A Plethora of Fossil Possums

View Images

In many technical papers describing fossil creatures, artists give readers a glimpse of prehistory through reconstructions of what those animals might have looked like in life. Not so with Pucadelphys andinus. In a paper published this week in Nature, artist S. Fernandez depicted a pair of the possum-like creatures at a critical point in their individual histories – just after death but before their soft tissues were stripped away to leave behind only fossilized bones.

The reason why Fernandez restored the small, fuzzy mammals right up to the point of death has everything to do with the way the creatures died. For the most part, the fossil record of small mammals consists of teeth and bits of jaw. Complete skeletons are exceptionally rare, and the few scraps of tiny mammal bones are often found in isolation. But Pucadelphys andinus, a marsupial that inhabited prehistoric Bolivia not long after the end-Cretaceous mass extinction 65 million years ago, was different. Thirty five individuals of this extinct mammal of different age groups were all found in close proximity, paleontologist Sandrine Ladevèze and colleagues report, and this collection of tiny bones may provide the oldest direct evidence of mammal social behavior yet known.

Tiupampa, Bolivia was very different during the time of Pucadelphys. Though the location was at roughly the same latitude then as it is now, Ladevèze and co-authors note, the sites at Tiupampa were actually lower in altitude and represent a lush tropical rainforest akin to the modern day Amazon. Dead animals are often rapidly recycled in these environments – bodies are often destroyed by insects, fungi, and bacteria before there is a chance of them being buried. How, then, did so many well-preserved Pucadelphys fossils wind up in so small an area?

Instead of being busted apart into a mass of disarticulated bones and skeletal fragments, the Pucadelphys individuals held together and were mixed among each other in two small sites about ten feet apart from each other. The mammals appear to have been rapidly killed and buried in a single event – perhaps when a local pond or lake flooded. (The discovery of an intact crocodile nest and the semi-complete, articulated skeletons of other small animals nearby reinforces the idea that the small-scale catastrophe rapidly buried animals in place.) Exactly what this aggregation means for the biology of these animals is another matter.

View Images
The skulls and jaws of inferred female (left) and male (right) Pucadelphys andinus as seen from the top, bottom, and the side. From Ladevèze et al., 2011

The hypothesis of Ladevèze and colleagues is that the two Pucadelphys sites represent a social group consisting of adult males, adult females, and sub-adults. They arrived at this conclusion by reviewing possible explanations for why these animals might be found together, especially since similar, modern-day marsupials such as possums and dasyurids are often solitary and territorial. While it is conceivable that the Pucadelphys came together to breed, the presence of immature animals conflicts with this hypothesis and no living possum species is known to come together in such numbers to mate. Nor does it seems likely that these animals were nesting together to share body heat and stay warm – the tropical habitat would have been plenty hot. The rapid burial of a large number of animals of different ages would seem to indicate that these animals were living in some sort social group when they died.

Frustratingly, this hypothesis is difficult to confirm. In their contribution to the book on assemblages of fossil animals – simply called Bonebeds – paleontologists Donald Brinkman, David Eberth, and Philip Currie wrote “[E]very bonebed provides an opportunity to glimpse some aspect of paleobiology.” Exactly what glimpses those aggregations offer depends on how the bonebed came together. A bonebed might represent a single catastrophic event or might have accumulated over time, and just because a group of animals is found in close association does not necessarily mean that they were a cohesive social group when they died. “[R]esearchers should also be critical of using the presence of the death assemblage itself as the primary evidence for normal or day-to-day aggregation paleobehavior,” Brinkman and colleagues cautioned, adding “Paleoenvironmental stresses such as floods, fire, drought, volcanic eruptions, and even disease often result in temporary and atypical aggregations and mass-death events among modern animals, and the same is likely to have occurred in the past.”

Mating season and nesting behavior can be ruled out as reasons for the Pucadelphys assemblage based upon inferences taken from living marsupials, but testing for environmental stresses that may have brought the animals together is difficult to accomplish. The thirty five animals were in close association when they perished – the way the fossil site was created testifies to that – but might their be alternative explanations for their closeness?  If flooding killed and buried these animals, for example, perhaps raised water levels swamped the local Pucadelphys burrows and forced them into one of the few remaining dry patches shortly before their deaths. Even if this alternative is false, the fact remains that it is extremely difficult to distinguish whether a bonebed indicates consistent social behavior or is the result of a single, unusual event. Discoveries of additional Pucadelphys sites elsewhere containing multiple individuals would throw support to the idea that these marsupials were gregarious, but, given the nature of fossil preservation, such localities may be difficult to find. Bonebeds, as spectacular are they can be, are often some of the most puzzling fossil mysteries.

Top image: A restoration of two dead Pucadelphys individuals based upon two partially-complete, articulated skeletons. By S. Fernandez and from Ladevèze et al., 2011.


Brinkman, D.; Eberth, D.; Currie, P. 2007. From Bonebeds to Paleobiology: Applications of Bonebed Data, in Rogers, R.; Eberth, D.; and Fiorillo eds., Bonebeds. Chicago: University of Chicago Press. pp. 221-263

Ladeveze, S.; de Muizon, C.; Beck, R.; Germain, D.; Cespedes-Paz, R. (2011). Earliest evidence of mammalian social behaviour in the basal Tertiary of Bolivia Nature, 474, 83-86 DOI: 10.1038/nature09987