Abstract
Deltatheroidans are primitive metatherian mammals (relatives of marsupials), previously thought to have become extinct during the Cretaceous mass extinction. Here, we report a tiny new deltatheroidan mammal (Gurbanodelta kara gen. et sp. nov.) discovered at the South Gobi locality in China (Xinjiang Province) that is the first Cenozoic record of this clade and renders Deltatheroida a Lazarus taxon (with a new record 10 million years younger than their supposed extinction). The vertebrate fauna associated with Gurbanodelta is most similar to that from the slightly older late Paleocene Subeng locality in Inner Mongolia. The upper molars of Gurbanodelta exhibit a broad stylar shelf with one prominent cusp (stylocone), and a paracone that is sharp and significantly taller than the metacone. The lower molar tentatively assigned to Gurbanodelta has a very small talonid without an entoconid. This combination of these features is known only in deltatheroidans. Phylogenetic analysis places Gurbanodelta as the sister taxon of the North American latest Cretaceous Nanocuris. Gurbanodelta is the smallest-known deltatheroidan, and roughly the same size as the smallest living marsupial. It is likely that the Gurbanodelta lineage dispersed between Asia and North America as part of known intercontinental mammalian dispersals in the late Paleocene, or possibly earlier.
Similar content being viewed by others
Introduction
The Cretaceous-Paleogene (K-Pg) transition is marked by one of the largest mass extinctions, and its aftermath saw a reorganization of global ecosystems. One of the best-known post-Mesozoic changes is the rapid diversification of placental mammals, and that Cenozoic radiation is coincident with a dramatic decline among metatherians in the Northern Hemisphere, which also originated in the Mesozoic1,2,3,4. Deltatheroidans are basal metatherians (the clade including marsupials and all mammals more closely related to them than to placentals or monotremes) and known only from the Cretaceous of Asia and North America1,2,5,6,7,8,9. Given the published fossil record, it was thought that deltatheroidans became extinct at the end of the Cretaceous1,9 with the youngest Mesozoic records (Nanocuris) from the latest Cretaceous sediments of western North America10. However, we describe here a new deltatheroidan taxon from the late Paleocene of China. Morphological comparison and phylogenetic analysis suggest that this new species is more closely related to the North American latest Cretaceous Nanocuris than to other Asian deltatheroidans. This discovery is the first-known evidence demonstrating that deltatheroidans survived the Cretaceous mass extinction, and furthermore, that the group dispersed between North America and Asia late in their history.
In deltatheroidans, the last molariform deciduous premolar is not replaced in adulthood, and the tooth position is traditionally referred as M1/m11,2,8,11. This morphology is shared with marsupials, but not placentals11. The upper molars of deltatheroidans are characterized by a broad stylar shelf, a small and mesiodistally compressed protocone, a particularly strong postmetacrista, small conules and stylar cusps, a paracone taller than the metacone, a high preprotocrista extending to buccal side and a short postprotocrista not extending buccally past the metacone. The lower molars have a trigonid that is much taller and broader than the talonid, a lingually positioned paraconid and a carnassial notch on the paracristid. The salient postmetacrista and paracristid form a postvallum/prevallid shearing mechanism, a feature adapted for a carnivorous diet. The preprotocrista and preparacrista form a surface for the double-rank prevallum/postvallid shearing mechanism. This combination of dental features supports the monophyly of Deltatheroida1,9,12,13.
Geological background
The new deltatheroidan fossils reported here were discovered in an unnamed early Cenozoic lithological unit at the South Gobi locality in the Gurbantunggut Desert in the Northeastern part of the Junggar Basin in Xinjiang Province, China (Fig. 1A). This unnamed unit is a set of sediments consisting of reddish-brown mudstones embedded with a few grayish-green or yellowish-gray fluvial sandstone beds or lenses (Fig. 1B). The upper part of the unit is a reddish-brown sandy mudstone bed containing abundant calcareous concretions. Below that bed is a thick reddish-brown mudstone containing a large amount of gypsum crystals, and the weathered surface of this layer is pale-reddish in color. The concretion-rich layer and the gypsum-rich layer are laterally extensive and can be traced transversely over tens of kilometers (Fig. 1C,D). Fossils were discovered in the sandstone lenses near the top of the whole section (Section C, Fig. 1B), the concretion-rich mudstone above the gypsum-rich layer (Section B, Fig. 1B), and the sandy mudstone with sandstone lenses below the gypsum-rich layer (Section A, Fig. 1A). The stratigraphically highest fossiliferous layer produced petrified wood and turtle carapace fragments. The middle fossiliferous layer (the concretion-rich mudstone layer), yielded only two rodent jaws, identified as Advenimus hubeiensis and Tamquammys robustus. Both species are typical of the early Eocene Bumbanian Asian Land Mammal Age (ALMA)14,15. The lowest layer produced the majority of fossils, including the new metatherian.
The non-mammalian vertebrates from the lowest fossiliferous layer include bone fragments of osteichthyans, amphibians, chelonians, squamates and crocodylians. There are many small fish specimens, and based on the size and morphological differences among the vertebrae, those specimens likely represent a minimum of two species. The frog specimens derive from relatively small sized frogs, and likely are from a minimum of two species. There are several carapace fragments from a relatively large trionychid turtle, and the morphology of those specimens is consistent with a single taxon. The lizard material is composed of scales, jaw fragments, and a humerus fragment. There are dentaries with acrodont and non-acrodont dentitions, and the tooth morphologies on those specimens vary, suggesting the occurrence of up to five species. The morphology of the lizard scales and some jaw fragments suggests that they derive from a non-glyptosaurine anguioid lizard. Anguioids are well known in the late Cretaceous and Paleogene of Asia and North America16. The morphology of the anguioid specimens from the South Gobi locality is almost identical to the unnamed anguioid from the late Paleocene Nomogen Formation at the Subeng locality in Erlian Basin of Inner Mongolia, China17. The acrodont jaw fragments, including one specimen with tricuspid teeth, suggest that at least one species represents an iguanian. The crocodylian material comprises several fragments of large osteoderms and several small laterally compressed teeth. The osteoderms are consistent with that of dyrosaurids, and would appear to be from individuals much larger than those that would likely have produced the teeth (that are only a few millimeters across). Overall, this non-mammalian vertebrate fauna, along with abundant gyrogonites (from charophyte algae) support a paleoenvironment with abundant aquatic habitats.
The matrix collected from the sandstone lens embedded in the lowest fossiliferous layer was screen-washed, and it produced small mammalian fossil remains representing eight species, including the deltatheroidan specimens reported here (Fig. 2). Among these fossils are species of insectivorans (Bumbanius ningi and Asionyctia guoi), Glires (Tribosphenomys minutus and Neimengomys qii), plesiadapiforms (Subengius mengi) and multituberculates (Mesodmops cf. tenuis), species that were previously only known from the Nomogen Formation at the Subeng locality (Fig. 2)18,19. The arctostylopid from the South Gobi locality is much larger than Palaeostylops iturus from the Subeng locality, and approaches the size of Anatolostylops zhaii from the early Eocene Nomogen III Fauna of the Bumbanian ALMA (Fig. 2)20,21.
Biostratigraphic and magnetostratigraphic data suggest that the Paleocene-Eocene Boundary in the Erlian Basin lies in or just below the “Gomphos” bed20,22,23. The mammalian fossils from the “Gomphos” bed comprise the Nomogen III Fauna of the Bumbanian ALMA, dominated by specimens of Gomphos. No specimens of Gomphos have been found at the South Gobi locality. The age of mammalian fauna from the Subeng locality is early Gashatan ALMA (late Paleocene), roughly equivalent to the late Tiffanian North American Land Mammal Age (NALMA)20,22,23. The presence of Anatolostylops sp. aff. A. zhaii at the South Gobi locality indicates a somewhat younger age than that of the Subeng Fauna, and the absence of the common Gomphos fossil from the South Gobi locality may suggest an older age than the Nomogen III Fauna. The Paleocene multituberculate Lambdopsalis bulla is the most common element in the Nomogen III Fauna, and is absent at the South Gobi locality. Given those biostratigraphic data, we correlate the mammalian fauna from the South Gobi locality to the late Gashatan ALMA, roughly equivalent to the Clarkforkian NALMA. Faunas of this age were previously unknown in Asia, and this locality and its fauna are approximately 10 million years younger than the Cretaceous mass extinction.
Result
Class Mammalia Linnaeus, 1758
Subclass Tribosphenida McKenna, 1975
Infraclass Metatheria Huxley, 1880
Order Deltatheroida Kielan-Jaworowska, 1982
Family Deltatheridiidae Gregory & Simpson, 1926
Gurbanodelta kara gen. et sp. nov.
Etymology
The genus name is derived from the Gurbantunggut Desert where the holotype was found and delta, a common component of deltatheroidan names. The specific epithet is from ‘kara,’ meaning black in the local Kazakh language.
Holotype
Institute of Vertebrate Paleontology and Paleoanthropology (IVPP), V 22802 (Fig. 3F–J), a right M2.
Hypodigm
IVPP V 22801, a right M2; IVPP V 22803, a right M3; IVPP V 22804, a right m1.
Locality and horizon
South Gobi mammalian fossil locality, Kalabulegen Town, Xinjiang Province, China (Fig. 1), late Gashatan ALMA in the late Paleocene.
Diagnosis
Smallest known deltatheroidan (Table 1), much smaller than all other deltatheroidans. Differs from all other deltatheroidans in having a stylar shelf narrower than the half of the width of the tooth, and a paraconid smaller and lower than the metaconid on m1. Differs from all deltatheroidans (except Atokatheridium) in lacking a buccal cingulum on the upper molars, and having a fully developed postmetacrista on M3. Differs from all deltatheroidans (except Tsagandelta and Deltatheroides) in having a narrow talonid that occupies half of the molar’s width, and differs from Tsagandelta, Sulestes and Deltatheroides in lacking a sharp mesial keel below the paraconid.
Description
The upper teeth (identified as M2–3) all have a triangular crown with three trenchant, buccally leaning cusps (the protocone, paracone and metacone). The mesial border of the tooth is straight. The buccal border is concave between the paracone and metacone (i.e., the ectoflexus is deep). Distally, the border between the protocone and metacone also has a shallow indentation. The protocone is a trenchant cusp, and the cusp is quite small relative to the tooth size, as in all deltatheroidans. The cusp is mesiodistally very compressed and buccolingually quite narrow. Both the preprotocrista and postprotocrista of the protocone are very sharp and strong. The former extends all the way to the buccal side and terminates at the large parastyle. The latter is very short, and it extends to the lingual side of the metacone only, not passing the base of the metacone. The pre- and postprotocristae connect to each other at the tip of the protocone at a narrow angle. The buccal surface of the protocone (bordered by those two strong ridges) is therefore narrow and concave. The lingual surface of the protocone is very narrow and forms a round ridge. There is no cingulum developed on the mesial, lingual or distal surfaces of the protocone. The paracone is smaller, but taller than the protocone. It has a convex and ridged lingual surface, and has a narrow and concave buccal surface. The preparacrista of the paracone is very long. It extends mesiobuccally for a short distance, and makes a sharp turn to the buccal side, becoming parallel to the buccal part of the preprotocrista. Eventually, the preparacrista terminates at the stylocone, distal to the parastyle. The postparacrista is short and low, and extends distally to meet the premetacrista of the metacone. The metacone is much smaller and lower than the paracone. The premetacrista of the metacone is short and low, and has the same form as the postparacrista. The postmetacrista is very strong, with its buccal part even longer and higher than the preparacrista; this feature is related to the postvallum-prevallid shearing mechanism.
The protocone, paracone and metacone enclose a very narrow but deep trigon basin. Buccal to the paracone and metacone, the tooth has a very broad stylar shelf. Buccolingually, the shelf occupies more than one-third of the tooth’s total width. The bottom of the stylar shelf is very smooth and bowl shaped, and its buccal edge has a poorly developed cingulum.
The conules and styles of the upper molars are moderately developed. Both the paraconule and metaconule are present as small swollen nodules on the preprotocrista and postprotocrista, and they do not project more ventrally than the two cristae. The paraconule is larger than the metaconule. A weak postparaconule crista extends to the base of the paracone. The metaconule does not have cristae. On the buccal side, only the parastyle and stylocone are relatively well developed, and they are present as twinned cusps located at the mesiobuccal corner of the tooth, barely projecting above the cristae connected to them. Near the parastyle, the buccal part of the preprotocrista has a tiny swollen nodule that forms a small cusp-like structure. Two similar tiny nodules also are present on the buccal border of the stylar shelf, and may be equivalent to the stylar cusps in other metatherians.
The three upper molars are quite similar to each other. IVPP V 22801 and V 22802 are identified as M2s, because the protocones of the two teeth are mesiodistally symmetrical. The distobuccal ends of the postmetacristae in the two teeth are elevated. In buccal view, the elevated part looks like a metastyle. V 22801 is smaller than V 22802. The twinned parastyle and stylocone of V 22801 are very close to each other, whereas the two cusps in V 22802 are well separated. V 22803 is slightly larger than the M2s and identified as an M3. The protocone of this M3 is mesially tilted, and as a result, its distal surface is slightly bigger than its mesial surface. The ectoflexus of the buccal tooth border of M3 is shallower than that of M2. The twinned parastyle and stylocone of M3 are well separated as in M2 (V 22802, but not V22801). The postmetacrista of M3 is quite straight, and its distobuccal end is not elevated.
The tentative attribution of the lower molar (IVPP V 22804) to Gurbanodelta kara is based on its very small size, and its combination of a triangular trigonid, well developed and widely separated paraconid and metaconid, trenchant protoconid, and long but narrow talonid. The eutherian insectivores from the same locality, such as Bumbanius ningi and Asionyctia guoi, are much larger than G. kara. The molars of these taxa have a more fully developed talonid with a large hypoconid and entoconid, and a broad talonid basin. The premolars of A. guoi and similar eutherian mammals are roughly similar to V 22804 in size, but the premolars of these mammals have a much weaker paraconid and metaconid, and a much shorter talonid than V 22804. Some eulipotyphlan placentals, such as Plagioctenodon, have molariform premolars. In these eutherians, the paraconid and metaconid of p4 are quite large, but differ from V 22804, with the metaconid more mesially positioned, and the paraconid and metaconid closer to each other. The carnassial notch between the protoconid and metaconid is absent in those eulipotyphlans. The p4 talonid is proportionally wider, and the difference in height between the trigonid and talonid is lower than in V 22084 and the molars of other deltatheroidans. In some basal marsupialiformes metatherians (= “traditional” definition of Marsupialia, e.g. in ref. 1), such as Alphadon and Peradectes, the trigonid of the lower molars has large, well-separated paraconids and metaconids, and notched shearing crests between the paraconid and protoconid and between the protoconid and metaconid. However, the talonids in these marsupialiform metatherians are proportionally much broader. The premolars in Alphadon or Peradectes-like marsupialiform are buccolingually narrow, with very weak paraconids and metaconids.
As a molar of a deltatheroidan, the small size, very open trigonid and low paraconid of V 22804 suggests that this tooth probably is an m1. The trigonid of the tooth is very tall. The protoconid is shaped like a triangular pyramid. The paraconid and metaconid are conical in shape, and all located near the lingual border of the tooth. The paraconid is smaller and lower than the metaconid. The enamel of the paraconid is chipped off; as a result, the cusp appears smaller than it would have been when intact. Both the paraconid and metaconid are much lower than the protoconid. The paraconid is widely separated from the metaconid, leaving the lingual side of the trigonid completely open. The distolingual side of the paraconid and the mesiolingual side of the metaconid are smooth, and no ridge connects the two cusps. Two blunt crests (the paracristid and protocristid) run down from the tip of the protoconid and connect to the paraconid and metaconid, respectively. The paracristid, linking the protoconid and paraconid, is long and straight. In mesiobuccal view, the paracristid is shaped like a checkmark. At the bottom of the checkmark, a rudimentary carnassial notch is developed (a lower molar feature related to the postvallum/prevallid shearing mechanism). The protocristid is short and has a shallow carnassial-notch-like structure. Along the distolingual margin of the metaconid, a weak but long distal metacristid is present. This ridge extends distally and continues to the lingual margin of the talonid.
The talonid of the lower molar is much narrower, shorter and lower than the trigonid. The hypoconid is the dominant cusp, and it barely projects more dorsally than the very short cristid obliqua. The cristid obliqua is low and weak. It extends from the mesial side of the hypoconid to connect to the distal wall of the metaconid. A tiny swollen nodule is developed on the lingual end of the hypocristid, and it can be interpreted as the rudimentary hypoconulid. The talonid basin is small and shallow with its lingual side is completely open. Many non-marsupial metatherians have mesial cingulid cusps at the base of the paraconid and protoconid, known as cuspid e and f. The enamel on the mesial side of V 22804 is broken, but the remaining dentine is smooth, and so it is unlikely that a cingulid and cingulid cusp were present.
The hypertrophy of postvallum-prevallid shearing in deltatheroidans results in a paraconid that is taller and larger than metaconid. A deep carnassial notch is always present on the paracristid. The talonid in deltatheroidans is small, but the hypoconid and hypoconulid are usually prominent. The lower molar (V 22804) referred to Gurbanodelta, has a large paraconid, but it is lower than the metaconid. The carnassial notch on the paracristid of this tooth is very shallow. The talonid is proportionally as small as those unequivocal deltatheroidans, but with a less projecting hypoconid and hypoconulid.
Discussion
The tiny mammal Gurbanodelta kara reported here exhibits many characters diagnostic of deltatheroidan metatherians, and they strongly support the placement of Gurbanodelta within Deltatheroida, despite its Cenozoic late Paleocene age. The combination of the broad stylar with only one prominent stylar cusp (stylocone), a mesiodistally compressed protocone, poorly developed paraconule and metaconule, a strong and buccally extended preprotocrista combined with a very short postprotocrista, a sharp paracone significantly taller than the metacone, a long and strong postmetacrista, a very small lower molar talonid and an absent entoconid is not known in any other mammals other than deltatheroidans. A few Mesozoic and Cenozoic mammals share some of the characters mentioned above, but not all in combination. For example, the early Eocene Didelphodus from North America and Europe is thought to be the most plesiomorphic placental mammal2,24, and it retains a relatively broad stylar shelf, salient postmetacrista, a paracone higher than the metacone and talonid much lower than the trigonid. However, Didelphodus has a relatively large protocone with a well-developed paraconule and metacone. The buccal part of its preprotocista is low and narrow, and its postprotocrista is long and extends to the distal side of the metacone. Furthermore, its lower molar has a large talonid with a well-developed entoconid. Basal marsupialiform metatherians, such as Alphadon and Peradectes, share a broad stylar shelf and strong preparacrista and postmetacrista with Gurbanodelta and other deltatheroidans. These basal marsupialiform metatherians differ from deltatheroidans in having a paracone that is usually lower than the metacone, a non-compressed protocone, and a large talonid (both buccalingally and mesiodistally) on the lower molars.
In early eutherians, the postprotocrista of the upper molars extends buccally past the base of the metacone, creating a so-called “double-rank postvallum/prevallid shearing” mechanism with the postmetacrista12,25,26,27. In deltatheroidans, the postprotocrista is very short, and the postmetaconule crista and postcingulum are absent. The postvallum/prevallid shearing mechanism is formed only by the salient postmetacrista12,25,26,27. This shearing mechanism is convergently emphasized in many phylogenetically distant carnivorous mammals, such as deltatheroidans, borhyaenoids, stagodontids, dasyuroids, creodonts, carnivorans, and Prionogale28. However, deltatheroidans are strikingly different from these other carnivorous mammals in having a very broad stylar shelf and a paracone saliently larger and taller than the metacone. As in some basal therians (and differing from eutherians and marsupial metatherians), the preprotocrista in deltatheroidans extends buccally and forms the double-rank prevallum/postvallid shearing12,25,26,27. The molars of Gurbanodelta clearly show the double-rank prevallum/postvallid shearing and the strong postvallum/prevallid shearing. However, the postvallum/prevallid shearing in Gurbanodelta is not very strongly developed, because the postmetacrista-paracristid is not greatly stronger than the preparacrista-protocristid. This situation is unlike other deltatheroidans except Nanocuris, probably being related to the very small size of the teeth of Gurbanodelta, and a transition to a more insectivorous lifestyle from a more carnivorous ancestry.
Our identification of Gurbanodelta as a deltatheroidan metatherian is based on detailed morphological comparisons (Supplementary Information), but is also by phylogenetic analyses (Fig. 4 and S-Fig. 1 in Supplementary Information). Parsimony analyses based on the dataset of Rougier et al.9, which includes all the deltatheroidan genera, reveal that Gurbanodelta is the sister species of the North American latest Cretaceous Nanocuris, and an analysis using a published broader sampling of therian mammals27 also places our new taxon among delatheroidans at the base of Metatheria, and not among any other clade of mammals (Supplementary Information). The two deltatheroidan taxa are joined by Atokatheridium to form a monophyletic group, with Asian Cretaceous deltatheroidans forming successive stems (Fig. 4). Given the inferred ~10 million ghost lineage leading to Gurbanodelta from its North American sister Nanocuris, the occurrence of Gurbanodelta in the late Paleocene of China (after the group’s supposed extinction at the end of the Cretaceous) would render it a Lazarus taxon29.
Gurbanodelta and its phylogenetic position within Deltatheroida demonstrates that at least one lineage of this metatherian group survived the K-Pg mass extinction. The body mass of Gurbanodelta was very small. Its molar area (less than 1 mm2) is only about 1/20th of that in Nanocuris. By using the predictive formulae from regression analyses of tooth area and body mass in extant, dentally conservative marsupials30, the body mass of Gurbanodelta is estimated at 4.3 g, suggesting that Gurbanodelta is roughly the same size as the long-tailed planigale (Planigale ingrami), the smallest living marsupial, and one of the smallest of all mammals31. The inferred diet of deltatheroidans, focusing on insectivory/carnivory, and their small body mass may explain their survival into the Cenozoic. It has been suggested that smaller-bodied, less specialized carnivores and animal-dominated omnivores had a better chance of survival32. Since Nanocuris was a part of the latest Cretaceous mammalian fauna in North America32, it seems highly probable that deltatheroidans survived there locally, but have yet to be sampled. A deltatheroidan similar to Atokatheridium and Gurbanodelta may be lurking somewhere undiscovered in North American Paleogene sediments. Further effort to examine small-sized mammalian teeth in the early Cenozoic likely will produce a fuller record of deltatheroidans, and potentially individuals of the Gurbanodelta lineage.
The K-Pg event is the most recent of the five largest mass extinctions33,34,35. Although most of the Mesozoic mammals became extinct by the end of Cretaceous, a few Mesozoic lineages are now known to have survived into the Cenozoic3,32,36. Those Mesozoic survivors on the northern continents include the stem therian multituberculates, a stem marsupial lineage leading to Peradectes, stem placental cimolestids, placental lineages leading to archaic ungulates (Protungulatum), a member of Zhelestidae and now a deltatheroidan3,37. On the southern continents, Mesozoic lineages, such as the australosphenidan monotremes, the gondawanatherians and the meridiolestidans are known to have been coeval with the Cenozoic eutherians and metatherians for millions of years36. The presence of those diverse Mesozoic survivors at the beginning of Cenozoic suggests that those archaic mammals probably still occupied a broad spectrum of ecological niches.
When the geographic distribution of the deltatheroidans is mapped onto the phylogenetic tree and ancestral states are reconstructed using maximum parsimony, the results indicate that an Asian or a North American origin for the deltatheroidan lineage represented by Gurbanodelta are equally parsimonious (Fig. 4). Gurbanodelta from the late Paleocene creates a gap in the deltatheroidan fossil record of nearly 10 million years, and a gap in the Asian fossil record of the group of over 20 million years. Given the temporal distribution of the youngest fossils (late Paleocene and Late Cretaceous) and their hypothesized phylogenetic relationships, it seems more likely that the Gurbanodelta lineage dispersed from North America to Asia (in the latest Cretaceous or Paleocene), or that the Nanocuris lineage dispersed to North America (with a cryptic Gurbanodelta lineage present in Asia) in the Cretaceous. The dispersal of the Cenozoic deltatheroidan lineage, from North American to Asia, or from Asia to North America in the Mesozoic, occurred prior to the well-known mass interchange of mammals and other taxa at the Paleocene-Eocene Boundary38,39,40,41,42. This metatherian dispersal, if Paleocene rather than Cretaceous in age, may have been penecontemporaneous with known Asian-North American intercontinental dispersals of the middle to late Paleocene that included the movement of tillodonts, rodents, plesiadapiforms, and other vertebrates38,39,41,42,43.
Methods
Measurements
Specimens were measured under the ZEN Pro 2012 system stored with a Zeiss stereo-microscope (Discovery V20), and were calibrated from the caliper. The results are listed in Table 1.
Phylogenetic analysis
We added Gurbanodelta kara to the dataset of Luo et al.27 to examine the systematic position of Gurbanodelta relative to Deltatheroida within a broader mammalian sample (Supplementary Information), and the recent dataset of Rougier et al.9 to examine the phylogenetic relationships between Gurbanodelta and other deltatheroidans more specifically. The deltatheroidan matrix is based on Rougier et al.44 and recently revised by Averianov et al.45, Wilson and Riedel10, and Rougier et al.9. In total, 53 taxa were scored for 156 characters. We followed the suggestion of Rougier et al.9 and excluded the taxa that have an uncertain tooth locus (i.e., Slaughteria) or that are known from a single tooth (i.e., Aegialodon, Comanchea, Trinititherium, Falepetrus, and Zygiocuspis) from the analysis. Stem zatherians, as represented by a compound taxon of Amphitherium plus dryolestids, were utilized as the outgroup.
Parsimony analysis was undertaken using TNT (Tree analysis using New Technology), a parsimony analysis program subsidized by the Willi Hennig Society46. The parsimony searching strategy of Ni et al.47 was used. We ran multiple replications, using sectorial searches, drifting, ratchet and fusing combined. Random sectorial search, constraint sectorial search and exclusive sectorial search were set on. Ten cycles of tree drifting, 10 cycles of ratchet and 10 cycles of tree fusing were performed in the search. Default parameter settings for random sectorial search, constraint sectorial search, exclusive sectorial search, tree drifting, ratchet and fusing were used. The search level was set as 10 for 103 (Luo’s dataset)27 and 47 (Rougier’s dataset)9 taxa. Optimal scores were searched with 10000 replications. Eleven characters are set as ordered. All characters have equal weight.
Additional Information
How to cite this article: Ni, X. et al. A late Paleocene probable metatherian (?deltatheroidan) survivor of the Cretaceous mass extinction. Sci. Rep. 6, 38547; doi: 10.1038/srep38547 (2016).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
Kielan-Jaworowska, Z., Cifelli, R. L. & Luo, Z. Mammals from the age of dinosaurs: structure, relationships, and paleobiology. 1–630 (Columbia Univeristy Press, 2004).
Rose, K. D. The Beginning of the Age of Mammals. 1–428 (The Johns Hopkins University Press, 2006).
Clemens, W. A. Were immigrants a significant part of the earliest Paleocene mammalian fauna of the North American western interior? Vertebrata PalAsiatica 48, 285–307 (2010).
Black, K. H., Archer, M., Hand, S. J. & Godthelp, H. In Earth and Life, Global Biodiversity, Extinction Intervals and Biogeographic Perturbations Through Time (ed Talent, John A. ) 983–1078 (Springer, 2012).
Butler, P. M. & Kielan-Jaworowska, Z. Is Deltatheridium a marsupial? Nature (London) 245, 105–106 (1973).
Kielan-Jaworowska, Z. & Nessov, L. A. On the metatherian nature of the Deltatheroida, a sister group of the Marsupialia. Lethaia 23, 1–10 (1990).
Rougier, G. W., Wible, J. R. & Novacek, M. J. Implications of Deltatheridium specimens for early marsupial history. Nature 396, 459–463 (1998).
Luo, Z.-X., Ji, Q., Wible, J. R. & Yuan, C.-X. An Early Cretaceous tribosphenic mammal and metatherian evolution. Science 302, 1934–1940 (2003).
Rougier, G. W., Davis, B. M. & Novacek, M. J. A deltatheroidan mammal from the Upper Cretaceous Baynshiree Formation, eastern Mongolia. Cretaceous Research 52, Part A, 167–177 (2015).
Wilson, G. P. & Riedel, J. A. New specimen reveals deltatheroidan affinities of the North American Late Cretaceous mammal Nanocuris. J. Vert. Paleontol. 30, 872–884 (2010).
O’Leary, M. A. et al. The placental mammal ancestor and the post–K-Pg radiation of placentals. Science 339, 662–667 (2013).
Cifelli, R. L. In Mammal Plylogeny. Vol. 1. Mesozoic Differentiation, Multitubercula′tes, Monotremes, Early Therians, and Marsupials (eds Szalay, Frederick S., Novacek, Michael J. & McKenna, Malcolm C. ) 205–215 (Springer-Verlag, 1993).
Kielan-Jaworowska, Z. & Cifelli, R. L. Primitive boreosphenidan mammal (?Deltatheroida) from the Early Cretaceous of Oklahoma. Acta Palaeontologica Polonica 46, 377–391 (2001).
Li, Q. & Meng, J. New ctenodactyloid rodents from the Erlian Basin, Nei Mongol, China, and the phylogenetic relationships of Eocene Asian ctenodactyloids. Am. Mus. Novit. 3828, 1–58 (2015).
Tong, Y., Wang, Y. & Li, Q. Subdivision of the Paleogene in Lingcha area of Hunan Province and early Eocene mammalian faunas of China. Geological Review 52, 153–162 (2006).
Alifanov, V. R. In The Age of Dinosaurs in Russia and Mongolia (eds Benton, M. J., Shishkin, M. A., Unwin, D. M. & Kurochkin, E. N. ) 368–389 (Cambridge University Press, 2000).
Van Itterbeeck, J. et al. Woodland in a fluvio-lacustrine environment on the dry Mongolian Plateau during the late Paleocene: Evidence from the mammal bearing Subeng section (Inner Mongolia, P. R. China). Palaeogeography, Palaeoclimatology, Palaeoecology 243, 55–78 (2007).
Meng, J. et al. New material of Alagomyidae (Mammalia, Glires) from the late Paleocene Subeng locality, Inner Mongolia. Am. Mus. Novit. 3597, 1–29 (2007).
Missiaen, P. & Smith, T. The Gashatan (late Paleocene) mammal fauna from Subeng, Inner Mongolia, China. Acta Palaeontologica Polonica 53, 357–378 (2008).
Wang, Y. et al. Early Paleogene stratigraphic sequences, mammalian evolution and its response to environmental changes in Erlian Basin, Inner Mongolia, China. Science China Earth Science 53, 1918–1926 (2010).
Wang, Y., Meng, J., Ni, X. & Beard, K. C. A new early Eocene arctostylopid (Arctostylopida, Mammalia) from the Erlian Basin, Nei Mongol (Inner Mongolia), China. J. Vert. Paleontol. 28, 553–558 (2008).
Sun, B. et al. Magnetostratigraphy of the early Paleogene in the Erlian Basin. Journal of Stratigraphy 33, 62–68 (2009).
Tong, Y., Li, Q. & Wang, Y. An introduction to recent advance in the study of the continental early Paleogene stages in China. Journal of Stratigraphy 37, 428–440 (2013).
McKenna, M. C. & Bell, S. K. Classification of Mammals, above the species level. (Columbia University Press, 1997).
Crompton, A. W. & Kielan-Jaworowska, Z. In Development, Function and Evolution of Teeth (eds Butler, P. M. & Joysey, K. A. ) 249–287 (Academic Press, 1978).
Fox, R. C. Molar structure and function in the Early Cretaceous mammal Pappotherium: evolutionary implications for Mesozoic Theria. Can. J. Earth Sci. 12, 412–442 (1975).
Luo, Z.-X., Yuan, C.-X., Meng, Q.-J. & Ji, Q. A Jurassic eutherian mammal and divergence of marsupials and placentals. Nature 476, 442–445 (2011).
De Muizon, C. & Lange-BadrÉ, B. Carnivorous dental adaptations in tribosphenic mammals and phylogenetic reconstruction. Lethaia 30, 353–366 (1997).
Wignall, P. B. & Benton, M. J. Lazarus taxa and fossil abundance at times of biotic crisis Journal of the Geological Society 156, 453–456 (1999).
Gordon, C. L. A first look at estimating body size in dentally conservative marsupials. J. Mammal. Evol. 10, 1–21 (2003).
van Dyck, S. & Strahan, R. The Mammals of Australia. 3rd edition, 1–887 (New Holland Publishers, 2008).
Wilson, G. P. & Riedel, J. A. Mammals across the K/Pg boundary in northeastern Montana, USA: dental morphology and body-size patterns reveal extinction selectivity and immigrant-fueled ecospace filling. J. Vert. Paleontol. 30, 872–884 (2013).
Raup, D. M. & Sepkoski, J. J. Mass extinctions in the marine fossil record. Science 215, 1501–1503 (1982).
Jablonski, D. Mass extinctions and macroevolution. Paleobiology 31, 192–210 (2005).
Krug, A. Z. & Jablonski, D. Long-term origination rates are reset only at mass extinctions. Geology 40, 731–734 (2012).
Rougier, G. W., Wible, J. R., Beck, R. M. D. & Apesteguía, S. The Miocene mammal Necrolestes demonstrates the survival of a Mesozoic nontherian lineage into the late Cenozoic of South America. Proc. Natl. Acad. Sci. USA 109, 20053–20058 (2012).
Archibald, J. D., Zhang, Y., Harper, T. & Cifelli, R. L. Protungulatum, confirmed Cretaceous occurrence of an otherwise Paleocene eutherian (Placental?) Mammal. J. Mammal. Evol. 18, 153–161 (2011).
Beard, K. C. East of Eden: Asia as an important center of taxonomic origination in mammalian evolution. Bulletin of Carnegie Museum of Natural History 34, 5–39 (1998).
Beard, K. C. East of eden at the Paleocene/Eocene boundary. Science 295, 2028–2029 (2002).
Ni, X., Hu, Y., Wang, Y. & Li, C. A clue to the Asian origin of euprimates. Anthropological Science 113, 3–9 (2005).
Beard, C. K. et al. Paleocene Hapalodectes (Mammalia: Mesonychia) from Subeng, Nei Mongol: further evidence of “East of Eden” dispersal at the Paleocene-Eocene boundary. Vertebrata PalAsiatica 48, 375–389 (2010).
Smith, T. et al. First Clarkforkian equivalent Land Mammal Age in the latest Paleocene basal Sparnacian facies of Europe: fauna, flora, paleoenvironment and (bio)stratigraphy. PLoS ONE 9, e86229, doi: 10.1371/journal.pone.0086229 (2014).
Wang, Y. & Jin, X. A new Paleocene tillodont (Tillodontia, Mammalia) from Qianshan, Anhui, with a review of Paleocene tillodonts from China. Vertebrata PalAsiatica 42, 13–26 (2004).
Rougier, G. W., Wible, J. R. & Novacek, M. J. New specimen of Deltatheroides cretacicus (Metatheria, Deltatheroida) from the late Cretaceous of Mongolia. Bulletin of Carnegie Museum of Natural History 36, 245–266 (2004).
Averianov, A. O., Archibald, J. D. & Ekdale, E. G. New material of the Late Cretaceous deltatheroidan mammal Sulestes from Uzbekistan and phylogenetic reassessment of the metatherian-eutherian dichotomy. Journal of Systematic Palaeontology 8, 301–330 (2010).
Goloboff, P. A., Farris, J. S. & Nixon, K. C. TNT, a free program for phylogenetic analysis. Cladistics 24, 774–786 (2008).
Ni, X. et al. The oldest known primate skeleton and early haplorhine evolution. Nature 498, 60–64 (2013).
Maddison, W. P. & Maddison, D. R. Mesquite: A Modular System for Evolutionary Analysis. Version 3.03, http://mesquiteproject.org (2015).
Acknowledgements
This project has been supported by the Strategic Priority Research Program of Chinese Academy of Sciences (CAS, XDB03020000), the National Basic Research Program of China (2012CB821904), the CAS 100-talent Program, the National Natural Science Foundation of China (41472025). We are grateful to Yongchun Yang, Jie Ye, and Zhilu Tang for their assistance in the field. We thank Yemao Hou for their help during CT scanning. Guizhen Wang, Ran Li, Ge Li and Tingyu Zhao sorted the concentrates of the screen-washing.
Author information
Authors and Affiliations
Contributions
X.N. designed the study, analyzed the data and wrote the paper. T.S. studied the non-mammal vertebrate fossils, contributed to the discussion of geographic distribution and edited the manuscript. J.M. contributed to the mammalian fauna identification. Q.L., L.L. and X.L. contributed extensively to the work presented in this paper and edited the manuscript.
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Electronic supplementary material
Rights and permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Cite this article
Ni, X., Li, Q., Stidham, T. et al. A late Paleocene probable metatherian (?deltatheroidan) survivor of the Cretaceous mass extinction. Sci Rep 6, 38547 (2016). https://doi.org/10.1038/srep38547
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/srep38547
- Springer Nature Limited
This article is cited by
-
Tibetan Plateau: An evolutionary junction for the history of modern biodiversity
Science China Earth Sciences (2020)
-
Paleogene mammalian fauna exchanges and the paleogeographic pattern in Asia
Science China Earth Sciences (2020)