An early date for a Diprotodon (Marsupialia; Diprotodontidae) from the Papua New Guinea highlands

A project undertaken at the University of Papua New Guinea, and supervised by J. Menzies

The vertebrate fossil record of New Guinea is still poorly known even though the first discoveries were made more than 60 years ago (Anderson, 1937).  Since that time a trickle of discoveries, all believed to be of Pleistocene age, have been made but, even so, the number of known sites that have yielded vertebrate fossils is less than one dozen.  Flannery (1994) reviewed all the mammal fossils and fossil sites known in New Guinea while Menzies & Ballard (1994) gave a detailed account of new megafauna discoveries made since the time of Flannery’s review.  The total number of extinct mammal species now recovered includes nine from two Pliocene sites and eight from Pleistocene sites, though none has been accurately dated.  Even the taxonomic allocation of the New Guinea diprotodontids is unclear and Flannery (1994) tentatively assigned all known fossils to the subfamily Zygomaturinae.  This necessitated removal of Nototherium watutense of Anderson (1937) from the Nototherinae but left its generic disposition in doubt.  Given the overall similarity of Nototherium watutense, Zygomaturus nimborensia and the New Guinea species of Kolopsis and Kolopsoides Flannery synonymised the Sentani Zygomaturus nimborensia with Nototherium watutense under the name ‘Kolopsis’ watutense.  All specimens except the Sentani ‘Kolopsis’ have been recovered from highland sites and nearly all the diprotodontid fossils so far discovered have been recovered from river or lake deposits and cannot be accurately dated from the recent sediments that they lie in.

The recent discovery of a fossil jawbone in the Yonki area of the Eastern Highlands of Papua New Guinea was particularly welcome because it offered the possibility of potassium-argon dating although the strata of the site itself are recent.  Unfortunately, the specimen is in poor condition as both anterior and posterior ends are damaged, the natural bony surface is eroded and few molar teeth remain.  The interesting aspect is that the jawbone is embedded in a nodule of tuffaceous groundmass, although it lay in recent (<60,000 year) lake sediments that generally lack tuff.  The intimate association of fine volcanic tuff with the fossil mandible, both as a coating on the outside and as an infilling of cavities in the bone, strongly suggest entrapment of the marsupial in a fall of volcanic debris. The recrystallisation and loss of primary texture in some of the bone could be a result of heating.  Subsequent to this event the fossil was redeposited in Quaternary lake sediments.

Initial analyses using the feldspar component of the tuff gives a date of some 12.1 (± 0.6) Ma (Miocene). If this is confirmed, the fossil history of the marsupials in New Guinea is pushed back a considerable period.  Attention will now be turned to the hornblende component of the groundmass in an attempt to confirm, or add precision to, this date.


Anderson, C. (1937).  Palaeontological notes no. 4; fossil marsupials from New Guinea.  Records of the Australian Museum 20, 73-76.

Flannery, T.F. (1992).  New Pleistocene marsupials (Macropodidae, Diprotodontidae) from subalpine habitats in Irian Jaya.  Alcherringa 16, 321-331.

Flannery, T.F. (1994).  The fossil land mammal record of New Guinea: A review.  Science in New Guinea 20, 39-48.

Menzies, J.I. & Ballard, C. (1994).  Some new records of Pleistocene megafauna from New Guinea.  Science in New Guinea 20, 113-139.