The northern region of Beringia is ecologically and biogeographically significant as a corridor for biotic dispersals between the Old and New Worlds. Large mammalian predators from Beringia are exceedingly rare in the fossil record, even though carnivore diversity in the past was much higher than it is in this region at present. Here we report the first fossils of cursorial hyenas, Chasmaporthetes, in Beringia and north of the Arctic Circle. Two isolated teeth recovered in the Old Crow Basin, Yukon Territory, Canada, were identified amongst over 50,000 known fossil mammal specimens recovered from over a century of collecting in the Old Crow Basin. These rare records fill an important intermediary locale in the more than 10,000 km geographic distance between previously known New and Old World records of this lineage. The Pleistocene age of these fossils, together with its Arctic Circle occurrence, necessitate a rethinking of the role of large-bodied hunter-scavengers in Ice Age megafaunas in North America, and the implications of lacking an important energy flow modifier in present day North American food webs.
Pleistocene vertebrate faunas from Beringia – the unglaciated region between northeast Asia and northwest North America straddling the Bering Isthmus – are of particular importance for our understanding of intercontinental dispersals (Hopkins et al., 1982). These faunas record the movement, persistence, and extinction during the Pleistocene and establishment of present-day guilds and food webs. Members of the family Hyaenidae (part of Order Carnivora, which also includes wolves, cats, seals, weasels, and relatives) have an extensive fossil record across the Old World beginning in the early Miocene, with >60 extinct hyaenid taxa having been described. They are often the most abundant predators at fossil localities across Eurasia, and are considered to have played key ecological roles as mesopredators and scavengers (Turner et al., 2008). They were, however, absent from North America until dispersal (presumably via the Beringian land bridge) during the Blancan North American Land Mammal Age (NALMA, Figure S1), ~4.7 Ma (mega-annum) ago. The hitherto northernmost record of the hyaenid Chasmaporthetes in North America is from Meade County, Kansas at approximately 37°N latitude. In this report, we describe the first known fossil records of hyaenids in the Arctic latitudes of Beringia, represented by two isolated teeth of Chasmaporthetes cf. C. ossifragus, a member of the cursorial hyaenid clade, adding to the exclusively southern United States and Mexico fossil records of hyenas in North America.
Geographic Region and Geologic Context—The Old Crow Basin, in the unglaciated region of northern Yukon Territory, Canada, has been the focus of paleontological, paleoenvironmental, and stratigraphic investigations for over a century (Harington, 1989, Harington, 2011, Irving and Beebe, 1984). Numerous exposures of unconsolidated, perennially frozen alluvial sediments are present along the Old Crow River and its tributaries, and reveal rich records of Quaternary biotic communities and environmental change. However, most fossils are recovered from numerous alluvial point bars, where admixture of anachronistic fossils from various Pleistocene and Holocene stratigraphic units makes geologic age constraint of the re-worked fossils difficult. One particular point bar near the downstream end of the basin known as Locality CRH 11A has been a focus for palaeontologists and is, by far, the richest and most diverse (Harington, 2011). Stratigraphic investigations on river bluffs, including biochronology, tephrochronology, and magnetostratigraphy suggest that exposed sediments in the basin are no older than early Pleistocene, ~1.4 Ma; thus, these re-worked fossils can be considered to span the Irvingtonian to Rancholabrean NALMA (Figure S1) (Westgate et al., 2017, Kuzmina et al., 2014).
Over 50,000 vertebrate fossils, representing at least 80 species from over 200 bluffs and point bar localities along Old Crow River, have been amassed in the collections of the Canadian Museum of Nature and Yukon Government Palaeontology Program. These faunas are dominated by common Quaternary Holarctic herbivores such as Mammuthus, Bison, Equus and Rangifer, whereas remains of large carnivorans including the bear Arctodus, and the felids Homotherium and Panthera are relatively rare (Harington, 2011). However, it is the exceedingly rare specimens from the Old Crow Basin that have proved most important. For example, a single fossil of the flat-headed peccary Platygonus compressus extended the known range for this species northward by over 3000 km (Beebe, 1980). Rare specimens of the canid Xenocyon lycaonoides provide a glimpse of a brief dispersal of this typically Eurasian carnivoran into the North American Arctic (Tedford et al., 2009). Rare specimens of the large bodied camel cf. Paracamelus provide an intermediary fossil locale for this taxon, which ranged from the Canadian High Arctic to western Europe (Rybczynski et al., 2013). Although the chronological uncertainties with these re-worked faunas are difficult to resolve, these proverbial “needle-in-a-hay-stack” rare fossils from the Old Crow Basin have provided major advances in our knowledge of Quaternary biotic history and biogeography.
CMN, Canadian Museum of Nature, Ottawa, Canada; F:AM, Frick fossil mammals collection, Division of Paleontology, American Museum of Natural History, New York, USA; IGM, Geological Institute, Mongolian Academy Sciences, Ulan Baatar, Mongolia; IPM, Paleontological Institute, Russian Academy of Sciences, Moscow, Russia; IVPP, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, Beijing, China; MAN, Museo Archeologico, Nuoro, Sardinia, Italy; MNHN, Muséum National d’Histoire Naturelle, Paris, France; NMT, National Museum of Tanzania; OGU, Odessa I.I. Mechnikov State University, Ukraine; PIU, Paleontology Collection, Museum of Evolution, Uppsala University, Sweden; PPM, Panhandle Plains Historical Museum, Canyon, Texas, USA; SAM, Iziko South African Museum, Cape Town, South Africa; Sk, Swartkrans Collection, Ditsong National Natural History Museum, Pretoria, South Africa; TM, Toros-Menalla locality, National Museum of Chad; UCB, Claude Bernard University, Lyon, France; UF, University of Florida Museum of Natural History, Gainesville, Florida, USA; USNM, United States National Museum (Smithsonian Institution)/National Museum of Natural History, Washington D.C., USA.
|Specimen No.||CMN 24958||CMN 38053|
|Width (anterior, mm)||9.94||13.12|
|Width (posterior, mm)||11.58||8.24 (partial)|
|L, mac (mm)||4.97||–|
|L. main cusp (mm)||10.38||–|
|Height, main cusp (mm)||13.34||–|
|L, dac (mm)||4.92||–|
|L. trigonid (mm)||–||22.25|
|L, talonid (mm)||–||6.41|
|H, paraconid (mm)||–||13.4|
|H, protoconid (mm)||–||15.55|
|HSB (b, m, t)||u, a, ?||u, a, a|
Referred specimens—CMN (Canadian Museum of Nature, Ottawa, Canada) 24958, isolated right p3; collected by Richard Harington and team from the Canadian Museum of Nature in 1973. CMN 38053, isolated left m1, collected by William Irving and Brenda Beebe of the University of Toronto in 1977 (Irving and Beebe, 1984).
Locality—Old Crown Basin CRH (C. Richard Harington) Locality 11A, Yukon Territory, Canada. The locality is a point bar along the lower Old Crow River (67.819° N, 139.904° W). The area is very remote, only accessible by plane to the village of Old Crow, and from there either on the river or by helicopter.
Age—Approximately Irvingtonian to Rancholabrean NALMA, or from about 1.4 Ma to 0.012 Ma (Harington, 2011). We interpret the most likely age of the Old Crow Chasmaporthetes to be no younger than the end of Irvington I (approximately 0.85 Ma; see discussion), likely closer to 1.4 Ma.
CMN 24958, right p3: The main cusp is slightly worn, and has distinct, smooth ridges running down the mesial and distal faces. A well-developed mesial accessory cusp (MAC) is separated from the main cusp by a pronounced notch. No mesial cingulum is visible. A well-developed distal accessory cusp (DAC) is similar in size to the MAC, with a similarly notched boundary with the main cusp. Bulging cingula are developed on the buccal and labial sides of the DAC. The distal face of the DAC forms a depression on the distal edge of the tooth, separating the lingual and labial cingula. Vertical surface furrows are visible throughout the tooth, being most pronounced on the surface of the main cusp.
CMN 38053, left m1: A moderate shear facet is developed on the carnassial blade, exposing dentine within a rim of enamel. The protoconid is taller than the paraconid; both cusps are only slightly worn at the apices. There is no metaconid. A low cingulum is present around the base of the crown. The talonid has a single centrally situated cuspid, the hypoconid, with a raised ridge connecting its apex to the base of the protoconid forming a notch at their contact point. There are three small accessory cuspids formed on the lingual to distal rim of the talonid basin. The labial edge of the talonid is broken, so the possible presence of additional cuspids cannot be determined. For additional details see Electronic Supplemental Material (ESM).
CMN 24958 has an overall black sheen. Small regions of milky yellow to white preservation allows Hunter-Schreger bands to be observed under a dissection microscope at 30–40× magnification. The specimen is broken off at the base of the crown. Two roots are visible, with the distal being larger than the mesial one. Preservation of CMN 38053 is similar to CMN 24958. There are two broken roots at the base of the crown, with the mesial being the larger. Both CMN 38053 and CMN 24958 exhibit transitional, acute-undulating Hunter-Schreger Band enamel microstructure, indicating moderate capability to resist abrasive and/or hard foods (Tseng, 2011).
The nearly symmetrical arrangement of p3 MAC and DAC around the main cusp, the linear alignment of the three cusps on p3 in occlusal view, and the centrally situated and trenchant hypoconid in the talonid basin are all characteristic of Chasmaporthetes, and are present on the referred specimens from Old Crow. Compared to specimens of C. ossifragus and other species of Chasmaporthetes, the p3 from Old Crow is relatively wide for its length (Figure 2A; Table S1). Lower first molar (m1) dimensions of the Old Crow specimen fall in line with the proportions observed in the global Chasmaporthetes sample (Figure 2B; Table S1). However, the m1 trigonid of the Old Crow specimen is relatively broad and short compared to the majority of the Chasmaporthetes measured (Figure 2C; Table S1). The MAC on CMN 24958 is relatively larger than the MAC on PPM (Panhandle-Plains Historical Museum) 2343 (C. ossifragus, Cita Canyon, Texas) and UF (Florida Museum of Natural History, University of Florida) 18088 (C. ossifragus, Inglis 1A, Florida), whereas the m1 trigonid and talonid proportions are similar among all three specimens, as well as to the isolated m1 from Inglis 1A (UF 18089).
Because there are only two teeth from the Old Crow Basin in this report, other characteristics such as degree of curvature of the dentition and p4 to m1 length ratios cannot be used to assign the specimens to an existing species, nor are there enough characters to warrant naming a new species. The main dental feature that has been described as characteristic of C. ossifragus (Berta, 1981) is consistent with our observations of the Old Crow specimen: a well-developed MAC is present in the Old Crow p3 (CMN 24957), similar to C. ossifragus specimens from Florida, and relatively larger in size compared to p3 MAC in other Chasmaporthetes species, especially Old World species such as C. lunensis. Following previous reviews of dental (Kurtén and Werdelin, 1988), and lacking more specific evidence to unambiguously assign the Old Crow fossils to any one of existing species of Chasmaporthetes other than from geographic delineation, we reserve judgment on species designation and instead identify the new fossils as Chasmaporthetes cf. C. ossifragus.
New records from the Old Crow Basin extend the geographic range of Chasmaporthetes in North America northward by more than 4,000 km (~2,500 miles) from the hitherto known localities ranging from the southern United States to central Mexico. Furthermore, the Old Crow locality is roughly 6,500 km (~4,000 miles) from the nearest Old World Chasmaporthetes locality in Shamar, Mongolia (Tseng et al., 2013, Qiu, 1987), dramatically reducing the distance between the Old and New World occurrences of the genus (Figure 2D). The occurrences from Old Crow provide the first physical evidence of the long-speculated Beringian dispersal route of Chasmaporthetes from northern Asia into North America.
The Nearctic occurrence of Chasmaporthetes during the Pleistocene suggests that the genus may have ranged throughout much of North America from its initial arrival during the Pliocene, to the Early Pleistocene. Although the majority of the fossil records of North American Chasmaporthetes are Blancan in age, the Mexican locality of El Golfo is Irvingtonian. Given the well-documented Middle and Late Pleistocene fossil records in North America, and the absence of any hyaenid specimens younger than Irvingtonian even at extremely well-sampled sites such as Rancho La Brea, or in Rancholabrean age faunas of central Yukon and Alaska (Guthrie, 1990, Harington, 2011), it is unlikely that the geologic age of the Old Crow Chasmaporthetes is much younger than around 1.4 Ma.
An analysis of stratigraphic ranges of similarly sized and/or coeval species that have been interpreted as hunter-scavengers suggests that potential competitors of Chasmaporthetes in North America include the canids Borophagus, Canis armbrusteri, and Xenocyon lycaonoides, and the short-faced ursid Arctodus simus (Figure 2E). Of these, Borophagus had the longest temporal overlap with Chasmaporthetes. Based on the timing of last occurrences of Borophagus and Chasmaporthetes, and the first occurrences of Canis armbrusteri and Arctodus simus, we hypothesize that competition between these two pairs of predators were in part responsible for the extinction of Borophagus and Chasmaporthetes.
Blancan Competitors—Although postcranial remains of Chasmaporthetes suggest that they were likely the most cursorial of the hyaenids (Berta, 1981, Tseng et al., 2013), skull biomechanical traits indicate that some species were capable bone-crackers (Tseng et al., 2011). This would have placed Chasmaporthetes into potential competition with the morphologically robust bone-cracking and bone-consuming borophagine canids from 4.7 to ~1.4 Ma (Tseng and Wang, 2011, Tseng and Wang, 2010, Wang et al., 2018, Werdelin, 1989) (Figure 2E). For most of the Blancan NALMA (4.7 to 1.4 Ma), Chasmaporthetes and the canid Borophagus were both found in the southern U.S. and into Mexico (Berta, 1981, Wang et al., 1999), with Borophagus going extinct in the late Blancan (Tseng and Geisler, 2016). This suggests that Chasmaporthetes eventually outlasted and/or outcompeted borophagine canids in North America. However, the two lineages coexisted in North America for almost three million years, so any potential competition may have been dampened by finer scale niche differentiation or lack of extensive geographic overlap. Out of ten Blancan-aged localities with Chasmaporthetes fossils, four of them (Benson, AZ, Cita Canyon, TX, Dry Mountain, AZ, and Santa Fe, FL) also contain Borophagus fossils (Berta, 1981, Wang et al., 1999).
Irvingtonian Competitors—The first large-bodied, wolf-like species to appear in North America was Canis armbrusteri, a possible predecessor to the extinct dire wolves (C. dirus) and/or extant gray wolves (C. lupus). Chasmaporthetes became extinct in North America within about 0.6 million years after the first occurrence datum of C. armbrusteri (Tedford et al., 2009). Shortly after the appearance of C. armbrusteri during the beginning of Irvingtonian I (~1.4 to ~0.85 Ma), the giant short-faced bear Arctodus simus appeared (Figure 2E). Whereas potential competition between C. armbrusteri and Chasmaporthetes is possible given comparable body sizes, A. simus would have completely dominated any carcass or kill site because of their very large size. Furthermore, evidence from functional morphology studies suggests that A. simus may not have been a specialized scavenger as previously inferred (Figueirido et al., 2010). Canis armbrusteri is known from early Irvingtonian sites in Arizona, but is not found in Chasmaporthetes localities. Arctodus localities are known throughout North America (including in Old Crow and Alaska), but co-occurrence with Chasmaporthetes in the Old Crow cannot be confirmed because temporal constraints on the appearance of Arctodus in the Arctic are not resolved (Harington, 2011). The hunting dog-like Xenocyon lycaonoides made a brief appearance in the New World during the latter half of Irvingtonian I as an immigrant from Eurasia; its occurrence in Old Crow is suggestive of potential sympatry with Chasmaporthetes, but its record in Beringia is too poor for confident conclusions (Tedford et al., 2009). Two common elements of later Pleistocene and Holocene megafaunas, Canis dirus and Canis lupus, appeared too late (during Irvingtonian II) to have been in potential contact or competition with Chasmaporthetes.
Lastly, as recently suggested (Tseng et al., 2013), the relative rarity of complete Chasmaporthetes fossil materials even in some of the richest carnivoran fossil localities, is indicative of relatively low population densities. The presence of only two Chasmaporthetes specimens from the Old Crow Basin establishes this record as a critically important new datum in the Arctic for understanding the evolution of Hyaenidae and Pleistocene turnover patterns leading to present-day ecological communities. Together with the extinction of bone-cracking borophagine canids prior to the end-Pleistocene megafaunal extinctions, the Early Pleistocene extinction of Chasmaporthetes contributed to a permanent change in the profile of the North American predator guild. The effects of losing these specialized bone-cracking hunter-scavengers likely had an impact on energy flow in North American food webs (DeVault et al., 2003, Wilson and Wolkovich, 2011), perhaps in a way similar to trophic cascades in African ecosystems following extirpation of large predators there (Atkins et al., 2019). It remains to be quantified to what extent potential successors (Arctodus and Canis) played comparable ecological roles in North America after bone-crackers disappeared, or whether this loss of a key ecological player was detrimental to ecosystem stability leading up to the megafaunal extinctions.
All data used in the study are included in the publication.
Kieran Shepherd and Margaret Currie provided access to CMN collections and microscope and photography support; X. Arura Tseng assisted with collection study notes; we thank the Vuntut Gwitchin First Nation for their longstanding and continued support of paleontology research in the Old Crow Basin; we are grateful to Richard “Dick” Harington, Gerry Fitzgerald and Charlie Thomas for the discovery of CMN 24958 and to Brenda Beebe and William Irving for collecting CMN 38058; we thank editor Hanneke Meijer and reviewers Manuel Salesa and Julie Meachen for constructive comments on earlier versions of the manuscript, and Josh Samuels for tracking down possible additional fossil occurrences; The Florida Museum of Natural History made photographs of Florida Chasmaporthetes specimens available online for reference.
The authors have no funding source to declare.
The authors have no competing interests to declare.
ZJT, GZ, LW designed the study and wrote the manuscript; ZJT made the figures and table; all authors approved the final version of the manuscript for publication.
Atkins, JL, Long, RA, Pansu, J, Daskin, JH, Potter, AB, Stalmans, ME, Tarnita, CE and Pringle, RM. 2019. Cascading impacts of large-carnivore extirpation in an African ecosystem. Science, eaau3561. DOI: https://doi.org/10.1126/science.aau3561
Beebe, BF. 1980. Pleistocene peccary, Platygonus compressus LeConte, from the Yukon Territory, Canada. Canadian Journal of Earth Sciences, 17: 1204–1209. DOI: https://doi.org/10.1139/e80-126
Berta, A. 1981. The Plio-Pleistocene hyaena Chasmaporthetes ossifragus from Florida. Journal of Vertebrate Paleontology, 1: 341–356. DOI: https://doi.org/10.1080/02724634.1981.10011905
Devault, TL, Rhodes, J, Olin, E and Shivik, JA. 2003. Scavenging by vertebrates: behavioral, ecological, and evolutionary perspectives on an important energy transfer pathway in terrestrial ecosystems. Oikos, 102: 225–234. DOI: https://doi.org/10.1034/j.1600-0706.2003.12378.x
Figueirido, B, Pérez-Claros, JA, Torregrosa, V, Martín-Serra, A and Palmqvist, P. 2010. Demythologizing Arctodus simus, the ‘short-faced’ long-legged and predaceous bear that never was. Journal of Vertebrate Paleontology, 30: 262–275. DOI: https://doi.org/10.1080/02724630903416027
Guthrie, RD. 1990. Frozen fauna of the mammoth-steppe, the Story of Blue Babe. Chicago: University of Chicago Press. DOI: https://doi.org/10.7208/chicago/9780226159713.001.0001
Harington, CR. 1989. Pleistocene Vertebrate Localities in the Yukon. In: Carter, LD, Hamilton, TD and Galloway, JP. (eds.), Late Cenozoic History of the Interior Basins of Alaska and the Yukon. US Geological Survey.
Harington, CR. 2011. Pleistocene vertebrates of the Yukon Territory. Quaternary Science Reviews, 30: 2341–2354. DOI: https://doi.org/10.1016/j.quascirev.2011.05.020
Kurtén, B and Werdelin, L. 1988. A review of the genus Chasmaporthetes Hay, 1921 (Carnivora, Hyaenidae). Journal of Vertebrate Paleontology, 8: 46–66. DOI: https://doi.org/10.1080/02724634.1988.10011683
Kuzmina, S, Froese, DG, Jensen, BJL, Hall, E and Zazula, GD. 2014. Middle Pleistocene (MIS 7) to Holocene fossil insect assemblages from the Old Crow basin, northern Yukon, Canada. Quaternary International, 341: 216–242. DOI: https://doi.org/10.1016/j.quaint.2013.10.025
Rybczynski, N, Gosse, JC, Richard Harington, C, Wogelius, RA, Hidy, AJ and Buckley, M. 2013. Mid-Pliocene warm-period deposits in the High Arctic yield insight into camel evolution. Nature Communications, 4: 1550. DOI: https://doi.org/10.1038/ncomms2516
Tedford, RH, Wang, X and Taylor, BE. 2009. Phylogenetic systematics of the North American fossil Caninae (Carnivora: Canidae). Bulletin of the American Museum of Natural History, 325: 1–218. DOI: https://doi.org/10.1206/574.1
Tseng, ZJ. 2011. Variation and implications of intra-dentition Hunter-Schreger band pattern in fossil hyaenids and canids (Carnivora, Mammalia). Journal of Vertebrate Paleontology, 31: 1163–1167. DOI: https://doi.org/10.1080/02724634.2011.602161
Tseng, ZJ, Anton, M and Salesa, MJ. 2011. The evolution of the bone-cracking model in carnivorans: cranial functional morphology of the Plio-Pleistocene cursorial hyaenid Chasmaporthetes lunensis (Mammalia: Carnivora). Paleobiology, 37: 140–156. DOI: https://doi.org/10.1666/09045.1
Tseng, ZJ and Geisler, JH. 2016. The first fossil record of borophagine dogs (Mammalia, Carnivora) from South Carolina, USA. Journal of Vertebrate Paleontology, 36: e1062022. DOI: https://doi.org/10.1080/02724634.2015.1062022
Tseng, ZJ, Li, Q and Wang, X. 2013. A new cursorial hyena from Tibet, and analysis of biostratigraphy, paleozoogeography, and dental morphology of Chasmaporthetes (Mammalia, Carnivora). Journal of Vertebrate Paleontology, 33: 1457–1471. DOI: https://doi.org/10.1080/02724634.2013.775142
Tseng, ZJ and Wang, X. 2010. Cranial functional morphology of fossil dogs and adaptation for durophagy in >Borophagus and Epicyon (Carnivora, Mammalia). Journal of Morphology, 271: 1386–1398. DOI: https://doi.org/10.1002/jmor.10881
Tseng, ZJ and Wang, X. 2011. Do convergent ecomorphs evolve through convergent morphological pathways? Cranial shape evolution in fossil hyaenids and borophagine canids (Carnivora, Mammalia). Paleobiology, 37: 470–489. DOI: https://doi.org/10.1666/10007.1
Turner, A, Antón, M and Werdelin, L. 2008. Taxonomy and evolutionary patterns in the fossil Hyaenidae of Europe. Geobios, 41: 677–687. DOI: https://doi.org/10.1016/j.geobios.2008.01.001
Wang, X, White, SC, Balisi, M, Biewer, J, Sankey, J, Garber, D and Tseng, ZJ. 2018. First bone-cracking dog coprolites provide new insight into bone consumption in Borophagus and their unique ecological niche. eLife, 7: e34773. DOI: https://doi.org/10.7554/eLife.34773
Werdelin, L. 1989. Constraint and adaptation in the bone-cracking canid Osteoborus (Mammalia: Canidae). Paleobiology, 15: 387–401. DOI: https://doi.org/10.1017/S009483730000957X
Westgate, JA, Pearce, GW, Preece, SJ, Schweger, CE, Morlan, RE, Pearce, NJG and Perkins, TW. 2017. Tephrochronology, magnetostratigraphy and mammalian faunas of Middle and Early Pleistocene sediments at two sites on the Old Crow River, northern Yukon Territory, Canada. Quaternary Research, 79: 75–85. DOI: https://doi.org/10.1016/j.yqres.2012.09.003
Wilson, EE and Wolkovich, EM. 2011. Scavenging: how carnivores and carrion structure communities. Trends in Ecology & Evolution, 26: 129–135. DOI: https://doi.org/10.1016/j.tree.2010.12.011