Welcome to the blog of the Mesozoic vertebrates research group of the Bayerische Staatssammlung für Paläontologie und Geologie

Thursday 26 July 2012

Coniophis, the origin of snakes, and long ghost lineages

Although snakes might not be the favourite critters of most people, these animals are certainly among the most amazing land-living vertebrates today. Legless, with greatly elongate bodies, and astonishing feeding adaptions, such as venomous fangs and hyperextensible jaws, snakes are also among the most successfull vertebrate groups, with almost 3000 species known today. However, the fossil record of snakes is still rather poor, and this is especially true for their origin. Thus, the evolution of the remarkable anatomical adaptations of snakes, and their functional and ecological context remain enigmatic.

Skull reconstruction of Coniophis and a modern snake (modified from Longrich et al. 2012, and Pough et al. 2004)
Nick Longrich and colleagues have now published new material of the latest Cretaceous snake Coniophis from North America that has some bearing on these questions. Although the taxon has been known for 120 years, only isolated vertebrae had been described so far. Longrich and colleages now described skull remains and additional vertebrae that throw new light on the anatomy and ecology of this early snake. The new remains show that the skull of Coniophis had the typical hook-shaped teeth and intramandibular joint of snakes, which help to expand the gape to swallow larger prey. However, the upper jaw was firmly attached to the other skull bones, much as in lizards, and thus not allowed the extensic movements between differetn skull bones that we see in modern snakes. Thus, Coniophis presents an interesting mosaic of characters that helps understanding how the feeding apparatus of snakes evolved.
However, one other aspect of the new paper I find somewhat more problematic. There is a long-standing debate whether snakes evolved from aquatic or land-living ancestors, and, if the latter was the case, whether these animal were terrestrial, tree-living, or burrowing. On the basis of the environment that Coniophis was found in and the features of the vertebrae, Longrich and colleagues deduce that this was a land-living and most probably burrowing animal. So far so good, but the next step goes further: on the basis of these findings, the authors suggest that the ancestor of snakes was probably terrestrial and burrowing. However, Coniophis comes from the Maastrichtian stage of the Cretaceous, some 65 million years ago, whereas the oldest known certain snakes date back to the late Early Cretaceous, some 100-120 million years ago. Now, that leaves us with two possibilites: First, the apparently primitive morphology of Coniophis is actually secondarily derived and has evolved from a "typical" snake, which seems unlikely. Second, Coniophis is a member of a primitive lineage of snakes which, according to its relationship with more dervied snakes, must reach back to the late Early Cretaceous at the least and thus have a ghost lineage (an evolutionary lineage without fossil record) of at least 35 million years. Given the fallacies of the fossil record in general and the poor fossil record of snakes in particular, the second explanation seems much more likely, and is also the one favoured by Longrich and colleagues. However, 35 million years are a very long time; indeed, this is more than half the length of the entire Cenozoic, the "age of mammals". The interpretation that a randmom member of such a long-lived evolutionary lineage that lived such a long time after its origin represents the original ecology of the clade is maybe not impossible, but nevertheless rather questionable. Assuming we would not have any good fossil record for either group, would you assume that elephants originated from marine animals, because their closest relatives, the Sirena, are marine today? Thus, although Coniophis provides exciting new insights into the evolutionary history of snake anatomy, I would be careful in interpreting its importance for the question of the ecological origin of snakes.

Longrich, N.R., Bhullar, B.-A.S. & Gauthier, J.A. 2012. A transitional snake from the Late Cretaceous period of North America. Nature, published online.


  1. Nice post and some valid points you make about the "original" snake ecology. Given the versatility of snake environmental adaptations we see today- burrowing, arboreal, aquatic, terrestrial- and the plasticity with which some species can go between all of the above environments we may never know the "one" true original snake modus operandi. And maybe that is ok...what we can say is that once snakes stumbled upon their design it suited them well in a variety of terrain.


  2. At least some basal proboscidians are regarded as semi-aquatic and I recall a paper in PNAS a few years ago that posited at least semi-aquatic origins for the group.