Jumping the Evolutionary Assembly Line

Evolutionary biologists are like puzzle-solvers. That’s true for any of the sciences, of course, but there’s one kind of puzzle that evolutionary biologists particularly like to solve: the order of assembly puzzle. Here’s how this puzzle works. Take a complex system that works very well. Now, break it down into its component parts and figure out how they work together (sometimes, this step is done by the physiologist in the next lab over).  Finally, figure out the order in which the parts were originally assembled. But there’s a catch: every time you add a part, the system has to form a working whole. It doesn’t have to have the same function as the finished system, but it does have to be a working system. You can’t break an old system until you have a new system in place.

Ironically, this puzzle was popularized by a creationist: Michael Behe. He called the puzzle “irreducible complexity” and he said it was impossible. Not only was he wrong, he was missing out on all the fun. Briefly, he said that evolution can’t be real, because it seems impossible to evolve certain complex systems from nothing. His best example was the vertebrate eye. In effect, he said that half of an eye was useless. Boy was he wrong! Not only can we show that each step in the evolution of the eye  is potentially useful (and an improvement on the step before), we can also show that for each step, there is a living animal that uses that exact system to great effect. Plus, we can show that complex eyes have actually evolved numerous times, and in multiple ways. But that’s not what I’m here to talk about; I came to talk about frogs.

If you think about it, frogs are a great example of a very successful complex system, and they function very differently than their non-froggy ancestors. It all has to do with locomotion.

The ancestor of all land vertebrates – including frogs, but also including salamanders and lizards, and birds and us mammals – walked with a gait that we call “lateral undulation”. That’s where the animal walks by bending the body to the left to reach forward, and then bends the body to the right to take a step, and reach forward again, and then repeats the whole cycle. Here’s what it looks like as a series of stick figures:


This is the same gait we see in modern salamanders, newts and lizards. We also see something similar in fish, because the ancestor of the land vertebrates used it for swimming. In fact, modern newts use it for swimming as well; they just fold up their legs, and the lateral undulation makes their tail swish from side-to-side very effectively.

Frogs do not undulate. Their entire pelvis has been re-engineered for jumping and hopping: motions in which both hind legs kick out at the same time and deliver a great deal of force all at once. Some frogs use the same motion for swimming, or have modified this structure further for the demands of burrowing. Some have even loosened it up for a sort of walking, but the general plan of the pelvis is pretty constant. Here are images of a frog skeleton (left) and a salamander skeleton (right), for comparison:

Frogs have other modifications for this mode of locomotion as well. Their hind legs are especially long, and they have fused many bony elements in their feet and loosened up the joints between other elements, in an effort to make strong but flexible limbs – like the coils of a spring. They have also modified their vertebral column and back substantially. They’ve greatly reduced the number of vertebrae between their arms and their legs and made the whole structure stiffer than it is in salamanders and newts. They’ve lost their ribs. And they’ve fused their tail vertebrae into a single rod, called the urostyle, that sits between the elongated bones of the pelvis and helps to translate forces from the pelvis to the body.

So, how did they put all these pieces together? How did they evolve from lateral undulation to jumping? It turns out that we have a snapshot from the middle of that process, and scientists have been arguing about it for decades. A recent study by Andrés I. Lires, Ignacio M. Soto and Raúl O. Gómez (Paleobiology, April 2016) sought to make sense of this nifty fossil. It’s called Triadobatrachus massinoti (or T. massinoti for short) and it has a fascinating suite of features, half-way between salamander and frog. The following figure shows a fossil of T. Massinoti from the top and bottom.

T. massinoti has a reduced number of vertebrae, but not as few as a frog has. It has an enlarged pelvis, but not as large as a frog’s either. It also has the beginnings of a urostyle, positioned correctly, but not yet fused into a single bone. Its tail was reduced but not absent. There is little or no doubt about where it belongs on the evolutionary tree: it’s on the branch leading to frogs and toads, but it’s not quite there yet. So did it use lateral undulation, or did it jump? In other words, it has parts of the modern system, but not the whole thing. How did its system function? How did the pieces fit together? How did one function take over from the other?

Lires et al. used a comparative method to address this question. Basically, they looked at known undulators and known jumpers and asked, “Is T. massinoti more like the former or more like the latter?” They focused on measurements of the relative lengths of the parts of the legs and feet, because several previous studies had shown that these measures were useful in predicting the mode of locomotion of several species of frog and toad – e.g. jumping, swimming, hopping, and walking. They collected specimens and X-rays of many extant (still living) species of frogs, toads, salamanders and lizards and made four measurements on each.


The raw data from Lires et al’s study. Fe=femur length, TF=Tibiofibular length, Tar=length of tarsus (in the foot), Hu= humeral length, RU=radioulnar length. All values were standardized against an average length, to remove the effects of the over-all size of the animal.

I can hear some of you asking, “Why lizards?” Good question! Basically, they were trying to avoid a problem: all of their species were amphibians, and they didn’t want that to bias their results. They needed a non-amphibian undulator as a control.

Anyway, they ran a statistical analysis on these leg and foot measurements and found that the combinations of numbers naturally grouped themselves into four modes of locomotion: Jumping, Swimming, Lateral Undulation and a combination of Hopping and Walking. Jumping and hopping are similar: a jump is just a big hop. Toads tend to either walk – one leg at a time – or make short hops, and it’s hard to tell the skeletal structure of hoppers from the skeletal structure of walkers. The upshot is that they had to merge the hopping and walking groups into one Hopper/Walker group, but the other modes were quite distinct.

Then, they performed the same measurements on our friend T. massinoti. Lo and behold, our proto-frog turned out to be closer to the lateral undulators than the other groups! However, it was near the edge of the lateral undulators, just starting, as it were, to reach out toward a new way of moving.


This is a graph of the discriminant analysis that they did. In this kind of statistical analysis, four different variables were sort of “squashed down” to two more significant values, called Root 1 and Root 2. Think of them as summaries of larger groups of data.

So, in terms of the order of assembly of frog locomotion, we now have a piece of the puzzle in place: the intermediate steps of expanded pelvis and reduced vertebrae are not enough to make this animal a jumper. Presumably, its lateral undulation was not inhibited – at least that’s what these data suggest.

Of course, there’s still the entire rest of the puzzle to work out. For example, if it wasn’t jumping, why did it have the expanded ilium? Why did it have a reduced tail and shortened vertebral column?

There are two ways to go about answering these questions. The first is to do a biomechanical analysis of these parts of the body. How much flexion exactly could we expect from the spine? How much force could the legs deliver? But what we really need is the second approach: we need more pieces of the puzzle. More intermediate steps. More fossils. Then we can ask more focused questions: Which happened first: the pelvis expanding or the back getting shorter? What came next? If we could do multiple biomechanical studies on each of the snapshots in the series – always being careful to put them in the right order – we could solve the puzzle.

But regardless of how we go about it, we have a great example here of the fun of studying evolution. Being able to watch the order of assembly, like a movie, unfolding over time. Not knowing how it ends just makes it like a mystery story. If we had rejected evolution, the movie would be over before it ever had a chance to start. It would be like throwing away the puzzle before we even look at the pieces.

 Image Credits:

Frog Skeleton: from George A. Boulenger – The fauna of British India including Ceylon and Burma. Reptilia and Batrachia. Taylor and Francis, London (fig. 127, p. 434 therein) (Via Wikipedia)

Salamader X-ray: Franco Andreone via Wikipedia and http://calphotos.berkeley.edu/cgi/img_query?query_src=photos_index&where-photographer=Franco+Andreone

All other images (except my lateral undulation drawing) are taken with permission (and with thanks from me!) from:
Andrés I. Lires, Ignacio M. Soto and Raúl O. Gómez Walk before you jump: new insights on early frog locomotion from the oldest known salientian. Paleobiology, Available on CJO 2016 doi:10.1017/pab.2016.11


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s