Diario de milewski

@simontonge

(For a concise version please see https://www.inaturalist.org/journal/milewski/66782-why-is-the-giant-panda-black-and-white-concise-version#)

Why is the giant panda (Ailuropoda melanoleuca, https://www.inaturalist.org/taxa/41659-Ailuropoda-melanoleuca) so boldly black and white?

Why - in contrast to other bears - is its colouration so consistent, i.e. lacking individual variation?

And why does it tend to be so unexpressive?

(The main function of the checker-like pattern of the giant panda is unlikely to be camouflage in a patchily snow-covered landscape, as a few authors have previously suggested.)

Could this plausibly be an example of warning colouration?

Is it even possible that the giant panda is the largest-bodied carnivore that possesses aposematic colouration at the scale of the whole body?

The most dangerous predators - apart from humans - historically affecting the giant panda were

• the tiger (Panthera tigris amoyensis, https://www.inaturalist.org/taxa/41967-Panthera-tigris) and
• the dhole (Cuon alpinus hesperius/lepturus, https://www.inaturalist.org/taxa/42101-Cuon-alpinus).

Also playing a part were

• the leopard (Panthera pardus delacouri/japonensis, https://www.inaturalist.org/taxa/41963-Panthera-pardus) and
• the snow leopard (Panthera uncia uncioides, https://www.inaturalist.org/taxa/74831-Panthera-uncia).

The giant panda was in all likelihood vulnerable to predation because of its noisy eating habits and relatively slow reproduction.

All members of the Carnivora defend themselves with their teeth, and many show their teeth to would-be attackers.

However, the giant panda seldom adopts a fang-baring expression or even opens its mouth in fear/anger.

Furthermore,

• the canines are relatively small (https://www.shutterstock.com/nb/image-photo/panda-sits-growls-turning-her-head-1560141806 and https://www.quora.com/Are-pandas-aggressive-or-calm) and would show poorly against the whitish muzzle eve if such an expression was adopted, and
• the claws are short and unimpressive compared with other bears, particularly the sun bear (Helarctos malayanus, https://www.inaturalist.org/taxa/41655-Helarctos-malayanus), which has extremely dangerous claws according to injured zookeepers.

What seems to have been overlooked about the giant panda is that its premolars are unique among carnivores in their shape and arrangement (https://animaldiversity.org/accounts/Vertebrata/specimens/collections/contributors/skulls/ailuropoda/a._melanoleuca/panda_ltr/?f=subject::specimen::teeth::lower%20tooth%20row and https://commons.wikimedia.org/wiki/File:Animaldentition_ailuropodamelanoleuca.png).

Also see https://upload.wikimedia.org/wikipedia/commons/4/45/Giant_Panda_Skull.JPG and https://boneclones.com/product/giant-panda-skull-adult-BC-324 and https://www.skeletonsandskullssuperstore.com/product/giant-panda-female-skull/). These premolars may not look intimidating (https://www.alamy.com/closeup-of-giant-panda-s-mouth-and-teeth-wolong-nature-reserve-sichuan-image106119.html), but would-be predators would be wise to beware.

The giant panda has the most powerful bite of any extant species of bear (https://www.researchgate.net/publication/224630206_Finite_Element_Analysis_of_ursid_cranial_mechanics_and_the_prediction_of_feeding_behaviour_in_the_extinct_giant_Agriotherium_africanum_the_bare_facts and https://zslpublications.onlinelibrary.wiley.com/doi/full/10.1111/j.1469-7998.2011.00862.x). The bite is powerful even at the premolars, because

• the jaws are short and the cheek-teeth are crowded right behind the canines, and
• the masseter muscles are extremely developed, in conjunction with a high sagittal crest (https://en.wikipedia.org/wiki/Sagittal_crest) and wide zygomatic arches (https://en.wikipedia.org/wiki/Zygomatic_arch).

See https://phys.org/news/2021-07-evolution-giant-panda-temporomandibular-joint.html and https://news.cgtn.com/news/3d3d674d7763444d32457a6333566d54/index.html and https://www.researchgate.net/figure/Adaptive-morphological-characteristics-of-giant-pandas-A-pseudothumb-denoted-by-the_fig1_266570673.

The giant panda has the shortest face of all bears. Ironically, the same short muzzle that makes the giant panda rather appealing as a 'human face' also houses its main weaponry.

The compact dentition of the giant panda can best be illustrated in contrast with the polar bear, the species most specialised for carnivory.

Please compare https://www.mindenpictures.com/stock-photo-giant-panda-ailuropoda-melanoleuca-skull-adult-length-mm-naturephotography-image80161382.html and https://www.zoochat.com/community/media/skull-giant-panda.469645/ with https://beautifullybony.wordpress.com/archive/bone-of-the-month/the-polar-bear/the-skull-biomechanics-of-the-polar-bear/ and http://digimorph.org/specimens/Ursus_maritimus/male/ and https://www.westend61.de/en/imageView/MW000023/skull-of-polar-bear-ursus-maritimus-in-front-of-black-background-side-view and https://www.acornnaturalists.com/bear-polar-skull-replica.html.

Among living bears, it is the least carnivorous species, i.e. the giant panda, that has

• the strongest bite, and
• the largest premolars.

Bamboo, a woody form of grass, is the staple diet of the giant panda.

An adaptive reason for the powerful bite at the level of the premolars may be an ability to slice bamboo, preparatory to grinding it with the molars (https://www.researchgate.net/figure/Segmented-maxillary-teeth-of-the-living-bear-species-analysed-in-this-article_fig3_328232826 and https://www.natureinstock.com/search/preview/giant-panda-ailuropoda-melanoleuca-skull-adult-close-up-of-molar/0_11300923.html and https://animaldiversity.org/accounts/Carnivora/specimens/collections/contributors/skulls/ailuropoda/a._melanoleuca/panda_lm/).

In particular, premolars 2 and 3 on the upper jaw, and 2, 3 and 4 on the lower jaw, form a tightly occluding set, operating by means of powerful jaw-muscles. (I have described these teeth in detail in the comments below.)

Although the dentition of the giant panda may be adaptive primarily to its diet, the same premolars are 'mutilars', capable of mutilating any animal attacking the giant panda. This is in conjunction with clamping by the jaws and hugging with the forelegs - in contrast to the piercing bite-and-release and scratching inflicted by felids.

The key to understanding the individually consistent black-and-white pattern of the giant panda as warning colouration may be to realise that the bamboo-crushing premolars of this species are quasi-carnassials, and that they are

• even more dangerous than the canines of most other carnivores, but
• intrinsically difficult to display directly for the hazard that they are.

According to this explanation, the giant panda may have evolved true aposematic whole-body colouration, as a way of announcing its hidden defensive capability.

In its strict sense, 'aposematic' refers to warning colouration in those organisms in which the defensive capabilities are not only extraordinary and specialised, but also non-apparent as such.

If the giant panda really is aposematic at the scale of the whole body, then it may possibly be the largest-bodied aposematic organism on Earth.

How do the patterns of dark/pale contrast differ between the giant panda and other bears?

I will discuss the eyes and ears in part 2. However, in the case of the chest, other species of bears show certain trends in colouration that can be extrapolated to the extremes shown in the giant panda.

More than any other family of carnivores, bears express two patterns consistently:

• all bears except for the brown bear (Ursus arctos) tend to be conspicuously dark or pale overall, and
• in the dark species there tend to be conspicuously pale marking in various configurations on the chest, exposed when the animal rears up in confrontation, and possibly functioning as warning signs.

The giant panda is odd in having large patches of black on an otherwise whitish coat. Other bears tend to have either dark or pale tones dominant, without gross variegation.

Pale bears are the polar bear (Ursus maritimus) and the pale morph of the American black bear (Ursus americanus). No predominantly pale bear has dark markings on its fur.

The retention of brownish on its posterior (https://www.istockphoto.com/photo/the-giant-panda-or-bamboo-bear-gm601399942-103438177 and https://www.shutterstock.com/nb/video/clip-1021033759-giant-panda-ailuropoda-melanoleuca-rubbing-tail-against and https://www.shutterstock.com/nb/video/clip-32900209-panda-forest and https://pandathings.com/learn-about-the-giant-panda/do-pandas-have-tails/) suggests that the giant panda is not unique among bears, just extreme in its differentiation into dark versus pale.

An explanation for the relatively plain coats of other bears is that these formidable and non-gregarious mammals, all of which rely on smell and none of which hunts by stealth, need neither cryptic colouration nor social colouration that advertises sex, age, or maturity.

The dark chest of the giant panda is unusual among bears in lacking the ominous 'pectoral flags' seen in the sloth bear, Asian black bear, and sun bear, in upright confrontation (https://www.reddit.com/r/NatureIsFuckingLit/comments/ogont1/sloth_bear_stands_its_ground_against_a_tiger_india/ and https://en.wikipedia.org/wiki/Asian_black_bear#/media/File:Ursus_thibetanus_3_(Wroclaw_zoo).JPG and https://www.bigstockphoto.com/image-330810433/stock-photo-asiatic-black-bear-standing-and-relax-in-the-summer-black-bear-with-chest-the-v-shape-is-white-woo and http://www.bearsoftheworld.net/sun_bears.asp and https://www.alamy.com/stock-photo-a-sun-bear-helarctos-malayanus-standing-up-on-it-hind-legs-at-rare-89813043.html).

Why does the giant panda seems to differ categorically from other bears of similar body size in this way?

A possible interpretation is that it has effectively expanded the insignia on chest and face, to convert the whole front to black-and-white (http://www.ecns.cn/m/visual/hd/2018/01-14/151448.shtml and http://www.china.org.cn/environment/2015-04/08/content_35268385_4.htm).

Please note that, when the giant panda rears up in confrontation,

• the blackish of the chest, shoulders and forelegs contrasts sharply with the mainly whitish face and belly, and
• any blurred pattern on the hindquarters is out of view..

Thus the whole black-and-white front of the giant panda is plausibly an extension of – to the point of substituting for – the 'pectoral flags' of other bears.

to be continued in https://www.inaturalist.org/journal/milewski/66777-does-the-colouration-of-the-giant-panda-hint-of-anti-predator-defence-by-means-of-mutilating-premolars-part-2#...

...continued from https://www.inaturalist.org/journal/milewski/66693-does-the-colouration-of-the-giant-panda-hint-of-anti-predator-defence-by-means-of-mutilating-premolars-part-1#

Please see https://explorebioedge.com/2023/10/18/biological-expositions-panda-copy/

Now let us examine the colouration around the eyes, in the giant panda, other bears, and the red panda (Ailurus fulgens, https://www.inaturalist.org/taxa/41653-Ailurus-fulgens) - which belongs to a different family but has a similar diet of bamboo and similar habitat in the same region of Asia.

The ocular patches of the giant panda are starker than those of any other bear.

However, the blackness of the fur around these eyes is only the starkest expression of a pattern shared in subtle form with other bears – particularly the partly bamboo-eating spectacled bear (Tremarctos ornatus, https://www.inaturalist.org/taxa/41657-Tremarctos-ornatus) of the Andes.

What is the function of the ocular exaggeration seen in the giant panda?

Dark ocular patches, bands and stripes are more common in carnivorous mammals than in herbivorous mammals. Depending on the details, they can hypothetically aid either crypsis or conspicuousness.

Few non-noxious species of the carnivore order combine an eye-mask with cryptic colouration. In the case of the suricate (Suricata suricatta, https://www.inaturalist.org/taxa/41943-Suricata-suricatta), which overall has cryptic rather than conspicuous colouration and is associated with salinas, the ocular patches possibly function to reduce glare.

And where they aid conspicuousness, this is usually indirectly by drawing attention away from the eyes towards a whole-body pattern of contrasting dark and pale, as typified by skunks (Mephitidae, https://www.inaturalist.org/observations?taxon_id=41871).

In Mustelidae, various species with warning colouration seem to mask their eyes in various designs. Perhaps it serves the untrustworthy image of bad-smelling members of the Carnivora to hide their eyes even as they advertise their noxiousness by means of dark-and-pale flags. This must seem sinister or at least confusing to a large antagonist.

Based on my interpretation of eye-masks in Carnivora, it might be consistent to suggest that pandas hide their eyes in combination with otherwise conspicuous colouration, as a form of warning.

However, I suggest that the giant panda is one species – and possibly the only one in its order – that uses ocular patches in a different way,

Instead of hiding its eyes, it emphasises them in a false stare.

This is a subtle difference from the mustelid pattern, but would also be configured to confuse enemy species in confrontation.

As observed by humans, the dark ocular patches of the giant panda seem not to mask but rather to accentuate the eyes as 'larger-than-life'.

This inadvertently panders to our biases by being rotund and furry, with a broad face and ocular patches that face forward as in humans and simulate the proportionately large eyes of youth. This anthropomorphic impression is of course psychological, because the real eyes of the giant panda, like those of all bears, are disappointingly small and 'piggy'.

However, as observed by the relevant predators:
In light too dim to show the real eyes but adequate to show the white face, the dark ocular patches of the giant panda would exaggerate rather than hide the eyes.

A large felid would probably not see the eyeballs of the giant panda - which are inconspicuous even in bright light, because the iris is dark and the eye-whites (scleras) are hardly exposed. However, the staring ocular patches, seeming to be the eyes of a larger-than-life adversary, might sway the risk-assessment of the predator.

When a large-bodied felid confronts the giant panda, the glaring dark orbs on a stark, intimidating face are a plausible form of confusion/intimidation.

Such encounters would seldom have occurred, even prehistorically. This is because of the natural scarcity of the giant panda, compared to common prey such as sambar deer (Cervus unicolor), takin (Budorcas taxicolor), and wild pig (Sus scrofa).

Hence the predators would tend not to be experienced enough to overcome their hesitancy.

Is there any other mammal worldwide – apart from the related spectacled bear – that shows a similar pattern for similar reasons, but by independent evolution?

Possible candidates occur among South American anteaters (https://www.inaturalist.org/observations?taxon_id=53537 and https://www.inaturalist.org/journal/milewski/59344-why-even-the-jaguar-fears-a-mere-anteater# and https://www.theguardian.com/environment/radical-conservation/2017/feb/21/jaguar-giant-anteater-amazon-rainforest-camera-trap-brazil).

Anteaters, like pandas, flee and reproduce slowly owing to an energy-poor diet. They are therefore likewise vulnerable to predation.

Although they lack teeth, their foreclaws are formidable because of the extraordinary power needed to excavate their main diet of social insects from wood and baked earth. Although small-brained and small-eyed, anteaters can be lethal in a vice-like clinch.

The giant anteater (Myrmecophaga tridactyla, https://www.inaturalist.org/taxa/47107-Myrmecophaga-tridactyla) again presents large felids, viz. the jaguar (Panthera onca) and the puma (Felis concolor), with a dark patch on a whitish background. In this case the ominous pattern is located on the wrist.

This pattern warns the would-be predator that sluggish does not necessarily mean vulnerable (https://www.dreamstime.com/royalty-free-stock-photo-giant-anteater-image25662955 and https://www.dreamstime.com/royalty-free-stock-photography-giant-anteater-detail-strolling-image39072067 and https://www.dreamstime.com/stock-photo-giant-anteater-adult-strolling-grass-image48467708 and https://www.dreamstime.com/giant-anteater-myrmecophaga-tridactyla-giant-anteater-myrmecophaga-tridactyla-also-known-as-ant-bear-image137426284 and https://www.dreamstime.com/royalty-free-stock-photography-giant-anteater-adult-strolling-grass-image39071997 and https://www.researchgate.net/figure/Segmentation-of-the-giant-anteaters-Myrmecophaga-tridactyla-body-showing-the-anterior_fig5_269667258 and https://www.aboutanimals.com/mammal/giant-anteater/ and https://www.dreamstime.com/stock-image-giant-anteater-image24206581 and https://www.dreamstime.com/anteater-cute-animal-brazil-giant-anteater-myrmecophaga-tridactyla-animal-long-tail-log-muzzle-nose-pantanal-brazil-image113787045 and https://www.dreamstime.com/stock-photography-giant-anteater-image20613402 and https://en.wikipedia.org/wiki/Giant_anteater#/media/File:Myresluger2.jpg).Is it mere coincidence that the dangerous foreleg of the giant anteater resembles the head of the giant panda? (https://www.boredpanda.com/anteaters-front-legs-look-like-pandas/?utm_source=google&utm_medium=organic&utm_campaign=organic and https://www.youtube.com/watch?v=kVuYGcs3OMA and https://www.youtube.com/watch?v=l6ngv3HGyjk and https://communicatescience.com/zoonomian/2009/01/29/antz-meets-kung-fu-panda/ and https://www.reddit.com/r/pics/comments/nmx1wi/have_you_ever_noticed_that_giant_anteater_legs/)

Now let us examine the red panda. This species is relatively small-bodied, but shows certain patterns in common with the giant panda, including slow reproduction and thus vulnerability to predation.

The red panda shows convergent colouration in its

• dark legs,
• noticeably pale face and dark markings on the ears and below the eyes, and
• difference between the conspicuous anterior and inconspicuous posterior, with a brown rump and tail.

The red panda differs from bears in two relevant ways, as follows:

• its starkly dark-and-pale ears are more conspicuous, relative to body size, than in any bear, and
• it takes a pattern of darkness under the eyes that is present in subtle form in most bears and expresses it more clearly than in any bear apart from the giant panda.

Ear pinnae with dark and pale fur, arranged in a striking pattern, occur in various Carnivora. However, of all the 280 or more species in this order, the red panda has among the most conspicuous ears relative to the rest of the body and head.

This is because the ears

• project sharply from the silhouette,
• have an eccentric lower tassel,
• have whitish rims and points, and
• present a contrast of dark versus pale on both the front and the back surfaces (https://www.timeout.com/melbourne/things-to-do/red-panda-encounters and https://kids.nationalgeographic.com/animals/mammals/facts/red-panda and https://www.alamy.com/stock-photo-red-or-lesser-panda-head-from-behind-showing-pale-ear-tips-china-30826963.html).

Indeed, the red panda takes all the various ways in which ears can be conspicuous across the spectrum of other carnivores, and combines them in one species.

The resulting ‘auricular flag’ sets the red panda particularly apart from bears other than the giant panda, all of which lack any particular auricular colouration.

The giant panda and the red panda, although independently evolved, happen to converge in having conspicuous ears by different patterns. For among all bears, it is the giant panda that has the most conspicuous of ears. And the pattern in the red panda once again hints that the black upstanding ears on a white head of the giant panda aid conspicuousness rather than crypsis.

The fur around the eyes presents a parallel to the fur of the ears. Although the dark patches associated with the eyes of the giant panda are more defined than those of the red panda, there is a similarity in orientation and width there.

The following show that both the giant panda and the red panda have similarly oriented dark patches associated with their eyes:

https://www.dreamstime.com/portrait-panda-bear-close-up-cute-china-animals-close-up-view-panda-s-head-portrait-shot-portrait-panda-bear-close-up-image183719243https://www.dreamstime.com/stock-images-panda-image11134394https://www.dreamstime.com/stock-image-cute-panda-bear-walking-park-image36431751https://www.dreamstime.com/royalty-free-stock-photos-red-panda-close-up-image10652028https://www.dreamstime.com/cute-red-panda-exploring-portrait-close-up-portrait-curious-red-panda-walking-grass-exploring-image134067169.

(writing in progress)

The giant panda reproduces slowly owing to its nutrient-poor diet of bamboo, and it is on this that its aposematism (see previous posts) is partly based.

The point of this Post is to show that this species has has a peculiarly long adolescent period.

Please see basic information at http://www.4panda.com/panda/pandatips/reproduction.htm and http://wwf.panda.org/what_we_do/endangered_species/giant_panda/panda/panda_life_cycle/

Since humans reach sexual maturity early in the teenage period, I now see that the thinking I used below was rather fuzzy. Using humans as a model, ‘teenage’ refers to a period mainly after, not before, sexual maturity, but before full body mass is attained. In female humans the teenage period, thus defined, would be only 13-17, because girls certainly reach full body mass by 17 years old. So I guess (bearing in mind that the female is the standard gender) that female ‘teenagers’ in the human species are really 13-17 years old, a period of 5 years.

Coming back to the giant panda:

I’ve so far not found information on when the female giant panda reaches full body mass. However, assuming that sexual maturity (which occurs at 4-8 years old but usually at ca 6 years old, about half the human value) occurs before full body mass is reached (which is a common pattern in mammals), then let’s assume that the female giant panda grows in body mass until about age 8 years. If so, then ‘teenager’ in females of the giant panda would be defined as 2-8 years old, a period of 6 years, which is impressively long for a member of a non-primate and as long as in the human species.

Please note that the giant panda achieves independence from its mother at about age 18 months (after being weaned by about 9 months), so we can certainly regard any age above 2 years as ‘teenage’, not so?

Out of date?So the answer is: the giant panda is adolescent at about 2-6 years old, which is a relatively long period for a member of the Carnivora. A growing giant panda can be described as adolescent for at least 3 years of its life whereas a growing wolf could be described as adolescent for hardly more than one year of its life, and the lion is more similar to the wolf than to the giant panda in this respect.

So, the bottom line, on second thoughts, seems to be this:

The human species has a female teenage period of about 5 years, ages 13-17.

The giant panda has a female teenage period of similar length, ages 2-8 years.

Despite being faster-growing and shorter-lived than the human species, the giant panda manages to have a prolonged adolescent period.

Although felids and canids are different, in growing faster and having brief adolescent periods of only ca 2 years, another lineage of Carnivora that would be worth looking at for comparison with the giant panda is the hyenas, which live long and may grow more slowly than cats and dogs. And of course we already know of the fossa (Cryptoprocta) of Madagascar, which has a prolonged ‘tween’ period, in its case pre-sexual maturity. I suspect that, although giant panda and fossa both take many years to mature, the difference is that the giant panda reaches sexual maturity earlier (relatively and possibly absolutely) than the fossa, hence has a longer ‘teenage’ period than the fossa.

As you can see, ‘teenage’ female humans are more or less always sexually mature, ‘teenage’ females of the fossa are not yet sexually mature, and ‘teenage’ females of the giant panda straddle sexual maturity. So the word ‘teenage’ hardly stands up to scrutiny as a useful way of referring to the growth stage of the giant panda that you observed in China, being too hard to define satisfactorily, not so? Nonetheless, I can see where people are coming from in applying this term to the giant panda, because (compared with felids and canids) it does have an extended period of playful adolescence.

Humans are adolescent at 13-19 years old, a period of 7 years of life.

but what does ‘teenage’ mean in terms of the life history of the giant panda, a species that lives at most 30 years and more usually only 20 years? And why describe the giant panda at this age as ‘teenage’ given that such terms are not usually applied to, for example, the wolf?

Well, sexual maturity in the female giant panda is at ca 6 years old, and sometimes as early as 4 years old.

whereas it is at <3 years old in the lion (Panthera leo) and <2 years old in the wolf (Canis lupus). Although weaning occurs within the first year in the giant panda, the animals grow for another >5 years before starting to breed. The giant panda certainly grows more slowly than does the lion or wolf, and possibly lives longer. I presume that even at 6 years old, the time of sexual maturity, the body is still not fully grown in the giant panda. I’ve not heard the term ‘teenager’ applied to adolescents of the wolf, presumably because that canid grows so fast that ‘teenagers’ would be limited to the age range of 1-3 years at most.

(writing in progress)

@maxallen @aguilita @marcelo_aranda @biohexx1 @jwidness @zarek @seasav @douglasriverside @napoleon1799 @oviscanadensis_connerties @walkwithbears @thebirdnerd @gwark @mc1991 @ursusont @chewitt1 @pelagicgraf @jayras @rangertreaty50 @williamwisephoto @bobby23 @frostfox @kmccartney3521 @suepemberton @ptexis @botswanabugs

(For details see https://www.inaturalist.org/journal/milewski/66693-does-the-colouration-of-the-giant-panda-hint-of-anti-predator-defence-by-means-of-mutilating-premolars-part-1#)

The black-and-white pattern of the giant panda (Ailuropoda melanoleuca) - which is remarkably uniform regardless of age or sex - was first interpreted by Desmond Morris (https://en.wikipedia.org/wiki/Desmond_Morris) as warning colouration half a century ago.

However, nobody seems to have taken that suggestion seriously.

This bamboo-chewing bear has an exceptionally strong bite. However, all carnivores defend themselves by biting, and teeth can be displayed directly by facial expression.

So, was it far-fetched to suggest that the giant panda has evolved a conspicuous, skunk-like pattern on its coat just to warn potential predators of teeth that have not particularly evolved for self-defence?

(The most relevant predators, formerly occurring - albeit in spare populations - throughout the habitat of the giant panda, were the tiger (Panthera tigris) and the dhole (Cuon alpinus, https://www.inaturalist.org/taxa/42101-Cuon-alpinus), the latter being far smaller-bodied than the giant panda but attacking gregariously.)

As it turns out, the giant panda does possess a hidden defensive capability, previously overlooked even by Desmond Morris. This species is unique among Carnivora in possessing 'mutilar', quasi-carnassial teeth. These are clipping/mangling premolars, which function both to cut and peel bamboo and to mutilate attackers.

The quasi-carnassials of the giant panda, namely premolars 2-4 on both the upper and the lower jaw, combine three features unusual among Carnivora:

• tight occlusion between upper and lower counterparts, in contrast to the non-/partial occlusion of the premolars in most other Carnivora including other bears (Ursidae), which tend to have a diastema (https://en.wikipedia.org/wiki/Diastema),
• placement just behind the canines, close enough to the front of the mouth to allow direct biting, and
• cusps sharp and sturdy enough to cut/break flesh and bone, driven by massive jaw muscles on a robust skull adapted to chew woody food (bamboo).

In most other Carnivora, the only premolar used for shearing/slicing is the fourth upper premolar; this upper carnassial tooth is used for processing food, not fighting in self-defence. What distinguishes the giant panda is that this and five other premolars on each side (left or right) have been converted into a set of quasi-carnassials. These are not as sharp as the carnassials of the polar bear, felids, and canids, but are as large and driven by equally strong musculature, while also being less fragile.

This may be in a sense absolutely as well as relatively the largest/strongest set of premolars of any member of the Carnivora. However, its significance has been overshadowed by the exceptional molars of the giant panda – which are unsuited to self-defence.

Crucial to explaining the colouration of the giant panda is that the hazard of the quasi-carnassials is not self-evident. As a result this species has been overlooked as an animal uniquely capable of mutilating an attacker’s body, with a clipping/mangling mechanism evolved for self-defence as much as for eating.

Instead of fang-baring or showing its (modest) claws in confrontation, the giant panda instead maintains a menacingly expressionless face - partly compensated for, in its social life, by a complex repertoire of vocalisations.

In its strictest definition, aposematism (https://en.wikipedia.org/wiki/Aposematism) is not just warning colouration, but a flagging - at the scale of the whole body - of a hidden defensive capability. Inasmuch as it conforms, the giant panda may be the largest-bodied aposematic organism on Earth.

(writing in progress)

The Tasmanian devil and the wolverine may seem worlds apart. One is a marsupial isolated in Tasmania while the other, surviving the rigours of Siberia and Alaska, is related to weasels. However, the two species are remarkably similar owing to the process of convergent evolution. And understanding the differences that remain between them can deepen our understanding of the Tasmanian devil in ways not possible by comparisons within Australia.
 
Apart from a certain ‘appealingly diabolical’ image in the public mind, the Tasmanian devil and the wolverine resemble each other in many ways. Both species:

• weigh about 10 kg, with male bigger than female
• have notoriously strong jaws for their body size
• have small eyes considering that they forage mainly in darkness
• retain a social life despite foraging as solitary individuals
• combine versatile predation with scavenging and an ability to eat bones
• forage over great distances despite running with a lumbering canter rather than the neat trot used by canids, felids, and wombats
• retain fore feet dexterous enough to handle food objects, and an ability to climb despite foraging mainly on the ground
• have similar colouration: dark with erratic white markings on the chest, and an individually variable pattern of pale stretching from the shoulders along the flanks to the rump.

Evolutionary convergence is a favourite concept in biology textbooks. However, the examples usually given – such as kangaroos being Australia’s answer to deer, or echidnas being analogous to the aardvark – can seem rather far-fetched. One of the most convincing of these supposed ‘mirror-images’ is Tasmanian devil vs wolverine, but even in this case we will see that the differences can be more revealing than the similarities. Explaining certain disparities between these two, beyond the marsupial-eutherian difference or the polarities in their climates, may give us new understanding of the largest surviving marsupial carnivore.
 
So why have the Tasmanian devil and the wolverine evolved to be so similar despite being ancestrally unrelated and having such different reproductive modes?
 
Both mammals live in poor environments: nutrient-poor, fire- and flood-prone Australia and underlain by permafrost and covered by snow for most of the year in the subarctic. In both places, prey is sparse and carnivorous mammals cannot afford to be too demanding. In adaptation to these conditions, both the Tasmanian devil and the wolverine combine relatively small bodies – and therefore appetites - with outsize abilities to bite any large item that does present itself. The body size and leg length in both cases are finely balanced to limit demands while at the same time to allow enduring locomotion in search of what little food there is. When confronted, both forms display their impressive teeth. The odd colouration shared by Tasmanian devil and  wolverine combines general concealment in the dark with social self-advertisement: the whitish markings are like exclamatory insignia, allowing individuals to recognise each other in occasional encounters at distances just at the limits of olfaction and the vision of relatively small eyes. The patterns of colouration thus reflect societies so stretched, in these poor environments, that they seem to consist of loners in human terms.
 
Given these similarities between two independently evolved mammals, the disparities between the Tasmanian devil and the wolverine become all the more revealing. These include both morphological and behavioural differences.
 
Morphologically:

• the head - and particularly the jaws - are far larger, proportional to the body, in the Tasmanian devil than in the wolverine, but the brain is smaller
• the dentition of the Tasmanian devil remains typical of dasyurid marsupials while that of the wolverine remains typical of eutherian Carnivora
• the feet are more powerful, with more prominent claws and rubbery pads, in the wolverine than in the Tasmanian devil
• the eyes are emphasised by pale skin in the Tasmanian devil but hidden by a dark mask in the wolverine, and differ greatly in the orientation of the pupil (vertical in the Australian form but horizontal in the subarctic form)
• whiskers (facial vibrissae) are far more prominent in the Tasmanian devil than in the wolverine.

Functionally:

• the Tasmanian devil eats gregariously, calling in unrelated individuals to share large carcasses, whereas the wolverine eats alone, caches any leftovers, and tolerates only family members anywhere near its food
• the Tasmanian devil, although not territorial, maintains communal latrines for olfactory communication, whereas the wolverine does not.
• the fang-baring display of the Tasmanian devil – despite extremely wide opening of the mouth – is essentially bluff whereas that of the wolverine means business
• in adulthood, the wolverine retains a far greater ability to climb trees than does the Tasmanian devil
• the rates of metabolism and growth are far greater in the wolverine (which does not hibernate) than in the Tasmanian devil
• longevity in the Tasmanian devil is half that in the wolverine, balanced by the greater number of infants per average litter (four) than in the latter (two).

In explaining these disparities, we can first discount the results of incongruities in climates. For example, there is no mystery as to why the wolverine is by far the furrier form, with feet broad enough to act as snowshoes. The difference in the orientation of the pupils may possibly be explained by excessive reflection of ultraviolet from snow: the wolverine has adapted by restricting this glare to a horizontal slit during daylight.
 
An important difference in the environments is that the wolverine risks occasional encounters with predators far larger than itself: the wolf and up to three species of bears. Although the Tasmanian devil formerly coexisted with the thylacine, the difference is that the wolf can fight gregariously whereas the thylacine fought alone or perhaps in pairs. Furthermore, there is competition between the wolverine and both the wolf and the brown bear for bones, whereas the dentition of the thylacine suggests that it left bones to the Tasmanian devil. These different predatory environments explain why the wolverine retains an ability to take refuge rapidly up trees, using the purchase of rubbery treads and sharp claws. Also thus explained is the subtlety of facial displays in the wolverine, which not only warns of a powerful bite but escalates the fang-baring expressions while at the same time hiding the eyes and thus the precise direction of attention of the confronted wolverine – which can also use its fore feet in self-defence.
 
The Tasmanian devil has, proportionately to its body size, the largest head of any marsupial or eutherian carnivore worldwide. The according differences in the dentitions amount to a case of quantity in the Tasmanian devil versus quality in the wolverine: the marsupial uses brute force of its molars to break bones whereas the wolverine uses fine adjustments of the jaw joints to apply precise pressure on its premolars and separately on specialised, small molars oriented at odds to the rest of the toothrows.
 
The extreme development of the facial whiskers in the Tasmanian devil remains to be fully explained, but it seems to be related to both the exceptionally large size of the head, and a habit of intraspecific ‘jaw-fencing’, a behaviour not seen in the wolverine or other eutherian carnivores. When the Tasmanian devil eats socially, the combination of whiskers and glaring eyes seem to help in maintaining sufficient personal distance.
 
Although the Tasmanian devil has been intensively studied, one aspect of its life history is seldom appreciated. This is the combination, in common with other dasyurids such as quolls and antechinuses, of a combination of reduced longevity and limited rates of reproduction (needs elaboration). At first glance it may seem that the Tasmanian devil reproduces more rapidly than does the wolverine, in keeping with the difference in longevity. However, this does not stand up to scrutiny because in fact the marsupial grows more slowly from conception to maturity than does any eutherian carnivore of similar body size – even one as beset by a difficult environment as the wolverine. The limited pressures from other predators in Australia have allowed the Tasmanian devil to breed relatively slowly (because of its relatively slow growth and despite the many newborns per litter), while relatively rapid turnover of generations have allowed it to respond to fluctuations in the availability of food according to cycles of climate and fire in Australia.
 
There is no real evolutionary convergence between the dentition of Gulo and that of Sarcophilus.

The dentition of Sarcophilus is essentially still that of a typical dasyurid marsupial, albeit scaled-up impressively. The dentition of Gulo is essentially still that of a typical member of the Carnivora (of which the most familiar examples are felids), but achieves its bone-crushing strength by being scaled DOWN in such a way that it is so compact that the teeth are unlikely to break no matter how much force is applied to them.
 
This concept is what seems to have been missed in previous comparisons of Sarcophilus with Gulo: that in a sense they not only fail to show any evolutionary convergence in dentition, but if anything have evolved in ‘opposite’ directions, with Sarcophilus going for increased size of jaws and teeth and compensating for the resulting breakage of the cusps by retaining the limited lifespan of dasyurids, whereas Gulo has gone for decreased size of the jaws and teeth so that its carnassials can be efficient at both shearing flesh and breaking bones, while its upper molar (which has no counterpart in Sarcophilus) can be brought to bear by precise sideways movement in the jaw joint, something apparently not used by any carnivorous marsupial.
  
I have found no suitable photos of the skeleton of Gulo, but this drawing does show how small the skull is relative to the body – the opposite of what we’ve seen in Sarcophilus. Because of the proportions shown by head to body, Sarcophilus looks like a ‘mighty mouse’ rather than a hyena, whereas Gulo looks like a small bear rather than a giant marten. This is relevant to our topic of bone-crushing jaws because it immediately raises the question ‘if the wolverine has such a proportionately small head, how does it crack large bones?’
 
https://s-media-cache-ak0.pinimg.com/originals/79/b5/76/79b57686b6a14a0c268dae5384993e3e.jpg
 
The following, again, is a drawing, but it is useful for its clarity. What is immediately noticeable in comparison with Sarcophilus is the development of the massive premolar on the upper jaw and its corresponding molar on the mandible. The cheek-teeth are far more differentiated and specialised in Gulo than in Sarcophilus. And the outer incisor on the upper jaw, i.e. that closest to the canine, is far larger than that in Sarcophilus; indeed Gulo has a kind of clamp in which the lower canine fits into a slot between upper canine and the nearby upper incisor, whereas Sarcophilus lacks this feature despite having more upper incisors (2 X 4) than Gulo (2 X 3).
 
http://img09.deviantart.net/00b7/i/2009/111/f/0/wolverine_skull_by_revelation_six.jpg
 
https://web.archive.org/web/20080527021506/http://www.wolverinefoundation.org/dentition.htm

The following shows how specialised the last upper premolar is in Gulo. This is a carnassial tooth typical of Carnivora, but it is also so stout that it may be used for breaking bones, without much risk of fracture of the cusps. And note the biggest difference of all from Sarcophilus: there is only one upper molar (compared with what is usually stated to be 4 in Sarcophilus), and this molar of Gulo looks nothing like those of Sarcophilus, being a) blunt and suitable for crushing bones, and b) set at right angles to the rest of the cheek-tooth row, providing broadened purchase or traction. What this means is that, although Gulo has fewer teeth than does Sarcophilus, its dentition is geometrically the more complex and by inference the more efficient relative to jaw size.
  
The following shows the two lower molars in Gulo: the small posterior-most tooth and the stout carnassial anterior to it. Compare this with the upper jaw above and note that there is no way for the upper molar (which projects inwards from the rest of the tooth row) to occlude any molar on the mandible unless the jaw is moved considerably sideways. What I infer from this is that, in order to crush bones or to grasp frozen meat/hide firmly, Gulo has to use considerable lateral mobility in its mandible. If so, this is, in a sense, the ‘opposite’ of the specialisation shown by Sarcophilus. The Australian mamma specialised in an extremely wide gape but there is no suggestion of unusual lateral mobility in the jaw-joint mechanism. Gulo is not known to gape widely but, as I infer from its dentition, it also has a kind of loose joint of the mandible with the rest of the skull, allowing it to apply a relatively small pair of jaws to large items with great force.
  
In Sarcophilus, the orbital socket is located about halfway between the front of the skull and the back of the skull. In Gulo the orbit is far closer to the front than to the back, and this is because the cranium is far longer in Gulo than in Sarcophilus. The point is that the skull shows how much larger the brain of Gulo is than that of Sarcophilus, relative to skull size. I suspect that this applies also relative to body size as a whole: Gulo is the brainier mammal.
 
http://wolverinefoundation.org/wp-content/uploads/2010/12/skull.gif
 
The following four photos show the skull of Gulo for direct comparison to that of Sarcophilus. This shows the clamping mechanism at the front of the jaw, and the far greater development of the carnassial mechanism at the rear of the jaw, in Gulo.
 
http://www.skullsunlimited.com/userfiles/image/variants_large_4328.jpg

https://www.nature-watch.com/images/products/large/806xx.jpg

http://www.boneroom.com/uploads/4/8/1/1/48118243/s521972503441136676_p1363_i1_w640.jpeg

http://www.educationalbiofacts.com/images/R-S302-Wolverine.jpg

The following two photos show the lower incisor row clearly. Please note that the lower incisors have a naturally staggered formation in Gulo, which means that the lower incisor row can maintain its strength and durability by a kind of ‘bracing’ in which the incisors are compacted into a short row. I have seen nothing like this kind of complexity in the lower incisor row of Sarcophilus.

What this means is that Gulo has a relatively narrow mouth compared with Sarcophilus, but when it latches on to an object with the front teeth, there is an extremely strong clamping mechanism because a) the lower canine fits between two large upper teeth, and b) the lower incisors are unlikely to break because they are arranged with such compactness and such mutual bracing.
 
http://www.skullsunlimited.com/userfiles/image/category5_species_1990_large_3.jpg

https://www.skullsunlimited.com/userfiles/image/category5_species_6600_large_3.jpg

 
The following several photos show the skull with the mouth open. The dentition resembles that of Felidae more than that of Sarcophilus, but is stronger than that of any cat, relative to skull size, because the dentition combines compactness with stoutness while at the same time retaining sharpness in a) the upper and lower canines, and b) the carnassial shear provided by the sliding occlusion of the last upper premolar against the first lower molar.
 
https://s-media-cache-ak0.pinimg.com/originals/72/e1/86/72e1861883ac7877f1b2505b2ecb3b13.jpg

http://img01.deviantart.net/9c7a/i/2004/260/a/0/wolverine_skull_1_by_rgstock777.jpg

http://img02.deviantart.net/b158/i/2010/342/5/2/wolverine_skull_by_zasalamell-d34hedj.jpg

http://images.hideandfur.com/inventory/wolverskullc.jpg

https://s-media-cache-ak0.pinimg.com/564x/bb/53/cf/bb53cf6f906043f31b92e87c0d643b52.jpg

http://img01.deviantart.net/0a74/i/2010/070/6/b/wolverine_skull_reference_2_by_lamelobo.jpg

To be continued...

(writing in progress)

(writing in progress)

See additional white markings in various individuals of Sarcophilus shown in https://www.youtube.com/watch?v=I761Bv-Y-gQ .
 
The following shows something approaching the maximal expression of pale markings in that otherwise blackish species, the Tasmanian devil: http://farm1.static.flickr.com/125/398925241_aabb266df5_o.jpg.
 
In this specimen, as you can see, the white ‘collar’ is about as broad as it ever gets in S. harrisii, and extends about as far on to the shoulders as it does in any individual. Furthermore, the corresponding mark on the rump is also about as broad and extensive as I’ve seen on any individual.
 
Another point, seldom remarked in the literature, is that S. harrisii also has pale around the eyes. However, this is not produced by depigmentation of the pelage. It is produced by the pale skin (flesh-coloured) showing through owing to the extreme sparsity of pelage on and near the eyelids. The same applies to the supraorbital tuft of facial vibrissae, although that is not particularly apparent in this photo.
 
So which other pale markings could any individual of S. harrisii possibly have, in addition to those seen below?
 
Well, it’s sometimes stated in the literature that individuals may have a white tip to the tail, but Nick Mooney has corrected that notion by pointing out that in all his extensive experience with the species he has never seen pale tipping on the tail. So I think we can cross that one off the list.
 
The most important pale marking not expressed in the individual below is something I’ve seen in a few individuals of S. harrisii: an isolated white irregular spot/blotch about mid-flank, at the same level as the white band on the shoulders below. Although the white bands on shoulder and rump are never joined in any individual of S. harrisii, the occasional individual hints at such continuousness by having that white mark at ‘mid-point’ along the flanks.
 
Several photos show individuals with irregular, untidy flecking of white on the body, almost like a faint ‘echo’ of the spotting seen in Dasyurus maculatus. Although such markings are too faint to seem functional in S. harrisii, the interesting thing about them is that they indicate the close relationship of Sarcophilus to quolls (Dasyurus), and then hint at how the bold white markings, illustrated below, have arisen evolutionarily: from a quoll-like ancestral colouration, most spots were lost but some were retained and expanded.
 
The trouble with the process I’ve just invoked is that no individual of any species of quoll (Dasyurus) that I’ve ever seen had any kind of tonal contrast on its chest. In Dasyurus, the chest is unremarkably coloured (nondescript rather pale pelage) and lacks either a dark ground-colour or any white spotting/blotching. So although Sarcophilus may have arisen from a spotted, quoll-like ancestor its development of tonal contrast on the chest is novel and significant. And, notwithstanding any resemblance to quolls, it remains true that Sarcophilus is the only marsupial worldwide that has converged evolutionarily with a pattern of colouration that crops up again and again among eutherian carnivores: dark pelage with a whitish insignia on the chest or front-of-neck.
 
The particularly interesting thing about the individual shown above, w.r.t. my comparisons with Gulo, is that this maximal patterning in Sarcophilus extends the evolutionary convergence. If it’s true that merely having white insignia on the chest or front-of-neck shows a general evolutionary convergence in colouration with various bears, mustelids, and even canids, then the extension of the pale areas along the side, to the rump, shows convergence with a DIFFERENT pattern, found unusually in Gulo, in which – quite independent of the pale insignia on the chest and front-of-neck – there is a pale ‘banding’ along the flank, from the shoulders to the root of the tail.
 
So, the individual above shows two distinct convergences evolutionarily. The pale on its chest shows convergence not only with Gulo but with many other Carnivora including particularly bears. But the pale on its shoulders and rump – despite being so contiguous/congruent with the chest in Sarcophilus – shows a quite separate convergence with a different system of pale pelage in Gulo.
 
Finally, what about those pale ‘spectacles’ around the eyes in Sarcophilus? Are they convergent with any pattern seen in eutherian carnivores? That’s a question I’ll have to explore further, but offhand I can’t recall anything similar in Carnivora. The general trend in Carnivora is the ‘opposite’ one: various families show a pattern of masking in which the eyes are obscured. Although the eyes are emphasised in some eutherian carnivores as well as some other marsupials, this is usually by pale patches of pelage at the position of the ‘eyebrow’ as seen in the typical appearance of the black-and-tan form of the kelpie breed of domestic dog. In the way it accentuates its eyes, Sarcophilus strikes me as odd compared with Carnivora, and that may warrant further thought.
 
Why is it that no bear, mustelid, canid, felid, raccoon, mongoose, civet, etc. etc. seems to have pale accentuation of the eyelids? Or have I overlooked some species that does?

(Never seen a white tail tip in a devil.Nick)
 
We’ve seen that white insignia on the dark chest are typical of bears and certain genera of mustelids.
 
It’s interesting that these kind of insignia seem quite absent from all of the melanistic morphs of wild felids.
 
When it comes to canids, there are few melanistic morphs in the wild. However, one species that does possess such a morph is Vulpes vulpes, at least in North America. This is the so-called ‘silver fox’, which as you know is taxonomically identical to the other, coexisting morphs of the same species but has a colouration similar to that of Sarcophilus.
 
Now, one of the clear differences between melanistic V. vulpes and Sarcophilus is that the former seems always to have a generous white tip to the blackish tail, whereas a white-tipped tail is rare in Sarcophilus and small even where it does occur.
 
The purpose of this photo-email is to show an interesting convergence between the melanistic morph of V. vulpes and Sarcophilus: in a small percentage of individuals of the melanistic morph of this species of fox, there is indeed an irregular white patch, an echo of what we’ve seen in the coexisting species Ursus americanus but even more intriguing because it is asymmetrical.
 
I get the impression that this white marking on the chest is associated with juveniles rather than adults, in melanistic V. vulpes.
 
The following is my tentative interpretation.
 
When felids go in for melanism, they tend to blacken the entire pelage, including parts normally conspicuous such as the tail tip (e.g. Panthera pardus) and back-of-ear (e.g. Leptailurus serval). I cannot recall ever seeing a melanistic wild felid with a white patch on its chest. My explanation is that felids are so specialised for crypsis and camouflage that white insignia on the chest would compromise their overall colouration too much; and because they are so solitary there is no considerable social imperative for mutual recognition of individuals by sight.
 
Bears show a quite different approach: they use overall dark colouration for inconspicuousness, but they retain bold pale insignia on the chest for self-advertisement because they do not stalk prey in the way felids do.
 
The various other families of Carnivora have various intermediates between these extremes. In the case of canids the whole phenomenon of melanism is minor, and the only dark forms are colour-morphs that function as much to confuse enemy taxa (i.e. to delay species-identification) as to hide the canids from their prey.
 
In the melanistic morph of V. vulpes, I find it significant that, quite unlike melanistic felids and of course also unlike dark bears (which lack noticeable tails) it is the tail tip that is the main glaringly pale ‘insignia’ – albeit not of much use for individual identification. This tail tip can be raised into view even while approaching a conspecific, making it visible at night. Foxes do not eat socially as Sarcophilus does, and so individual insignia matter little to them. So the occasional occurrence of white patches on the chest of the ‘silver fox’ is interesting in the sense that it represents convergent evolution among families and other higher-order taxa but a minor phenomenon in the scheme of things carnivoran. And I would still like to know whether in foxes these insignia are particularly associated with juveniles.
  
All the following photos show Vulpes vulpes.
 
http://www.wildernessshots.com/wp-content/uploads/2013/12/Red-and-Black-Fox-Ft-Collins-Colorado.jpg

 
https://s-media-cache-ak0.pinimg.com/originals/56/c5/d1/56c5d14ac39700f4b416ace41291d6d8.jpg

 
http://orig14.deviantart.net/5fbb/f/2012/011/f/b/black_fox_pup_in_the_wild_by_witchdoctortim-d4lzv5f.jpg

 
https://media.mnn.com/assets/images/2016/07/silver-fox.jpg

 
https://ih1.redbubble.net/image.5264791.8243/flat,1000x1000,075,f.jpg

 
https://photos.smugmug.com/Nature/North-America/Pacific-Northwest/i-CKVg9jd/6/S/RN8I4124ax1-S.jpg

 
http://hdblackwallpaper.com/wallpaper/2015/08/black-and-silver-fox-27-free-wallpaper.jpg

(writing in progress)

(writing in progress)
 
Although much has been written about the bite-force of Sarcophilus, and its role as bone-crunching scavenger, there is an obvious aspect to the dentition of this marsupial that seems to have been underplayed. And this is pertinent to our current comparison with Gulo.
 The overlooked/underplayed aspect is that there seems to be nothing specialised about this dentition as such, w.r.t. bone-crunching. This stands in contrast to the dentition of e.g. Crocuta, which certainly is specialised for bone-crunching.
 
Sarcophilus achieves its formidable status as a bone-cruncher in three ways, which I have not seen stated clearly before:
 - Large head and teeth relative to body size, which naturally boosts bite-force
 - Extremely wide gape, ensuring that bones can be placed between the posterior-most molars, thus using leverage in a similarly crude way
 - Short lifespan, which means that wear (breakage) of dental cusps matters relatively little because the dentition is not designed to last long.
 
At the risk of caricaturising Sarcophilus, let me put it this way:
 
Sarcophilus has remained unspecialised, because just as there has never been a niche for a large scavenger in Australia there has also never been a niche for a scavenger committed to a diet of bones. Essentially, Sarcophilus is just a scaled-up dasyurid with a particularly scaled-up head, but few qualitative modifications to the typical dasyurid design. By retaining the typical dasyurid pattern of short lifespan, it has got around the need to make its teeth durable in crushing bones. It instead just uses a crude tactic, biting hard on bones with teeth that are not particularly designed for such application, and accepting the inevitable damage to the teeth.
 
And by retaining a relatively small body and thus appetite, Sarcophilus is adapted to a continent with sparse and unreliable food owing to its nutrient-poor soils. A combination of small body (and thus limited food-demands) with large head (and thus versatility in taking large items as well as smaller ones) makes sense in the Australian context.
 
You will see from the photos below that Sarcophilus possesses neither the flat molars seen in Canis and Ursus, nor the conical premolars seen in Crocuta. Not only are its molars unremarkable, but they even retain the sharp cusps of a typically insectivorous mammal, scaled up to a size where they are far too big to be suited to insectivory but are instead used in a crude way on large items including hard ones. Sarcophilus does not even have a particular carnassial dentition along the lines of felids or Canis; instead its whole cheek-tooth row is carnassial but in a relatively imprecise and thus poorly-specialised way. It is consistent with this ‘crude’ approach to dentition that Sarcophilus does not bother to have a deciduous dentition while growing to maturity; it just crudely subjects its part-grown adult teeth to wear from the start, with little regard to keeping the teeth intact for as long as possible as most eutherians do.
 
In this way, Sarcophilus is functionally hyena-like ‘by default’, having adapted not its teeth as such but rather the crude parameters of the rest of the body and the pace of life.
 
By contrast, bone-crushers such as Crocuta have proportionately small heads and teeth, but use a far more ‘technically sophisticated’ solution to applying force, using ‘precision-tooling’ rather than brute force. In accordance with this, Crocuta has a great lifespan, the opposite of what is seen in Sarcophilus. And the simplicity of the approach of Sarcophilus, compared with the ‘artfulness’ of the approach of Crocuta, correlates with the great difference in relative brain size between these mammals.
  
The following several photos show how large the skull of Sarcophilus is relative to body size. The disproportionate size of the cheek-teeth is the main reason for its bite-force.
 
https://upload.wikimedia.org/wikipedia/commons/thumb/d/d4/Tasmanian_devil_skeleton.jpg/220px-Tasmanian_devil_skeleton.jpg

https://s-media-cache-ak0.pinimg.com/originals/e5/71/b8/e571b8e2f005a4dce30ca33dce84317e.jpg
 
https://www.wired.com/images_blogs/wiredscience/2009/05/3-tasmanian-devil-fossil_rm.jpg
 
https://upload.wikimedia.org/wikipedia/commons/5/55/Tasmanian_Devil_skeleton.jpg
 
http://www.nhc.ed.ac.uk/images/collections/mammals/marsupials/devilskel.jpg
 
The following four photos show that all of the cheek-teeth have cusps that are initially sharp in youth, before being broken by wear.
 
http://animaldiversity.org/collections/contributors/anatomical_images/family_pages/dasyuromorphia/dasyuridae/medium.jpg

https://upload.wikimedia.org/wikipedia/commons/thumb/a/a0/Animaldentition_sarcophilusharrisii.png/220px-Animaldentition_sarcophilusharrisii.png
 
https://i0.wp.com/upload.wikimedia.org/wikipedia/commons/6/64/Cambridge_Natural_History_Mammalia_Fig_076.png
 
https://c1.staticflickr.com/6/5086/5316470850_d387a102a0_b.jpg

The following many photos show the whole skull. The great size of the canines certainly does show impressive specialisation for carnivory, but has little bearing on bone-scavenging. Look in particular for any sign of ‘crowding’ of the lower incisors, which as we’ll see in part 2 is a special feature of Gulo.
 
https://boneclones.com/images/store-product/product-608-main-original-1454455009.jpg
 
http://www.skullsunlimited.com/userfiles/image/variants_large_4392.jpg
 
http://lrrpublic.cli.det.nsw.edu.au/lrrsecure/sites/web/gondwana/Animal_Fossils_of_Gondwana/lo/fossils_06/graphics/skull4_wh_sm.jpg
 
http://img01.deviantart.net/68da/i/2015/239/a/5/tasmanian_devil_skull_by_lizardman22-d94j8wd.jpg
 
http://tasmaniantimes.com/images/uploads/skull1mooney.jpg
 
https://www.scammellauctions.com.au/catalogue/shop_image/product/595/thumb-13ae4099fd5e4af8bdefee83e0a08c4f.jpg
 
https://s-media-cache-ak0.pinimg.com/originals/e5/71/b8/e571b8e2f005a4dce30ca33dce84317e.jpg
 
https://animalians.wikispaces.com/file/view/TasmanianDevil.jpg/235106478/TasmanianDevil.jpg

http://www.mammalogy.org/uploads/imagecache/library_image/library/1285.jpg

The only evidence of caching I have seen in dasyurids is captive spotted tailed quolls I briefly cared for hiding rats, once in a water dish. People who extensively manage them in captivity would  have much more experience. nick
  
One of the behavioural differences between Sarcophilus and Gulo is that the latter habitually stores food in caches, whereas such behaviour seems unrecorded in the marsupial.
 
It is easy to dismiss this difference along the lines that ‘because the wolverine lives in a refrigerator, obviously it is going to use the cold to preserve any surplus food’ under the snow. However, this does not stand up to scrutiny as a satisfactory explanation. This is partly because any piece of food is quite easy to preserve just by burying it. Contrary to popular assumption organic material does not quickly rot when buried; instead the mere act of burial is similar to refrigeration in retarding the process of decomposition, for the simple reason that the oxygen supply to bacteria and invertebrates is largely cut off.
 
Nor would it be satisfactory to claim that the main reason why Gulo is so avid in its caching of food is that it is the superior digger of the two. It is true that the fore foot of Gulo is far more powerful than that of Sarcophilus, with far longer and stouter claws. However, Gulo depends on the sharpness of its claws for climbing to escape its major enemies, which are the wolf and the brown bear. Gulo is quite unlike e.g. Mellivora because Gulo seems reluctant to dig in earth whereas Mellivora lives by digging. Gulo may look like a ‘badger’ but it is not a badger in the sense of being a habitual digger; instead if functions more like a giant marten.
 
In order to understand the difference between the two forms, I suggest that the most important thing to realise is that Sarcophilus is effectively the ‘top scavenger’ in its ecosystem, whereas Gulo is in competition with both the wolf and two species of bear.
 
Although everyone knows that the brown bear (Ursus arctos), like the wolf (Canis lupus), ranges as far north as Alaska and Siberia, what may be more easily overlooked is that even the black bear (U. americanus) also overlaps extensively with Gulo in natural distribution. Please see the maps below.
 
The black bear, which weighs about 5-fold the wolverine and is omnivorous with some ability to scavenge and even to break bones in its jaws, actually occurs across most of the distribution of Gulo in North America. Admittedly bears are not major competitors for carrion or bones because they hibernate in the cold season. However, they may help to explain why Gulo habitually breaks up carcases and scatters the pieces while at the same time burying them to some degree, a pattern of behaviour of which I’ve read nothing in the case of Sarcophilus.
 
In the case of Sarcophilus, there is no competitor for bones and there seems never to have been.
 
In addition to the principle of competition for bones there is a second, possibly even more important principle involved: that in guilds of carnivorous mammals there is fighting like cats and dogs on a level that has nothing proximately to do with food. All the larger Carnivora will gladly kill each other just for the hell of it. I.e. the wolf is keen to kill the wolverine simply for existing, and even where food is not part of the immediate equation; and indeed when the wolf kills the wolverine this is usually neither because it was competing directly for food nor because it is going to eat the dead wolverine. This is hatred on a systematic scale among guild-members in Carnivora and it shapes the behaviour of the smaller members of these guilds. For example, the main reason why smaller spp. of felids have camouflage colouration is probably their fear of their larger relatives, not any need to hide from their own prey.
 
With all of the above in mind:
 
Whereas the approach of Gulo seems to be to disperse and hide any surplus bones as rapidly as possible as a individual forager, the approach of Sarcophilus seems to be to call in as many conspecific individuals as possible to share the bones as rapidly as possible.
 
The result is that, while food-caching is a pronounced aspect of the behaviour of Gulo, it is sociable eating which is the pronounced aspect of the behaviour of Sarcophilus. The mindset of Sarcophilus seems to be: ‘I can’t possibly finish off all the bones of this ‘roo because I’m already satiated; so let’s give them to a friend or relative so that the favour will be returned at some time in the future.’ This kind of altruism would not work well in the habitat of Gulo because calling conspecifics to the carcase would attract the competing (and downright murderous towards Gulo) wolf plus, at least in spring, summer and autumn, two species of bears.

Original distribution of Ursus americanus:
https://www.bear.org/website/images/stories/images/images-new/historic_black_bear_range.jpg
 
Original distribution of Gulo gulo:
http://admin.scirecordbook.org/images/species/M65.jpg

I’ve now read Owen & Pemberton (2005) carefully on the topic of Sarcophilus. I’ve found as many questions as answers, and here I’ll list some of these in no particular order. Sarcophilus is worth characterising more precisely and comparatively than anyone seems to have attempted so far.
 
I get the impression that Sarcophilus is in line with the semelparous smaller dasyurids, but not semelparous mainly because it is too big. It still has that kind of life history strategy but there is no precise term for its particular strategy because it does not qualify categorically as semelparous. My point is that Sarcophilus cannot be understood without considering its short lifespan and its reliance on rapid replacement of generations rather than survivorship. The proneness to cancer is interesting in the context of this limited lifespan, and fits the notion that Sarcophilus evolved essentially free of predation and limited instead by the availability of resources.
 
The cheek vibrissae are a special feature of Sarcophilus, in great contrast to Gulo. I’ll investigate the details further.
 
Sarcophilus is both smaller and less sexually dimorphic than Gulo. As a result, the largest male Gulo is five-fold the body mass of the mean female Sarcophilus, i.e. about 30 kg vs about 6 kg. Even so, I note that the largest male Gulo resembles a bear not because it is massive as much as because it has such a proportionately small head and small eyes.
 
Sarcophilus has pale skin (and a small patch of rather pale fur) around its eyes, which I find significant but unremarked. Whereas Gulo hides its eyes by means of a dark mask, Sarcophilus  manages to advertise its eyes - despite the co-darkness of its face and its iris – to a degree not seen in related dasyurids.
 
If blushing of the ear pinnae occurs in social excitement, can this be seen by Sarcophilus, i.e. does it have colour vision for pink? If not, why have this physiological reaction?
 
The authors state repeatedly that Sarcophilus is easy to capture, but I didn’t get exactly how and why. There is a slight hint of ‘playing dead’ (as in ‘playing possum’) in Sarcophilus. Can a man likewise run down Gulo on flat ground?
 
If the tarsus has a bare posterior surface in Sarcophilus but not in canids, surely this is worth particular mention (see page 79).
 
Does Gulo ever use the fore feet to wash its face as Sarcophilus does? And does Sarcophilus uses its fore feet to pick up/handle food, particularly bare bones that it tries to crush?
 
Sarcophilus seems about equivalent to Gulo in digging abilities, which are modest in both and about equivalent to those of canids. In the light of this, the lack of caching of food by Sarcophilus seems significant.
 
If the mother in Sarcophilus really does flee with pouch-young dangling from the body and being bumped about on the ground or obstructions, shouldn’t this be stated clearly and explained? Does Sarcophilus ever carry its juveniles in its mouth, as Carnivora do? The mother in Sarcophilus carries your offspring around, after bearing them, for >15 weeks. How does the period combining gestation and the carrying of offspring in Sarcophilus compare to the gestation period in Gulo?
 
How do juveniles of Sarcophilus grasp the pelage of their mother? Do they use only the fore feet or the hind feet as well? Is any real dexterity involved? I get the impression that Sarcophilus uses digital grasp of the fore digits when climbing trees, rather than the traction of claws. If so, is the same true of its hind feet? Do juveniles ever ride on their mother in Gulo?
 
Sarcophilus seems to differ considerably from hyenas, all of which have long necks, in having ‘virtually no neck’. Why the difference?
 
The description of how the mouth and teeth are displayed by Sarcophilus when it reacts defensively is too vague. Is there any real fang-baring similar to that seen in the wolf or in certain felids?
 
It is interesting that Sarcophilus both swallows and defecates in large pieces. Is this anterior-posterior analogy a mere coincidence? I suspect that the large size of the faecal sausage is related to social marking; i.e. the animal ‘saves up’ its faeces so that it can deposit as much as possible in one go when returning to the communal latrine. Also, digestion does not seem as thorough in Sarcophilus as in hyenas. With hyenas, the faeces usually consist of paste (turning into powder) in which few if any objects can easily be discerned; with Sarcophilus, this book reports the incidence of whole recognisable paws and feet.
 
Is it true that Gulo is territorial despite its vast home ranges?
 
Underplayed in this book is that the locomotion of Sarcophilus seems specialised for flat surfaces and clear paths, and avoidance of rocks. Both contrast greatly with Gulo.
 
Sarcophilus presents an odd combination of inconspicuousness and noisiness. The spotted hyena, although spotted in keeping with camouflage, is not inconspicuous in this way, perhaps simply because it is so much bigger. But of which animal does Sarcophilus need to be so scared? I infer that the function of its blackness and general inconspicuousness is mainly vs its prey, i.e. for ‘hunting’. This function of colouration seems implausible in the case of hyenas as well as Gulo. 
 
Is the gregarious eating a way of dismembering carcases by tugging, much as happens with crocodilians? It seems hard to explain any other way, because if the main function is to share food to promote the species at the level of the population, then this could be achieved by serial vocalisation AFTER satiation. This puzzle seems to have been poorly thought-out by previous authors.
 
Why does nobody mention whether Sarcophilus trots or not? Can Sarcophilus sit like a dog, or not?
 
Why does Sarcophilus soil its nest/den?
 
There seems no doubt that Sarcophilus is far less brainy than comparable canids, and as lacking in encephalisation as the average marsupial. What a contrast with Mellivora, which is so brainy that it is the only member of the Carnivora capable of using tools.
 
I intend to photo-compared the dentitions of Sarcophilus and Dasyurus, and possibly also Antechinus.
(writing in progress)

(writing in progress)

Rates of metabolism and mortalities:
 
Obviously, Sarcophilus is a marsupial and Gulo is a eutherian. However, this needs interpretation to be meaningful.
 
What I don’t think has been pointed out before as such is that although it is Gulo that has the faster pace of life in the metabolic sense it is Sarcophilus that has the greater turnover in terms of populations. This may seem both obscure and somewhat contradictory, but I’ll begin to elaborate it here.
 
Most marsupials are inferior to comparable eutherians in body temperature and metabolic rate. Although I don’t have the figures to hand, I have no doubt that Sarcophilus and Gulo conform to this trend. I suspect that the body temperature of Gulo is about 38.5 degrees C whereas that of Sarcophilus is about 36 degrees C. The comparison of metabolic rates is complicated by the fact that Sarcophilus has a smaller body than Gulo, and by the fact that the environment of Gulo is far colder, on average, than that of Sarcophilus. However, I have little doubt that, in general, a given 100 g of the body cells of the eutherian convert chemical energy to ‘heat’ more rapidly in Gulo than in Sarcophilus. And this is perhaps the most basic aspect of ‘pace of life’.
 
The great difference in brain size between the two forms is consistent with the difference in metabolic rates, just as the fact that kangaroos have smaller brains than those of like-size ruminants is consistent with the unquestionably lesser rate of metabolism of the ‘roos in that comparison. Marsupial mammals generally tend to run relatively cool and to require relatively little energy; and although there are various exceptions to this trend (e.g. the koala runs hotter than sloths) I see no reason why Sarcophilus would be an exception.
 
I also get the impression, by the way, that Sarcophilus digests its food less thoroughly than does Gulo.
 
Given this ‘running cooler’ on the part of Sarcophilus, it would be easy to expect the marsupial to be the longer-lived of the two forms. However, the opposite is the case and this is where the comparison gets a bit paradoxical.
 
The truth is that Sarcophilus is much shorter-lived than Gulo, in terms of both life expectancy and longevity. In that sense it is the marsupial that has the greater ‘pace of life’, i.e. its populations turn over more rapidly than those of the eutherian in the comparison.
 
Sarcophilus, although about the same body size as Carnivora such as Felis catus that can live to 15 years, is so short-lived that by 7.5 years it has naturally senesced even if it has experienced no injury or mishap. It just ‘burns out’ by an age half that expected for the senescent ‘burnout’ of comparable eutherians. And in keeping with this ‘hard-wired’ pattern of frequent mortality, the natality of Sarcophilus is correpondingly greater, in various ways, than that of Gulo.
 
For example, whereas each adult female of Gulo tends to reproduce only every other year, that of Sarcophilus can reproduce every year. Whereas each litter of Gulo never consists of >4 neonates, that of Sarcophilus can consist of as many as 50. And whereas Gulo seldom weans more than two juveniles, Sarcophilus routinely weans four. Although I have no reason to believe that the actual rate of somatic growth of infants and juveniles is any greater in Sarcophilus than in Gulo, the rate at which individuals turn over in natality/mortality is clearly greater in Sarcophilus than in Gulo. For a start, by far the majority of every litter is doomed in Sarcophilus, because the mother bears >20 neonates but only has four teats and each neonate can survive only if it manages to attach permanently to a teat. Then, because the pouch is so small that the suckling infants hang haphazardly from the groin of their mother, they tend to get ‘beaten up’ during the normal locomotion of their mother. Then each female Sarcophilus can wean several juveniles per year but given that all the habitat tends already to be occupied it is the fate of most of the newly-independent juveniles to die before finding a home for themselves. And finally: even if everything goes well and an individual finds and home and reproduces successfully, the best it can expect is to die ‘of old age’ by about 6 years old, after only about four years of reproductive attempts.
 
Although it is simplistic to put it this way, Sarcophilus is ‘r-selected’ whereas Gulo is ‘K-selected’. This is because Gulo lives relatively long, reproduces relatively infrequently, bears a relatively small number of neonates at each birth, and leaves the infants securely in a carefully hidden den during the suckling period; the mother makes no attempt to carry the infants or juveniles around on her body while foraging, which inevitably exposes them to risk. And even after weaning, there is a long period in which the adolescent Gulo remains with or near her mother, learning from its parents.
 
Do you see that few of these intriguing differences are explained, per se, by the fact that Sarcophilus is a marsupial while Gulo is a eutherian?

(writing in progress)
 

(writing in progress)

Here’s an aspect that seems to have been quite overlooked by all previous writers on the topic of the Tasmanian devil: the relationship between scavengers and large, bone-gnawing rodents.
 
My main point is that it is impossible to understand the role of Sarcophilus as a bone-eater without the context that Australia is unique among large land-masses in lacking any large, fully terrestrial rodent that competes for the consumption of bones. Put another way: Australia lacks not only hyena-like eutherians but also bone-gnawing rodents of any considerable size or abundance, and this is why Sarcophilus, although of modest size and with only modest ability to break bones in its jaws, was the ‘top scavenger’ on the vast landmass comprising Australia, New Guinea and Tasmania.
 
In our comparisons with Gulo, it is easy to invoke the wolf and two species of bears as competitors with the wolverine for carcases and bones. However, it takes some lateral thinking to notice another competitor in the same environment: the North American porcupine Erethizon dorsatum.
 
Consider for a moment that this porcupine, although so easily overlooked in the boreal guild of bone-eating scavengers, is itself far more massive than Sarcophilus. The mean body mass for E. dorsatum is about 9 kg, compared with only about 6.5 kg for Sarcophilus. And, incidentally, it is far longer-lived as well, with a maximum lifespan of up to 30 years instead of 7.5 years.
 
Unlike bears, E. dorsatum does not hibernate, and so it potentially continues to consume bones through winter.
 
Now, one of the remarkable features of Australia – and one which is anomalous relative to the general theme that islands favour large rodents in place of other lineages of mammals – is that there are no large rodents indigenous to this continent. The largest non-aquatic species is Mesembriomys gouldii (about 0.6-0.65 kg) but this is restricted to tropical woodlands no farther south than the Gulf of Carpentaria. Hydromys chrysogaster can exceed a body mass of 1 kg and lives to this day in Tasmania, but is tied to water. The largest rodent in southern Australia is Leporillus conditor (about 0.35 kg), which reached no farther southeast than the northwestern tip of Victoria and was absent from Tasmania. The largest terrestrial rodent indigenous to Tasmania are Mastacomys fuscus and Rattus lutreolus (both about 120 g) and even these are restricted to the sort of dense vegetation which Sarcophilus – with its dependence on paths – tends to avoid. An indigenous rodent of about 0.2 kg, namely Conilurus albipes, occurred on the southeastern Australian mainland, but it did not occur in Tasmania and is now extinct.
 
Recent studies have shown how squirrels consume bones: https://www.bu.edu/research/articles/squirrel-bone-gnawing/ .
 
The following article shows that Erethizon dorsatum is frequently attacked by the wolverine: http://journals.sfu.ca/cfn/index.php/cfn/article/view/1727 .
 
So one cannot really understand the biology of Gulo fully without mentioning the largest rodents in its habitat, on which it preys and which compete with it for bones. What this means is that the main competitors for bones with Gulo are the wolf, two species of bears (warm season only) and the North American porcupine (possibly more in winter than in summer).
 
We can thus characterise Australia, in terms of bone-eating scavengers, in at least two different ways. Firstly, it is the only landmass – other than perhaps Madagascar where I suspect that extinct hippos filled this niche – on which the largest bone-eating scavenger weighed only <10 kg. Secondly, it is the only landmass on which the largest fully terrestrial, incidentally bone-eating rodent weighed <1 kg, with most of the landmass (i.e. the whole extratropical zone) having no such rodent >0.5 kg, and with the largest living rodent in the remaining habitat of Sarcophilus being only <150 g.

(writing in progress)

maternal behaviour of Sarcophilus.
 
Kangaroos have a pattern of ‘maternal jettisoning’ in which a mother, stressed by pursuit by a predator, will jettison her pouch-young. This in itself may be understandable even to a species with a placental mindset, such as the human species. But what is incomprehensible to most persons who know about this tactic is that, when reunited with her pouch-young once the danger has passed, the mother often rejects it, thus condemning it to death. As you know, wildlife carers are universally familiar with this problem in kangaroos and their relatives.
 
Similar, albeit less easily described, maternal ‘lapses’ are known in bandicoots, and I suspect that this is a general pattern among marsupials. From the placental point of view it seems to show a weak maternal instinct, but on reflection it makes sense where the offspring are so small that there is a premium on preservation of the life of the mother, if necessary at the expense of ‘cheap’ offspring in which a limited amount has been invested, and in the context of a continent with limited pressure from predators. Indeed, at an even broader perspective, it makes sense that mammals that ‘gestate’ their offspring in a pouch rather than in a womb should retain the option to sacrifice their offspring; and because marsupials are generally less brainy than placentals it makes sense that mothers should tend to lack the cognitive versatility to ‘realise’ that there is sometimes an unnecessary loss when their hard-wiring causes them to ‘fail’ as mothers once reunited with jettisoned, but fortuitously surviving, offspring.
 
Now, I don’t know much about such behaviour in dasyurids, but I did indeed read, in a newspaper article published on 28 Aug. 200 about the breeding program for Sarcophilus in Monarto Zoo, that “Veterinarian Ian Smith said the joeys were given their first health checks only yesterday because adult females often turned on or deserted their young if they were separated. ‘During the breeding season, we try to maintain a hands-off approach. We’re more concerned that they might leave them and less so that they might eat them,’ Dr Smith said.”
 
Do you see the implication here, i.e. that Sarcophilus is similar to ‘roos and bandicoots in a tendency to lapse in the maternal instinct after temporary separation from pouch-young associated with stress imposed by a potentially predatory species (in this case human)?
 
I have not read anything specific about maternal behaviour in Gulo, and perhaps there are few observations of this anywhere in the world. However, it seems fair to assume that the maternal bond is stronger in this eutherian than in some marsupials, not so? I.e. I suspect that, if suckling, denned offspring were to be lost to a mother and then fortuitously reunited with her unharmed, she would accept them instantly.
 
So my question to you is this: do you think that this is indeed a true difference, i.e. a kind of evolutionary non-convergence or divergence, between Sarcophilus and Gulo? Or on the other hand do you think that Sarcophilus differs in this respect from ‘roos and bandicoots, and that its maternal behaviour is more similar to the eutherian behaviour than is true in some other marsupials? A corollary question is: have you ever heard of any evidence that a mother Sarcophilus carrying relatively large pouch-young, when chased and stressed by a potential predator such as the domestic dog, actually jettisons her pouch-young?