A reconstruction of Thylacoleo carnifex (based upon a drawing by Peter Schouten).  Thylacoleo is thought to have been primarily an arboreal (tree dwelling) animal which possessed a strong grip.  Note the powerful hand with robust digits.  Another feature of the hand was the presence of a large, hooded claw on the thumb which was apparently used as a weapon.

Few extinct mammals have awakened as much curiosity and speculation as to their way of life and their relationship to other forms as the remarkable Marsupial 'Lion' Thylacoleo, described in 1859 by the distinguished British palaeontologist Sir Richard Owen as 'the fellest and most destructive of predatory beasts'.     This explanation was soon disputed in 1871 when Owen wrote in response: 'several eminent and experienced investigators of Fossil Remains have endeavored to determine to which of the groups specified the Thylacoleo was closely or immediately allied.  Some have been led to the belief of it having been a Kangaroo, some have deemed it a Potoroo or Rat-Kangaroo, others would rank it with the arboreal Phalangers or Koalas; but all concur in repudiating its carnivorous character....and have sought with more or less ingenuity, to invalidate the conclusions which I had been led to deduce from the parts of the fossilized remains....'

        Hence, Owen began the defence of his theory; with the passing of 140 years and more than 40 scientific papers, the general view of Thylacoleo now agrees with Owen's original interpretation.  To understand why the argument ensued we must investigate the teeth of this amazing mammal.  The prominent front two teeth, the incisors, are characteristic of marsupials of the order Diprotodonta, a group which includes the possums, koalas, wombats and kangaroos.  Most of these animals have an essentially herbivorous diet.  Thylacoleo's cheek-teeth are dominated by enormous, blade-like third premolars, which work in a way similar to the flesh shearing carnassial teeth of mammals of the placental order Carnivora.  The molar teeth are quite small and reduced to only one upper and two lowers on each side.  The canines are reduced to mere pegs, and apparently serve little function.  Owen (1871) asserted that in the living carnivores the relative enlargement of the carnassial teeth was a good indicator of the ferocity of the species.  On that basis, Thylacoleo must have been an animal of unprecedented ferocity!  Moreover, the great reduction in size of the canine teeth to mere stubs in Thylacoleo was more than compensated for by the extreme enlargement of the first incisors, which were 'adapted for piercing, holding and lacerating, like the canine of a carnivore' (Owen 1871).  Truly, as Broom (1898) noted, much of the opposition to Owen's interpretation was based on the presumption that 'diprotodonts can't be carnivors', because all others were herbivorous.  This belief in turn had led to notions of more fancied diets such as ripe fruit or eggs (Cope, 1882, 1884).  However, Broom was not successful in putting the matter to rest, as in 1929 Anderson drew attention to a paper by Charles De Vis (1883) of the Queensland Museum which claimed that Thylacoleo was a marsupial 'hyena' rather than a 'lion'.  Anderson pointed out that hyenas crushed and ate bone and that such animals 'have broad strong teeth with several cusps, and a well-developed cingulum (a low shelf around the base of the tooth) on the molars and premolars, the purpose of which is to protect the gums from injury by splinters of bone.  The teeth of Thylacoleo are not of this type; nor is a cingulum to be found on them'.
 

After demonstrating that Thylacoleo was not a 'hyena' Anderson proceeded to take an illogical step; dismissing the carnivore theory completely he returned to the presumption of a specialized plant-eater which fed on native curcubits or cycad nuts, even though he was aware that 'there are difficulties in accepting the view that Thylacoleo could obtain adequate food from these sources'.     Additional problems with the carnivore hypothesis arose from an examination of the large incisor teeth set together at the front of the jaw, forming a structure similar to the beak of a parrot (Gregory, 1951).  In this position they may have been capable of piercing and lacerating but 'a living animal caught in such incisors could spin about as if on a pivot' (Gill, 1954a) and would be 'particularly unsuitable for catching and holding anything alive and struggling' (Marett Tims and Hopewell Smith, 1914).

The skull of Thylacoleo.  This one was recovered from a Pleistocene cave deposit.  Fossils such as this are quite fragile, and should be treated with a protective sealant such as polyvinyl butyral to guard them from the deteriorating effects of humidity and temperature extremes.  Note the large, projecting incisors in both the upper and lower jaw.

    Over the passage of time since Owen's description many skull and jaw fragments were discovered in cave and swamp deposits throughout Australia, but it was not until 1956 that any firm image of what the animal's postcranial skeletal anatomy was like became possible, when, during quarrying at Naracoorte in South Australia, an associate skeleton was found.  During the following five years additional partial skeletons were unearthed in South Australian caves (Daily, 1960) and also in the sand dunes of Lake Menindee (Tedford, 1967).  In 1966, a discovery of immense significance was made at Moree in New South Wales, where workers at a loam pit uncovered a nearly complete Thylacoleo skeleton which was missing only its hind feet and the tip of the tail.  When the matrix was cleared away from the skeleton, it was found to contain the fragmentary remains of a pouch young (Finch, 1971).  In 1969 the discovery of the fossil chamber in Victoria Cave, Naracoorte, lead to the recovery of the remains of an additional 14 individuals, including well-preserved and articulated fore and hind feet.  (Wells, 1974;  Wells and Nichol, 1977).  A clearer image of this unique marsupial is now forming.  Finch (1971) has portrayed Thylacoleo as 'an animal the size of a leopard with comparatively long limbs for its size and a large, heavy head supported on a thick muscular neck... The paws were strong, heavily clawed, and probably used in striking prey and for tearing it'.
 

A composite Thylacoleo skeleton (cast) assembled from bones found in Victoria Fossil Cave, Naracoorte, South Australia.

Wells' and Nichol's (1977) study of the hands of Thylacoleo showed that the thumb is robust, capable of moving quite independently of the other digits and possesses a very formidable, hooded claw.  The remaining fingers are long and slender, bear small claws and only limited palmar flexion.  When standing, the weight of the animal would have been borne by the ends of the toes in a similar way as that seen in cats.  In spite of this, Thylacoleo's hand would have had a powerful grip through opposition of the thumb, not to the other digits as in primates, but to a flattened bone (the pisiform) in the wrist.

    The structure of the hind foot implies that the weight of the animal was distributed along the side of the foot (a plantigrade stance).  Indeed, the hind foot is very similar in design to that of the Brush-tailed possum, even to syndactyly of digits II and III (i.e. the two toes are positioned very close to each other, enclosed by a sheath of skin), evidence if its diprotodont ancestry.  Wells and Nichol (1977) were unable to note few significant anatomical differences in the hands of terrestrial and arboreal carnivores, and proposed that a hand well-adapted for climbing could be adapted equally well for holding prey.

    Apart from the data now available regarding the skeleton of Thylacoleo, examinations of recently found skulls and teeth have enabled further insights in feeding habits, settling the old controversies that once raged about this marsupial.