Walruses, Sea Lions and Fur Seals

Adaptation to a primarily aquatic lifestyle has happened numerous times within mammals, but some groups have radiated more in this environment than others. One particularly well-known group of marine mammals is the pinnipeds, the seals and sea lions.

Australian sea lions Neophoca cinerea on a beach on Kangaroo Island, copyright Diver Dave.

Pinnipeds are highly modified for life in the water, with streamlined bodies and all four limbs modified into flippers. When I was young, many of the animal books that I read referred to pinnipeds as their own distinct order within the mammals. However, it has long been recognised that pinnipeds are derived from within the Carnivora and these days they are almost universally treated as a subgroup of the latter. Modern pinnipeds are divided between three families: the Phocidae ('true' seals), Otariidae (fur seals and sea lions) and Odobenidae (which has only one living species, the walrus Odobenus rosmarus). While some morphological analyses have argued for a relationship between the walrus and the Phocidae, the majority view treats the walrus and the Otariidae as together forming a clade Otarioidea, commonly referred to as the eared seals. There has historically also been some argument about whether the pinnipeds represent a single clade; some have argued for two separate origins, Otarioidea being related to bears whereas Phocidae were supposed to be closer to otters and weasels. However, the current majority supports a single origin for the group.

Northern fur seals Callorhinus ursinus, photographed by M. Boylan.

Eared seals differ from true seals in the possession of small external ears, and the ability to turn the hind flippers back under the body so that they can still function (if somewhat awkwardly) as feet when moving on land. I have seen Australian sea lions on coastal islands near Perth (there are boat tours that will take you to see them) and I can confirm that they can run along the beach at a surprising speed when they wish to. True seals have the hind flippers permanently directed behind them and so are forced to awkwardly belly-flop along when not swimming (doubtless as a result of this, true seals also differ from eared seals in that males lack an external scrotum). In the water, the hind flippers provide the main source of propulsion in true seals whereas eared seals get more of their thrust from the fore flippers (sea lions have been said to swim like penguins). As an aside, eared seals are also apparently unusual among mammals in that their milk completely lacks lactose. The lactose intolerant among you need not be denied dairy, you need only milk a walrus.

Mounted skeleton of Allodesmus sp., copyright Momotarou2012.

The earliest eared seals are known from the Miocene when they appear to have originated in the northern Pacific. Two extinct families from this place and period, the Enaliarctidae and Desmatophocidae, are commonly included in the Otarioidea, though it remains possible that either of these families should be placed outside the pinniped crown group, or closer to the true seals. The early Miocene Enaliarctidae differ from other otarioids in retaining differentiated premolars and molars (later forms have the cheek teeth uniform in appearance) and may well represent the ancestral form of the group. The mid- to late Miocene Desmatophocidae combined a rather Phocidae-like skull with a more Otarioidea-like post-cranium; the best-known genus Allodesmus had larger eyes than other otarioids and may have hunted in deep waters. One species of desmatophocid, Allodesmus sinanoensis, may have reached a length approaching five metres, making it larger than a modern walrus and rivalling the elephant seals in size. I highly recommend a series of posts on Allodesmus written a few years back by Robert Boessenecker (1, 2, 3, 4) that cover just about everything you might want to know about this animal.

Skull of Gomphotaria pugnax, from Robert Boessenecker.

Though only one walrus species is generally recognised in the modern fauna, the family was much more diverse in the past. However, most fossil Odobenidae lacked the tusks of a modern walrus and would have been more similar at a glance to sea lions. These early odobenids would have probably been generalist fish-feeders (Boessenecker & Churchill 2013). The modern walrus, in contrast, feeds primarily on bivalves. They don't crush the clam's shell but grab it with their lips and then suck powerfully enough that the meat is ripped out. Other than the tusks, the teeth of a modern walrus are small and weak; one close fossil relative, the Pliocene Valenictus chulavistensis, went so far as to lose the non-tusk teeth entirely. The tusks themselves are usually thought to function in display and the like rather than having any prominent role in feeding. However, it is an intriguing detail that the fossil whale Odobenocetops that converged in its feeding biology with walruses also possessed a large tusk. The non-tusk teeth were still used in feeding in the fossil clam-feeding walrus genera Dusignathus and Gomphotaria, which had a pair of large forward-directed tusks in both the upper and lower jaws.

Suckling South African fur seals Arctocephalus pusillus, copyright Robur.q.

The majority of living eared seals belong to the Otariidae, which have been divided in the past between the fur seals and sea lions. Fur seals tend to be smaller than sea lions and possess a dense layer of underfur. However, more recent phylogenetic studies (particularly molecular ones) have thrown this distinction out the window (e.g. Higdon et al. 2007). Instead, the northern fur seal Callorhinus ursinus of the north Pacific is probably the sister species to all other living otariids. Even the southern fur seals, generally placed in a single genus Arctocephalus, may not be monophyletic relative to the New Zealand sea lion Phocarctos hookeri (as a result, some authors have suggested resurrecting the genus Arctophoca for all southern fur seals other than the South African fur seal Arctocephalus pusillus). The South American fur seal Otaria flavescens may also be associated with this latter group. The two north Pacific sea lions, Steller's sea lion Eumetopias jubatus and the Californian sea lion Zalophus californianus, form a clade outside the southern otariids. The remaining species is the Australian sea lion Neophoca cinerea whose position has been harder to pin down: some analyses place it close to the New Zealand sea lion but others position it well away from all other southern otariids, possibly even outside all other otariids except the northern fur seal.

Walruses Odobenus rosmarus crowded on shore, from here.

Fur seals and sea lions were heavily hunted in the past for pelts and oil and some species remain endangered. Climate change poses a particular threat to cold-water species; for instance, recent years have seen significant contractions in walrus ranges, leading to dramatic crowding in the locations remaining. Conversely, the Antarctic fur seal Arctocephalus gazella, once feared extinct, has apparently exhibited a population explosion in recent decades, perhaps because lowered whale populations have led to more food being available for seals.

REFERENCES

Boessenecker, R. W., & M. Churchill. 2013. A reevaluation of the morphology, paleoecology, and phylogenetic relationships of the enigmatic walrus Pelagiarctos. PLoS One 8 (1): e54311.

Higdon, J. W., O. R. P. Bininda-Emonds, R. M. D. Beck & S. H. Ferguson. 2007. Phylogeny and divergence of the pinnipeds (Carnivora: Mammalia) assessed using a multigene dataset. BMC Evolutionary Biology 7: 216.

Repenning, C. A., & R. H. Tedford. 1977. Otarioid seals of the Neogene. Geological Society Professional Paper 992: i–vi, 1–93, 24 pls.

Doe, it's Deer

Marsh deer Blastocerus dichotomus, copyright Jonathan Wilkins. An animal that just screams out, "Am I wearing the Chanel boots? Yes, I am."

I hardly need to explain what deer are, do I? Deers (Cervidae) are generally recognised as the second most diverse family of hoofed mammals (after bovids) in the modern fauna. Their most recognisable feature, of course, is the possession of antlers: bony cranial appendages that are shed and regrown every year rather than being permanently in place like the horns of a bovid. When antlers first grow, they are covered with a layer of skin (the velvet) that supplies them with blood, but this skin is later shed to expose the bare bone. In most species, antlers are only grown by males whose use them in conflicts during the mating season. The only genus of deer that grows antlers in both sexes is Rangifer, the reindeer. There is also one living species that lacks antlers, the Chinese water deer Hydropotes inermis; instead of antlers, males of this species possess large, dagger-like canines. In the majority of deer species, antlers are subcylindrical and often branched but broad palmate antlers have evolved on multiple occasions within the family. Antler morphology is generally significant in distinguishing taxa but it should be noted that variation within species is not unknown. For instance, the few recorded males of the small, now possibly extinct population of moose introduced to the south of New Zealand lacked the large palmate antlers generally associated with the species, probably due to poor nutritional conditions. Instead, they had more slender antlers that only became moderately palmate distally, like those of a fallow deer Dama dama.

One of the few photographs of moose from New Zealand, from here. I think this might be the one shot at Herrick Creek in 1952 but I could be wrong.

The first antlered deer are known from Europe back in the early Miocene, about 17 million years ago. They are not known from North America until some time later in the Pliocene, about five mya (Pitra et al. 2004), though these days they are every bit as diverse in the Americas as in Eurasia. They never made much inroad into Africa, only extending into the northernmost part of the continent, and they never made it into Australasia under their own steam, though a number of species have been dispersed to various parts of the world by humans. For instance, at least half a dozen species have become established in New Zealand, and until recently reindeer might be found wandering among penguin colonies in South Georgia.

Reindeer and king penguins on South Georgia, from here.

Recent decades have seen some pretty wild swings in cervid taxonomy, with the number of subfamilies recognised varying from two to seven, and some authors recognising a much larger number of genera and species than others. However, our general understanding of cervid interrelationships is pretty good these days, with many differences between systems being a question of ranking more than anything else. Recent studies have agreed that modern deer can be divided between two primary lineages that may be called the Cervinae and Capreolinae (Gilbert et al. 2006). The Cervinae include the majority of deer species in the Old World with a single species (the wapiti Cervus canadensis) extending its range into the New World. The remaining New World deer all belong to the Capreolinae, which also includes four genera (Rangifer, Hydropotes, the roe deer Capreolus and the moose Alces) found in Eurasia.

Male tufted deer Elaphodus cephalophus, copyright Heush.

The Cervinae can be divided between two tribes, the Muntiacini and Cervini. Muntiacini include the muntjaks of the genus Muntiacus and the tufted deer Elaphodus cephalophus. These are small deer native to southern and eastern Asia. Antlers are small and simple in all Muntiacini: muntjaks have antlers with only a single short anterior branch whereas the tufted deer has unbranched antlers that are barely visible under the large tuft of hair that this species has on top of the head. Muntiacini also resemble Hydropotes in their possession of large canines in the males. The other tribe, Cervini, includes larger deer species with more complex, multi-branched antlers. Some authors have historically placed all species of Cervini within a single genus Cervus; others may recognise nine distinct genera. Numbers of recognised species have also varied, largely due to phylogenetic studies finding that taxa previously recognised as conspecific subspecies may be more distantly related to each other or may not form monophyletic units. For instance, the wapiti has often been regarded as a subspecies of the red deer Cervus elaphus but recent studies have suggested that it is more closely related to the sika C. nippon and the white-lipped deer Przewalskium or Cervus albirostris, two east Asian species (Pitra et al. 2004). Difficulties in elucidating cervin phylogeny are probably best exemplified by the case of Père David's deer Elaphurus or Cervus davidianus, originally native to southern China but now only surviving in captivity. Molecular phylogenies associate this species closely with the brow-antlered deer Cervus eldi but it has many morphological features indicating a close relationship with C. elaphus, and it is widely suspected that Père David's deer originated from a hybridisation event between the two latter species.

Pudu (I think a northern pudu Pudu puda), copyright Neil McIntosh.

The Capreolinae can be divided between three main lineages. One comprises the roe deer and water deer; another comprises the moose (again, I'm making a point of referring to genera rather than species because the number of recognised species may differ between authors). Note that the position of the water deer suggests that the antler-less state of this species represents a secondary loss rather than retention of a primitive state. The majority of capreolines belong to the third lineage, commonly recognised as the tribe Odocoileini. Except for the reindeer, the species of this lineage are restricted to the New World, with the higher diversity in South America. The Odocoileini are perhaps the most taxonomically uncertain section of the deer family. There appears to be no question, at least, that Rangifer represents the sister group of all other Odocoileini. The remaining odocoileins have generally been divided between six genera: Odocoileus (including the mule deer O. hemionus and white-tailed deer O. virginianus), Mazama (brockets), Pudu (pudus), Hippocamelus (guemuls), the marsh deer Blastocerus dichotomus and the pampas deer Ozotoceros bezoarticus. However, except for the two monotypic genera, monophyly of all these taxa was placed in question by a recent molecular phylogenetic study of the group by Gutiérrez et al. (2017). The issue is particularly marked for the brockets, small deer with unbranched antlers, as not only the genus as a whole but also species within the genus have been indicated as non-monophyletic. Recent years have, as a result, seen something of a burst of new brocket species being described. It is quite probable that a similar taxonomic explosion may be in line for the genus Odocoileus, with both of the currently recognised species including a number of subspecies and each being of suspect monophyly. Matters are further complicated by the possibility of hybridisation between the two 'species'.

REFERENCES

Gilbert, C., A. Ropiquet & A. Hassanin. 2006. Mitochondrial and nuclear phylogenies of Cervidae (Mammalia, Ruminantia): systematics, morphology, and biogeography. Molecular Phylogenetics and Evolution 40 (1): 101–117.

Pitra, C., J. Fickel, E. Meijaard & P. C. Groves. 2004. Evolution and phylogeny of Old World deer. Molecular Phylogenetics and Evolution 33: 880–895.

The European Blackbuck

Horn cores of Gazellospira torticornis hispanica, from here.

From one -spira genus to another, somewhat different one. Gazellospira is a genus of spiral-horned gazelles known from the Pliocene and early Pleistocene of Europe and northern Asia (the reference to Miocene on the Wikipedia page for this genus looks like it might be an error). Most of the known specimens of Gazellospira have been assigned to a single species, G. torticornis, but a second species G. gromovae has been named from the lower Pleistocene of Tadzhikistan. Specimens from the Upper Pliocene of Spain have also been assigned to a distinct subspecies G. torticornis hispanica on the basis of their smaller size than G. torticornis from elsewhere (Garrido 2008).

Gazellospira is a close relative of the modern blackbuck Antilope cervicapra of southern Asia and would have resembled it in overall appearance. The most obvious difference between the two is probably the horns which diverge at a much greater angle in Gazellospira than in the blackbuck (at an eyeball estimate, the angle between the horns in Gazellospira looks to be close to 90° versus closer to 45–60° in Antilope). Like the blackbuck, Gazellospira was probably a more or less mixed feeder, alternating between browsing and grazing as the seasons required.

Gazellospira's eventual extinction was probably connected to the cooling climate of the Pleistocene (the related genus Gazella, which survives to the present in more southerly regions, disappeared from Europe at about the same time). It seems to have been gone from the greater part of Europe by the end of the Pliocene (Crégut-Bonnoure 2007), persisting into the Pleistocene as remnant populations in Iberia, Italy, Greece and central Asia.

REFERENCES

Crégut-Bonnoure, E. 2007. Apport des Caprinae et Antilopinae (Mammalia, Bovidae) à la biostratigraphie du Pliocène terminal et du Pléistocène d’Europe. Quaternaire 18 (1): 73–97.

Garrido, G. 2008. Lu muestra más moderna y completa conocida de Gazellospira torticornis (Bovidae, Artiodactyla, Mammalia en el Plioceno superior terminal de Europa occidental (Fonelas P-1, Cuenca de Guadix, Granada). Cuadernos del Museo Geominero 10: 413–460.

Hyopsodontids: Little Slinkers of the Palaeogene

The oft-repeated quote about mammalian palaeontology is that it tends to be focused on "the tooth, the whole tooth, and nothing but the tooth". This is primarily the result of pragmatic constraints: because they are much harder than the other bones of the mammalian skeleton, teeth are much more likely to be preserved in the fossil record. There are a great many fossil mammals for which the teeth remain pretty much the only part of the animal known. However, there is no question that this focus tends to limit our understanding of mammalian evolution. On the one hand, the complex morphology of many mammalian teeth means that they provide a wealth of characters for analysis. On the other, the morphology of teeth is heavily influenced by their bearer's diet and lifestyle, meaning that phylogenetically informative features are probably outweighed by the products of ecological convergence.

Reconstruction of Hyopsodus from Savage & Long (1978), via here.

All of which is pretty important background to keep in mind for any discussion of the Hyopsodontidae, a group of small (mostly rat- or weasel-sized) mammals recognised from the Palaeogene, the early part (Palaeocene and Eocene epochs) of the Caenozoic era. Hyopsodontids are generally assigned to the 'Condylarthra', a group of mammals that has long been recognised as one of the classic examples of a 'wastebasket taxon'. Condylarths were originally united as primitive relatives of the ungulates, the hoofed mammals. However, the individual condylarth families themselves have not got much in common otherwise, and (particularly with the current acceptance that 'ungulates' are probably not a monophyletic group) it is hard to come up with a definition for 'condylarths' that amounts to much more than 'medium-sized, unspecialised Palaeogene placentals'.

The hyopsodontids have been recognised as one of the longest-lived groups of 'condylarths', with assigned members extending from close to the start of the Palaeocene right up to near the end of the Eocene. Here again, though, we come up against the question of definition. The majority of taxa that have been aligned with the hyopsodontids are among those known only from teeth. Features of the hyopsodontid dentition include fairly simple incisors and premolars, small canines, and molars that are more or less bunodont (that is, the cusps are rounded or conical and clearly separate from each other rather than being connected by lophs). The problem is that these are all primitive, unspecialised features. Hyopsodontids are therefore defined more by their lack of alternate specialisations than anything positive, making them something of a wastebasket within a wastebasket.

Until recently, the only hyopsodontid known from much in the way of postcranial material was the type genus Hyopsodus, a number of species of which are known from an extended period of the Eocene. Indeed, Hyopsodus was one of the most abundant mammalian genera of its time, accounting for over a quarter of mammalian remains in a number of deposits where it is found (Rose 2006). These remains combine to give a picture of Hyopsodus as a long, low-bodied animal that has been compared in its proportions to a dachshund, a weasel, or a prairie dog. Hyopsodus would have been a ground-hugging slinker of an animal, build for concealment rather than speed. Short claws on the forelegs may be consistent with a certain degree of digging ability, whether in search of buried tubers or to scrape shallow burrows. Overall, Hyopsodus was probably a generalist, able to make a living wherever it may find itself: the real rat of the Eocene.

It was only relatively recently that Penkrot et al. (2008) provided further descriptions of limb-bone material from two other genera associated with the hyopsodontids, Apheliscus and Haplomylus. And despite the dental similarities between these genera and Hyopsodus, their postcranial anatomy indicates a quite different animal. Though small, the apheliscines were relatively long-legged, speedy runners: sprinters rather than slinkers. Penkrot et al. interpreted the apheliscines as relatives of the modern elephant shrews of Africa; whether or not there was a valid phylogenetic connection, the two would have certainly been ecologically similar.

Both Hyopsodus and Apheliscus were included in the broad-scale analysis of early placental phylogeny by Halliday et al. (2017). The results of the analysis corroborate the implications of the postcranial anatomy: despite dental similarities, the 'hyopsodontids' in the broad sense are not a monophyletic group. Over a dozen other genera are known of candidate hyopsodontids, but so long as they are known only from dental characters their true position remains uncertain. Without postcranial data, it seems, we can't handle the tooth.

REFERENCES

Halliday, T. J. D., P. Upchurch & A. Goswami. 2017. Resolving the relationships of Paleocene placental mammals. Biological Reviews 92 (1): 521–550.

Penkrot, T. A., S. P. Zack, K. D. Rose & J. I. Bloch. 2008. Postcranial morphology of Apheliscus and Haplomylus (Condylarthra, Apheliscidae): evidence for a Paleocene Holarctic origin of Macroscelidea. In: Sargis, E. J., & M. Dagosto (eds) Mammalian Evolutionary Morphology: A Tribute to Frederick S. Szalay pp. 73–106. Springer.

Rose, K. D. 2006. The Beginning of the Age of Mammals. JHU Press.

Where Do You Put Your Camels?

A dromedary Camelus dromedarius dares you to say what you make of it. Copyright John O'Neill.

In any discussion of the conflicts that may exist between morphological and molecular data in phylogenetic analysis, hippos and whales are bound to come up sooner or later. The claim in the late 1990s that these are each other's closest living relatives (and hence, that whales are nested fairly deeply within the artiodactyls, or even-toed hoofed mammals) was greeted with amazement, incredulity and more than a little skepticism. The story even caught the interest of the general public through news stories like this one, meaning that many people who are otherwise unfamiliar with the trivia of mammalian phylogeny may have picked up this detail. Since then, the whale-hippo relationship has been tested, re-tested and examined again, using every data source available. But the insertion of the whales was not the only way that molecular data mixed up the artiodactyl family tree. There was also the question of where one put the camels.

Based on anatomical data, it had previously been generally agreed that camelids (including camels and llamas) were most closely related to the ruminants, the group including such artiodactyls such as cattle, deer or giraffes. Both camelids and ruminants regurgitate cud pellets from the stomach back to the mouth in order to break their food more efficiently*, and both camelids and ruminants have a stomach divided into chambers with food only travelling to the rear section of the stomach after it has been re-chewed. They do differ in that whereas ruminants have the stomach divided into four distinct chambers, camelids only possess three; the rear two chambers (the abomasum and omasum) are not clearly differentiated in camelids. They were also united by features of the dentition, such as the presence of distinctly crescent-shaped cusps on the rear teeth. This latter feature lead the camelid+ruminant grouping to commonly be referred to as the Selenodontia (the 'moon-teeth').

*Yes, giraffes do chew cud. Yes, the cud does travel all the way between the stomach and the mouth each time.

However, the advent of molecular analyses cast doubt on this long-accepted arrangement. Instead of supporting the expected Selenodontia clade, molecular analyses placed camelids as the sister group to all other artiodactyls, with the ruminants instead being sister to the whales+hippos clade (with a pigs+peccaries clade the next clade out). This implied that the shared features of camelids and ruminants had arisen convergently (or else all other artiodactyls had reverted to a state considered more plesiomorphic for the group as a whole). In support for such a proposition, one might point to the ecological similarities in play. Camelids and ruminants are more specialist browsers than the non-selenodont artiodactyls, which are commonly more omnivorous (pigs and peccaries) or even carnivorous (whales).

Alternative phylogenies of artiodactyls, based on morphological (left) and molecular (right) data, from Spaulding et al (2009).
However, even if one is willing to credit that the 'selenodont' characters may have been the result of similar dietary pressures, one must also consider the issue that there are a number of fossil artiodactyl groups with selenodont or quasi-selenodont features. Examples of these include the Protoceratidae, a North American group that has commonly made some sort of appearance in popular books on fossil animals due to the weird home arrangements of some species, and the semi-bipedal Anoplotherium. In an influential morphological study of artiodactyl relationships, Gentry & Hooker (1988) referred to the possibility that some of these groups might be members of the selenodont stem, sitting outside the exclusive camelid+ruminant clade. Obviously, if selenodonts were not monophyletic, fossil 'selenodonts' might be aligned to either camelids or ruminants, but they couldn't be connected to both. Most studies that posited selenodont polyphyly, however, looked at living taxa only and did not consider extinct groups.

The most detailed study that I've found so far that considers the relationship between data from fossil taxa and from molecular sources in artiodactyl phylogeny is that published by Spaulding et al. (2009). This combined analysis of both morphological and molecular data produced results that were largely concordant with the latter, generally supporting placement of camelids as the sister group to all other artiodactyls (it's worth noting, mind you, that the size of the molecular data set used was considerably larger than that for the morphology, and an analysis of their morphological data only resulted in selenodont monophyly). The various 'proto-selenodonts' were scattered to the stems of various Recent clades. Protoceratids, for instance, were associated with ruminants rather than with camelids*. There are still a number of groups that remain yet to be analysed, but they've made a start.

*Another result of their analysis that is not directly relevant to the selenodont question, but cannot go unremarked upon, is that their tree indicates that Andrewsarchus, a lead contender for the title of largest terrestrial mammalian carnivore ever, might some sort of giant entelodont. I don't know how much I should read into this—not all of Andrewsarchus' potential relatives were included in Spaulding et al.'s analysis—but that's the sort of result that one just wants to be true.

REFERENCES

Gentry, A. W., & J. J. Hooker. 1988. The phylogeny of the Artiodactyla. In: Benton, M. J. (ed.) The Phylogeny and Classification of the Tetrapods vol 2. Mammals pp. 235–272. Clarendon Press: Oxford.

Spaulding, M., M. A. O’Leary & J. Gatesy. 2009. Relationships of Cetacea (Artiodactyla) among mammals: increased taxon sampling alters interpretations of key fossils and character evolution. PLoS ONE 4(9): e7062. doi:10.1371/journal.pone.0007062

The Gordian Mouse

North American deer mouse from near Santa Fe, New Mexico, possibly Peromyscus sonoriensis. Copyright J. N. Stuart.

The North American deer mouse Peromyscus maniculatus has been recognised as one of the most widespread mammal species in North America. It has been recorded from most of the continent, with the main areas of absence being northern Canada and Alaska, the south-eastern corner of the United States (where it is replaced by a closely related species, the smaller Oldfield deer mouse Peromyscus polionotus) and coastal regions of Mexico. They are highly adaptable animals, eating a wide range of foods, and their tendency to gnaw just about anything available for food or nesting has not always endeared them to their human compatriots. In recent years, they have also received their fair share of unwelcome attention as vectors for pathogens such as the Sin Nombre hantavirus*, which causes a devastating (and commonly fatal) pulmonary disease in humans. It is therefore not surprising that the deer mouse has become one of the most extensively studied mammal species out there (not quite in house mouse or black rat territory, maybe, but still definitely up there). A search for 'Peromyscus maniculatus' on Google Scholar brings back well over 16,000 results. Surely with this degree of attention, this is at least one taxon for which I cannot deploy my usual 'the taxonomy is uncertain'?

*'Sin nombre' is, of course, Spanish for 'without name'. The reason for this coy appellation is that the virus in question was first labelled the 'Four Corners' virus after being identified from patients in that region of the United States in 1993, but was renamed after residents of the region protested.

Of course it isn't. Quite the opposite, in fact. With a range this large, it is not surprising that a fair amount of variability has been recognised in the North American deer mouse over the years. Over sixty subspecies of Peromyscus maniculatus have been recognised from various regions. This variability has also been part of the deer mouse's appeal as a study animal, as evolutionary scientists have examined the relationships between subspecies. The problem comes when you realise that in some places you may find two distinct 'subspecies' of deer mouse in a single locality. If a 'subspecies' is defined as a geographical sub-unit of a reproductively coherent species (as it traditionally has been), then that ain't right. Two taxa occupying a single range and not interbreeding should be separate species, surely? But travel to another location nearby, and you'll find the two 'species' interbreeding and merging into one. One case that has been cited as a classic example of a 'ring species' is around the Rocky Mountains, where the subspecies P. maniculatus artemisiae and P. m. nebrascensis do not interbreed with each other, but both interbreed with other subspecies to the north and south.

Part of the problem, of course, is that the 'species' is just a horrendously messy concept. There are a wealth of different species concepts out there, but the essential ideal underlying most is that a 'species' represents the point at which the reticulating relationships between interbreeding individuals become less important for understanding relationships than the branching relationships between population lineages. Which is not a point at all. Lineages diverge at multiple, quasi-independent levels in the process of speciation—they separate geographically, behaviourally, genetically, morphologically—and there is no magic point at which they suddenly change from 'one species' to 'separate species'. Nevertheless, as the recent trend has been to recognise species on a finer level than in the past, it seems likely that the future will see a subdivision of the current Peromyscus maniculatus.

Figure showing distributions of lineages of Peromyscus maniculatus from Kalkvik et al. (2012).

Over much of its range, Peromyscus maniculatus can be divided between two main morphotypes: a forest form with larger ears and a longer tail, and an open-country form with smaller ears and a shorter tail. Where the two forms are found in one region, they maintain their ecological distinctiveness. However, because of the aforementioned hybridisation between geographically adjacent populations, it is unlikely that any subdivision of P. maniculatus will be directly between these two morphotypes. One split that has already been widely accepted was proposed by Hogan et al. (1993), who found that certain populations in coastal north-west North America were genetically quite distinct from other P. maniculatus, and recommended the recognition of a separate species P. keeni. Peromyscus keeni is also generally larger and longer-tailed than P. maniculatus. However, more recent studies of the phylogeography of North American deer mice by Dragoo et al (2006) and Kalkvik et al. (2012) have found that even with the removal of P. keeni, P. maniculatus remains paraphyletic to both that species, to the south-west species P. polionotus (whose distinction from P. maniculatus has not generally been questioned), and possibly to the Mexican P. melanotis. Both the later studies identified six major lineages within P. maniculatus, and Dragoo et al. (2006) suggested that it may need to be divided between at least three species. The 'true' Peromyscus maniculatus, under this scheme, includes the two lineages found in the north-east of North America (the original type locality of the species being in Labrador). A coastal lineage found in the south-west was identified by both studies as related to P. keeni, and it remains to be seen whether it would be better included in that species or recognised as its own separate species. A lineage identified in southern New Mexico could be recognised under the name of Peromyscus blandus. The remaining two lineages, one found around the Rocky Mountains and one in the Great Plains, formed a clade that Kalkvik et al. (2012) identified as the sister lineage of P. polionotus. Dragoo et al. suggested that the name Peromyscus sonoriensis was available for this clade. However, Kalkvik et al. recognised that the Rocky Mountain lineage was a forest form and the Great Plains lineage a open-country form, so there may be grounds for their recognition as distinct species.

While the correlation between these studies appears promising, it must be stressed that both were analysing the same gene (cytochrome b) and it remains to be seen whether the lineages they identified continue to be supported by other sources of data. It also needs to be seen whether they stand up to the inclusion of further populations: western Canada and Mexico stand out as poorly sampled areas in both studies. Peromyscus maniculatus in its current form may represent one species, it may represent four, or it may yet refer to even more than that.

REFERENCES

Dragoo, J. W., J. A. Lackey, K. E. Moore, E. P. Lessa, J. A. Cook & T. L. Yates. 2006. Phylogeography of the deer mouse (Peromyscus maniculatus) provides a predictive framework for research on hantaviruses. Journal of General Virology 87: 1997-2003.

Hogan, K. M., M. C. Hedin, H. S. Koh, S. K. Davis & I. F. Greenbaum. 1993. Systematic and taxonomic implications of karyotypic, electrophoretic, and mitochondrial-DNA variation in Peromyscus from the Pacific Northwest. Journal of Mammalogy 74 (4): 819-831.

Kalkvik, H. M., I. J. Stout, T. J. Doonan & C. L. Parkinson. 2012. Investigating niche and lineage diversification in widely distributed taxa: phylogeography and ecological niche modeling of the Peromyscus maniculatus species group. Ecography 35: 54-64.

The Litopterns: Macrauchenia and More

Much has been made of the "splendid isolation" of South America for a large part of the Cenozoic. Finding itself girt by sea, South America became home to a number of endemic groups of animals: the 'terror-birds' of the Phorusrhacidae, notoungulates that were something like a rhino and something like a rabbit, and giant armadillos and anteaters. Among these uniquely South American animals were the subjects of today's post, the litopterns.

Digital reconstruction of Macrauchenia by Deskridge.

Litopterns are one of those groups of animals that tend to be represented in recent popular media by a single example, which many of you may recognise in the picture above. This was Macrauchenia patachonica, one of the latest surviving litopterns (like many other South American taxa, litopterns did not fare well during the so-called Great Faunal Interchange when South and North America became connected). But as with so many other under-represented groups, the popular exemplar is not necessarily a prime example. Macrauchenia was not only one of the last litopterns, it was also one of the largest, and the litopterns came in a whole range of appearances.

The earliest litopterns are known from the late Palaeocene. The basalmost members of the group are classified as the Protolipternidae, but the members of this family are united by primitive characters only. It is generally accepted that litopterns were closely related to two Palaeocene families of South American 'condylarths', the Didolodontidae and Sparnotheriodontidae, and there is a certain degree of arbitrariness about whether or not these families should also be treated as litopterns. Those who would exclude the didolodontoids from the litopterns do so on the basis of the latter's tarsal morphology, which has become adapted for a more cursorial lifestyle. The protolipternids bridge the gap between didolodontoids and other litopterns in that they possessed a litoptern-like tarsus, but retained teeth more like the didolodontoids. Protolipterna also retained five toes on the feet, while this number was reduced in later litopterns (Bastos & Bergqvist 2007). The other noteworthy feature of protolipternids was that they were not very big. Rose (2009) includes an illustration of partial upper and lower jaws of the protolipternid Asmithwoodwardia whose scale bar indicates that the complete skull must have been less than five centimetres in length, or about the size of a brown rat. Cifelli (1983) suggested on the basis of their small size that these animals may have been more leapers than runners, a suggestion not directly supported but not entirely ruled out by Bastos & Bergqvist (2003).

Reconstruction of Thoatherium minusculum by Charles R. Knight.

The remaining litopterns mostly belong to the families Proterotheriidae, Adianthidae and Macraucheniidae, united by specialisations of the dentition and reduction of the number of toes to three (a fifth family, the late Palaeocene Notonychopidae, are represented by dental remains only). The Palaeocene to Pleistocene Proterotheriidae have attracted a reasonable amount of interest in the past because of their convergences with the horses in the Northern Hemisphere. Like horses, proterotheriids centred locomotion on the middle toe only, with the toes on either side being reduced. In the Miocene proterotheriid Thoatherium, the side toes were almost completely lost, reduced to splints even smaller than those of the modern horse (being by this measure more horse-like than an actual horse, Thoatherium has also been a popular subject for books on evolution). In other respects, however, proterotheriids were not so horse-like. With relatively low-crowned teeth, proterotheriids and other litopterns were browsers rather than grazers, and they may have preferred more wooded terrain rather than grasslands. Ecologically, proterotheriids were probably more like deer or small antelopes than horses, and they resembled small antelopes in size. Only one proterotheriid survived into the Pleistocene, Neolicaphrium recens, and only in Uruguay and northern Argentina (Ubilla et al. 2011).

The Adianthidae were small litopterns (though not as small as the protolipternids) known from the Eocene to the Miocene. Most adianthids are known only from dental remains and/or jaw fragments, though some limb bones are known from the Miocene Adianthus godoyi (Cifelli 1991). These indicate a gracile form, probably more similar to proterotheriids than to macraucheniids, though Cifelli noted the similarities to the former were likely related to size rather than indicative of any deeper affinity.

Reconstruction of the cramaucheniine macraucheniid Theosodon garretorum, with the carnivorous metatherian Borhyaena tuberata, by Charles R. Knight.

The Macraucheniidae retained three functional toes, with the middle toe not substantially larger than the two side ones. They also differed from the proterotheriids in the development of a longer neck, and have usually been compared to camels in appearance (the name 'Macrauchenia' was originally coined to effectively mean 'big llama', in the mistaken belief that it represented an ancestor of that animal). They have been divided between to subfamilies, the Oligocene to Miocene Cramaucheniinae and the late Miocene to Pleistocene Macraucheniinae, though the latter are undoubtedly descended from the former. The cramaucheniines retain a plesiomorphic anterior nasal opening, but in the Macraucheniinae the nasal bones are reduced and the nasal opening has moved posteriad on the skull (Dozo & Vera 2010). It is this dorsal position of the nasal opening that has lead to the interpretation of Macrauchenia as having some form of proboscis, like that of a tapir. The combination of a long neck and a proboscis is, however, an unusual one, and I've wondered if it may have been more of a prehensile upper lip. The macraucheniids did better in the Pleistocene than the proterotheriids, with three species described from a large chunk of the continent, but eventually they two went the way of the toxodont.

REFERENCES

Bastos, A. C. F., & L. P. Bergqvist. 2007. A postura locomotora de Protolipterna ellipsodontoides Cifelli, 1983 (Mammalia: Litopterna: Protolipternidae) da Bacia de São José de Itaboraí, Rio de Janeiro (Paleoceno superior). Anuário do Instituto de Geociências 30 (1): 58-66.

Cifelli, R. L. 1983. Eutherian tarsals from the Late Paleocene of Brazil. American Museum Novitates 2761: 1-31.

Cifelli, R. L. 1991. A new adianthid litoptern (Mammalia) from the Miocene of Chile. Revista Chilena de Historia Natural 64: 119-125.

Dozo, M. T., & B. Vera. 2010. First skull and associated postcranial bones of Macraucheniidae (Mammalia, Litopterna) from the Deseadan Salma (late Oligocene) of Cabeza Blanca (Chubut, Argentina). Journal of Vertebrate Paleontology 30 (6): 1818-1826.

Rose, K. D. 2009. The Beginning of the Age of Mammals. JHU Press.

Ubilla, M., D. Perea, M. Bond & A. Rinderknecht. 2011. The first cranial remains of the Pleistocene proterotheriid Neolicaphrium Frenguelli, 1921 (Mammalia, Litopterna): a comparative approach. Journal of Vertebrate Paleontology 31 (1): 193-201.

Why Are There So Many Avahis?

Western woolly lemur Avahi occidentalis, photographed by Axeltelford.

A few years back, I wrote a post on the lemur family Indriidae: the indri, the avahis, the sifakas. One thing I briefly mentioned in that post is that recent years have seen an apparent avalanche of new indriid species being described. But why has this happened, and how sturdy are these new distinctions?

In 1982, Tattersall provided an overview of Malagasy lemurs that recognised just four species of indriid: the indri Indri indri, the avahi Avahi laniger, Verreaux's sifaka Propithecus verreauxi and the diademed sifaka P. diadema (Tattersall 2007). A fifth species was added in 2008, the golden-crowned sifaka P. tattersalli. But the real explosion has come in only the last ten years or so. Recent workers have proposed the recognition of seven species of sifaka (Mayor et al. 2004), and no less than nine species of avahi (Zaramody et al. 2006, Andriantompohavana et al. 2007, Lei et al. 2008). Each of the species within a genus is generally geographically separated from its congeners, and some species are recorded only from very small ranges.

In the case of the sifakas, none of the new 'species' is actually a new taxonomic entity per se. With the exception of P. tattersalli, all were previously recognised previously as subspecies of either P. verreauxi or P. diadema. The most obvious differences between the various varieties of sifaka is coloration. As noted in the earlier post linked to above, popular depictions of sifakas are heavily biased towards P. [verreauxi] verreauxi, found in the south-west of Madagascar, with a white body and black skull-cap (photo below by Jouan & Rius):

However, the sifakas are much more varied than you might think from watching David Attenborough documentaries alone. As well as the red-and-black Propithecus [diadema] diadema illustrated in the earlier post, sifakas vary from the almost entirely black P. [diadema] perrieri of the far north of Madagascar (photograph by Pete Oxford):

to the almost entirely white north-eastern P. [diadema] candidus (photo by Kevin Schafer):

The various sifaka subspecies were analysed by Mayor et al. (2004), who identified them as genetically distinct as well as distinct in appearance, and therefore recommended treating them all as separate species. However, other authors such as Tattersall (2007) have pointed out that morphological distinctions between populations may become less clear when overall variation is considered.

The above photos, from Rakotonirina et al. (2013), show variation in sifakas at a single site in central-west Madagascar, near the boundary between the ranges of Propithecus [verreauxi] deckeni and P. [verreauxi] coronatus and including individuals that might be assigned on grounds of coloration to either taxon.

In the case of the avahis, things are even more convoluted than for the sifakas. While the diurnal sifakas may vary noticeably in external appearance, the nocturnal avahis keep to a more or less basic brown. There are some slight differences between avahis on the western and eastern sides of Madagascar that had lead to the recognition of two separate subspecies, Avahi laniger laniger in the east and A. l. occidentalis in the west. A. laniger and A. occidentalis were subsequently treated as separate species on the basis of differences in their karyotypes. Each has been further subdivided into multiple species largely on the basis of genetic data alone (though vocalisation data was also a factor in separating A. unicolor from A. occidentalis). What is more, the genetic distinctions have mostly been made on the basis of mitochondrial data only, and some 'species' have only been represented in analyses by data from a few individuals. Markolf et al. (2011) suggested that genetic species could not be distinguished reliably on the basis of such small samples because of the increased risk of confusing individual variation for species-level distinctions. In the majority of cases, differences in mitochondrial genes between Avahi samples have correlated with geographical separation, but there is at least one notable exception. The central-east Malagasy location of Ranomafana has provided samples that fall into three distinct haplotype clusters. Though recognised as a single species A. peyrierasi on the basis of their common distribution, these three clusters do not form a monophyletic group in phylogenetic analyses, and the geographically separate taxa A. betsileo, A. meridionalis and A. ramanantsoavana are all nested between the A. peyrierasi haplotypes (Lei et al. 2008).

Eastern woolly lemur Avahi laniger, photographed by Inaki Relanzon.

None of the Avahi species as currently recognised overlap in range. However, in a landmass that has lost four-fifths or more of its original forest cover, it is worth asking how much of this isolation is original, and how much man-made relictualism. As always in questions of scientific research, we are left noting that further investigation is required.

REFERENCES

Andriantompohavana, R., R. Lei, J. R. Zaonarivelo, S. E. Engberg, G. Nalanirina, S. M. McGuire, G. D. Shore, J. Andrianasolo, K. Herrington, R. A. Brenneman & E. E. Louis Jr. 2007. Molecular phylogeny and taxonomic revision of the woolly lemurs, genus Avahi (Primates: Lemuriformes). Special Publications, Museum of Texas Tech University 51: 1-59.

Lei, R., S. E. Engberg, R. Andriantompohavana, S. M. McGuire, R. A. Mittermeier, J. R. Zaonarivelo, R. A. Brenneman & E. E. Louis. 2008. Nocturnal lemur diversity at Masoala National Park. Special Publications, Museum of Texas Tech University 53: 1-41.

Markolf, M., M. Brameier & P. M. Kappeler. 2011. On species delimitation: yet another lemur species or just genetic variation? BMC Evolutionary Biology 11: 216.

Mayor, M. I., J. A. Sommer, M. L. Houck, J. R. Zaonarivelo, P. C. Wright, C. Ingram, S. R. Engel & E. E. Louis Jr. 2004. Specific status of Propithecus spp. International Journal of Primatology 25 (4): 875-900.

Rakotonirina, L. H. F., F. Randriantsara, A. H. Rakotoarisoa, R. Rakotondrabe, J. Razafindramanana, J. Ratsimbazafy & T. King (in press, 2013). A preliminary assessment of sifaka (Propithecus) distribution, chromatic variation and conservation in western central Madagascar. Primate Conservation.

Tattersall, I. 2007. Madagascar's lemurs: cryptic diversity or taxonomic inflation? Evolutionary Anthropology 16: 12-23.

Zaramody, A., J.-L. Fausser, C. Roos, D. Zinner, N. Andriaholinirina, C. Rabarivola, I. Norscia, I. Tattersall & Y. Rumpler. 2006. Molecular phylogeny and taxonomic revision of the eastern woolly lemurs (Avahi laniger). Primate Report 74: 9-23.

The Wolf in Time

Black-backed jackal pup Canis mesomelas, photographed by Blake Matheson.
The dogs of the genus Canis include some of the most familiar of all mammals: the wolf Canis lupus, the coyote C. latrans, and of course the domestic dog Canis familiaris. I have already discussed in an earlier post how these three, together with the golden (Canis aureus) and the Simien (C. simensis) jackals, form a cluster of closely related species (that I'll refer to as the 'wolf group') that are not always clearly separated. Today, I'll take things a bit further and look at the fossil history of the genus Canis.

Coyote Canis latrans, from Ryan Photographic.
The earliest taxa assigned to the genus Canis are known from the late Miocene, about six million years ago (Tedford et al. 2009). Early Canis have been identified in both Europe (C. cipio) and North America (C. ferox), though there is some uncertainty about whether the European C. cipio should be treated as Canis or assigned to the related, slightly earlier fossil genus Eucyon. Whatever the case, it doesn't appear to have been long before Canis populations were well and truly established on both continents. The North American Canis ferox was, as far as I can tell, probably not dissimilar to a modern coyote in appearance, and early Canis species probably also resembled coyotes in being fairly generalist predators. In the evolutionary analysis by Tedford et al. (2009), C. ferox was suggested to have begat C. lepophagus at the beginning of the Pliocene, which in turn begat two lineages: one leading to the modern wolf group, the other leading to three North American Plio-Pleistocene species (C. thooides, C. feneus and C. cedazoensis) that were smaller than their ancestor and probably similar in appearance to modern jackals. It is somewhat unfortunate that Tedford et al.'s analysis did not include the African side-striped (C. adustus) and black-backed (C. mesomelas) jackals, which molecular and morphological analyses have generally agreed lie outside the wolf group. Biogeography alone suggests that the North American 'jackals' were probably convergent rather than directly related to the modern African species, but it would be nice to know.

Mounted skeleton of dire wolf Canis dirus, from lora_313. This species probably weighed between 50 to 80 kg, which is comparable in size to a very large dog such as a bullmastiff or great dane.
The modern wolf group diversified in the late Pliocene, including a number of fossil species as well as the modern. The rate of diversification and spread of wolf-group Canis was such that palaeontologists refer to their appearance in the fossil record as the 'wolf event', and use it as a marker of the development of the colder tundra climate of the Pleistocene ice ages. Higher diversity in Eurasia suggests that it was probably the centre of diversification, with North American species derived from repeated colonisation. Significant among these was the relatively large C. armbrusteri, a close relative of the grey wolf C. lupus. Canis armbrusteri is notable as the probable ancestor of the late Pleistocene dire wolf C. dirus, made famous by its appearances in the works of Robert E. Howard* and similar authors. As well as being a dominant predator in North America, the dire wolf spread into northwestern South America. A similar large Canis species, C. nehringi, is also known from the same time in Argentina, but the analysis of South American canids by Prevosti (2010) was unable to clearly determine whether C. nehringi was a southern relative of C. dirus or a convergent relative of the Xenocyon lineage.

*A man who spent far too much time thinking about oiled chests if ever there was one.

Dholes Cuon alpinus, from Rajnish Pradhan.
Xenocyon is itself relevant to the history of Canis: first appearing in the late Pliocene, Xenocyon lycaonoides is probably the ancestor of the modern African hunting dog Lycaon pictus and the Asian dhole Cuon alpinus, forming a hypercarnivorous lineage specialised for collaborative hunting of large prey. Phylogenetic analyses of modern taxa have varied as to whether Lycaon and Cuon are the sister group of modern Canis, or whether they are in fact more closely related to the wolf group than are C. adustus or C. mesomelas, rendering Canis paraphyletic. Removal of the latter two species from Canis into separate genera as Schaeffia adusta and Lupulella mesomelas to preserve monophyly has been suggested, but almost universally ignored (as well as failing to resolve the status of the non-wolf-group fossil Canis species). Tedford et al. (2009) even nested the Xenocyon lineage within the wolf group itself, as sister to the Canis lupus-C. dirus group, but one might suspect the influence of convergences to large size and hypercarnivory. Prevosti (2010) placed Lycaon and Cuon in a more standard position just outside the wolf group, but did not consider as many fossil Canis species as Tedford et al.

Remains of Cynotherium sardous (plus some smaller mammal), from here.
The Xenocyon lineage was undoubtedly Eurasian in origin, but the primarily Eurasian X. lycaonoides did spread into northern North America, and a second species X. texanus was found in the Pleistocene of (surprisingly) Texas. The modern dhole Cuon alpinus was also present in North America in the latest Pleistocene, with remains of at least four individuals found in a cave in northeastern Mexico, as well as being found in Europe (Tedford et al. 2009). Also a member of the Xenocyon lineage was the Pleistocene Cynotherium sardous, found on the Mediterranean islands of Sardinia and Corsica (which were a single island when the Mediterranean sea level was lower). Though descended from hypercarnivorous ancestors, Cynotherium became adapted in its island habitat to hunting smaller prey (such as the Sardinian lagomorph Prolagus sardus). Though it retained the simplified dentition of a hypercarnivore, it became smaller and the skull became less reinforced, as befits an animal no longer wrestling down large ungulates (Lyras et al. 2006).

REFERENCES

Lyras, G. A., A. A. E. Van Der Geer, M. D. Dermitzakis & J. De Vos. 2006. Cynotherium sardous, an insular canid (Mammalia: Carnivora) from the Pleistocene of Sardinia (Italy), and its origin. Journal of Vertebrate Paleontology 26 (3): 735-745.

Prevosti, F. J. 2010. Phylogeny of the large extinct South American canids (Mammalia, Carnivora, Canidae) using a "total evidence" approach. Cladistics 26: 456-481.

Tedford, R. H., X. Wang & B. E. Taylor. 2009. Phylogenetic systematics of the North American fossil Caninae (Carnivora: Canidae). Bulletin of the American Museum of Natural History 325: 1-218.

Beaver Fever

Eurasian beaver Castor fiber, from here.

Beavers are one of those animals that are familiar even to people who do not live in parts of the world where you can find beavers. The two living species of beaver are semi-aquatic rodents with one species each native to Eurasia (Castor fiber) and North America (C. canadensis) (though the North American beaver has been introduced to several parts of Europe). Differences between the two are slight: the Eurasian beaver is generally larger (up to 35 kg) and has a somewhat longer skull and a less rounded tail. Beavers are best known, of course, for their construction of elaborate subaquatic nests and dams*. Dams are generally about fifteen to seventy metres across, but have been recorded over 600 metres across (Rybczynski 2008). Beavers may also dig burrows connected to their dams, and construct canals over one hundred metres long (Rybczynski 2008).

*The original text here has been edited following Howard's comment below.

Phylogenetically, beavers are somewhat remote from other rodents, and represent the last survivors of a once more diverse lineage. First known from the late Eocene, the members of the beaver family Castoridae are divided in the most recent treatments between five subfamilies (Korth 2001, 2004). The plesiomorphic subfamilies Agnotocastorinae and Anchitheriomyinae are not well known, and the Agnotocastorinae in particular may be non-monophyletic (Rybczynski 2007). The remaining subfamilies fall into two distinct lineages: one containing the Palaeocastorinae (Oligocene-Miocene), the other the Castoridinae (Oligocene-Pleistocene) and Castorinae (Oligocene-present). Of these two lineages, only the latter are known to have been semiaquatic: the Palaeocastorinae are strictly terrestrial.

Preserved Daimonelix burrow in the American Museum of Natural History, with specimen of Palaeocastor fossor in the presumed nesting chamber, photographed by Inazakira.

The palaeocastorines, a strictly North American lineage, were specialised burrowers. Their incisors, which have rounded faces in modern beavers, became flattened and adapted for digging. Their burrows were distinctive helicoidal structures, described as trace fossils under the name of Daimonelix ('devil's spiral'), that could reach over 2.5 metres in depth and twenty centimetres in diameter. These burrows were constructed in 'towns' with multiple burrows in close proximity. Though each burrow was independent, without connections between adjacent burrows, such close positioning suggests that palaeocastorines may have had a well-developed social structure (Hugueney & Escuillié 1996). However, though the palaeocastorines were much more diverse at their apogee than the castoroidine-castorine lineage, they became extinct after a relatively short period.

Reconstructed skeleton of Castoroides ohioensis alongside that of (I presume) a modern beaver in Earlham College, from here.

The Castorinae and Castoroidinae may never have achieved the diversity at any one point in time of the palaeocastorines, they were more successful over the long haul: the more diverse of the two subfamilies, the Castoroidinae, only became extinct fairly recently. Castoroidines are commonly referred to as the 'giant beavers', and while not all castoroidines were giant (many, if not most, were smaller than modern beavers), the largest of them certainly were: the North American Castoroides reached an estimated size of about 100 kg, and would have been as large as a small bear. Whether the giant Castoroides produced similarly gigantic dams, however, is uncertain. Evidence of wood-chopping behaviour like that known for modern beavers (in the form of preserved wood bearing identifiable tooth marks, in association with beaver remains) is only well supported for one fossil species, the castoroidine Dipoides (suggested evidence for wood-chopping in Castoroides is more equivocal) (Rybczynski 2008). Phylogenetic bracketing between Dipoides and modern beavers would suggest that wood-chopping arose at the base of the castoroidine-castorine clade; alternatively, the absence of direct evidence of such behaviour may suggest convergence between these two species. Also, Dipoides was a less efficient wood-cutter than modern Castor, cutting with the rounded edges of its incisors while Castor uses the flattened ends, and if it used chopped wood to construct nests then they would have probably been correspondingly more simple (beavers also use chopped wood for food, eating the leaves and bark, so wood-chopping does not automatically indicate dam-building). There are other indications that fossil beavers may not have been as specialised aquatically as the modern species: the early castorine Steneofiber, for instance, did not possess the flattened tail of Castor (a flattened tail has been indicated for Castoroides but Castor and Castoroides probably developed such tails independently) (Hugueney & Escuillié 1996).

Reconstruction of Trogontherium cuvieri, from Fostowicz-Frelik (2008).

Perhaps the primary enigma among fossil beavers is the European Pleistocene Trogontherium. Although also referred to as a 'giant beaver', and often implied to be a European parallel to Castoroides, Trogontherium was a quite different animal. Fostowicz-Frelik (2008) argued that leg proportions and other features indicate that Trogontherium was a more terrestrial, cursorial animal than other beavers (in particular, its narrowed rather than flattened toe bones suggest that it lacked the webbed feet of modern beavers). The phylogenetic analysis of beavers by Rybczynski (2007) placed Trogontherium as closely related to Castoroides, but certain plesiomorphies in its tooth morphology lead Rybczynski to suggest that this position was probably an artifact of convergences due to large size, and that Trogontherium should perhaps be in a much more basal position.

REFERENCES

Fostowicz-Frelik, Ł. 2008. First record of Trogontherium cuvieri (Mammalia, Rodentia) from the middle Pleistocene of Poland and review of the species. Geodiversitas 30 (4): 765-778.

Hugueney, M., & F. Escuillié. 1996. Fossil evidence for the origin of behavioral strategies in early Miocene Castoridae, and their role in the evolution of the family. Paleobiology 22 (4): 507-513.

Korth, W. W. 2001. Comments on the systematics and classification of the beavers (Rodentia, Castoridae). Journal of Mammalian Evolution 8 (4): 279-296.

Korth, W. W. 2004. Beavers (Rodentia, Castoridae) from the Runningwater Formation (Early Miocene, early Hemingfordian) of western Nebraska. Annals of Carnegie Museum 73 (2): 1-11.

Rybczynski, N. 2007. Castorid phylogenetics: implications for the evolution of swimming and tree-exploitation in beavers. Journal of Mammalian Evolution 14: 1-35.

Rybczynski, N. 2008. Woodcutting behavior in beavers (Castoridae, Rodentia): estimating ecological performance in a modern and a fossil taxon. Paleobiology 34 (3): 389-402.

Groundhogs, Woodchucks and Other Big Squirrels

Thirteen-lined ground squirrel Ictidomys tridecemlineatus, photographed by Phil Myers.

The Holarctic ground squirrels of the Marmotini were the subject of one of my earliest posts at this site, before I really knew what I was doing*. So I'll have a go at improving it now.

*Not, of course, that I know what I'm doing now.

The Arctic ground squirrel Urocitellus parryii, photographed by Ianaré Sévi.

Marmotini is the clade of squirrels that includes ground squirrels (Spermophilus), antelope ground squirrels (Ammospermophilus), marmots (Marmota) and prairie dogs (Cynomys). Authors seem to differ on whether to also include the chipmunks (Tamias), but the question is somewhat semantic: agreement seems to be universal that the chipmunks represent the sister group to the remaining marmotins (Herron et al. 2004), so the only real question is how inclusive one wishes to make the term. The Chinese rock squirrels Sciurotamias may also belong to the Marmotini (Steppan et al. 2004). Except for the semi-arboreal chipmunks, marmotins are largely terrestrial in habits. They nest in underground burrows (including chipmunks), and some species form quite complex societies.

Père David's rock squirrel Sciurotamias davidianus, from here.

Ground squirrels previously assigned to the genus Spermophilus* have a wide range through Eurasia and North America. However, both morphological and molecular data indicate that Cynomys is derived from within 'Spermophilus', and molecular data indicate that Ammospermophilus and Marmota are as well (Herron et al. 2004). Helgen et al. (2009) divided the former Spermophilus between eight genera. Six of these genera are found in North America, one (Spermophilus proper) is found in Eurasia, and only one (Urocitellus) spans the divide between northeast Asia and North America. Whether the Marmotini as a whole are Eurasian or North American in origin is equivocal: of the three basalmost branches, Sciurotamias is definitely Eurasian, Tamias could be either (the Siberian chipmunk Tamias sibiricus is the sister to the remaining North American species) and the Spermophilus clade is probably North American in origin, with dispersals back to Eurasia in Marmota, Urocitellus and Spermophilus (Herron et al. 2004).

*Particularly in the European literature, it was not uncommon in the past to find the name Citellus being used in place of Spermophilus. Citellus Oken 1816 is indeed an older name than Spermophilus Cuvier 1825; however, the publication that the former derives from was not one that used the binomial system, and hence it has been declared invalid as a source of names (International Commission on Zoological Nomenclature 1956).

The woodchuck Marmota monax, from here.

Marmotins were the dominant squirrel group in North America during the Neogene; tree squirrels, though present, were exceedingly rare (Emry et al. 2005). The Pliocene Paenemarmota was the largest of all marmotins, reaching the size of a large beaver (Repenning 1962).

REFERENCES

Emry, R. J., W. W. Korth & M. A. Bell. 2005. A tree squirrel (Rodentia, Sciuridae, Sciurini) from the Late Miocene (Clarendonian) of Nevada. Journal of Vertebrate Paleontology 25 (1): 228-235.

Helgen, K. M., F. R. Cole, L. E. Helgen & D. E. Wilson. 2009. Generic revision in the Holarctic ground squirrel genus Spermophilus. Journal of Mammalogy 90 (2): 270-305.

Herron, M. D., T. A. Castoe & C. L. Parkinson. 2004. Sciurid phylogeny and the paraphyly of Holarctic ground squirrels (Spermophilus). Molecular Phylogenetics and Evolution 31: 1015-1030.

International Commission on Zoological Nomenclature. 1956. Opinion 417. Rejection for nomenclatorial purposes of volume 3 (Zoologie) of the work by Lorenz Oken entitled Okens Lehrbuch der Naturgeschichte published in 1815–1816. Opinions and Declarations Rendered by the International Commission on Zoological Nomenclature 14: 1–42.

Repenning, C. A. 1962. The giant ground squirrel Paenemarmota. Journal of Paleontology 36 (3): 540-556.

Steppan, S. J., B. L. Storz & R. S. Hoffmann. 2004. Nuclear DNA phylogeny of the squirrels (Mammalia: Rodentia) and the evolution of arboreality from c-myc and RAG1. Molecular Phylogenetics and Evolution 30: 703-719.