Snake scales are really cool!

Whew! Six years of work paid off last week with the publication of a big paper in the prestigious journal Proceedings of the National Academy of Sciences. It was widely covered by the media, and already the author of an influential textbook on vertebrate biology has reached out to us to include our images and results in the next edition. Many of us aspire to the “textbook level” of attention for our research—i.e., when one’s work is considered foundational and winds up being featured in relevant textbooks, right alongside Darwin’s finches, etc. This is my first such experience and I am beyond proud! So, I thought I would share the story of the project with you.

Our collaborative Georgia Tech–Zoo Atlanta research group led by Dan Goldman (Georgia Tech) and myself has been working on various aspects of sidewinding and other locomotion in snakes for years, and the group has been fantastically productive. The work is just plain fun as well, because I have yet to get tired of watching snakes, especially sidewinders, move with such fluid grace. One day in my lab back in 2016, I was looking for something in the literature and stumbled across an odd paper in the engineering, not biology, literature that concerned the microstructure of the belly scales of snakes. A bit more searching and I realized that a small group, mostly in Germany, had been studying frictional properties of snake scales in attempts to bio-mimic them to improve the function of engineered artificial surfaces of devices that operate in high-friction environments.

I was fascinated by how complex were the microscopic textures on snakes’ bellies. The general pattern is of tiny, tiny little rearward facing spikes. The engineers were measuring and interpreting these as structures that reduced friction as the snake moves forward, and increases friction if the snake starts to slip backward. In this interpretation, the spikes were operating kind of like those terrifying metal claws at the car rental place (“Do not reverse! Severe tire damage!”). That made sense to me, but as I was reading my mind kept turning to sidewinders (which that group did not sample) and I conjured that this rear-facing spike anatomy would not seem to be relevant to a sidewinder (because they don’t move forward) and may even cause frictional problems as the snakes naturally moved sideways. I brought it up at our next sidewinder-team meeting and the group agreed. So, I began a campaign to collect shed skins from a wide variety of snakes. You don’t need the whole snake for this work, just the shed skins. This is very convenient and allowed me to include species that we don’t happen to have at Zoo Atlanta.

We’ve all heard of Scanning Electron Microscopy and seen the amazing images of fly eyeballs, etc., that the technology can produce. The engineers in Germany had been using that technique. Dan Goldman put me in contact with a researcher at City University of New York who is a master of a very different technique called Atomic Force Microscopy. I will be completely honest that I thought he was playing a snipe-hunt-type practical joke on me. I had never heard of this, and the name sounded like some goofy contraption in a James Bond movie. Well, it is real and produces both amazing images and also the ability to make measurements of these microstructures on an atomic scale. Whoa! So, we worked through my growing collection of shed skins, focusing on vipers so as to include the closest relatives of sidewinders for purposes of comparison, and patterns started to emerge. Typical vipers had the rear-facing spikes similar to what the German teams had found, except now we had atomic-scale measurements of the structures.

Once we felt we had the system mastered, we anxiously ran our first scan of a sidewinder skin from the Zoo. My hypothesis was that sidewinders would be fundamentally different, but I honestly had no idea what “different” would look like or just how different they would be. The first sidewinder image to emerge was mind-blowing. The spikes were technically present but reduced to tiny nubs that we predicted would be functionless. The surface also was cratered by many large, shallow pits the likes of which no one had ever seen. Instantly, I knew that we needed samples from some of the species of Africa, especially the Sahara Desert, that also sidewind. While they are vipers, they are not pitvipers, and so they are only distantly related to our U.S. sidewinder (which is a species of rattlesnake). We don’t have those African species at our Zoo, so I called in favors from other AZA institutions. The results were stunning—the African species were very similar to the sidewinder, but even more drastically modified in that the remnants of the spikes were missing altogether.

This story instantly got a lot bigger when we realized we had just documented a really remarkable example of convergent evolution. Convergent evolution is when two or more unrelated species independently evolve very similar anatomies or behaviors as they live ecologically similar lifestyles, often on opposite sides of the globe. The coloration and resting behaviors of green tree pythons and emerald tree boas are a famous example, as are the spike-shaped leaves in cacti and euphorb plants.

But we hit a wall because while we had described the anatomy, we had no idea how the spike-less and cratered textures might function in the context of sidewinding locomotion and sand. Dan Goldman introduced me to one of his post-doctoral researchers, Dr. Jennifer Rieser, who does not work on snakes but has amazing analytical skills in developing mathematical models of how things function. What a great addition to the team! We met and I explained the system and the snakes to her and she disappeared for a few months and emerged with a series of phenomenally complex, but really elegant, models as to how the spiked and unspiked/cratered textures function in sidewinding, and she went further to show how they would function in more typical snake locomotion (specifically a mode called lateral undulation, which is what you think of if you think of a typical snake flowing through the grass). She largely confirmed the functions of the spikes that the German teams had found, but also landed upon the significant new discovery that the spikes really function far more importantly in the “lateral” aspect of lateral undulation. In other words they were functioning kind of like a speed-skater where, by pushing mostly sideways, the skater shoots forward. Her models revealed that the modified textures in sidewinding species were miserable for lateral undulation—note that in 10 years of intensely studying sidewinders, we have never seen them attempt lateral undulation. Correspondingly, the typical spiked texture really compromised functional sidewinding.

Our discoveries further expanded the notion that sidewinders are fundamentally specialized for their weird mode of locomotion in behavior, and we discovered that their specialization extends to the atomic scale of their belly scales. And this has happened at least twice in two unrelated groups of highly specialized snakes. Wow! A last-minute addition to the paper was graduate student Jessica Tingle at University of California Riverside, who has been studying the evolution of sidewinding. She had developed a biogeographic historical timeline for the evolution of sidewinding, and her work was a crucial addition because it showed that the African sidewinding vipers are evolutionarily very much older that the US sidewinder. This allowed us the interpretation that the reduced, but not absent, spikes in U.S. sidewinders evidently are a step toward the complete eventual evolutionary loss of those structures, as seen in the older African species.

What a fun ride this project has been. Next steps! We are going to sample more African sidewinding species, even though I now expect that we can fully predict what we will find. And we are starting to investigate the function of the spikes in other forms of snake locomotion, and also branch out to sample legless lizards—I honestly have no idea what we will find there. And, finally, while we now understand how the spikes function, we do not understand how those crater-like pits function.

Come visit our now-famous sidewinders at the Zoo, and you may never look at them quite the same way again!

Publication: Rieser, J. M., T.-D. Li, J. L. Tingle, D. I. Goldman, and J. R. Mendelson III. 2021. Functional consequences of convergently evolved microscopic skin features on snake locomotion. Proceedings of the National Academy of Sciences USA 118 No. 6 e2018264118                        https://doi.org/10.1073/pnas.2018264118

Joe Mendelson, PhD
Director of Research