The many babirusa species: laissez-faire lumping under fire again

In the previous post we looked at babirusas, covering their phylogenetic position, distribution, as well as a bit of their behaviour, biology and morphology. While babirusas are famous for the bizarre upper canines that emerge from the dorsal surface of the snout in males, the function of these teeth remains uncertain. As we saw in the previous post, it has been proposed that they function in display, in fighting, or in helping the animal to push its way through dense vegetation. The latter idea is least likely and is unsupported by observations. The fighting idea might seem logical and indeed MacKinnon (1981) proposed that a fighting male might hook one of his upper canines over one of the lower canines of his opponent, thereby both disarming the opponent and allowing unprotected access to his throat and face. Incidentally, the picture above is one of Charles Tunnicliffe's paintings from his Asian Wild Life Brooke Bond picture card set (go here for the whole set, and for previous blog comments see my post on the most fantastic jerboa)

The idea that the teeth function in fighting is not supported however by the fact that the teeth are brittle and easily broken, and only shallowly rooted in the socket and unsuited for withstanding large leverage forces (Macdonald 1993, Macdonald et al. 1993). Furthermore, fighting male babirusas don’t use their upper canines at all, rather they stand bipedally and ‘box’ and prance round each other, each animal trying to impale the other on its lower canines. Clayton (2003) reported that one combatant may be lifted clear of the ground during one of these fights, and that severe wounds may be inflicted on the neck of the lifted individual. It presently appears most likely that the upper canines function in sexual display, though I don’t know of any studies testing this empirically [adjacent head shot is from Lynn Clayton’s Babirusa.Org site].

The bipedal ‘boxing’ behaviour of babirusas is odd, but arguably odder is a unique sort of ‘ploughing’ behaviour they have recently been shown to practise. On being presented with an area of soft sand, captive babirusas (mostly males) have been noted to kneel down and push their head and chest forward through the sand, the result being a deep furrow. One obscure report from the 1970s suggests that Sulawesi people associated babirusas with the creation of straight-line furrows, and in 2002 possible babirusa furrows were reported from south-eastern Sulawesi, but this behaviour has otherwise gone unreported from the wild, and was first documented among captive individuals during the 1990s.

After studying babirusas kept at the Royal Zoological Society of Antwerp, Leus et al. (1996) found that males performed the behaviour most vigorously when placed in the enclosure of another male, and as they ploughed they made various snorting and growling noises, and produced lots of foamy saliva. The animals were also seen to ‘mouth’ sand, the implication being that they were testing it for sensory clues left behind by other babirusas. Ploughing behaviour almost certainly, therefore, has a scent-marking function (analysis of babirusa saliva does not reveal the presence of sexual pheromones, as is the case in wild boar, but an unknown substance suggested to serve this role was discovered). A viscous fluid discharged from an orifice near the eye is also known to be produced by babirusas, and while it again likely functions in sexual behaviour its exact function remains mysterious (Leus et al. 1996).

Ploughing behaviour is apparently unique to babirusas among Suidae, and its discovery implies that babirusas in captivity might have their behaviour enriched if they are provided with suitable areas of soft, plougable sand or soil.

Observant readers will have noticed that, strangely, I’ve refrained thus far from using a scientific binomial for babirusas, plus I’ve consistently (I think) referred to them in the plural, and not as a single species. What gives? Well, the proverbial cat is already out of the bag, but the traditional taxonomy of referring all babirusas to the single species Babyrousa babyrussa is now defunct and there are good reasons for recognising several species. Babirusa taxonomy was reviewed by Groves (1980) who recognised four subspecies, one of which (B. babyrussa bolabatuensis from south-eastern Sulawesi) was named for Holocene fossils and apparently extinct. A fifth subspecies, B. babyrussa beruensis, is larger than the more recent forms and known only from Pleistocene fossils.

In more recent publications however (Groves 2001, Meijaard & Groves 2002a, b), it has been argued that most of the supposed subspecies are distinct enough to be recognised as distinct species, being as different from one another as are universally recognised species among other artiodactyl groups. Whether taxa are awarded subspecific or specific rank is of course subjective, but not only is it misleading to include all of these taxa within a single species, there is also the practical problem that species are always regarded as more deserving of conservation effort than subspecies. If this is news to you, or seems odd or just stupid, here is what William Oliver recently had to say about the subject…

'I am not really sure why, but species are generally regarded as being somehow
more important than subspecies and, hence, merit higher conservation priority. I
confess I can’t swallow this view without a good deal of choking and spluttering
– particularly if one considers that a polytypic species – being an abstracted
sum of its parts – doesn’t actually exist. In fact, the adoption of this view
can be (and often is) positively unhelpful, in that subspecies are all too often
ignored and species’ conservation priorities are thus ‘loaded’ in favour of the
least threatened subspecies or populations – a patently ridiculous (and arguably
irresponsible) situation’ (Oliver 2001, p. 4).

Whether there are political reasons for raising ‘subspecies’ to specific status or not, it is still clear that the named babirusa taxa are quite distinct from one another. The species are…

-- B. babyrussa Linnaeus, 1758: a small species, with long, thick body hair and a well developed tail tuft. The common names Golden babirusa and Hairy babirusa have been used lately for this species. The upper canines of males are short and slender and cross the lower canines in lateral view. The upper canines tend to diverge or be subparallel, but they may be weakly convergent. B. babyrussa is from the Sula Islands (east of central Sulawesi) and Buru (just south-east of the Sula Islands). Most workers think that the babirusas were introduced to these islands, and if this is so we don’t know what their true place of origin was. Unfortunately babirusas from the adjacent eastern and south-eastern parts of Sulawesi remain unknown, so it is not possible to test whether or not they are particularly similar to B. babyrussa.

-- B. togeanensis Sody, 1949: the largest species, it has sparser, shorter body than B. babyrussa and, in contrast to B. celebensis, the tail tuft is well developed. The upper canines of males are this species’ most distinctive feature: they are short, slender, rotated forwards and always converge. It is endemic to the Togian (or Togean) Islands, an island group that has been separated from the adjacent eastern arm of Sulawesi for c. 12,000 years. This suggests that the babirusas from the eastern arm of Sulawesi should be closely related to the Togian animals, but an absence of specimens makes this impossible to test (Meijaard & Groves 2002b). It cannot be ruled out that Togian babirusas were introduced to the Togian Islands by people. The adjacent photo of a Togian babirusa, taken in 1997 by M. Akbar on Malenge Island, is reproduced from Meijaard & Groves (2002b).

-- B. celebensis Deninger, 1909: this is the babirusa we are all familiar with (see adjacent image of male and female). It has sparse or absent body hair, a nearly hairless tail tuft, and long, relatively thick upper canines that emerge vertically, converge slightly and curl dorsally in a circle. B. celebensis is the only species kept in captivity (so well done Filipe, this is the specific identity of the animal I keep going to see at Marwell: see How big is a white rhino?). B. celebensis is the commonest babirusa and the only one for which widely available photos are available. B. celebensis is only definitely known from the northern arm of Sulawesi.

On mainland Sulawesi, babirusas are certainly not restricted to the northern arm: historically they occurred across the island, and those of central Sulawesi at least are still extant. Unfortunately their specific status remains uncertain as not enough data is known – while populations might be continuous with B. celebensis and hence referable to that species, a skull from near Kulawi in central Sulawesi differs from other babirusas in being particularly small, in having proportionally long lower premolars, and in various cranial proportions. Meijaard & Groves (2002b) thought that the skull had its strongest affinities with Togian babirusas, and they concluded that it probably represented another taxon. More data is needed to confirm this. A babirusa skull photographed on sale in south-central Sulawesi has also proved enigmatic, being quite different from the Kulawi skull and unidentifiable based on the data available (Meijaard 2003).

Babirusas are also known from southern Sulawesi. Wiles & Mustari (2002) reported a skull, a tooth, and some eyewitness accounts of live animals from south-east Sulawesi, but in south-west Sulawesi babiruas are extinct so far as we know (though this needs confirmation). Hooijer (1950) named B. bolabatuensis (though originally as a subspecies of B. babyrussa) for teeth from the south-western arm of Sulawesi, though it has most recently been concluded that the material from this region cannot be reliably identified taxonomically (Meijaard & Groves 2002b). The Kulawi skull has at times been suggested to represent a modern example of B. bolabatuensis. The adjacent map, showing the distribution of the different babirusa taxa, is from Meijaard & Groves (2002b).

What with the taxonomically indeterminate babirusas of central Sulawesi, the unconfirmed extinction of those of southern Sulawesi, and the rumoured presence of babirusas on various small islands around Sulawesi, they are a ripe area for discovery and further research. Reports, photos and specimens are badly needed from these regions, and we have some way to go before babirusa diversity and biogeography is properly understood.

What perhaps interests me most about all of this is that we yet again see a case where what is supposedly a single diverse ‘species’ has now been split up into several, thus reversing somewhat the trend of the early 20th century where multiple large mammal species were lumped together, typically with little or no justification. It now seems that at least some of this lumping was over-zealous and in fact rather lazy, and it’s for this reason that I refer to this procedure as laissez-fair lumping. This opinion is not my own: I have mostly been inspired by senior mammalogists such as Colin Groves, Peter Grubb, Jeheskel Shoshani and Esteban Sarmiento. In arguing that the two African elephant species Loxodonta africana and L. cyclotis can clearly be separated as distinct species, Grubb et al. (2000) wrote that…

‘Biodiversity of large mammals is severely underestimated. The existence of a narrow hybrid zone among large mammals can be detected in casual field surveys, which is not the case for small mammals and other animals that have to be trapped for close investigation. This simple fact has led to the downgrading of perfectly distinct, diagnosable species to a level where they become taxonomically ‘invisible’ and thus lost to biodiversity studies. There are many examples of large mammal genera in which single species are currently supposed to extend through forest and savannah zones (as in the elephant case treated here), and this series of case studies might be a place to start testing the proposition that their biodiversity has been underestimated’ (p. 3).

Besides African elephants, this sort of taxonomic lumping has recently been asserted for giraffes (see Giraffes: set for change), the Potamochoerus bushpigs and river hogs (Grubb 1993, Groves 2000), the wild pigs of the Philippines (Oliver 1995, Groves 1997), warthogs (Grubb 1993, Randi et al. 2002), the African buffaloes traditionally all placed together in Syncerus caffer (Grubb 1972, Groves 2000), the bushbucks traditionally grouped together as Tragelaphus scriptus (Grubb 1985, Groves 2000), the red deer and wapiti traditionally grouped together as Cervus elephus (Geist 1999), gorillas (Groves 1996, Sarmiento & Butynski 1996, Sarmiento & Oates 2000), orangutans (Xu & Arnason 1996) and others. As I mentioned in the giraffe post, while Victorian zoologists were in the habit of naming multiple large mammal species, it was mostly 20th century workers who decided to arbitrarily lump these multiple forms together.

As I’ve tried to explain before (go here), while biologists tend to be harshly critical of workers who erect new taxa based on only small apparent differences (such workers become known as ‘splitters’), there is far less criticism of those who indulge in laissez-faire lumping: it’s almost as if the lumpers are regarded as those ones who have the most realistic handle on species-level biodiversity. But do they? Well, as African elephants and the various other examples cited above indicate, no they don’t in all cases and we should be as sceptical of lumping as we are of splitting.

Finally for now, if anyone knows of any employment opportunities that would suit me, please let me know. Seriously. For the latest news on Tetrapod Zoology do go here.

Refs - -

Clayton, L. 2003. Tusk master. BBC Wildlife 21 (1), 52-57.

Geist, V. 1999. Deer of the World. Swan Hill Press, Shrewsbury.

Groves, C. P. 1980. Notes on the systematics of Babyrousa (Artiodactyla, Suidae). Zoologische Mededelingen 55, 29-46.

- . 1996. Do we need to update the taxonomy of gorillas? Gorilla Journal 12, 3-4.

- . 1997. Taxonomy of wild pigs (Sus) of the Philippines. Zoological Journal of the Linnean Society 120, 163-191.

- . 2000. What are the elephants of west Africa? Elephant 2, 7-8.

- . 2001. Mammals in Sulawesi: where did they come from and when, and what happened to them when they got there? In Metcalfe, I., Smith, J. M. B., Morwood, M. & Davidson, I. (eds) Faunal and Floral Migration and Evolution in SE Asia-Australia. A. A. Balkema Publishers (Lisse, The Netherlands), pp. 333-342.

- . & Grubb, P. 1993. The Eurasian suids Sus and Babyrousa. In Oliver, W. L. R. (ed) Pigs, Peccaries and Hippos Status Survey and Action Plan. IUCN/SSC Pigs and Peccaries Specialist Group & IUCN/SSC Hippos Specialist Group (Gland, Switzerland), pp. 107-111.

Grubb, P. 1972. Variation and incipient speciation in the African buffalo. Zeitschrift für Säugetierkunde 37, 121-144.

- . 1985. Geographical variation in the bushbuck of eastern Africa (Tragelaphus scriptus; Bovidae). In Schuchmann, K.-L. (ed) Proceedings of the International Symposium on African
Vertebrates: Systematics, Phylogeny and Evolutionary Ecology. Zoologisches Forchungsinstitut und Museum Alexander Koenig (Bonn), pp. 11-27.

- . 1993. The Afrotropical suids: Potamochoerus, Hylochoerus and Phacochoerus. In Oliver, W. L. R. (ed) Pigs, Peccaries and Hippos Status Survey and Action Plan. IUCN/SSC Pigs and Peccaries Specialist Group & IUCN/SSC Hippos Specialist Group (Gland, Switzerland), pp. 66-75.

- ., Groves, C. P., Dudley, J. P. & Shoshani, J. 2000. Living African elephants belong to two species: Loxodonta africana (Blumencah, 1797) and Loxodonta cyclotis (Matschie, 1900). Elephant 2, 1-4.

Hooijer, D. A. 1950. Man and other mammals from Toalian sites in south-western Celebes. Verhandelingen der Koninklijke Nederlandsche Akademie van Wetenschappen, Afd. Natuurkunde. Tweede sectie 46, 7-164.

Leus, K., Bland, K. P., Dhondt, A. A. & Macdonald, A. A. 1996. Ploughing behaviour of Babyrousa babyrussa (Suidae, Mammalia) suggests a scent-marking function. Journal of Zoology 238, 209-219.

Macdonald, A. A. 1993. The babirusa (Babyrousa babyrussa). In Oliver, W. L. R. (ed) Pigs, Peccaries and Hippos Status Survey and Action Plan. IUCN/SSC Pigs and Peccaries Specialist Group & IUCN/SSC Hippos Specialist Group (Gland, Switzerland), pp. 161-171.

- ., Bowles, D., Bell, J. & Leus, K. 1993. Agonistic behaviour in captive Babirusa (Babyrousa babyrussa). Mammalian Biology 58, 18-30.

MacKinnon, J. 1981. The structure and function of the tusks of babirusa. Mammal Review 11, 37-40.

Meijaard, E. 2003. Note on a photo of two babirusa skulls from Rante-pao, south-central Sulawesi. Asian Wild Pig News 3 (1),12.

- . & Groves, C. 2002a. Proposal for taxonomic changes within the genus Babyrousa. Asian Wild Pig News 2 (1), 9-10.

- . & Groves, C. 2002b. Upgrading three subspecies of babirusa (Babyrousa sp.) to full species level. Asian Wild Pig News 2 (2), 33-39.

Oliver, W. L. R. 1995. The taxonomy, distribution and status of Philippine wild pigs. Ibex J.M.E. 3, 26-32.

Randi, E., d’Huart, J. P., Lucchini, V. & Aman, R. 2002. Evidence of two genetically deeply divergent species of warthog Phacochoerus africanus and P. aethiopicus (Artiodactyla: Suiformes) in East Africa. Mammalian Biology 67, 91-96.

Sarmiento, E. & Butynski, T. 1996. Present problems in gorilla taxonomy. Gorilla Journal 12, 5-7.

- . & Oates, J. F. 2000. The Cross River gorillas: a distinct subspecies, Gorilla gorilla diehli Matschie 1904. American Museum Novitates 3304, 1-55.

Wiles, R. & Mustari, I. A. H. 2002. Records of babirusa and warty pigs in SE Sulawesi. Asian Wild Pig News 2 (2), 31-32.

Xu, X. & Arnason, U. 1996. The mitochondrial DNA molecule of Sumatran orangutan and a molecular proposal for two (Bornean and Sumatran) species of orangutan. Journal of Molecular Evolution 43, 431-437.

The deer-pig, the Raksasa, the only living anthracothere… welcome to the world of babirusas

Among the most unusual and exotic of big mammals that I’ve seen in captivity is the Babirusa: as discussed in a previous post (How big is a white rhino?), Marwell Zoo has a few individuals, and one of these at least is an apparently friendly individual who comes to greet you as you approach his enclosure (see photo below). What with my several recent blog posts about peccaries (go here and here), I thought I might continue the suiform theme, and in between doing other things I’ve been reading a lot about babirusas lately. What have I been doing lately you ask? That would be telling, but recent adventures have involved a buzzard skull, sika deer, those green lizards again (go here for background) and Megamuntiacus. In fact expect a lot on deer here in future. I’m pretty much giving up on dinosaurs. Anyway, back to the babirusas.

You might be surprised to learn that babirusas have been known to westerners for a comparatively long time, having been named Sus babyrussa by Linnaeus in 1758 (the generic name Babyrousa was first coined by Perry in 1811). In fact babirusas were first mentioned in the European literature as early as 1658, and there have even been claims that the Romans knew of babirusas in the 1st century AD. Linnaeus wrongly identified Borneo as the babirusa homeland, and over the following years other authors misidentified Sumatra, Amboina and elsewhere as the place where babirusas came from.

Today we know that they are endemic to Sulawesi, the adjacent Togean (or Togian) Islands, the Sula Islands (just east of central Sulawesi) and Buru (south-east of the Sula Islands, and the most westerly of the Moluccas). What might also surprise you is that the babirusas found across these islands are not all alike, and in fact the sort of babirusas we are all familiar with – those with naked skin and upper canines that curve in a circle – are by no means representative of these animals as a whole. Historically, babirusas were probably present across the whole of Sulawesi, but by the 19th century they had disappeared from the south-western peninsula. As of 1990, they were still present on Buru and two of the Sulu Islands (Mangole and Taliabu), but have become extinct on others (Macdonald 1993). Thanks to logging, habitat destruction and illegal hunting, babirusas are under pressure across their range.

Sula Island and Buru babirusas have often been regarded as introduced, and while this is likely (we know that Sulawesi people kept babirusas and traded them far and wide because of their remarkable appearance and good eating) these populations represent distinct taxa, the biogeographical origins of which remain obscure. It is also conceivable however that these colonisations occurred naturally, as babirusas are very strong swimmers and well able to make short sea journeys. On Sulawesi, babirusas have been observed swimming across the 10-km-wide Lake Poso (Melisch 1994). Adjacent photo from here.

While babirusas look pig-like and are classified as part of Suidae, they are distinctive and unusual. Combining rather slender legs with a barrel-shaped body, they can exceed 1 m in length and have a shoulder height of c. 30 cm. Some individuals weigh as much as 100 kg. Babirusas are odd in having particularly remarkable canines (more about those in a minute), but less well known is that they differ from other pigs in several details of their anatomy, and in fact resemble peccaries and other artiodactyls in a few features. The tendons of their feet and some of their throat muscles are strikingly peccary-like, and they resemble ruminants (though only superficially) in a few details of their pectoral musculature, and in having a complex, multi-chambered stomach. Furthermore, the snout of babirusas is not as specialised as that of other pigs.

In a curious parallel to this combination of anatomical features, the word babirusa combines babi, meaning pig, with rusa, meaning deer. It is supposed that Sulawesi people chose this name as the large canines of babirusas recall antlers, but it is also possible that the name reflects the amalgamation of deer-like slender legs and a multi-chambered stomach with the pig-like traits of the animal. Sulawesi people have always known of pigs other than babirusas, as Sulawesi is also home to the Sulawesi warty pig Sus celebensis*. Incidentally, you may recognise Rusa as the old generic name for the Sambar Cervus unicolor and other deer of Indonesia, the Philippines and adjacent islands. The Sunda sambar or Timor deer C. timorensis occurs on Sulawesi, but was probably introduced there from Java or Bali.

* A large extinct Pleistocene suid, Celebochoerus heekereni, was also endemic to Sulawesi, but so far as we know was not related to either babirusas or Sus celebensis.

In view of the divergent anatomy of babirusas, most artiodactyl specialists agree that they represent an ancient lineage, Babyrousinae, which branched off from the rest of Suidae early in its evolution (Thenius 1970). This is supported by chromosome data, as several autosomes present in babirusas have no equivalent in other suids. Unfortunately babirusa fossils only go back as far as the Pleistocene, but in theory we should expect to find babyrousines going back to the Oligocene. Is it possible that babirusas aren’t part of Suidae? Such a view was favoured by Deninger (1909) who argued that babirusas descended from anthracotheres like Merycopotamus. Anthracotheres, a widespread extinct group that appeared in the Eocene and survived into the Pleistocene in eastern Asia, are probably ancestral to hippos (and thus perhaps not close to suids at all), and while Deninger’s idea has been mostly dismissed it was viewed favourably by Groves (1981). He thought it at least possible that babirusas might really be extant anthracotheres, which is a pretty radical thought. For the record, no, this has not been accepted and babirusas are universally regarded as suids.

Despite the ancient divergence of babyrousines from other suids, a male babirusa recently hybridised with a female domestic pig at Copenhagen Zoo (this was announced in the news last week). Of the five resulting piglets, one has died but the others are apparently ok. While the successful mating has surprised some biologists, we should remember that successful hybridisation can occur between species that are only distantly related, so it actually doesn’t mean much.

Like many (but not all) pigs, babirusas are omnivorous, and are said to eat invertebrates whenever they find them. They have also been reported to eat fish on occasion, to catch small mammals, and even to catch and eat the juveniles of other babirusas (Leus & Morgan 1995). They eat all kinds of plant material, including fruits, leaves, flowers, berries, nuts, bark and tubers, and they not only browse and dig to obtain such items, they are also surprisingly good at standing bipedally (without support) to feed on leaves. This again is a curious parallel with deer, in particular Sambar.

So what’s with the bizarre curving tusks? Present only in males (females lack canines entirely), they grow continuously throughout life, and their growth, anatomy and function are all odd. The lower canine is normal in position and anatomy, it’s just that it becomes particularly long during growth, overlapping the outside edge of the snout as it grows. The upper canine is another story. Initially growing downwards – like any normal mammalian upper canine – it is then rotated as the alveolus itself turns to force the tooth upwards, and it eventually emerges from the dorsal surface of the snout. The most anterior part of the spiral parallels the long lower canines. As mentioned earlier, we are mostly familiar with those babirusas where the upper canines curl in a circle as they grow, forming a spiral over the animal’s forehead. As we’ll see in the next post, spiralling upper canines of this sort are not present in all kinds of babirusas.

Regardless, in those babirusas with spiralling tusks, some authors say that, if the animal lives long enough, the tusks grow fatally into the face (Irven 1996). However, in the old male skulls that I’ve seen (see accompanying images: the woodcut is from Alfred Russel Wallace’s 1869 The Malay Archipelago), the tips of the upper canines begin the anterodorsal part of their curvature a short distance dorsal to the upper surface of the skull, so if they were to continue to grow they would harmlessly curl upwards. Furthermore, so far as I can tell from the literature, no-one has ever found a babirusa skull in which the upper canines have bored into the bone.

A Balinese demon with curling tusks that emerge from its cheeks – the Raksasa – might have been inspired by stories or sightings of babirusas (Groves 1980).

Famously, people on Celebes once supposed that babirusas hung from trees with their tusks, and stay there in wait for passing females. This seems to be the one ‘fact’ about babirusas that everyone knows, as it’s mentioned in just about every article, paper and book that discusses them. It’s often stated that the tusks might be used in display or fighting, but there is also the old idea that the tusks allow males to push their way through dense stands of ratten cane, thereby allowing tusk-less females and juveniles to follow in single file behind. Well, maybe the tusks can be used in this way, but they can’t of course have evolved for this purpose given that the intermediate stages leading up to this ‘end’ condition wouldn’t have been at all useful. More on tusks in the next post.

Perhaps surprisingly in view of their sensitivity to cold, babirusas have fared quite well in zoological collections, having first been kept in Europe at ‘la Ménagerie du Roi’ in Paris during the 1840s, and having bred at London Zoological Garden as early as 1884. In 1995, 29 zoos worldwide held babirusas. Several individuals have survived in captivity for more than 20 years, with the record holder being an animal kept at Chicago which, on its death in 1920, was 21 years and 4 months old. Paul Irven (1996) wrote that captive babirusas are ‘sensitive and responsive … with an endearing character’. They are also said to exhibit excitement and enthusiasm on greeting familiar people, engaging in tail wagging, head shaking and jumping and running about. This friendly disposition makes them quite different from many other non-domesticated suids.

And that’s that for now: more on babirusas in the next post. Coming soon: Up-close and personal with Red deer, Why Draco volans is boring, Temnospondyls for beginners, Kinglets and the passerine supertree, Fake Chinese turtles, Steep Holm and the biggest slow-worm ever, Naish’s guide to Rhinogradentia, Really flying lemurs, The habits of storks, The probing guild, Recently extinct island-dwelling crocodilians and Our lost tree frogs. And for the latest news on Tetrapod Zoology do go here.

Refs - -

Deninger, K. 1909. Über Babirusa. Ber. Naturf. Ges. Freiburg 17, 179-200.

Groves, C. P. 1980. Notes on the systematics of Babyrousa (Artiodactyla, Suidae). Zoologische Mededelingen 55, 29-46.

- . 1981. Ancestors for the Pigs: Taxonomy and Phylogeny of the Genus Sus. Technical Bulletin 3, Department of Prehistory, Research School Pacific Studies, Australian National University.

Irven, P. 1996. The Babirusa. Mainly About Animals 29, 5-7.

Leus, K. & Morgan, C. A. 1995. Analyses of diets fed to babirusa (Babyrousa babyrussa) in captivity with respect to their nutritional requirements. Ibex J.M.E. 3, 41-44.

Macdonald, A. A. 1993. The babirusa (Babyrousa babyrussa). In Oliver, W. L. R. (ed) Pigs, Peccaries and Hippos Status Survey and Action Plan. IUCN/SSC Pigs and Peccaries Specialist Group & IUCN/SSC Hippos Specialist Group (Gland, Switzerland), pp. 161-171.

Melisch, R. 1994. Observation of swimming babirusa in Lake Poso, central Sulawesi, Indonesia. Malayan Nature Journal 47, 431-432.

Thenius, E. 1970. Zur Evolution und Verbreitungsgeschicht der Suidae (Artiodactyla, Mammalia). Zeitschrift für Säugetierkunde 35, 321-342.

The quest for normality

I recently returned home. See if - from this cryptic range of photos - you can work out where I was (and extra points to those who can spot the significance of the plaster cast). Posts to follow soon: babirusas first. For the latest news on Tetrapod Zoology do go here.

The Cupar roe deer carcass

Britain is home to exotic cats of several species, big and small. If this seems like a bold assertion to make, I should point out that the evidence for the existence of so-called alien big cats, or ABCs, is compelling and if you want to know how I’ve arrived at this conclusion you should start by reading a former blog post: British big cats: how good, or bad, is the evidence? Once more, ABCs are the subject of much discussion here in Britain, in part because the TV company Channel Five is advertising a ‘big cat search’ project (website here). It’s fronted by Nick Baker, one of the few TV naturalists whose knowledge and approach to the subject I respect.

British ABCs have also been in the news this year because Danny Bamping, founder of the British Big Cat Society, has reported his successful exhuming of a puma skull from north Devon in July 2005. Furthermore, the first formal conference devoted to British ABCs was held in March of this year at Market Harborough in Leicestershire. Because the conference was organized by a researcher who believes that ABCs are ghost-like entities from a parallel dimension I chose not to attend, but as luck would have it she wasn’t involved in the end and Jon Downes had to step in to handle the meeting. Jon is also organizing the cryptozoology conference that I’m attending later this week.

Among the pieces of evidence used by some to support the reality of ABCs have been livestock corpses. For many years farmers and other people have reported finding the carcasses of large mammals – mostly sheep but also calves, foals and other livestock – that seem to have been killed by ABCs (for photos see Brierly 1989, Francis 1983, 1993). Supposedly, the wounds present on these corpses, and the manner in which they have been gutted and/or eaten, are diagnostic of felid killers. But like many who have tried to examine this body of evidence impartially, I remain sceptical, and in virtually all cases it is never really clear that dogs can be excluded outright. But there is one exception that stands head and shoulders above all the others: the Cupar roe deer carcass.

On the night of June 16th 2001, journalist Ralph Barnett was driving home from Dundee to Cupar (north-east Fife, Scotland). As a journalist, Barnett has admitted familiarity with the subject of ABCS, and in particular with the ABCs of Scotland, but he had no special prior interest in the subject. On rounding a bend and coming out of a slight dip in the road, he switched his headlamps to full beam. What he took to be the headlamps of another car immediately ahead caused him to undertake an emergency stop, but it wasn’t a car in front of him, it was – so he reports – a big dark-coloured cat. It leapt away out of sight, and as it did Barnett realised that it had been feeding on the carcass of a Roe deer Capreolus capreolus, still lying there in the road.

Barnett called the local police on his mobile phone and they ‘attended in significant numbers – certainly more than would normally be available for a disturbance in Cupar town centre at that time on a Saturday night’. The police elected not to retain the carcass and it was unfortunately dumped at the roadside and left there, but Barnett took excellent photos, all of which have been posted on the Scottish big cats website. A detailed description of the carcass was posted to accompany the images, and after being asked questions about the carcass by several ABC investigators Barnett supplied further additional details.

As seen in the accompanying close-up, the deer seems to have been killed by asphyxiation. This is evidenced by bulging eyes, an open mouth with protruding tongue and clotted blood pooled on the side of the face. The eyeballs were ruptured and still moist. A series of sub-parallel lacerations on the side of the neck look exactly like claw marks (and were interpreted as such by Barnett): they were deep grooves incised into the neck.

The fact that the carcass was in the middle of the road suggests that it was dragged there (Barnett suggested that the cat was in the process of moving the carcass when he chanced upon it). In keeping with this the carcass had been eviscerated, and what appeared to be a sub-circular grip mark was present on one of its shoulders. The carcass was cold to the touch and without signs of decomposition, and both Barnett and a police officer agreed that it had been dead for less than 48 hours. The tip of one of the antlers was broken off, which would also be in keeping with the carcass having been dragged across the road surface. The entire carcass was split open along its ventral surface, the bones of its pelvis were partially dislocated, and its left hindlimb was defleshed right down to the bones. Its ribs had apparently been cleanly broken. Barnett reported that moist blood, tufts of deer hair and disturbed earth were present at the side of the road.

So far as I can tell – and this opinion is echoed by those who have investigated the details provided by Barnett – this is a pretty convincing big cat kill. The extensive trauma present on the carcass simply cannot have been caused by anything else. The good evidence for asphyxiation strongly suggests that the deer was killed by a conventional felid throat-hold: if anyone can come up with a better explanation for bulging, ruptured eyes, a protruding tongue and clotted blood massed on the side of the face I’d like to hear it. The only way you could fake this is by catching the deer live and strangling it to death by hand, and this doesn’t strike me as likely.

In the adjacent image the labels denote the following: A. SIGNS OF ASPHYXIATION; A1. Mouth open / tongue swollen; A2. Face congested with blood / eyes bulging; A3. Neck raked by teeth and claws. B. SIGNS OF BEING DRAGGED; B1 Bite or grip mark on back shoulder; B2. Broken antler tip. C. DAMAGE TO CARCASS; C1. Split from breastbone to groin; C2. All internal organs missing; C3. Pelvis dislocated; C4. Rear left leg stripped of flesh. With credit to the Scottish Big Cat Trust.

For me, this case is a big deal as it’s the only truly compelling British big cat kill: there are others, sure, but the evidence hasn’t been as well documented or reported, nor is it available. Whether Ralph Barnett really encountered a big cat crouching over that carcass is of course something that only he knows, though personally I see little reason to doubt the veracity of his account. However, whether he saw what he said he did or not is irrelevant as the photos speak for themselves. Given that the other lines of evidence we have for British ABCs – the hair, photographic evidence, and the dead bodies – already demonstrate that the animals are a reality, it is inevitable that genuine big cat kills would be discovered and documented eventually. In my opinion the Cupar roe deer carcass is the first good, well documented example, and as such it's an important piece of additional evidence for ABC reality. Comments - negative or otherwise - welcome!

Finally, it’s worth noting that Roe deer are ideal prey for big cats like pumas and leopards, and they are regularly predated upon by leopards where the two coexist.

Incidentally, Britain’s roe deer are usually thought of as native – in fact together with Red deer Cervus elephus they are always said to be our only truly native deer (and this is in a country with seven wild deer species). But it’s little known that Roe deer were in fact extinct across most of southern Britain by the 18th century and have since been restocked from elsewhere (mostly from Scotland). Nowak (1999) – that’s Walker’s Mammals of the World (Sixth Edition) – cited Christopher Lever’s The Naturalized Animals of the British Isles (Lever 1977) as the source for this, but Lever only mentions roe deer once and not in connection with this successful reintroduction. An excellent source on the history of roe deer in Britain is Richard Prior’s Living With Deer (Prior 1965). While a handful of English roe deer might be true natives, it’s only really those of Scotland that represent the original populations. More on alien deer in a future post.

Note that this post was promised a loooong long time ago: see British big cats: how good, or bad, is the evidence? and The bear-eating pythons of Borneo and Ichthyosaur wars and marvellous mixosaurs and Toys, toys, toys and How big is a white rhino. I may take my time, but I do keep my promises, see. For the latest news on Tetrapod Zoology do go here.

Refs - -

Brierly, N. 1988. They Stalk by Night – the Big Cats of Exmoor and the South West. Yeo Valley Publications, Bishops Nympton.

Francis, D. 1983. Cat Country. David and Charles, Newton Abbot.

- . 1993. The Beast of Exmoor and Other Mystery Predators of Britain. Jonathan Cape, London.

Lever, C. 1977. The Naturalized Animals of the British Isles. Hutchinson & Co, London.

Nowak, R. M. 1999. Walker’s Mammals of the World, Sixth Edition (two volumes). The Johns Hopkins University Press, Baltimore and London.

Prior, R. 1965. Living With Deer. Andre Deutsch, London.

How big is a white rhino?

Despite efforts to resist, I cannot help but post yet more Marwell Zoo photos. Mostly these photos were taken by Mark Witton: thanks Mark. Firstly, I'm sure you're wondering how big a White rhino Ceratotherium simum is. Well, here's the answer. Ok, so it's a wooden cut-out and not a real rhino, but I assure you it's life-size. Graeme (at rear) is over 6 ft tall: I'm 5'10.



And speaking of Graeme, here comes the long-awaiting, super-predictable photo entitled 'Graeme meets the folks'. He knew what we were up to when we took the photo. The artiodactyl in the photo is a charismatic and very friendly male Babirusa. Until recently I would have labelled the species Babirousa babyrussa, but - yet again - the species-level taxonomy of babirusas has recently been revised (for a previous post on over-zealous lumping in extant megamammals see Giraffes: set for change). A blog post is planned: if you can't wait until then check out Meijaard & Groves (2002a, b). Babirusa males have large curving upper canines that curve dorsoposteriorly as they emerge from the dorsal surface of the snout (yes, they emerge from the dorsal surface of the snout). This male is in the habit of covering his tusks with wet mud, thereby obscuring them from view. I have no idea why he does this and have never read of this behaviour.













Here's a photo of my back and the big male giraffe they have at Marwell. I have nothing interesting to say about it, but you can never get bored with giraffes can you.
















As mentioned in the ground hornbill post, we were 'frustrated by anteaters'. What did I mean by this? Well, on my previous visit to Marwell the anteater had remained asleep and curled up, tail folded over its body and head (for photo go here). And this time it was in exactly the same place and exactly the same position. Except for a few brief seconds when it raised its tail and lifted its head: Mark was quick enough to get this photo. I suppose some time it might get up and walk around, but as for whether or not I'll ever see this, I do not know.

Coming next: finally, the Cupar roe deer carcass (for background info see British big cats: how good, or bad, is the evidence?). For the latest news on Tetrapod Zoology do go here.

Refs - -

Meijaard, E. & Groves, C. 2002a. Proposal for taxonomic changes within the genus Babyrousa. Asian Wild Pig News 2 (1),9-10.

- . & Groves, C. 2002b. Upgrading three subspecies of babirusa (Babyrousa sp.) to full species level. Asian Wild Pig News 2 (2),33-39.

Bucorvids: post-Cretaceous maniraptorans on the savannah

With Dale Russell and Ron Séguin’s ‘dinosauroid’ in mind – a scaly humanoid suggested by them as a possible (and hypothetical) descendant of Cretaceous troodontid theropods (Russell & Séguin 1982) – I stared through the cage bars at the menacing, striding bipedal predator, strikingly coloured in black and brick red. Bored, it picked up a dead mouse and threw it around for fun. It picked up a stone in its mouth and repeatedly dropped it onto the little furry corpse. Russell & Séguin based their logic on the misguided principle that big-brained vertebrates would inevitably come to resemble hominids in posture and body shape, but if the speculations of some ornithologists are to be believed, then the awesome feathered dinosaur I watched at the zoo – a Southern ground hornbill Bucorvus leadbeateri – was a genuine sort of avian pseudo-hominid.

Yes I’ve been to Marwell zoo again (see What I saw at the zoo and More on what I saw at the zoo). Mark Witton and Graeme Elliott were in attendance: it was a sort of University of Portsmouth Palaeobiology Research Group Day Out. We were frustrated by anteaters, humbled by rhinos, awed by megabats, and salivated on by giraffes.

Ground hornbills are big birds, reaching 4 kg in weight and with a wingspan that can reach 2 m. They're entirely black except for their white primaries and for the bright red fleshy wattles that decorate their faces and throats (a second species in the genus Bucorvus [which means ‘big crow’], the Northern or Abyssinian ground hornbill B. abyssinicus, has red and blue facial skin in males and all-blue skin in females). Inhabiting the wooded savannahs of tropical southern African, Southern ground hornbills forage on the ground, picking up animal prey from among the grass or from low-growing vegetation, and grabbing everything from insects, snails, arachnids and worms to lizards, snakes, tortoises and even squirrels and hares. They may also excavate wasp and bee nests. Prey is grasped delicately with the forceps-like bill tips and snakes and other dangerous prey may be repeatedly squeezed at the jaw tips until dead. Prey is swallowed using a ‘throw and gulp’ action.

The jaws are laterally compressed and gently curved, and though the bill tips meet when the jaws are in occlusion, a gap is usually visible part way along the jaws, even when they’re closed. The size of the gap seems variable between individuals, and it’s much less pronounced than that seen in some other hornbills, like the red-billed and yellow-billed Tockus species. All hornbills share a heavily reinforced bill: the palatal surface forms a complete bony roof to the mouth, and all of the palatal bones are fused together. This is a modified version of the desmognathous condition and it presumably evolved to provide mechanical support for the bill: among so-called ‘higher landbirds’ it is present in birds that kill and consume active prey, as well as in forms that beat prey dead against a perch. Among this lot are motmots, bee-eaters and rollers. This isn’t the whole story behind the desmognathous palate incidentally, as it’s also present in waterfowl, ibises, spoonbills, pelicans and other groups (Huxley 1867).

Hornbills are also unusual in possessing a unique ligament – the quadratomandibular ligament – that connects the body of the quadrate with the inside surface of the lower jaw. In ground hornbills the anatomy of the quadratomandibular ligament makes it impossible for the upper jaw to be raised or lowered without an automatic lowering or raising of the lower jaw, and the ligament is so strong that the lower jaw stays connected to the skull even in dried skeletons (Burton 1984). The hornbill quadratomandibular ligament was only discovered in 1940, and as Walter Bock noted in 1964 it has been mostly overlooked and little studied. Oh, if all this talk of ‘raising or lowering the upper jaw’ confuses you, I should explain that a special hinge zone at the base of the bill allows hornbills (and other birds with the same sort of hinge zone) to raise and lower the entire upper jaw relative to the rest of the skull, an ability termed prokinesis.

Given the importance of the bill tips in handling prey, it would make sense if the birds could see the end of their own bill, and indeed work on the visual field perceived by the birds indicates that the bill tips do indeed intrude into the lower part of the hornbill’s binocular field. This is highly unusual among birds and it seems that the huge decurved bill and precision-grasping feeding technique that hornbills employ evolved in concert with their visual field (Martin & Coetzee 2004) [if you’re really interested in this particular subject do check out Graham Martin’s website]. Immense, stiff, flattened and widely spaced black eyelashes, the longest of which are 18 mm long, encircle the eye and overhang its upper part. The birds seem to use these as sunshades, deliberately positioning the head so that the cornea is shaded from bright light when appropriate (Martin & Coetzee 2004).

Several features of the hornbill head and neck seem specially suited to help support the big, heavy head. Ground hornbills have more cervical vertebrae than other hornbills (15 opposed to 14), but like other hornbills their atlas and axis (the two vertebrae closest to the skull) are fused together. This is unique among birds, with the exception of one-off freaks. While most birds have just one occipital condyle, hornbills have two, as there is an accessory one on the supraoccipital (the bone that forms the upper border to the foramen magnum). Some of their neck muscles have special accessory slips that insert further down the cervical column than is typical for birds, and presumably this assists in carrying and moving the large head (Burton 1984). What’s also odd about ground hornbill neck anatomy is that they have no cranial carotid arteries (Ottley 1879), and don’t ask me how that works.

What of course makes hornbills immediately distinct is the presence of the bony casque that sits on top of the bill. In ground hornbills this is a modest structure located dorsal to the proximal part of the bill, ramified internally by large spaces and sheathed in life by keratin (in the Northern ground hornbill the casque is much taller than that of the southern species). Basal hornbills, which probably include the ground hornbills (read on), have a small casque that is often little more than a mid-line dorsal ridge on the bill, but in more derived forms it is a large, mostly hollow structures supported internally by strut-like trabeculae (Kemp 1995): for images of sectioned casques go here. Casques probably function as identification devices as they differ in shape between species, sexes and growth stages, but an intriguing proposal is that they function as resonating devices used to amplify calls (Alexander et al. 1994). Those of you reading Bousfield & LeBlond’s (1995) Cadborosaurus monograph right now – you know who you are – will be amused to see Alexander et al.’s paper cited therein, I kid you not.

In terms of casque morphology, the most strongly modified hornbill is the Helmeted hornbill Rhinoplax vigil, a territorial fig specialist of SE Asia. The casque of this species is solid, looks like a block of ivory – it’s actually called hornbill ivory* – and, believe it or don’t, is employed in aerial jousting. During this behaviour flying males smack their casques together, the resulting CLACK being audible from at least 100 m away (Kinnaird et al. 2003). The blunt leading edge of the casque might actually result from this head-butting behaviour. If you want to know what a Rhinoplax casque looks like in section, visit the relevant page on Matt Wedel's site (here). Thanks to Matt for reminding me about this.

* It’s prized for carving by Chinese and Japanese artists, being used by the latter for netsuke sculptures.

Anyway, I digress. Compared to other hornbills, ground hornbills have elongate tarsometatarsi, and they also have short toes, though strangely with the hallux being the longest of the four (note that this is rarely depicted accurately in artwork). The scutes covering the tarsus look odd: large, non-imbricating scutes are scattered along the anterior surface like crazy paving, and not arranged in a row as is normally the case in birds with large tarsal scutes. Ground hornbills are reported to be able to run extremely quickly, apparently at speeds of up to 30 km/h, and though they do their foraging by striding around on the ground, they’re capable, though reluctant, fliers (see adjacent image, borrowed from Natural Encounters).

Their social behaviour is interesting in that non-breeding individuals, including immatures of both sexes and adult males, assist breeding pairs in raising their young. As many as 12 non-breeders may collaborate to assist a single breeding pair. All of these birds are related and know each other well, and they use intimidation, play, mutual grooming and other forms of interaction to cement social bonds. It is well known that hornbill females become walled into a cavity nest while they incubate their eggs and raise their chicks, but ground hornbills are unique among the group in not doing this. Natural cavities in rocks or trees are used, and of the 1-3 eggs laid only one chick ever survives. The ‘surplus’ chicks may perhaps be an insurance should the first-hatched chick die. Apparently no ground hornbill has ever been known to raise more than one chick, but if you look at the title of Kemp & Kemp (in press) below – a reference I borrowed from Alan Kemp’s website – you’ll see that some new data on this is soon to be published. Kemp is a leading expert on hornbills and has been involved in a long-term field study of ground hornbills.

Once chicks fledge, their mortality rates are low (31% survive to maturity), but they’re tremendously slow in maturing, taking about 6 years. In keeping with this, ground hornbills are very slow breeders, fledging on average one chick every 9 years. Pairs may go for as long as 18 years without breeding. Unsurprisingly, ground hornbills are highly vulnerable to human persecution and habitat loss, and because they rely on a few species of large trees for nesting, and even then require special trees that have natural cavities, the number of available nest sites may be a limiting factor on their distribution. As you would guess for all this K-strategy behaviour, they are long-lived: Kemp (1996) suggested that they might be among the longest-lived of birds, perhaps surviving into their fourth decade.

So in ground hornbills we seem to have slow-breeding, terrestrial, ground-feeding, carnivorous savannah-dwellers that belong to a clade of mostly arboreal, forest-dwelling frugivores. Kemp has noted that ground hornbills therefore seem to have followed a similar evolutionary path to us hominids (Kemp 1996). The complex social behaviour and apparent intelligence of the birds also makes them reminiscent of primates, though note that I don’t want to promote the horrible opinion that the only time non-mammals become interesting is when they somehow seem mammalian in their behaviour or biology.

Phylogenetic studies indicate that ground hornbills are most closely related to the SE Asian Buceros hornbills, all of which are arboreal inhabitants of tropical forests (Kemp 1995, 2001). It is also fairly well agreed nowadays that Murie (1873) was right in arguing that hornbills are close allies of hoopoes and wood-hoopoes, the Upupiformes (Burton 1984, Olson 1985, Mayr 2000, 2003, Mayr et al. 2003). In fact some workers now use the term Upupiformes for this hornbill + hoopoe + wood-hoopoe clade (Mayr 2003). If you’re more familiar with the inclusion of these birds within Coraciiformes – the name previously used for rollers, kingfishers, bee-eaters, todies, motmots, hoopoes and hornbills – then you’ll be interested to know that there is now good evidence indicating that this grouping is artificial. More about that another time.

Anyway, this phylogenetic data indicates that the ancestors of ground hornbills were woodland or forest birds, and that the unusual behavioural characters of ground hornbills – the carnivory, the terrestriality, the striding gait (other hornbills hop when on the ground), the savannah habitat – are derived novelties. Indeed it’s been suggested at times that the Bucorvus species are so unusual compared to other upupiforms that they should get their own ‘family’, Bucorvidae (Kemp 1995). This wasn’t followed in the most recent review of the group however (Kemp 2001). What’s more, we know that ground hornbills have been doing what they do for quite a long time, as a fossil member of the genus, Bucorvus brailloni, is known from the Miocene (Brunet 1971). It’s from north of the Moroccan Atlas Mountains, and hence well north of where ground hornbills occur today.

While there are good phylogenetic reasons for thinking that ground hornbills descended from arboreal forest-dwelling ancestors, it’s unfortunate that there aren’t yet any fossils to support this. A few alleged hornbill fossils have been reported from Eocene and Miocene Germany and France (Geiseloceros robustus, Cryptornis antiquus and Homalopus picoides) but Olson (1985) showed that none of them are really hornbills. However, hornbills almost certainly originated in the Eocene at least, as we now know that close relatives of hoopoes had appeared by this time (Mayr 2000). We are simply missing the fossils.

Of course you don’t need to compare ground hornbills with Cretaceous predatory dinosaurs, or with hominids or other primates, to make them interesting. But I can’t help coming back to Russell & Séguin’s ‘dinosauroid’ hypothesis. No, post-Cretaceous maniraptorans wouldn’t end up looking like scaly tridactyl plantigrade humanoids with erect tailless bodies. They would be decked out with feathers and brightly coloured skin ornaments; have nice normal horizontal bodies and digitigrade feet; long, hard, powerful jaws; stride around on the savannah kicking the shit out of little mammals; and in the evenings they would stand together in the trees, booming out a duet of du du du-du, a deep noise that would reverberate for miles around.

For the latest news on Tetrapod Zoology do go here.

Refs - -

Alexander, G. D., Houston, D. C. & Campbell, M. 1994. A possible acoustic function for the casque structure in hornbills (Bucerotidae). Journal of Zoology 233, 57-67.

Brunet, J. 1971. Oiseaux miocénes de Beni-Mellal (Maroc); un complement à leur etude. Notes et Mémoires Services Géoloques (Morocco) 31, 109-111.

Bousfield, E. L. & LeBlond, P. H. 1995. An account of Cadborosaurus willsi, new genus, new species, a large aquatic reptile from the Pacific coast of North America. Amphipacifica 1 (supplement 1), 1-25.

Burton, P. J. K. 1984. Anatomy and evolution of the feeding apparatus in the avian orders Coraciiformes and Piciformes. Bulletin of the British Museum of Natural History (Zoology) 47, 331-443.

Huxley, T. H. 1867. On the classification of birds; and on the taxonomic value of the modification of certain of the cranial bones observable in that class. Proceedings of the Zoological Society of London 1867, 415-472.

Kemp, A. C. 1995. The Hornbills. Oxford University Press, Oxford.

- . 1996. Hammer of the savannah. BBC Wildlife 14 (5), 32-36.

- . 2001. Family Bucerotidae (Hornbills). In del Hoyo, J., Elliott, A. & Sargatal, J. (eds) Handbook of the Birds of the World, vol. 3. Lynx Edicions, Barcelona, pp. 436-523.

- . & Kemp, M. I. In press. How often might Southern ground hornbills fledge two chicks? Data from the Kruger National Park, 1967-1999. In Proceedings 4th International Hornbill Conference, Mabula Game Lodge, Bela Bela.

Kinnaird, M. F., Hadiprakarsa, Y.-Y. & Thiensongrusamee, P. 2003. Aerial jousting by Helmeted hornbills Rhinoplax vigil: observations from Indonesia and Thailand. Ibis 145, 506-508.

Martin, G. R. & Coetzee, H. C. 2004. Visual fields in hornbills: precision-grasping and sunshades. Ibis 146, 18-26.

Mayr, G. 2000. Tiny hoopoe-like birds from the Middle Eocene of Messel (Germany). The Auk 117, 964-970.

- . 2003. On the phylogenetic relationships of trogons (Aves, Trogonidae). Journal of Avian Biology 34, 81-88.

- ., Manegold, A. & Johansson, U. S. 2003. Monophyletic groups within ‘higher land birds’ – comparison of morphological and molecular data. Journal of Zoological and Systematic Evolutionary Research 41, 233-248.

Murie, J. 1873. On the Upupidae and their relationships. Ibis 15, 181-211.

Olson, S. L. 1985. The fossil record of birds. In Avian Biology, Volume III, pp. 79-238.

Ottley, W. 1879. Bucorvus abyssinicus, on the vessels in its head and neck. Proceedings of the Zoological Society of London 1879, 461-467.

Russell, D. A. & Séguin, R. 1982. Reconstruction of the small Cretaceous theropod Stenonychosaurus inequalis and a hypothetical dinosauroid. Syllogeus 37, 1-43.

We flightless primates

Mention ‘flying primate’ and most zoologists will think you’re referring to the well known, controversial theory of John Pettigrew of the University of Queensland. Initially basing his theory on retinotectal organization (viz, the way in which data from the retina is processed in the brain), Pettigrew (1986) argued that megabats (the group that includes fruit bats) are not close relatives of microbats (the mostly small, mostly insectivorous bats that mostly use echolocation), but that they’re actually flying primates, of a sort (read on). If this is right, then Chiroptera [the proper name for the bat clade] is not monophyletic, and true flight evolved at least twice among mammals. Furthermore, Primates and their close relatives are more diverse than conventionally thought, and include an impressive radiation of hitherto-misclassified volant species.

In more detailed papers, Pettigrew (1991) and Pettigrew et al. (1989) marshaled evidence from eye, brain and spinal cord anatomy, fore- and hindlimb, finger and metacarpal proportions, and haemoglobin sequences, and again concluded that megabats and primates shared a common ancestor, and that microbats were not close relatives of megabats, but that their affinities lay elsewhere. Pettigrew et al. (1989) further argued that colugos (aka flying lemurs, or dermopterans) were also part of the megabat-primate clade, and essentially late-surviving relics which resembled the common ancestor of the megabat-primate clade. Pettigrew and colleagues weren’t the first to question bat monophyly: John E. Hill of the then British Museum (Natural History) had done this as early as 1976, Smith & Madkour (1980) argued that micro- and megabats were of separate origins, and Hill & Smith (1984), in one of the best and oft-cited overviews on bat evolution and biology, expressed scepticism of bat monophyly and a preference for megabat-primate affinities (p. 36).

Popularly, Pettigrew et al.’s notion that megabats are closer to primates than to microbats became characterized as the ‘flying primate’ theory, and Pettigrew et al. used this term in their papers. This characterization isn’t accurate however as Pettigrew et al. (1989) specifically stated that, within their favoured phylogenetic scheme, megabats would be outside of the clade Primates, and thus not primates in the true sense (p. 551).


I must admit that the ‘flying primate’ theory has a great deal of intuitive appeal, and this probably explains why it’s become both well known and much written about. Colin Tudge wrote about it in The Independent for example (his article was titled ‘That’s no bat, that’s my brother'), and it isn’t every day that problem areas within the higher-level phylogenetics of placental mammals make it into daily newspapers. Why is the ‘flying primate’ hypothesis intuitively appealing? Perhaps because it appears to have reasonable and easily understood character support and is exciting in contending that a complex and strongly modified bit of morphology – namely the bat wing, with its bizarre elongated fingers and patagial membranes – evolved independently more than once.

The accompanying image by Peter Schouten [click for larger version], commissioned by John Pettigrew, depicts the 'flying primate' hypothesis in graphic form: note that colugos and megabats branch off from the primate lineage, and aren't alongside microbats. This image is borrowed from Pettigrew's Neuroscience UQ site.

However, from the start most bat experts had a problem with the concept: Wible & Novacek (1988) showed how numerous skeletal and soft tissue features ‘strongly support the inclusion of megabats and microchiropterans within the single order Chiroptera’ (p. 1). Simmons et al. (1991) argued that Pettigrew’s entire approach to the bat monophyly question was flawed, Thewissen & Babcock (1992) argued that the detailed anatomy of wing musculature best supported bat monophyly, while Bailey et al. (1992) and Ammermann & Hillis (1992) found that genetic data better supported the case for bat monophyly than that for diphyly. Bailey et al.’s paper is even titled ‘Rejection of the ‘flying primate’ hypothesis’, and in an accompanying news piece in Science Ann Gibbons wrote how the ‘flying primate’ hypothesis was ‘heading for a crash landing’. A comprehensive overview of all the data supporting bat monophyly was provided by Nancy Simmons (1994).

However, DNA-based studies later produced by Pettigrew and his colleagues produced a new, even more surprising result: rhinolophoids (the horseshoe bats and their relatives) were consistently found to group together with megabats, the implication now being that Microchiroptera might be non-monophyletic (Hutcheon et al. 1998, Kirsch & Pettigrew 1998, Pettigrew & Kirsch 1998). The existence of a rhinolophoid-megabat clade has also been supported by other research teams (e.g., Teeling et al. 2000, 2002, Liu et al. 2001) but note that many of these studies (Hutcheon et al. 1998, Teeling et al. 2000, 2002, Liu et al. 2001) did not support Pettigrew's idea of chiropteran diphyly.

Let’s suppose for a moment however that Pettigrew and colleagues are right, and that neither Chiroptera nor Microchiroptera are monophyletic. Suppose also that bats are members of the placental mammal group Archonta – an idea that was quite widely accepted until recently (read on). Archonta was first named by William King Gregory in 1910* to house Scandentia (tree shrews), Chiroptera, Dermoptera and Primates (Gregory also included sengis/elephant shrews within Archonta, but they were later removed from it as their affinities clearly lie elsewhere). As the only mammalian clade whose constituent groups are all tied to a climbing lifestyle, it was always very nice to have the bats in there, given that an arboreal origin for bat flight has always been favoured by everyone who’s ever pondered the issue.

* It’s always been problematical that archontan monophyly has only ever been supported by two morphological characters: the anatomy of the penis, and the shape of one of the facets on the astragalus. By their genitals and ankles ye shall know them. If you’re wondering, Archonta means ‘ruling beings’. Yuck.

As you see from the little cladogram I’ve knocked up here [click for larger version], bat diphyly and archontan monophyly makes it at least possible – and phylogenetically parsimonious – that flight was primitive for the megabat-primate clade, or in other words that primates are secondarily flightless. That’s right: you and me, and other apes, and lemurs, tarsiers and monkeys all descend from flying, winged proto-primates. If I were George Olshevsky I might dub this the ‘bats come first’ theory.


But, no, it wasn’t to be. Bat diphyly is not currently favoured by the evidence, and even Archonta has now fallen by the wayside: DNA-based phylogenies now indicate that, while Scandentia, Dermoptera and Primates probably do form a clade – the newly dubbed Euarchonta – bats aren’t close relatives of euarchontans at all. Instead they go elsewhere within the placental mammal clade Laurasiatheria, being (most surprisingly) closest to carnivorans, perissodactyls and artiodactyls (e.g., Murphy et al. 2001, Liu et al. 2001). And what of the idea that microbats aren’t monophyletic? Alas, a recent review found that this arrangement was not as well supported by the data as is the traditional view of microbat monophyly (Jones et al. 2002).

As always the full story that I wanted to cover is even more complex than what’s related here. I haven’t even mentioned the paromomyids: a group of Eocene euarchontans proposed at one time to be colugo-like gliders and thus implicated in the issue of primate and colugo origins. Nor have I discussed the little nyctitheres, a poorly known and long-mysterious Eocene-Oligocene group that appear to have been primitive, scansorial relatives of euarchontans. In fact comparison of nyctitheres with bats and euarchontans led Hooker (2001) to restate the case for archontan monophyly, with Deccanolestes from the Late Cretaceous of India being the oldest member of the group according to his study. Genetic data clearly doesn’t favour archontan monophyly right now, but there is still at least some morphological support for it. A common problem.

Here’s the great irony of this post. I started writing about the whole ‘flying primate’ thing because I planned to use it as the introductory few paragraphs for a different topic. So while most zoologists think Pettigrew and megabats when hearing of ‘flying primates’, far less well known is that primates proper – and let me make it clear this time that I really do mean primates in the traditional sense (viz, without the megabats) – do however include species that fly. Well, fly sort of. More to come soon.

For previous posts on bats see Greater noctules: specialist predators of migrating passerines and Chewed bones and bird-eating microbats. For stuff on vampire bats please go here.

Update (added 5-8-2006): a new paper just published by Nishahara et al. (2006) has reported new genetic support for a laurasiatherian clade composed of bats, perissodactyls, carnivorans and pangolins. They name this clade Pegasoferae, a name derived by uniting Pegasus (in their view a sort of bat-perissodactyl combination) with Ferae (the name they use for the carnivoran + pangolin clade). Thanks to Stephen Bodio's Querencia and microecos.

Nishihara, H., Hasegawa, M. & Okada, N. 2006. Pegasoferae, an unexpected mammalian clade revealed by tracking ancient retroposon insertions. Proceedings of the National Academy of Sciences 103, 9929-9934. Pdf here.

Refs - -

Ammerman, L. K. & Hillis, D. M. 1992. A molecular test of bat relationships: monophyly or diphyly? Systematic Biology 41, 222-232.

Bailey, W. J., Slightorn, J. L. & Goodman, M. 1992. Rejection of the ‘flying primate’ hypothesis by phylogenetic evidence from the globin gene. Science 256, 86-89.

Hill, J. E. & Smith, J. D. 1984. Bats: a Natural History. British Museum (Natural History), London.

Hooker, J. J. 2001. Tarsals of the extinct insectivoran family Nyctitheriidae (Mammalia): evidence for archontan relationships. Zoological Journal of the Linnean Society 132, 501-529.

Hutcheon, J. M., Kirsch, J. A. W. & Pettigrew, J. D. 1998. Base-compositional biases and the bat problem. III. The question of microchiropteran monophyly. Philosophical Transactions of the Royal Society of London B 353, 607-617

Jones, K. E., Purvis, A., MacLarnon, A., Bininda-Emonds, O. R. P. & Simmons, N. B. 2002. A phylogenetic supertree of the bats (Mammalia: Chiroptera). Biological Reviews 77, 223-259.

Kirsch, J. A. W. & Pettigrew, J. D. 1998. Base-compositional biases and the bat problem. II. DNA-hybridization trees based on AT- and GC-enriched tracers. Philosophical Transactions of the Royal Society of London B 353, 381-388.

Liu, F.-G. R., Miyamoto, M. M., Freire, N. P., Ong, P. Q., Tennant, M. R., Young, T. S. & Gugel, K. F. 2001. Molecular and morphological supertrees for eutherian (placental) mammals. Science 291, 1786-1789.

Murphy, W. J., Eizirik, E., Johnson, W. E., Zhang, Y. P., Ryder, O. A. & O’Brien, S. J. 2001. Molecular phylogenetics and the origins of placental mammals. Nature 409, 614-618.

Pettigrew, J. D. 1986. Flying primates? Megabats have the advanced pathway from eye to mid-brain. Science 231, 1304-1306.

- . 1991. Wings or brains? Convergent evolution in the origin of bats. Systematic Zoology 40, 199-216.

- . & Kirsch, J. A. W. 1998. Base-compositional biases and the bat problem. I. DNA-hybridization melting curves based on AT- and GC-enriched tracers. Philosophical Transactions of the Royal Society of London B 353, 369-370.

- ., Jamieson, B. G. M., Robson, S. K., Hall, L. S., McAnally, K. I. & Cooper, H. M. 1989. Phylogenetic relations between microbats, megabats and primates (Mammalia: Chiroptera and Primates). Philosophical Transactions of the Royal Society of London B 325, 489-559.

Simmons, N. B. 1994. The case for chiropteran monophyly. American Museum Novitates 3103, 1-54.

- ., Novacek, M. J. & Baker, R. J. 1991. Approaches, methods, and the future of the chiropteran monophyly controversy: a reply to J. D. Pettigrew. Systematic Zoology 40, 239-243.

Smith, J. D. & Madkour, G. 1980. Penial morphology and the question of chiropteran monophyly. In Wilson, D. E. & Gardner, A. L. (eds) Proceedings of the 5th International Bat Research Conference. Texas Tech Press (Lubbock), pp. 347-365.

Teeling, E. C., Madsen, O., Van Den Bussche, R. A., de Jong, W. W., Stanhope, M. J. & Springer, M. S. 2002. Microbat paraphyly and the convergent evolution of a key innovation in Old World rhinolophoid microbats. Proceedings of the National Academy of Sciences 99, 1431-1436.

- ., Scally, M., Kao, D. J., Romagnoll, M. L., Springer, M. S. & Stanhope, M. J. 2000. Molecular evidence regarding the origin of echolocation and flight in bats. Nature 403, 188-192.

Thewissen, J. G. M. & Babcock, S. K. 1992. The origin of flight in bats. BioScience 42 (5), 340-345.

Wible, J. R. & Novacek, M. J. 1988. Cranial evidence for the monophyletic origin of bats. American Museum Novitates 2911, 1-19.

Why putting your hand in a peccary’s mouth is a really bad idea

In a recent post I discussed the apparent recent discovery of a new, fourth peccary species (Meet peccary # 4). While this case has now become relatively well known in the cryptozoological world, I’ve learnt from Matt Bille* that not all artiodactyl experts are ready to accept the animal, the so-called Giant peccary, as a valid new species. We await developments. But of course there’s lots more to peccaries than their species-level diversity, and here we’re going to look at a few other areas of interest: their former diversity and fossil history, and their fascinating cranial anatomy.

* Best known in the cryptozoology world for the newsletter he previously produced (Exotic Zoology) and for his books: Rumors of Existence and the new Shadows of Existence (see Hancock House website).

Peccaries today are entirely American, so you might be surprised to hear that fossil evidence indicates that they originated in the Old World, and in fact were still there until as recently as the Late Miocene. The oldest named peccary is Egatochoerus* from the Upper Eocene of Thailand (Ducrocq 1994), and similarly-aged taxa are known from southern China. If these forms are peccaries (read on), then they migrated into North America early in their history, as peccaries made their American debut during the Early Oligocene, and possibly even in the Late Eocene given that there are possible records of Thinohyus from this time.

* Incidentally this is one of those taxa whose name incorporates an acronym: in this case EGAT, the Electricity Generating Authority of Thailand.

At about the same time, peccaries also seem to have got from Asia into Europe as several Oligocene genera – including Doliochoerus and Propalaeochoerus – appeared suddenly in western Europe, yet without antecedents. European peccaries, mostly belonging to the clade Doliochoerinae Simpson, 1945, persisted until the end of the Miocene, and members of this group also inhabited Asia during the Miocene. It was also in the Miocene that peccaries invaded Africa. Morotochoerus from Middle Miocene Kenya seems to be ancestral to Schizochoerus (Pickford 1988), a genus that first appeared in Africa but had spread to Asia and Europe by the Late Miocene (Pickford 1978). Africa-Arabia was an island until the Early Miocene, incidentally, when a contact with Asia was formed, allowing African taxa to colonise Eurasia. This is known as the Proboscidea Datum Event as, obviously, it’s the point when African proboscidean groups make their first appearance in Eurasia. The temporal and geographical distribution of Morotochoerus and Schizochoerus indicates that peccaries invaded Africa from Asia during or after the Proboscidea Datum Event, with one lineage then leaving Africa later in the Miocene. In and out of Africa, in other words (which, incidentally, is the title of a paper on primate evolution by C.-B. Stewart and T. R. Disotell).

Some of these Old World peccaries were rather unusual. The doliochoerine Lorancahyus of Miocene Spain, for example, had tubulidentate teeth, and if that word sounds familiar that’s probably because you’ve seen it before as the group name for aardvarks: Tubulidentata. Tubulidentate teeth are traversed by tubules, hence the name, and they’re thought to be an evolutionary response to the ingestion of large abrasive particles (such as quartz grains) such as those taken into the mouth with soil. It’s even been suggested that some fossil teeth from Miocene Spain, originally described as belonging to aardvarks, may actually be from this lineage of peccaries (Pickford & Morales 1998). What were these peccaries doing with their aardvark-like teeth? As usual, we don’t know.

But are these Old World peccaries really peccaries? Well, perhaps not, as some workers have expressed scepticism, stating for example that the identification of Old World peccaries as peccaries proper is ‘not based on synapomorphy’ (Wright 1998, p. 389), the implication therefore being that these taxa might be suids that are convergently peccary-like in a few features. Van der Made (1997) proposed that Old World peccaries are in fact different enough from peccaries proper to be regarded as a different family, Palaeochoeridae Matthew, 1924. This is problematical however as the type genus for this group, Palaeochoerus from Oligocene-Miocene Africa and Europe, almost certainly is a suid proper and not a peccary, Old World or otherwise. Furthermore, Pickford (1998) argued that Van der Made’s concept of Palaeochoeridae is probably polyphyletic, including not only Old World peccaries and suids but also groups like the sanitheriids of the Old World Miocene. On the other hand other workers have described some of the Old World peccaries as ‘unambiguous Tayassuidae’ (Ducrocq 1994, p. 765).

While the debate isn’t over, Old World peccaries do at least share characters with unambiguous American peccaries that aren’t seen in suids: a vertical lower canine located close to the premolar row, a prominent trigonid on the fourth upper premolar, thin enamel, and other features (Pickford & Morales 1989). Old World and New World peccaries also lack many characters present in all suids, but these carry less weight as, theoretically, they could be retained plesiomorphies if Old World peccaries are suids and not peccaries proper. They include the lack of dentary symphyseal splaying around the canines and the lack of an obliquely oriented lower tooth row. But for the time being I’d say that Old World peccaries really do look like peccaries after all, not that I’m an expert.

Like pigs, peccaries use a specialised rhinarial disk for rooting in soil and their snout is specialised for this behaviour. The disk itself is an unusual novelty, the snout is proportionally lengthened compared to that of other artiodactyls and the nuchal muscles (which help support the head) are hypertrophied and with enlarged attachment areas. However, the muscles that operate the disk originate from different points in peccaries and suids.

Strangely, all of the skull sutures in adult peccaries are completely closed, and in fact even juveniles exhibit closure of certain of the sutures ordinarily open in young mammals. This obviously rules out the possibility of any sort of cranial kinesis. A research team led by Katherine Rafferty and Susan Herring have been looking at strain patterns in pig skulls, and they’ve found that during occlusion of the teeth, the snout bones are deformed and pull apart slightly at their sutures (Rafferty et al. 2003). I’m guessing that this is somehow relevant to the major fusion of the sutures seen in peccaries (maybe peccaries have evolved a novel solution to coping with strains built up during tooth occlusion), but I don’t know if this area has been studied. It’s surely relevant to stuff we’ll get to in a minute.

Peccaries are well known for having big, scary canines and, unlike suids, both the upper and lower canines of peccaries are used in biting. Also in contrast to those of suids, both canines in peccaries are vertically implanted: in suids the upper canine exits the maxilla anterolaterally, and then curves dorsally. In all placental mammals the lower canine bites ‘ahead’ of the upper canine (look at sloths and you’ll see that they differ – almost certainly because one of their canines isn’t actually a true canine), and in peccaries the almost total lack of enamel on the posterior surface of the lower canine means that it is constantly sharpened as it moves against the enamelled anterior face of the upper canine. Because the upper and lower canines tightly interlock, peccaries are virtually incapable of moving the lower jaw from side-to-side when the jaws are closed. A special mucosal pocket, bordered by a raised boss on the maxilla termed the canine buttress, houses the lower canine when the jaws are closed (Herring 1972).

The tight interlocking of the canines prevents jaw movement during full occlusion, but there’s more: bony stops around the jaw joint further prevent anteroposterior movement of the jaws when they’re closed. Consequently the jaws can only open and close in a simple hinge-like arrangement. Herring (1972, p. 502) suggested that ‘this action probably helps to guide the lower canine into its correct occlusal relationship, thus preventing injury to soft tissues’.

But exactly why do the canines interlock? Inspired by Herring’s study, Kiltie (1981) studied peccary teeth in detail and noted a correlation between tooth morphology and a diet of hard nuts and seeds. Kiltie didn’t propose that the canines were used to break open the food items, but that, like the bony stops around the jaw joint itself, their interlocking helped prevent dislocation of the joint when tremendous force was applied across the molar teeth. Several other features of the peccary dentition are in agreement with the idea that peccaries are specialised for breaking open rock-hard objects, as are behavioural studies. Captive animals are reported to often try to break open excessively hard nuts and seeds. I wonder if anyone has ever measured the bite strength of a peccary. Whatever, all the more reason not to put your hand in a peccary’s mouth, and to have those new fences up at Marwell Zoo (go here).

If Kiltie is right about the bracing function of the canines, then this would explain why – unlike many other mammals that use their canines as offensive weapons – peccaries are not sexually dimorphic in canine size. They aren’t dimorphic in body size, nor in head shape, either. But this isn’t the whole story, as fossil peccaries ordinarily are dimorphic, with many forms exhibiting significant (i.e., distinctly bimodal) differences in canine size (Wright 1998). In fact what’s almost bizarre is that fossil populations of the living species exhibit sexual dimorphism in canine size, meaning that the living populations lost dimorphism somewhere along the way.

So what gives? We don’t know why sexual dimorphism was lost in the group (decreasing need to avoid niche overlap, due to declining diversity in contemporaneous megafauna?), but comparison with related groups, and examination of the peccary fossil record, indicates that sexual dimorphism in canine size is primitive for the group (it’s the condition they inherited from their ancestors). Use of the canines as bracing structures therefore looks like an exaptation: a new use for a set of structures that were previously used for something else.

I did want to talk about the history of peccaries in the Americas, but I’ve run out of time. Coming next: probably those sexy tupuxuarids (go here for teaser). For the latest news on Tetrapod Zoology do go here.

Many thanks to Steve Bodio (of Querencia) for the Collared peccary Tayassu tajacu skull photos that appear here. The Collared peccary photo at top is from birdfotos.com and the White-lipped peccary T. pecari photo is from the Cites sites.

Refs - -

Ducrocq, S. 1994. An Eocene peccary from Thailand and the biogeographical origins of the artiodactyl family Tayassuidae. Palaeontology 37, 765-779.

Herring, S. W. 1972. The role of canine morphology in the evolutionary divergence of pigs and peccaries. Journal of Mammalogy 53, 500-512.

Kiltie, R. A. 1981. The function of interlocking canines in rain forest peccaries (Tayassuidae). Journal of Mammalogy 62, 459-469.

Pickford, M. 1978. The taxonomic status and distribution of Schizochoerus (Mammalia, Tayassuidae). Tertiary Research 2, 29-38.

- . 1998. A new genus of Tayassuidae (Mammalia) from the Middle Miocene of Uganda and Kenya. Annales de Paléontologie 84, 275-285.

- . & Morales, J. 1989. On the tayassuid affinities of Xenohyus Ginsburg, 1980, and the description of new fossils from Spain. Estudios Geologicos 45, 233-237.

- . & Morales, J. 1998. A tubulidentate suiform lineage (Tayassuidae, Mammalia) from the Early Miocene of Spain. Comptes Rendu de l’Academie des Sciences, Paris, Serie II 327, 285-290.

Rafferty, K. L., Herring, S. W. & Marshall, C. D. 2003. Biomechanics of the rostrum and the role of facial sutures. Journal of Morphology 257, 33-44.

Van der Made, J. 1997. Systematics and stratigraphy of the genera Taucanamo and Schizochoerus and a classification of the Palaeochoeridae. Proceedings of the Koninkliijke Nedderlandse Akademie voor Wetenschappen 100, 127-139.

Wright, D. B. 1998. Tayassuidae. In Janis, C. M., Scott, K. M. & Jacobs, L. L. (eds) Evolution of Tertiary Mammals of North America. Volume 1: Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals. Cambridge University Press, pp. 389-401.