Sci. 41, 561–603 (2013).
76. Head, J. J., Howard, A. F. & Muller, J. e Origin and Early Evolutionary History of Snakes (Cambridge University Press, 2022).
77. Rage, J.-C. Les Continents Péri-atlantiques au Crétacé Supérieur: Migrations des faunes continentales et problèmes paléo-
géographiques. Cretac. Res. 2, 65–84 (1981).
78. Chatterjee, S., Scotese, C. R. & Bajpai, S. e restless Indian Plate and its epic voyage from Gondwana to Asia: Its tectonic, pale-
oclimatic, and paleobiogeographic evolution. Geol. Soc. Am. 529, 1–147 (2017).
79. Borneman, N. L. et al. Age and structure of the Shyok suture in the Ladakh region of northwestern India: Implications for slip on
the Karakoram fault system. Tectonics 34, 2011–2033 (2015).
80. Andjić, G., Zhou, R., Jonell, T. N. & Aitchison, J. C. A single Dras–Kohistan–Ladakh arc revealed by volcaniclastic records. Geo-
chem. Geophys. Geosyst. 23, e2021GC010042 (2022).
81. Martin, C. R. et al. Paleocene latitude of the Kohistan–Ladakh arc indicates multistage India–Eurasia collision. Proc. Natl. Acad.
Sci. U.S.A. 117, 29487–29494 (2020).
82. Jin, S. et al. A smaller greater India and a Middle-Early Eocene collision with Asia. Geophys. Res. Lett. 50, e2022GL101372 (2023).
83. Reeves, C. e position of Madagascar within Gondwana and its movements during Gondwana dispersal. J. Afr. Earth Sci. 94,
45–57 (2014).
84. Yadav, R ., Bajpai, S., Maurya, A. S. & Čerňanský, A. e rst potential cordyliform (Squamata, Scincoidea) from India (uppermost
Cretaceous–lowermost Paleocene): An African lizard clade brings possible implications for Indo-Madagascar biogeographic links.
Cretac. Res. 150, 105606 (2023).
85. Golobo, P. A. & Morales, M. E. TNT version 1.6, with a graphical interface for MacOS and Linux, including new routines in
parallel. Cladistics 39, 144–153 (2023).
86. Bedford, G. S. & Christian, K. A. Standard metabolic rate and preferred body temperatures in some Australian pythons. Aust. J.
Zool. 46, 317–328 (1998).
87. Chappell, M. A. & Ellis, T. M. Resting metabolic rates in boid snakes: Allometric relationships and temperature eects. J. Comp.
Physiol. B 157, 227–235 (1987).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
18
Vol:.(1234567890)
Scientic Reports | (2024) 14:8054 | https://doi.org/10.1038/s41598-024-58377-0
www.nature.com/scientificreports/
Acknowledgements
e authors acknowledge with thanks the helpful comments, suggestions and a constructive critique of the manu-
script by the reviewers and Editor, Scientic Reports. Authors also thank Ritu Sharma, Debasis Das, Vivesh Vir
Kapur, N. Saravanan, Lisa Cooper, Lauren Stevens and Hans ewissen for help during eld work and, Aatreyee
Saha, Abhay Rautela and Poonam Verma for help and discussions. e Science and Engineering Research Board
(SERB) (Grant no. PDF/2021/00468 as National Post-doctoral Fellowshipto DD) and the Department of Science
and Technology (Project no. SR/S4/ES-222/2006 to SB), Government of India are acknowledged for nancial
support. DD would like to acknowledge IIT Roorkee for providing infrastructural facilities. SB would like to
acknowledge support obtained from IIT Roorkee as part of his Institute Chair Professorship.
Author contributions
S.B. and D.D. conceived the problem. S.B. collected the fossils. D.D. and S.B. analysed and interpreted the data
and wrote the manuscript. S.B. and D.D. were involved in further revisions.
Competing interests
e authors declare no competing interests.
Additional information
Supplementary Information e online version contains supplementary material available at https:// doi. org/
10. 1038/ s41598- 024- 58377-0.
Correspondence and requests for materials should be addressed to D.D.orS.B.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access is article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
© e Author(s) 2024
Content courtesy of Springer Nature, terms of use apply. Rights reserved
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
Content uploaded by Debajit Datta
Author content
All content in this area was uploaded by Debajit Datta on Apr 18, 2024
Content may be subject to copyright.
