Bears and Bamboo: The fossil record of giant pandas

Where do giant pandas come from? Of course, the proximal answer involves a male and female panda – and maybe some panda porn, if life in captivity dampens the mood – but I’m not talking about that. What I’m wondering about is the evolutionary origin of these bamboo-eating bears.

Until recently, there was little to be said about the prehistory of pandas. A few skulls, mandibles, and other assorted fragments from caves and fissures in southwestern Asia were all that had turned up. Prior to the origin of the modern panda, the larger species Ailuropoda baconi lived during the past 750,000 years, and was preceded by the poorly-known Ailuropoda wulingshanensis and a smaller species – Ailuropoda microta – which occupied China between 2 and 2.4 million years ago. Beyond that it gets a bit hazy. The earliest potential member of the giant panda lineage is the approximately seven million year old bear Ailurarctos, but there are not any solid points between it and the later pandas to draw together.

Notices of most of these fossil finds were tucked away in obscure journals or were only briefly mentioned in catalogs of specimens recovered during American Museum of Natural History expeditions. From the known parts – especially the teeth – the fossil bears did not seem all that different from the modern pandas. Thanks to a single discovery, though, paleontologists have begun to piece together a better understanding of how these bears changed over time.

The fossil that has spurred several new studies into the origins of pandas is the skull of the smallest and earliest giant panda species, Ailuropoda microta. Found in southwestern China’s Jinyin cave, this worn skull is considerably different from those of later species and looks rather puny compared next to them. Nevertheless, the 2007 description of the skull by Changzhu Jin and colleagues points out that this animal shared some tell-tale characteristics associated with the modern panda’s diet of coarse, fibrous bamboo. The cheek teeth of A. microta, though lacking the extra cusps seen in living pandas, were broad and well-suited to grinding, and the back of the skull was expanded for heavy chewing muscles. Overall, its skull was not as heavily-built as that of the largest fossil panda, A. baconi, but it appeared that at least some unique giant panda traits were already present about two million years ago and had just been tweaked a bit since then.

Exactly how these species relate to each other is unclear. The authors of the 2007 description interpreted them in a straight-line march from A. microta to A. wulingshanensis and onto A. baconi before a size reduction culminating in the modern A. melanoleuca. (A follow-up paper by Wei Dong on CT scans of the brain cavities of these bears showed that a reduction in brain size went along with the reduction in body size.) Given that we still know so little about these bears, however, an evolutionary march of the pandas cannot be confirmed, and better sampling will be needed to tell whether all these fossil species represent a lineage as straight as a bamboo stem or whether there were splits which led species to overlap in time with each other. There is much that remains unknown about the diversity of prehistoric pandas and their precise placement in time.

Even if the recent history of prehistoric pandas remains a bit fuzzy, the discovery of the A. microta skull has allowed paleontologists to identify some of the evolutionary trends that shaped this peculiar group of bears. In 2010 Borja Figueirido and co-authors looked at how many times the group of mammals which contains dogs, cats, and bears – called carnivorans – has evolved similar adaptations in their skulls to eating plants. Their hypothesis was that a combination of shared evolutionary constraints and similar pressures from natural selection determined the unique skull shapes of carnivorans that went vegetarian.

One of the prime examples of this kind of convergence comes from the two distantly-related modern pandas. There’s the panda bear, and then there’s the red panda (Ailurus fulgens), which last shared a common ancestor with the giant panda over 40 million years ago. Despite this distance, however, the red panda also feeds on bamboo, has enlarged molars for grinding tough food, and even has a specialized wrist bone (the sesamoid) that creates a jury-rigged, opposable “thumb“. These shared traits may have appeared independently as adaptations to a similar diet, although, as explained in a 2006 study led by Manuel Salesa, the fact that the fossil red panda Simocyon had a pseudo-thumb but lacked plant-crushing teeth suggests that red panda thumbs were initially adaptations to life in the trees and were only later co-opted for eating bamboo. The pattern of convergent evolution cannot be understood without knowing the evolutionary history of the groups being compared and what traits may have undergone a change in function thanks to natural selection.

But Figueirido and co-authors were not considering whole bodies. They focused their study on similarities of the skull. What they found was that specialized, plant-eating carnivorans – or species that get 95 percent of their intake from plant food – have broad, short skulls with deep jaws and stout molars. This package of traits generates high bite forces, and the only carnivorans with stronger bites are the hypercarnivorous species which specialize in taking down large prey. The reason for this may be that, compared to ungulates like antelope or deer, plant-eating carnivorans are not well-suited to eating plants. They lack the complex digestive systems of the hoofed mammals for breaking down plants, and the construction of their jaws prevents them from chewing as efficiently. In order to survive, they have to eat heaping helpings of plant food to make up for their general lack of efficiency, and so they were adapted to have very strong jaw muscles to keep working through all that browse. The evolutionary baggage the herbivorous carnivorans carried with them constrained what was possible, and the giant panda is the most famous example of this.

Just when giant pandas began to shift to an all-bamboo diet is another matter. On the basis of teeth alone it seemed that bamboo-eating was a long-held giant panda tradition, going back millions of years, but the discovery of the approximately two-million-year-old skull of A. microta has allowed paleontologists to get a better handle on the timing of the associated changes in anatomy. In a study just published in Naturwissenschaften by Figueirido, Paul Palmqvist, Juan Pérez-Claros, and Wei Dong, landmarks on the skulls of the known giant panda species were used to track changes during the group’s evolutionary history. The aim of this research was to determine whether giant pandas really have undergone minimal modification since the late Pliocene or whether the unique traits seen in their skulls developed more recently.

The results of the analysis showed that A. microta had a skull very much like that of the modern panda in profile, but it differed in some subtle ways. Its molar tooth row was shorter than in living giant pandas, its snout was comparatively longer, and its braincase was narrower, in addition to a handful of other differences. When looked at all together, the skull of A. microta was most similar to that of other giant pandas but was still intermediate between that of the panda bears and other species of living bears. Contrary to what was reported in the initial description of the skull, giant panda head shape did not remain static for the past two million years.

While it is difficult to be sure without the lower jaws and other parts of the skeleton, the skull anatomy of A. microta probably indicates that giant pandas were already bamboo specialists by two million years ago. Minor differences in their anatomy hints that they were not able to eat as much bamboo as their living relatives – their jaw forces were weaker, and they lacked an expanded second molar to grind down on bamboo stems – but their skull shapes are consistent with a diet of tough plants.

Frustratingly, though, paleontologists have only an extremely limited view of giant panda evolution. Of three potential fossil species, only two are known from relatively complete skulls, and the fossil teeth of Ailurarctos appear to indicate that the fossil lineage of giant pandas goes back seven million years or more. That leaves us with a five million year gap in panda evolution, and even the history of the more recent pandas is only partially known. In order to fill in those gaps, paleontologists will have to go back to the caves and fissures of Asia to uncover new clues.

Top Image: Tai Shan the panda cub while at the National Zoo in the spring of 2008. Photo by the author.

References:

Dong, W. (2008). Virtual cranial endocast of the oldest giant panda (Ailuropoda microta) reveals great similarity to that of its extant relative Naturwissenschaften, 95 (11), 1079-1083 DOI: 10.1007/s00114-008-0419-3

Figueirido, B., Palmqvist, P., Pérez-Claros, J., & Dong, W. (2010). Cranial shape transformation in the evolution of the giant panda (Ailuropoda melanoleuca) Naturwissenschaften, 98 (2), 107-116 DOI: 10.1007/s00114-010-0748-x

FIGUEIRIDO, B., SERRANO-ALARCÓN, F., SLATER, G., & PALMQVIST, P. (2010). Shape at the cross-roads: homoplasy and history in the evolution of the carnivoran skull towards herbivory Journal of Evolutionary Biology, 23 (12), 2579-2594 DOI: 10.1111/j.1420-9101.2010.02117.x

Jin, C., Ciochon, R., Dong, W., Hunt, R., Liu, J., Jaeger, M., & Zhu, Q. (2007). The first skull of the earliest giant panda Proceedings of the National Academy of Sciences, 104 (26), 10932-10937 DOI: 10.1073/pnas.0704198104

Salesa, M. (2006). Evidence of a false thumb in a fossil carnivore clarifies the evolution of pandas Proceedings of the National Academy of Sciences, 103 (2), 379-382 DOI: 10.1073/pnas.0504899102