Dec 04 2008

Turtle on the Half-Shell

I love it, on many levels, when dramatic new transitional fossils are discovered. Tiktaalik is a recent favorite – an important transition from fish to tetrapods. Ambulocetus (the walking whale) is also a beautiful transitional animal – it’s about as close as you can get to a half whale/half terrestrial mammal.

And, of course, my all time favorite is the classic archaeopteryx – bridging two major groups, dinosaurs and birds. Archeopteryx has teeth, clawed wings, and a bony tail, but also has feathers and could clearly fly. It lacks the keeled breast bone of modern birds for attaching powerful flight muscles, however, and also lacks a triossial canal, necessary to perform a wing flip maneuver for taking off from a standing start. This thing was half bird/half theropod dinosaur.

Now a new transitional fossil species has been discovered: Odontochelys semitestacea (the half-shelled turtle with teeth), which is 220 million years old.

The Evolution of Turtles

One of the ways in which I find these key transitional fossils so fascinating is that they tell us something about the particular pathway that the evolution of a certain group of creatures took.  Piecing together the evolutionary past is like putting together a jigsaw puzzle, except you have no picture of the final product to guide you. You have the satisfaction of clicking a new piece into place, and you also get a small clue as to what the final picture will look like.

Turtles are interesting because up until now all turtle ancestor fossils have had their shells completely intact. There were no fossils documenting how the turtle shell evolved (“how” meaning what phylogenetic path, not by what mechanism).  From a developmental perspective it looks as if the turtle shell forms out of the backbone and ribs. So we would expect to find turtle ancestors with partially expanded vertebra and/or ribs, but we could not be sure. Also, we did not know if turtles evolved on land, in the sea, if their top shell or bottom shell evolved first, and what purpose it served.

The turtle shell is a dramatic evolutionary adaption, and yet it appeared fully formed in the fossil record, so paleontologists could only speculate about its origins.

A report is about to be published in the journal Nature by authors Chun Li of the Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences in Beijing, Xiao-Chun Wu of the Canadian Museum of Nature in Ottawa, and  Olivier Rieppel, PhD, chairman of The Field Museum’s department of geology. The report will present Odontochelys.

This turtle ancestor has a partially extended backbone – a partial top shell or carapace. It also has a complete plastron protecting its belly (the underside of the shell that modern turtles have).

This means that the bottom shell evolved first. This suggests that the creature was water-dwelling and needed protection for its vulnerable belly. On land such creatures would have their belly close to the ground and so it would not need protection. Once the lower shell was in place, a defensive strategy would favor upper protection as well. Once this process started, the pressure for more and more complete protection seems obvious.

This fossil also kills a competing theory – that turtle shells evolved from dermal plates, which is not uncommon among reptiles (such as the hard plates on many dinosaurs).

Incidentally – this species also had teeth, which is a primitive feature in turtles who lost their teeth in favor of a beak, much like birds.

Transitional Species

This term causes some confusion, which is exploited by creationists (who are themselves the most confused). In a very real sense all species are transitional. All species connect to the evolutionary web of life, and so represent a connection between neighboring branches. All species also have the potential to be the ancestors to future descendant species, and so are a transitional link from their past to their future.

But all species are also simply adapted to their current existence. They are not in the process of becoming anything – because evolution is blind to the future. The reptile ancestors of Odontochelys did not evolve into Odontochelys on their way to becoming modern turtles – they simply evolved into Odontochelys by adapting to their immediate needs. We can only see that Odontochelys was transitional from its ancestors to modern turtled by looking back at what happened.

This may seem a trivial point, but creationists perpetuate the myth that by “transitional” biologists mean some improbable monster, a chimera with mismatched parts that is in the process of becoming something but is not something itself. It is important to understand that transitional species are only transition in retrospect, and they were just as adapted to their environment as any creature.

Therefore the “half wings” of some bird ancestors were not really “half wings”, but were fully formed for something else, like display, or trapping insects. The “half shell” of Odontochelys is only a half shell because we know its descendants will evolve into modern turtles (or the descendants of one of its close relatives, as it may not be on the direct line to turtles), but it is fully formed for Odontochelys’s purposes.

Evidence for Evolution

Of course, transitional species are powerful evidence for evolution, which is why creationists spend so much time and effort trying to (unsuccessfully) knock them down. There are always details they point to in order to attempt to cast doubt on the evidence – the species may not be in the direct line to the putative descendant, the time-line may not be obvious, and the quality of the specimen is never perfect.

But their big strategy is to simply deny that transitional species are transitional. They fall back on the unfalsifiable claim that God could have made a reptile with only the bottom half of a turtle shell – that doesn’t prove turtles evolved from this creature.

What they miss is that science is about making predictions, and the better any theory makes predictions the more confirmed it becomes. If all life on earth evolved, from a common ancestor, then all life shares a common descent. Therefore if you take any two branches on the “tree of life,” sometime in the past there must have been a common ancestor. Or, there must be a transition between any child group and its parent group.

Whales evolved from terrestrial mammals. Evolution predicts we should find specimens from a time when whale ancestors were part way between their terrestrial ancestors and modern aquatic whales. This prediction was validated by Ambulocetus and other specimens.

Birds as a group are most closely related to reptiles, which are a more ancient group. Therefore it is likely that we will find specimens from a group of reptiles that were bird-like but had not yet evolved all the characteristic of modern birds. This is exactly what we find.

We should find walking-fish who were ancestors to modern tetrapods – done.

And we should find turtle-like reptiles with only part of a modern turtle shell. Now we can check off that box as well.

Creation does not require that such species should exist at an appropriate time in the past. It does not predict them (actually, it predicts nothing, because it’s not a scientific theory). Common descent through evolution requires, and therefore predicts, and such specimens exist – and we are finding them at a steady rate, each one further validation of evolution.

Exploiting Legitimate Scientific Controversy

One of the tactics of creationist is to misinterpret legitimate discussion about how evolution unfolded as if it casts doubt on whether evolution happened.  This new find is a single fossil filling in a sparse area of the fossil evolutionary tree, so of course there are multiple ways to interpret its significance.

In the same issue of Nature Robert R. Reisz & Jason J. Head wrote an interesting comment:

Although this evolutionary scenario is plausible, we are particularly excited by an alternative interpretation and its evolutionary consequences. We interpret the condition seen in Odontochelys differently — that a carapace was present, but some of its dermal components were not ossified.

Very interesting. This is, of course, science at its best – considering all possibilities, presenting alternative interpretations of the data. This debate will likely continue until it is resolved by new specimens. Of course, both interpretations are evolutionary, and either way this specimen is an important primitive turtle – the most primitive yet discovered.

And, of course, the creationists are already exploiting this internal debate as if it casts down on evolution itself.

What creationists don’t do, of course, is make predictions. What fossil discoveries would invalidate creation? The answer is, none, because creation is not a falsifiable scientific theory. Evolution, on the other hand, would have crumbled if we kept finding fossils out of any possible evolutionary temporal sequence or morphological pattern. Horses in the Cambrian fauna would do it.


Odontochelys is the oldest, and most primitive turtle. It fits nicely into an evolutionary pattern – throwing one more piece of evidence onto the mountain that is already there.  The details and controversies will be worked out when further evidence comes to light. And I predict that it will.

The same thing happened with the evolution of whales – the first specimen was dismissed as an anomaly, but then a slew of whale ancestors were found fleshing out the whole sequence from land to sea.

There was a time when archaeopteryx was the only transitional species from reptiles to birds, and creationists claimed it was just some weird bird. Now, an entire group of feathered dinosaurs has been discovered, showing a variety of primitive, derived, and transitional features.

Answersingenesis has this to say about the fossil evidence for turtles:

The biblical account of Creation in Genesis —animals created to reproduce after their kinds—would mean that turtles should be instantly recognizable as turtles, with the shell and other unique features fully formed from the start, and no series of ‘pre-turtle ancestors’ should be found. It is obvious that the fossil record of turtles gives powerful support to biblical Creation, and stands opposed to the idea of evolution.

Hmm…but then the pre-birds, pre-whales, pre-terretrial vertebrates, pre-mammals, etc. should have already invalidated creationism. I will hold them to their statement, however. When more turtle ancestors are discovered will they stick to their prediction that according to creation they should not exist? Of course not. They will do what they did for whales and birds – make ridiculous arguments about the tiny details and miss the obvious big picture.

I predict that now that we know where to look, we will find further primitive turtles and we will flesh out a branching bush of creatures filling in the morphological space between reptile ancestors and more modern turtles.

This scenario has played itself out numerous times now. Each time creationists scream and yelp about the gaps in the fossil record. But when those gaps are filled, they never acknowledge it or adjust their conclusions – they just move over to some other gap.

The cognitive dissonance must be deafening.

21 responses so far