About the SA Blog Network

Tetrapod Zoology

Tetrapod Zoology

Amphibians, reptiles, birds and mammals - living and extinct
Tetrapod Zoology Home

The ‘Tree-Kangaroos Come First’ hypothesis

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

Top left: Common brushtail possum (Trichosurus vulpecula) (by wollombi, from wikipedia), a phalanger. Top right: Red kangaroo (Macropus rufus). Bottom: Goodfellow's tree-kangaroo (by Richard Ashurst). The bear-like tree-kangaroo might look more similar to the phalanger than to the Red kangaroo. What does it all mean?

One of my favourite groups of marsupials are the wonderful tree-kangaroos. There are presently ten recognised tree-kangaroo species; they occur exclusively on New Guinea, Umboi, New Britain and north-east Queensland (and it’s generally thought that they were introduced to Umboi by humans).

Tree-kangaroos first became known to Europeans in 1826 when crew of the Dutch sailing ship The Triton collected four ‘wangoerie’ specimens (kept as pets by local people) from the north coast of New Guinea. Salomon Müller published the name Dendrolagus for these animals in 1842. It means ‘tree hare’ – the ‘hare’ part of the name apparently being a reference to the game-like quality of tree-kangaroo meat.

Some tree-kangaroo species have only been discovered very recently. Two new taxa (Seri’s tree-kangaroo D. stellarum and the Tenkile D. scottae) were named in 1990. And in the same year, wildlife photographer Gerald Cubitt encountered a Dani tribesman in the Sudirman Range, Papua, who had recently captured individuals of a distinctive, mostly black tree-kangaroo. This animal was known to locals as the Dingiso and it seemed to be far more terrestrial than other tree-kangaroos. It was formally named as a new species (D. mbaiso) in 1993. The Golden-mantled tree-kangaroo D. goodfellowi pulcherrimus was also named in 1993. If you’re interested in knowing more about the history and discovery of tree-kangaroos, be sure to track down Tim Flannery et al.’s Tree Kangaroos: A Curious Natural History (Flannery et al. 1996) and Roger Martin’s Tree-Kangaroos of Australia and New Guinea (Martin 2005).

Bennett's tree-kangaroo (D. bennettianus), by J. Smit, from wikipedia.

Those boring stereotypes and crass mischaracterisations

As is the case with so many groups of animals, the same set of facts tend to be trotted out about tree-kangaroos whenever you read about them. One is that they are clumsy and overall poor at climbing. This really isn’t a fair criticism and I wonder if it’s laced with the implication that marsupial tree-climbers are inferior when compared with the wonderful squirrels and monkeys with their Magic Placenta Powers, combined with the notion that tree-climbing kangaroos are just absurd and clearly some sort of evolutionary joke. In fact, studies of climbing behaviour in tree-kangaroos show that they’re proficient at rapid vertical scaling of tree trunks (so long as there are enough boles, patches of rough bark and climbing plants to provide purchase); they’re also good at balancing on and traversing narrow horizontal supports, and they’re agile leapers in the tree-top environment.

I remember getting the impression – actually from the literature on bird origins – that tree-kangaroos are scarcely modified for their lifestyle; in other words that they lack obvious arboreal specialisations. This isn’t in the least bit accurate (Dececchi & Larsson 2011).

Bennett's tree-kangaroo forepaw, from Martin (2005).

Notable tree-kangaroo specialisations for a life in the trees include sharply curved, proportionally large hand claws, independently moveable fingers, a particularly mobile shoulder joint, massive, thick, broad, heavily tuberculated hand-, wrist- and foot-pads, and especially flexible ankles. Their forelimb and shoulder anatomy is specialised for grasping, gripping and mobility (Warburton et al. 2011).

In the upper arm, a large rotator cuff is present (Flannery et al. 1996): this is the name given to a linked set of muscles that allow a particularly wide range of forelimb movement. The long, slender, sometimes tasselled, tails of tree-kangaroos are used in maintaining balance while climbing and are rather different from the proportionally shorter, more muscular, more ‘springy’ tails seen in many other macropods.

Tree-kangaroo muscle mass is about one-third less that of terrestrial macropods (Grand 1990) [for the record, “macropod” = all members of the kangaroo radiation, from rat-kangaroos and potoroos to the many wallaby and true kangaroo lineages]. This could be interpreted as a weight-saving adaptation, but it’s also likely due to their proportionally short hindlegs and reduction in the muscles required for perpetual terrestrial bounding. Tree-kangaroo fur is peculiar in growing in whorls, supposedly an adaptation for rain-shedding (the hair on the neck grows towards the head, rather than toward the shoulders). In some tree-kangaroo species, the whorls are on the shoulders, in others they’re in the middle of the back, and in others they’re located at the base of the tail.

Foot sole of Bennett's tree-kangaroo, from Martin (2005).

Tree-kangaroo limb joints are obviously extremely able to absorb massive amounts of shock and compression (I have the idea that this is because they possess proportionally enormous cartilage pads and capsules, but I can’t find out where I ‘learnt’ this). I say this because tree-kangaroos engage in a pretty radical bit of defensive behaviour. When threatened while in the tree-tops they leap to the ground – even if it’s more than 15 m away. Martin (2005) noted that, while this might work fine when confronted by pythons, big owls, eagles, or even thylacoleonids (tree-kangaroos would almost certainly have faced predation from this group in the recent past), it isn’t so effective when the hunters are humans and their dogs. Obviously, humans can flush a tree-kangaroo from an arboreal refuge, making it an easy target after pursuit on the ground.

The 'Queensland tiger' as it was supposed to look. Ignore the other animals.

Once on the ground, tree-kangaroos hop bipedally in conventional macropod fashion, though usually with the long tail held up and over the body. They frequently get hit by cars, and I have to note the interesting suggestion that some sightings of ‘Queensland tigers’ might be fleeting glimpses of terrestrial tree-kangaroos, seen disappearing into bush (Martin 2005, p. 30). Also worthy of mention is the fact that tree-kangaroos are unique among macropods in being able to walk by moving their hindlimbs independently of one another (Windsor & Dagg 1971). Other macropods can’t do this: they can only move their hindlimbs synchronously and thus cannot ‘walk’ as we normally understand the term.

Anyway, Martin (2005, p. 98) refers to one particularly impressive vertical leap made by a Bennett’s tree-kangaroo during the night. He was able to return the next day and measure the height of the branch the kangaroo had leapt from. It was 22 m off the ground, yet the animal had bounded off, apparently without injury. Procter-Gray & Ganslosser (1986) studied the different forms of movement practised by Lumholtz’s tree-kangaroo D. lumholtzi and used the term ‘crash’ for these impressive vertical leaps. They also stated that, after performing a ‘crash’, an individual “landed on its feet or all fours, and was immediately able to begin bipedal leaping on the ground. Never did a tree-kangaroo appear injured or stunned after a crash” (p. 347). On landing on the ground, the animals produce (unsurprisingly) a loud thud. Jared Diamond once said “The loud thuds of falling tree-kangaroos as they hit the ground were among the distinctive sounds that I heard daily in the Foja Mountains” (Diamond 1997, p. 692).

Goodfellow's tree-kangaroo (D. goodfellowi), by Timmy Toucan, from wikipedia.

We should note that tree-kangaroos don’t always make incredible leaps like this when threatened, since individuals of some species are also reported to freeze when confronted by predators and thus hope to avoid attention by crypsis.

Whence Dendrolagus?

Long-time readers will know that I have a real fondness for what are, if I may use the most politically correct term, ‘non-standard’ hypotheses. Mammals evolving from aquatic humanoids or amphisbaenians, primates going through an ancestral gliding stagesmall, birdy tree-climbing archosaurs giving rise to all dinosaurs, whales evolving from ichthyosaurs… that sort of thing.

Kangaroos belong to a large group of Australasian marsupials called diprotodontians, and ancestral diprotodontians were – based on the anatomy and ecology of early fossil diprotodontians and of the distribution of climbing habits in living species – almost definitely able tree-climbers. Evidently, early members of the kangaroo clade became specialised for life on the ground. In the standard scenario, one kangaroo lineage then became specialised anew for scansorial and eventually arboreal life, giving rise to the tree-kangaroos. And then at least one tree-kangaroo lineage – the one that led to the Dingiso – seemingly became terrestrial again… or, at least, more terrestrial than other tree-kangaroos, anyway.

Authors have typically been rather vague with respect to where tree-kangaroos might fit within the macropod radiation. Taxonomic lists, dendrograms and classification sequences have generally placed tree-kangaroos (sometimes as their own ‘tribe’, Dendrolagini, within the advanced macropod ‘subfamily’ Macropodinae) outside the group that includes true kangaroos and wallabies, and sometimes close to the New Guinea forest wallabies (e.g., Raven & Gregory 1946, Nowak 1999, Kear & Cooke 2001). The implication from such a placement is that tree-kangaroos evolved from an early, primitive member of Macropodinae. Because members of this clade are known from the Middle Miocene, tree-kangaroos must have diverged by this time or earlier.

A dendrogram depicting possible relationships among macropods, from Raven & Gregory (1946). Note that tree-kangaroos are shown as being close to New Guinea forest wallabies (Dorcopsis and Dorcopsulus) and are 'basal' compared to most true kangaroos and wallabies.

‘Tree-Kangaroos Come First’?

But tree-kangaroos are really unusual compared to other macropods. They seem weirdly, well, ‘primitive’.

For one thing, their short-leggedness is obvious. And their legs aren’t just short: the proportions of their legs (talking hindlegs here) are odd for a macropod in that their shin and thigh bones are very similar in length (Kear et al. 2008). Similar proportions are seen elsewhere in climbing phalangers like the Common brushtail possum Trichosurus vulpecula. For comparison, the tibia is nearly twice as long as the femur in big, specialised saltatorial macropods like Macropus and Sthenurus.

Tree-kangaroos are also unusual compared to other macropods in lacking extensive contact between the distal ends of the tibia and fibula (Kear & Cooke 2001) and in the reduced amount of musculature they possess. Mentioned above as a possible weight-saving specialisation for their arboreal lifestyle, it means that tree-kangaroos are more similar in the general form of their musculature to climbing phalangers like Pseudocheirus than to normal kangaroos (Grand 1990).

Buergers tree-kangaroo (D. goodfellowi buergeri), photographed at San Diego Zoo by Justin Griffiths, from wikipedia.

A general characteristic of marsupials is the absence of a patella (with bandicoots being an annoying exception… the fact that bandicoots also have a placenta [yes, you read that right] means that some authors have actually seriously regarded bandicoots as placentals, not as marsupials). However, marsupials do have a fibrocartilaginous structure in the same place, generally termed a patelloid. Macropods mostly have what’s been termed a ‘type IV’ patelloid, but tree-kangaroos differ in having a proportionally much smaller, ‘type III’ patelloid that’s elsewhere seen in the potoroos (Potoroidae) (Reese et al. 2001). Potoroos are small macropods, thought to have diverged during the Late Oligocene – between 30 and 25 million years ago – from the rest of the macropod radition (Kear et al. 2008, Prideaux et al. 2010). Whether they should or should not be included within Macropodidae remains the topic of debate.

White-throated or Vogelkop tree-kangaroo (D. ursinus), from Gould, 1863.

These primitive character traits seen in tree-kangaroos – those phalanger-like limb proportions, the very un-kangaroo-like amount of musculature they possess, that small patelloid, and that small amount of tibia-fibula contact – have conventionally been interpreted as reversals. But could they show that tree-kangaroos are, in actuality, truly primitive with respect to macropodid macropods, and hence outside the clade that includes most or all other lineages within Macropodidae?

Exactly this was suggested by Paul Hopkinson of San Diego State University in 1991 (Hopkinson 1991). In a conference abstract titled ‘Systematic position of Dendrolagus (Marsupialia, Macropodidae) and its implications for the ancestral lifestyle of kangaroos’, he specifically stated “It is suggested that the present systematic position of Dendrolagus within the Macropodinae is incorrect and that Dendrolagus is the sister taxon of all other Macropodidae”.

Lumholtz's tree-kangaroo, by Joseph Smit, 1884.

I’ve been unable to find out whether Hopkinson’s abstract ever formed the basis for a technical paper (do say if you know otherwise), but it does seem that he submitted a 1994 thesis on the same topic (Hopskinson 1994). I haven’t been able to track this down. Anyway, reading between the lines, it seems very clear that Hopkinson was positing (1) a basal position within Macropodidae for Dendrolagus and the exclusion of Dendrolagus from the macropodid clade Macropodinae (sorry, you have to keep up with the names for this to make sense), (2) the presence of various primitive, phalanger-like morphological details in Dendrolagus, and (3) an arboreal ancestry for Macropodidae as a whole, with Dendrolagus serving as a model ancestor for the prototype macropod. If the title for this article has been lost on you, what I’m getting at is the idea that tree-kangaroos might be the ‘ancestral form’ for the terrestrial macropod radiation, in the same way that bird-like, climbing archosaurs are ancestral for dinosaurs in the non-standard (and non-parsimonious) ‘Birds Come First’ model.

Could Hopkinson have been right? Well, as usual with surprising ‘non-standard’ hypotheses… no, it doesn’t seem that the bulk of evidence supports the ‘Tree-Kangaroos Come First’ hypothesis. My apologies if I created the impression that a ‘basal position’ for tree-kangaroos is likely.

Those ‘primitive’ characters present in tree-kangaroos really must be interpreted as reversals since they’re substantially out-weighed by a more impressive list of other characters. These place tree-kangaroos within the macropodid clade Macropodinae (e.g., Kear et al. 2008, Prideaux et al. 2010), not “outside the group that includes true kangaroos and wallabies” as implied by traditional classifications (e.g,. Nowak 1998), and definitely not outside the rest of Macropodidae as per Hopkinson. Molecular data supports the nesting of tree-kangaroos within Macropodinae as well (e.g., Baverstock et al. 1989, Kirsch et al. 1995, Burk et al. 1998).

Macropod phylogeny, plotted against a timescale, from Prideaux & Warburton (2010). The dendrolagine branch is shown extending to the Middle Miocene. Click to enlarge.

What I find particularly interesting is that both morphological and molecular phylogenies have frequently (though not ubiquitously) found tree-kangaroos to be especially close to rock-wallabies (Petrogale). Rock-wallabies are small, nocturnal, group-living macropods that occur throughout much of Australia but aren’t found on New Guinea. Most recently, Prideaux et al. (2010) used the name Dendrolagini for the Dendrogale + Petrogale clade, and recovered it as the sister-taxon to Macropodini, the clade within Macropodinae that includes ‘true’ kangaroos and wallabies (again, I hope you’re following the fact that macropod, macropodid and macropodine all have separate meanings and are not interchangeable).

Yellow-footed rock-wallaby (Petrogale xanthopus), by Peripitus, from wikipedia.

Charles De Vis actually suggested as long ago as 1887 that tree-kangaroos might be close kin of rock-wallabies since he perceived a strong superficial similarity between the two, and others commented on the same possibility in later years. Rock-wallabies make very logical close relatives of tree-kangaroos given that they’re agile climbers, well able to hop about on rocks, cliffs and on sloping trees and low branches*. Rock-wallabies also recall tree-kangaroos in often being fairly brightly coloured, boldly patterned, and in possessing long, cylindrical, sometimes slightly bushy tails.

* Martin (2005) described how people who live alongside coastal populations of the Proserpine rock-wallaby P. persephone in northern Queensland actually believe these animals to be tree-kangaroos since they often see them climbing in trees.

Note also that the ‘Tree-Kangaroos Come First’ hypothesis would require that tree-kangaroos be pretty old, since both fossils and molecular data show that the macropod radiation was well underway by the Late Oligocene. However, all fossil tree-kangaroos are from the Pleistocene or perhaps the Late Pliocene. Furthermore, some of these fossil forms share skeletal characters with Petrogale (Prideux & Warburton 2008). And molecular clock estimates have led some authors to suggest that tree-kangaroos originated sometime within the last 8 million years (Campeau-Péloquin et al. 2001), in which case they aren’t old at all but actually a fairly recent invention.

One of the first tree-kangaroo images I can recall seeing: from Augusto Vigna Taglianti's 1979 book The World of Mammals (Sampson Low). Much of the art in the book is brilliant.

Is the idea that tree-kangaroos are ‘primitive’, morphologically and ecologically archaic macropods any more appealing than the idea that they’re actually highly specialised arboreal members of an otherwise terrestrial radiation? Frankly, who cares: we should be all about following the evidence, and the data putting tree-kangaroos deep within Macropodidae, and specifically within Macropodinae and close to rock-wallabies, is actually very good.

In fact, I personally think this idea is more interesting than Hopkinson’s alternative proposal. It means that a highly specialised, morphologically aberrant lineage is placed smack-bang in the middle of all those mostly terrestrial macropodine kangaroos. Would we ever think that macropodines, of all marsupials, would take to the trees, to learn how to drop safely from heights of over 20 m?

For previous Tet Zoo articles on marsupials and other metatherians (note that they’re horribly under-represented at Tet Zoo), see…

Refs – -

Baverstock, P. R., Richardson, B. I., Birrell, I. & Krieg, M. 1989. Albumin immunologic relationships of the Macropodidae (Marsupialia). Systematic Zoology 37, 38-50.

Burk, A., Westerman, M. & Springer, M. 1998. The phylogenetic position of the Musky rat-kangaroo and the evolution of bipedal hopping in kangaroos (Macropodidae: Diprotodontia). Systematic Biology 47, 457-474.

Campeau-Péloquin, A., Kirsch, J. A. W., Eldridge, M. D. B. & Lapointe, F.-J. 2001. Phylogeny of the rock-wallabies, Petrogale (Marsupialia: Macropodidae) based on DNA/DNA hybridisation. Australian Journal of Zoology 49, 463-486.

Dececchi, T. A. & Larsson, H. C. E. 2011. Assessing arboreal adaptations of bird antecedents: testing the ecological setting of the origin of the avian flight stroke. PLoS ONE 6(8): e22292. doi:10.1371/journal.pone.0022292

Diamond, J. 1997. Flying yellow kangaroos. Nature 385, 692.

Flannery, T.F., Martin, R. & Szalay, A. 1996. Tree Kangaroos: A Curious Natural History. Reed, Melbourne.

Grand, T. I. 1990. Body composition and the evolution of the Macropodidae (Potorous, Dendrolagus, and Macropus). Anatomy and Embryology 182, 185-192.

Hopkinson, P. 1991. Systematic position of Dendrolagus (Marsupialia, Macropodidae) and its implications for the ancestral lifestyle of kangaroos. Journal of Vertebrate Paleontology 11 (supp. 3), 36.

Hopkinson, P. J. 1994. The Systematic Position of Tree-Kangaroos (Dendrolagus: Macropodidae: Marsupialia) and its Implications for the Ancestral Lifestyle of Kangaroos. PhD Thesis, San Diego State University.

Kear, B. P. & Cooke, B. N. 2001. A review of macropodoid systematics with the inclusion of a new family. Memoirs of the Association of Australasian Palaeontologists 25, 83-101.

- ., Lee, M. S. Y., Gerdtz, W. R. & Flannery, T. F. 2008. Evolution of hind limb proportions in kangaroos (Marsupialia: Macropodoidea). In Sargis, E. J. & Dagosto, M. (eds). Mammalian Evolutionary Morphology: A Tribute to Frederick S. Szalay. Springer Science, pp. 25-35.

Kirsch, J. A. W., Lapointe, F.-J. & Foeste, A. 1995. Resolution of portions of the kangaroo phylogeny (Marsupialia: Macropodidae) using DNA hybridization. Biological Journal of the Linnean Society 55, 309-328.

Martin, R. 2005. Tree-Kangaroos of Australia and New Guinea. CSIRO Publishing (Collingwood, Victoria).

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

Prideaux, G., & Warburton, N. (2008). A new Pleistocene tree-kangaroo (Diprotodontia: Macropodidae) from the Nullarbor Plain of south-central Australia Journal of Vertebrate Paleontology, 28 (2), 463-478 DOI: 10.1671/0272-4634(2008)28[463:ANPTDM]2.0.CO;2

Prideaux, G. J. & Warburton, N. M. 2010. An osteology-based appraisal of the phylogeny and evolution of kangaroos and wallabies (Macropodidae: Marsupialia). Zoological Journal of the Linnean Society 159, 954–987.

Procter-Gray, E. & Ganslosser, U. 1986. The individual behaviors of Lumholtz’s tree-kangaroo: repertoire and taxonomic implications. Journal of Mammalogy 67, 343-352.

Raven, H. C. & Gregory, W. K. 1946. Adaptive branching of the kangaroo family in relation to habitat. American Museum Novitates 1309, 1-33.

Reese, S., Pfuderer, U. R., Bragulla, H., Loeffler, K. & Budra, K.-D. 2001. Topography, structure and function of the patella and the patelloid in marsupials. Anatomia, Histologia, Embryologia 30, 289-294.

Warburton, N. M., Harvey, K. J., Prideaux, G. J. & O’Shea, J. E. 2011. Functional morphology of the forelimb of living and extinct tree-kangaroos (Marsupialia: Macropodidae). Journal of Morphology 272, 1230-1244.

Windsor, D. E. & Dagg, A. I. 1971. The gaits of the Macropodinae (Marsupialia). Journal of Zoology 163, 165-175.

Darren Naish About the Author: Darren Naish is a science writer, technical editor and palaeozoologist (affiliated with the University of Southampton, UK). He mostly works on Cretaceous dinosaurs and pterosaurs but has an avid interest in all things tetrapod. His publications can be downloaded at He has been blogging at Tetrapod Zoology since 2006. Check out the Tet Zoo podcast at! Follow on Twitter @TetZoo.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 61 Comments

Add Comment
  1. 1. Finback 8:29 am 10/17/2011

    That last picture looks almost terrifying. It knows where you’re hiding.

    Also, on the 1946 diagram, I notice _Palorchestes_ – was that considered to be within the macropod clade at the time? It looks like it’s branching off from _Sthenurus_.

    Link to this
  2. 2. Hastley 8:49 am 10/17/2011

    Actually, the incredible distances tree kangaroos drop is perfectly consistent with a paper my adviser just published. Long story short, rapid stretching damages muscle, but a tendon can serve as a “power attenuator”‘, rapidly absorbing mechanical energy from the outside world, then allowing the muscle to lengthen at a more moderate rate. The experiment was in turkeys dropped from various heights, but it should work fine in anything else, and tree kangaroos probably have nice, big Macropod tendons.

    Link to this
  3. 3. THoltz 9:45 am 10/17/2011

    If I remember properly, Hopkinson was torn into by some senior marsupial worker after his presentation, and some of that may be related to why we haven’t heard about it much since. (Memory is somewhat blurry here, as I had just gone before him in the Romer Prize session, so I was still recovering from presentation jitters.)

    Link to this
  4. 4. llewelly 10:01 am 10/17/2011

    “One of the first tree-kangaroo images I can recall seeing: from Augusto Vigna Taglianti’s 1979 book The World of Mammals (Sampson Low)”

    Look at it again. I don’t believe that, and neither does anybody else. Your picture is obviously a cut-out from a poster advertising the late-night horror flick _Attack of the Tree-Kangaroos!_

    On a different topic, Poll: The Real Drop Bears are:
    (a) Tree Kangaroos
    (b) Falling eucalyptus branches.

    Link to this
  5. 5. jtheodor 10:34 am 10/17/2011

    I saw Hopkinson’s talk at SVP that year. If I recall it correctly, his talk was ok, but in the questions, Fred Szalay got up and ranted about the idiocy of cladistics if that was the result, because everyone knew that tree kangaroos were derived. And being Fred, it was quite the rant. Then Hopkinson said “I’m sorry, was there a question in there?”. Never saw a publication or heard of him again (if you hadn’t cited the name I would not have remembered it, but did remember Fred’s rant!)

    Link to this
  6. 6. Dartian 11:17 am 10/17/2011

    tree-kangaroos are unique among macropods in being able to walk by moving their hindlimbs independently of one another (Windsor & Dagg 1971). Other macropods can’t do this: they can only move their hindlimbs synchronously

    So how do kangaroos swim, then?

    Link to this
  7. 7. Hastley 1:22 pm 10/17/2011

    Dartian – Wallabies swim via alterating-leg dog-paddle, as seen here:

    Link to this
  8. 8. BrianL 1:41 pm 10/17/2011

    I have to say I like your quip about ‘Magic Placenta Powers’. It does often seem that marsupials are still considered as somehow inferior to placentals. I’d say that any sort of real or perceived marsupial ‘weakness’ is more the result of Australia and South America being fairly small, isolated, ‘safer’ continents during most of the Cenozoic rather than any inherent marsupial inferiority vis a vis placentals.

    Anyhow, a very minor correction here: You describe one of the illustrations as “White-throated or Vogelkopt tree-kangaroo”. Surely this must be a typo as it should be ‘Vogelkop’ (Bird Head) without an extra T. Of course you’re familiar with hamerkops, so undoubtedly that extra T was a slip of the keyboard.

    I’m glad I read this post of yours, because I too was always under the impression that tree kangaroos were actually poorly suited to living in trees!

    By the way, do we know if thylacoleonids actually hunted in trees?

    Link to this
  9. 9. kmkohler 2:51 pm 10/17/2011

    Does anyone know of any videos of crashing tree-kangaroos?

    Link to this
  10. 10. victorg 3:07 pm 10/17/2011

    Dartian – Wallabies swim via alterating-leg dog-paddle, as seen here:

    Huh? That seems to contradict the part Dartian quoted. Unless it was supposed to mean “they can only [walk by] mov[ing] their hindlimbs synchronously”?

    Link to this
  11. 11. Christopher Taylor 9:18 pm 10/17/2011

    Huh? That seems to contradict the part Dartian quoted. Unless it was supposed to mean “they can only [walk by] mov[ing] their hindlimbs synchronously”?

    The video demonstrates that they can move their hind legs diachronously, contrary to what Darren implied. I suspect that they might still be prevented from doing so on land simply by how long the feet are compared to the legs. A kangaroo couldn’t walk by the standard method of passing the foot under the heel, because the hips can’t move enough (not a limitation in water, because the feet can move downwards). They would have to lift the feet flat by lifting the entire leg at the hip.

    Link to this
  12. 12. Christopher Taylor 9:20 pm 10/17/2011

    And yeah, Palorchestes used to be considered a gigantic kangaroo before it was known from more than scrappy material. I think the name even means ‘ancient leaper’ or something similar.

    Link to this
  13. 13. Mike from Ottawa 11:14 pm 10/17/2011

    “One of my favourite groups of marsupials are the wonderful tree-kangaroos.”

    Are there any tetrapods that are not one of your favourite groups, Darren?

    Joking aside, is there any group of tetrapods that you find boring?

    Link to this
  14. 14. Mythusmage 12:41 am 10/18/2011

    So the bandicoot can be considered a transitional animal, a marsupial with placental features.

    BTW, don’t other marsupials develop a rudimentary placenta, for the short period of pregnancy before the fetus is developed enough to make it to the pouch or skin fold, depending on the species?

    Link to this
  15. 15. Andrew O 12:59 am 10/18/2011

    “whales evolving from ichthyosaurs…”

    O_o who was proposing that?

    And more importantly, did they do any reconstructions of hypothetical transitional forms? I’d love to see how that could have worked- three-lobed tails, maybe, like the stabilisers on a plane?

    Link to this
  16. 16. Dartian 2:10 am 10/18/2011

    That seems to contradict the part Dartian quoted.

    Yes, that was what I was getting at in my earlier comment – kangaroos can move their hindlimbs in ‘ordinary mammal’ fashion, at least under some circumstances.

    (Admittedly, however, although I’ve always assumed that kangaroos dog-paddle, I realised that I’ve never seen good footage of them doing that. There are plenty of videos on the internet showing swimming kangaroos, but the one linked to by Hastley was the first one I’ve seen where the macropodid is swimming in water so clear that you can actually see how the hindlimbs move. So thanks for that, Hastley!)

    don’t other marsupials develop a rudimentary placenta

    Yes, marsupials do, in fact, have placentas too, although (with the exception of the bandicoots’) they’re different from those of the eutherians. See:

    Freyer, C., Zeller, U. & Renfree, M.B. 2003. The marsupial placenta: a phylogenetic analysis. Journal of Experimental Zoology 299A, 59-77.

    Link to this
  17. 17. naishd 4:15 am 10/18/2011

    Many thanks to all for the very interesting comments, much appreciated. The recollections of Hopkinson’s talk and Szalay’s response are especially appreciated.

    And, yes, that tree-kangaroo from Taglianti’s 1979 book does look freaky which I why I wanted to include it. I did know even as a child, however, that tree-kangaroos looked more like teddy bears than long-tailed tree-demons.

    Finback (comment 1): Palorchestes was assumed to be a macropod (sometimes a sthenurine macropodid) in the years following Owen’s description of it in 1874, and this view was maintained right up to 1958 when Jack Woods re-identified it as a vombatiform. That Raven & Gregory 1946 paper even includes photos of a model of its reconstructed skull “16 inches long” where it’s made to look just like a giant kangaroo and there used to be two ‘Palorchestes‘ reconstructions in the Australian Museum: they were essentially 3-m-tall grey kangaroos. Flannery and Archer said that these reconstructions “probably did more to inspire graziers to give thanks for the end of the Pleistocene period than to give an idea of what Palorchestes azael really looked like”.

    Thanks, BrianL, for the correction on ‘Vogelkop’. A silly typo on my part.

    On macropod hindlimb actions, I had honestly thought (and this was confirmed by what I’ve seen and read) that macropods really were mechanically ‘constrained’ to synchronous movement of the hindlimbs. The video of the doggy-paddling animal – obviously displaying asynchronous movement of the hindlimbs – was therefore a surprise. As Chris says (comment 11), it may therefore be that macropods only move their feet synchronously because their foot length makes it ‘convenient’, not because they ‘have’ to. Ergo, any macropod that evolves shorter feet (as per Dendrolagus) can switch back to independent movement of the hindlimbs.


    Link to this
  18. 18. SRPlant 4:26 am 10/18/2011

    I’m another (non-scientist) who was under the impression that tree-kangaroos were poorly adapted to climbing – I even recall reading that this “fact” was backed up by these animals often having bones that showed damage previously sustained in falls. Is this bunkum?

    Link to this
  19. 19. naishd 4:43 am 10/18/2011

    A few more responses…

    Do we really know if thylacoleonids climbed in trees? (comment 8). Of course, we don’t know anything for sure about the behaviour of this group, and there’s certainly been some disagreement as to the lifestyle of Thylacoleo (authors have regarded it as everything from terrestrial cursor to arboreal specialist). Thylacoleo is an extreme member of its group due to its size, but even it has manual features indicative of a climbing ability. I agree with those who think that smaller thylacoleonids were scansorial and probably quite agile in the trees.

    Mike from Ottawa (comment 13) asks “Joking aside, is there any group of tetrapods that you find boring?”.

    Err, no.

    Mythusmage (comment 14) wrote “So the bandicoot can be considered a transitional animal, a marsupial with placental features.”

    Well, no; rather, members of Marsupialia have convergently evolved placentas similar to those of members of Placentalia. Bandicoots really are good, honest members of Marsupialia, not ‘transitional’. Note that complex placentas have evolved several times within Tetrapoda. Among squamates, skinks are well known for their mammalesque placentas. While I’m here, worth saying that the earliest members of the placental lineage had pouches. And that not all marsupials have pouches.


    Link to this
  20. 20. naishd 5:25 am 10/18/2011

    Oh, regarding the whole “whales from ichthyosaurs” thing (see comment 15), I’ll save it for another article.


    Link to this
  21. 21. Dartian 6:00 am 10/18/2011

    this “fact” was backed up by these animals often having bones that showed damage previously sustained in falls

    AFAIK, that claim might well be accurate, but it doesn’t necessarily mean that tree-kangaroos are therefore poor climbers. If you live in trees, you might fall down from there, and that can happen to the best of climbers. There are few if any mammals that are better adapted to arboreal living than gibbons – yet even these supreme climbers may fall and break their bones. Quite frequently too, in fact; in the early 20th century, the great Swiss primatologist Adolph Schultz analysed the skeletons of more than one hundred wild-shot gibbons, and discovered that about a third of them had bones that had fractured (and healed) during their lifetimes. (And it should be noted that Schultz’ sample only included gibbons that had survived falls; it didn’t include individuals that had fallen to their deaths.)

    Link to this
  22. 22. THoltz 9:10 am 10/18/2011

    Jess is spot on: it WAS Szalay, and that was indeed the nature of the exchange. Wow, 20 years ago…

    Link to this
  23. 23. Valerio Peverelli 9:40 am 10/18/2011

    “And it’s generally thought that they were introduced to Umboi by humans”: No, by flying Ropen!

    “Tree-kangaroo limb joints are obviously extremely able to absorb massive amounts of shock and compression (…). I say this because tree-kangaroos engage in a pretty radical bit of defensive behaviour. When threatened while in the tree-tops they leap to the ground – even if it’s more than 15 m away. Martin (2005) noted that, while this might work fine when confronted by pythons, big owls, eagles, or even thylacoleonids (tree-kangaroos would almost certainly have faced predation from this group in the recent past)”: And flyng Ropen!

    Don’t forget the Ropen!

    Ropen! Ropen! Ropen!

    Link to this
  24. 24. naishd 9:50 am 10/18/2011

    In my haste, I admit that I had indeed forgot the Ropen. Jonathan Whitcomb would be proud :)


    Link to this
  25. 25. SRPlant 10:25 am 10/18/2011

    Thanks, Dartian – one can only conclude that gibbons are poorly adapted to arboreal life!

    Link to this
  26. 26. vdinets 12:16 pm 10/18/2011

    Hastley: great video! In fact, I’ve read that macropods can’t use their hind feet independently in more than one book, popular and scientific alike.

    Link to this
  27. 27. Halbred 4:42 pm 10/18/2011

    Duckbills are pretty dull, even the lambeosaurines. Maybe that’s because they lived alongside ceratopsids, ankylosaurs, and pachycephalosaurs, all of which were significantly more awesome.

    Ceratopsids: horns and spikes and beaks!
    Ankylosaurs: Armor and clubbed tails and horns!
    Duckbills: We grew up really fast and bred like rabbits!


    Link to this
  28. 28. naishd 6:40 pm 10/18/2011

    Halbred – did you crash the wrong party?


    Link to this
  29. 29. victorg 8:10 pm 10/18/2011

    Darren: I thought so at first, but then I saw a link to Mike from Ottawa (comment 13) asks “Joking aside, is there any group of tetrapods that you find boring?”.

    Link to this
  30. 30. Christopher Taylor 8:14 pm 10/18/2011

    Joking aside, is there any group of tetrapods that you find boring?

    Maybe we can establish that by looking at what tetrapods Darren hasn’t ever written about. Aardvarks, it would seem, are really, really dull. Always going on about their plans to extend their burrow, and trying to show you photos of their calves.

    Link to this
  31. 31. Cameron McCormick 10:15 pm 10/18/2011

    trying to show you photos of their calves.

    So do hyraxes have calves, or is that term reserved for largish mammals?

    Link to this
  32. 32. Christopher Taylor 11:47 pm 10/18/2011

    I have no idea what a baby aardvark is actually called (a varklet?) ‘Calf’ just somehow sounded appropriate.

    Link to this
  33. 33. Cameron McCormick 12:10 am 10/19/2011

    It seems there are instances of juvenile Aardvarks and Hyraxes being called “calves”, but it seems “pup” is more common. Drat, so I guess calves aren’t a shared, derived characteristic of Afrotheria.

    Link to this
  34. 34. BilBy 3:19 am 10/19/2011

    Maybe not ‘varklet’: perhaps ‘babavark’ or ‘varketjie’; but this “…trying to show you photos of their calves” gave me a mental image of an aardvark (a deeply fascinating beast) with photos of the backs of its legs

    Link to this
  35. 35. naishd 4:00 am 10/19/2011

    I have been reliably informed that baby aardvarks are actually called cubs. It’s due to some sort of old idea about them being bear-like animals, and indeed males and females are meant to be called boars and sows, as is also the case with bears. Allegedly, a group of aardvarks is called an ‘aarmory’.


    Link to this
  36. 36. Jerzy New 4:16 am 10/19/2011

    Darren, you should make quantitative study of awesomeness vs. dullness in tetrapods.

    Link to this
  37. 37. naishd 4:24 am 10/19/2011

    Jerzy – that’s a good idea. Back when I was lecturing to students I used to draw up a table showing how giant pandas were fascinating and not – contra the opinion of some – failures that should be allowed to go extinct. Plus points for giant pandas included ‘high inherent cuteness factor’, ‘freaky colour scheme’ and ‘s-shaped penis’.


    Link to this
  38. 38. KarMannJRO 4:24 am 10/19/2011

    I thought of aardvark legs, too.

    As far as dull and boring goes, I would point out that cephalopods are certainly not exciting tetrapods. (Even PZ would agree with that!)

    Link to this
  39. 39. Jerzy New 4:27 am 10/19/2011

    BTW, if I correctly remember my physics lessons (which I don´t), then 10 kg tree kangaroo falling from 20 m generates equal force as 70 kg human falling from 40 cm. Or, more likely, aforementioned tree kangaroo feels like being jumped on by a 70 kg human from 40 cm.

    Either way, animals smaller than humans can survive very big falls simply because of their small weight.

    Link to this
  40. 40. Dartian 5:20 am 10/19/2011

    It’s due to some sort of old idea about them being bear-like animals

    Aardvark literally means ‘earth pig’, though (for comparison, warthog is vlakvark in Afrikaans). Thus, it would make (slightly) more sense to call baby aardvarks ‘piglets’.

    Link to this
  41. 41. naishd 5:22 am 10/19/2011

    Yeah, ‘earth pig’ for aardvark is well known, but the term ‘antbear’ used to be used for it too, right?


    Link to this
  42. 42. vdinets 12:21 pm 10/19/2011

    Afrikaans for piglet is varkie, so baby aardvark should be called aardvarkie.

    A study of perceived animal awesomeness done by questionnaires would be very interesting. All the age and gender differences…

    Link to this
  43. 43. Yodelling Cyclist 4:32 pm 10/19/2011

    Hello Jerzy New. Bluntly – you don’t ;-) , but the point is a good one.

    Force is equal to mass times acceleration. The key factors are how rapidly the falling body decelerates on contact with the ground, how much velocity there is to shed in total and the total mass. Note that objects falling under gravity accelerate at the same rate – “g” on earth (about 9.8 m/s^2). We could discuss air resistance, but I doubt it’s very significant to this problem (although it is to the classic mouse in a mineshaft problem). You have calculated a situation in which the momentum of the human and the kangaroo are equivalent on impact, but I suspect the kangaroo is going to decelerate much more slowly than the human – I defer (because I have no good understanding of these things)discussion of tendons and muscles and other gooey things to the biologists around here – but I suspect it’s going to decelerate at a lower rate, and so experience a lower force.

    Further – consider the kinetic energy involved, which goes with the second power of the velocity. The ‘roo is dissipating a hell of a lot more energy on impact than the 40 cm jump human- well about seven times more. That’s the same amount as a human falling from 2.8m, or 9 and a bit feet. Certainly that’s survivable.

    Jerzy New does make a good point though – these animals aren’t dissipating any more energy than the human body can take in an impact through the legs, and there’s good reason to suspect that the impact force is endurable for just about any 10kg mammal landing in its preferred orientation (i.e. not on the head/along the vertebrae etc.).

    (I do not condone throwing dogs out of windows to test this assertion!).

    Link to this
  44. 44. Yodelling Cyclist 5:43 pm 10/19/2011

    Thinking about it a bit more: if we forced humans to do many 9 foot drops, we would be getting quite an injury rate – certainly we’d be seeing sprained ankles as the falls added up. If tree kangaroos do these falls and just shrug them off, perhaps we should be thinking that these animals aren’t remarkable for surviving these crashes, but for doing them routinely.

    Link to this
  45. 45. Dartian 3:20 am 10/20/2011

    the term ‘antbear’ used to be used for it too, right?

    Yes, although increasingly rarely from circa the 1950ies onwards.

    Incidentally, the aardvark’s German name, Erdferkel, literally means ‘earth piglet’. (I don’t know, however, if that’s also the vernacular name that the German-speaking colonists in Africa in the late 19th-/early 20th century would have referred to this animal with; there is the German word Ameisenbär, which literally means ‘ant bear’, but that is – at least nowadays – reserved to the New World myrmecophagine anteaters.)

    Yodelling Cyclist:
    throwing dogs out of windows to test this assertion

    For what it’s worth, there is some (admittedly limited) published information about fall survival in wild chimpanzees (which are fairly large animals). In her 1986 book The Chimpanzees of Gombe: Patterns of Behavior, Jane Goodall presented data for a two-year period of observations which included 51 falls from trees. Not all of the heights were recorded, but in at least 13 instances, the chimps fell to the ground from a height of more than 10 metres – and, although they might have suffered injuries of varying severity, most survived; in fact, only two fall-related chimpanzee deaths occurred during this observation period (although, obviously, the number of falls that were unobserved by people is unknown).

    Getting back to the main subject, and citing Darren’s OP:

    What I find particularly interesting is that both morphological and molecular phylogenies have frequently (though not ubiquitously) found tree-kangaroos to be especially close to rock-wallabies

    I agree completely; it’s highly fascinating. It’s the kind of neat result that, once you start to think about it, ‘makes sense’. After all, if you’re adapted to hopping on rocks it probably isn’t really that great an evolutionary step – sorry, leap! – to adapt to hopping on sloping tree trunks (as witnessed by the fact that some extant rock-wallaby species occasionally ascend trees).

    Would we ever think that macropodines, of all marsupials, would take to the trees

    When I’ve taught students about evolution and ecomorphology, tree-kangaroos have been one of my favourite examples to demonstrate the inherent opportunism of natural selection; i.e., that even highly specialised lineages may, so to speak, ‘switch direction’ and adapt successfully to wholly new econiches and ways of life.

    Link to this
  46. 46. Jerzy New 8:22 am 10/20/2011

    Yodelling Cyclist thanks for correcting. Injury is indeed related to velocity, not just dynamic energy.

    Arboreal animals do have structural adaptations useful to survive falls, but I pointed that fall is generally much less a problem for smaller animal than humans assume.

    @Dartian, in Kibale NP wardens sometimes find healthy chimpanzees dead without any visible injury or sign around, as if stuck dead, and assume they fell from a tree. If I understand it well, Mrs Goodall noticed 2 deaths in 2 years in a community of 100 or so chimps. Not that few at all. Given that a chimp weighs about as much as a human, not surprising.

    Back to topic: aren’t hyraxes and peccaries boring?

    Link to this
  47. 47. Andreas Johansson 8:37 am 10/20/2011

    One might think that frequent fall injuries is indicative of good climbing abilities – an animal with poor such might be expected to keep on the ground and have little opportunity to fall farther than its own height (which is largely harmless unless you’re a fairly large critter).

    I know I’ve heard of the ichthyosaur -> whales “theory” before, but I can’t recall where. Grogdamn leaky memory!

    But it reminds me of another question that’s been sloshing around in my brain for a while – Darwin in (at least one edition of) The Origin of Species makes reference to Mesozoic whale bones that had supposedly then recently been found, as an example of even big aquatic animals with hard parts can be hard to find in the fossil record and absence of evidence therefore isn’t evidence of absence. Now, whatever Darwin was refering to clearly wasn’t actual Mesozoic whale bones, but anyone know the story behind it? Where they misidentified bones of marine saurians or the like? Or just misdated Cenozoic ones?

    Link to this
  48. 48. Spugpow 10:18 am 10/20/2011

    Here’s a video that includes footage of crashes:

    Here’s just the better crash:

    Link to this
  49. 49. Dartian 10:35 am 10/20/2011

    in Kibale NP wardens sometimes find healthy chimpanzees dead without any visible injury or sign around, as if stuck dead, and assume they fell from a tree

    What? If there are no visible injuries on the bodies, why on earth would they assume that the cause of death of those chimps was falling?!

    And would they have left scientifically valuable specimens just lying there in the bush, rather than collecting the carcasses for a detailed investigation (where the exact cause of death could almost certainly be established)? It’s standard practise at Gombe (and indeed at most if not all field sites where great apes have been habituated) to collect, whenever possible, any and all remains of dead apes that happen to be discovered. I find it very hard to believe that they wouldn’t routinely do that at Kibale, too.

    Or, in other words: please provide some citation for that claim of yours.

    Link to this
  50. 50. SRPlant 12:52 pm 10/20/2011

    @ Spugpow, excellent clip.

    Anyone any idea what the tit-like bird is at 5.45?

    Link to this
  51. 51. Zoovolunteer 2:11 pm 10/20/2011

    One other point – I recall reading a little while ago of some tree kangaroo bones found on the Nullarbor plain, which is now treeless desert. I understood that fires lit by native Australians eventually converted dry forest to the present state, exterminating dry-country adapted tree kangaroos and leaving only the rainforest forms that survive today – does anyone have any more on that?

    Link to this
  52. 52. naishd 4:13 pm 10/20/2011

    Nullarbor Plain tree-kangaroos: I’ve written some text on them (the Pleistocene species Bohra illuminata and B. nullarbora) – I originally planned to use it in the article above but had to prune it out. I’ll publish it as a separate article, stay tuned.


    Link to this
  53. 53. Halbred 6:31 pm 10/20/2011

    Darren wrote: “Halbred — did you crash the wrong party?” Naw, I was just responding to questions of what tetrapod groups, if any, are boring. ;-)

    Link to this
  54. 54. Dartian 1:51 am 10/21/2011

    Anyone any idea what the tit-like bird is at 5.45?

    It seems to be a fantail Rhipidura – but I don’t know which species (there are many).

    Link to this
  55. 55. SRPlant 4:30 am 10/21/2011

    @ Dartian, thanks again. I will start googling.

    Link to this
  56. 56. Jerzy New 7:40 am 10/21/2011

    I guess you need to read research from Kibale to be sure.

    BTW, if tree kangaroo chased by arboreal thylacoleonid jumped down – would thylacoleonid follow and chase it?

    Link to this
  57. 57. Dartian 8:04 am 10/21/2011

    I guess you need to read research from Kibale to be sure.

    If you make a claim, it’s your job to back it up (if requested) with a reference. Not mine.

    Link to this
  58. 58. David Marjanović 11:23 pm 10/22/2011

    Joking aside, is there any group of tetrapods that you find boring?

    Heh. I remember all too well how someone claimed on the Dinosaur Mailing list, many years ago, that ducks were “kinda boring”. Darren replied with a long, rambling paragraph on the many fascinating features of ducks. :-)

    the earliest members of the placental lineage had pouches

    Unknown. Plenty of non-placental eutherians clearly lacked the extended pregnancies of placentals, as shown by their epipubic bones, but where do you take evidence of pouches from? Do pouched and pouchless marsupials differ osteologically?

    Link to this
  59. 59. David Marjanović 11:26 pm 10/22/2011

    Oh, and…
    Ichthyosaurs? I only knew about mosasaurs-to-whales, and that was claimed by engineers in newspapers, not biologists in the scientific literature.
    I’m surprised you didn’t know that kangaroos swim with alternating leg movements. Even I knew that.

    Link to this
  60. 60. Yodelling Cyclist 10:51 pm 10/23/2011

    If anyone is interested, mosasaurs-to-whales is described here:

    It’s spectacular, but not in a good way. Longisquama-as-birds fans will probably find it laughable, it is just that …. I can’t think of the adjective.

    Link to this
  61. 61. David Marjanović 2:07 pm 10/24/2011

    Wow. What a gigantic argument from ignorance, personal incredulity, and more ignorance. And some ignorance on top.

    It’s not just about whales. Follow the link at the bottom.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article