Doug Leigh: Like the dismay some felt when Pluto’s status as a planet was taken away by astronomers in 2006, it may be disappointing to learn that – except for one particular species of the animal – paleontologists don’t call pterosaurs “pterodactyls” anymore. Ryan and I wondered: what other misunderstandings are prevalent about the reptiles.

Armita Manafzadeh: Common misconceptions about pterosaur include that they’re dinosaurs or that they’re flying lizards, but really pterosaurs are their own group of flying reptiles that lived during the age of the dinosaurs. Pterosaurs were around from the late triassic soar around 230 million years ago to the end of the Cretaceous surround 66 million years ago. When the non-avian dinosaurs, not the birds, also became extinct pterosaurs were actually the first group of vertebrates to evolve powered flight. They evolved flight independently from dinosaurs and they lived all around the world in a crazy range of sizes. So, there was a pterosaur called Quetzalcoatl that lived in Texas and had a wingspan the size of a school bus. They’re really just this diverse interesting group of animals and because the group of pterosaurs has no living descendants it makes it really challenging to understand them, unless we start to consider their position within the family tree of life and we start to look at birds and also crocodiles and their other living relatives. The first pterosaurs were discovered in the late 1700s in the limestone’s of Bavaria. These were Jurassic limestone synthase area and when people found these first pterosaurs they thought that they were reptilian bats, they really didn’t know what they were looking at. I mean imagine if you found the first pterosaur you’d have really no idea what you were looking at. They weren’t quite bird-like and they weren’t quite bat legs. They didn’t know what to make of them but the first drawings that were made of pterosaurs, the first reconstructions and restorations, both scientifically and publicly portrayed pterosaurs in this very bat-like way drawn with their hips swung out to the sides, drawn with some skin membranes that stretched from their wings down to their ankles. There wasn’t evidence for a lot of these bat-like characteristics not direct evidence but because people were trying to place pterosaurs in their mind and understand what they were, they were reconstructed in this way. So, even more and more pterosaur specimens were found really well preserved specimens, a variety of morphologies over the next two centuries. This bat-like imagery from those first reconstructions has stayed with us to the modern day and if you look at restorations of pterosaurs through time you can see that that Ballack imagery has really just persisted and it’s gone on to permeate every aspect of the pterosaurs we see around us. So, if you’re watching Jurassic Park, if you’re reading books about dinosaurs and prehistoric creatures, almost every pterosaur you see drawn still has bat-like morphology.

Ryan Watkins: Armita was interested in examining how pterosaurs’ ligaments affected their possible flying postures. But because such ancient tissue typically isn’t available to paleontologists, Doug and I were curious to learn how modern birds could serve as a proxy for studying the range of their possible motions.

Armita Manafzadeh: When you are studying mobility in a living animal, one of the things that doesn’t make a lot of sense to do is to start to peel back layers of tissues and then look at the mobility you get with every layer peeled back. So for example, if you take off all the skin and then look at what your mobility is, take off all the muscles and then look what your mobility is, take off the ligaments and look again. You’ll get different mobilities. Those shouldn’t be applied wholesale to any extinct animal because the shapes of those different tissues at each layer differ between even close living relatives and they’re extinct ancestors. So, what we’re kind of pushing for here is a more structure by structure approach. In our study, we find that this one specific ligament, a ligament on the bottom surface of the joint capsule that holds the thigh bone into the hip socket, is a ligament that prevents abduction so outward swinging of the leg and we know this because we’ve done a ligament simulation. We’ve looked at a fiber orientation of this ligament, we’re very confident that any ligament placed in this way will limit abduction and then we have to set it all right. Birds have this ligament and we’ve seen how it limits abduction in a bird and we’ve seen that it stops the bird from swinging its leg out to the side, but how can we be confident that it would have had the same effect on a pterosaur and how can we be confident the pterosaurs even had this ligament? So, again that’s where our family tree of related animals becomes really important. If we plot terrasaurs on their family tree and we look at who they’re related to there are two closest living relatives their closest living relative is birds and they’re also related to crocodiles and alligators and outside of that they’re related to the other reptiles and if we look at just sections of all of these living reptiles every living reptile that’s been studied it has the same ligament placed in this way in the joint capsule that holds its thigh bone into its hip socket so if all of these living reptiles have this ligament we can be very very confident that pterosaurs would have also had this ligament and that it’s a very ancient structure. So, if we’re confident of the pterosaurs had this structure it was placed in the same way and we understand the mobility effects of this particular structure. Then we can start to infer how the structure would have affected mobility and pterosaurs as well. So the difference is really that we’re not taking mobility data from a bird and saying that the mobility of a pterosaur would have been exactly the same, that wouldn’t make any sense. Pterosaurs aren’t birds and we know that what we’re doing is looking at a very specific structure and its effect in a bird which is a close relative and then inferring the presence of that structure and its effects in the pterosaur which is the extinct relative.

Doug Leigh: Ryan and I first read a story about Armita’s study in Atlas Obscura, an online magazine which began in 2009 as a catalogue of unusual travel destinations, but since has expanded into other types of content, including surprisingly good science reporting. The article mentions a rather creative approach that Armita first took in exploring the range of motion of pterosaurs’ joints, so we asked her to tell us about it.

Armita Manafzadeh: Yeah, that’s kind of a duh story on this project. We were wondering about the pterosaurs, we were wondering if they could get their hips into this bat-like pose, and the question became how can we test this right. Like that’s something we could tell stories about. But, how can we do a rigorous test of this question, and when we want to test something about an extinct animal. How to reconstruct it? How it would have moved? We really have to just recur to its phylogenetic position. So, we have to look at the family tree of animals and figure out who the living relatives of those extinct animals are. In the case of pterosaurs, their closest living relatives, birds. So, it was still an early idea we had to figure out how to start to collect some preliminary data and see if our suspicions that soft tissues would have prevented this pose was actually true. So, we thought what’s an easy source of birds and I thought about this for a long time. I realized, well, I could easily just walk to the grocery store up the street from campus, go pick up a bird and then start to figure out what these soft tissue structures even look like and how they’re limiting motion. So, the first problem was really that I knew there had to be ligaments surrounding the hip joint of this bird, these elastic tissues that connect bone to bone. I knew they had to be there but there weren’t any pictures of them in the literature. I couldn’t figure out exactly where they should have been. So, my first step I went on my first grocery store trip and I bought five chickens and I brought them back to the lab. I said all right five chickens, that’s ten hip joints I have. Ten chances to mess up and find where I think these ligaments are. The first four chickens or the first eight hip joints I went through I cut through the joint capsule every time. I just had no idea where these structures were and finally that last chicken it was a spring break of my senior year of undergrad. I said in the lab well after it was on spring break vacation on the beach I found that joint capsule. It was the best moment ever, I couldn’t have asked for a better spring break. But, I finally found those structures, so great I found them but now the problem was how to test the question. So, I had to be able to compare what these chickens hip joints could do with that joint capsule intact versus what I would think they could do from dry bones alone. That’s why I said great I found it I have no idea how to measure this. So, my first pass was taking a protractor that I found in a drawer in the lab and just measuring some simple angles in the three major axes of joint motion. How far forward and backwards can you swing? How far in and out can you swing to the side? and how can you rotate along the long axis of your thigh bone? By measuring some of those angles with a protractor, I was able to compare from my chicken with ligaments on and then by looking at dry chicken bones and doing the same saying that was my initial comparison of joint mobility.

Ryan Watkins: Armita’s article includes details on how she carried out the dissections of bird joints. This got us wondering whether she learned to dissect animals as a biology undergrad, or if she gained those skills in graduate school … as well as if it’s a typical competence for paleontologists to develop.

Armita Manafzadeh: I think one of the most important things you can do, as a paleontologist who’s interested in the function of extinct animals, is to familiarize yourself with living animals. Because the remains we have, fossil animals, are just such a small percentage of all the things that make up an animal in life. One of the things Kevin instilled in me, when I was an undergrad in this lab, and something that I’m continuing here at Brown is really understanding living animals. To secretin living animals, understanding how all of their tissues fit together, and then what we can infer from what. Little remains in the fossil record about the whole story. When I started this project back at Berkeley, I really had no dissection experience at all, not even a little bit. So, when I had those grocery store chickens, my first round of dissection was with a kitchen knife until I walked over to another lab and someone gave me a scalpel to use. So, I started just acting entirely with a scalpel which I now know is terrible dissection technique.Then finally, through a combination of self teaching, and ultimately coming here and taking human anatomy in a medical school and learning how to properly dissect, I definitely got a lot better to find dissection. But, these joint capsules, the ligaments themselves, are really strong and pretty hard to get through. But, there are areas of the joint capsule that are very very thin and if you happen to just nick it with a tool that is too sharp, you’ve destroyed the entire capsule and your study is pretty much useless. So, there were many attempts on grocery-store chickens and many nights of me coming up with new recipes for chickens, eat the rest of the bird that I didn’t use as an undergrad, from all my failed dissections.

Doug Leigh: Despite her innovative approach to measuring chickens’ range of motion, Armita soon discovered that her D.I.Y. methods weren’t going to be sufficient. So the study moved from grocery-store chicken to quail, since their hip structure is more similar to that of pterosaurs. And instead of using protractor measurements, Armita used a one-of-a-kind, million-dollar x-ray system at Brown University. Ryan and I asked how the project went from a decidedly low-tech endeavor to such a sophisticated one.

Armita Manafzadeh: This is another story I can tell you. I did all my measurements with chickens and protractors. I was preparing the work for presentation at a conference and that was around the same time I was applying to graduate programs to come work on my PhD. I met my PhD advisor Steve Gates II at one of these conferences. I told him about what I had been working on, I told him I was interested in range of motion because I knew that was one of his interests as well. He looked at me, he said well you know protractor measurements just aren’t going to cut it. In fact, one of his students at the time, Robert, and him were about to publish a paper specifically detailing why protractor measurements were insufficient and why joint mobility really had to be considered in three dimensions. So, I stood there and I thought great! you know, here’s this prospective PI who I’m really trying to impress by doing something that’s in his wheelhouse, and he’s telling me everything I’ve done is woefully wrong for all of these reasons and needs to be improved. But, ultimately it ended up working out really well. I realized I had a lot I could learn from Steve about how joints work and about how to improve our quantitative framework for addressing questions like this. So, I came to work with him for graduate school and that’s when I realized, alright, I can do this right now. I have the methods here at Brown University. We have a biplane or x-ray system, so we’re able to take two x-ray videos at once. That’s a technique that was developed here for reconstructing the bone motions in 3D.

Generally, that’s done on living animals but I was able to do that on dead quail by manipulating their bones. I switched over to quail because they’re less domesticated than chickens. I wanted to make sure they’re adults so their bones don’t have too much cartilage. At the end, that would mess with my mobility measures. So, I ended up redoing my measurements at Brown with quail and in three dimensions, using this biplane or x-ray setup. It was several years removed from my initial chicken and protractor measurements. The concept was the same, the ultimate goal was the same, but the methodology was much much more rigorous and ultimately led to a much better paper.

Ryan Watkins: One of Armita’s goals with her study was to introduce an approach to measuring joint mobility called “Range Of Motion” –  or ROM – mapping, as she explains after this short break.

This episode is sponsored by we share science when researchers are curious about what is happening in science they go to we share science to explore video abstracts uploaded by other researchers you can search their vast catalogue of video abstracts to learn about the latest scientific findings or you can share your research with the world whether your research is in progress or already published at we share science you can share your science and grow your impact explore the world’s research at wesharescience.org now back to parsing science.

Ryan Watkins: Here’s Armita Manafzadeh.

Armita Manafzadeh: If we’ve argued that protractor measurements are insufficient, that a bar graph for each degree of freedom in a joint rotation is insufficient, and we need to talk about joint rotations in all three dimensions, simultaneously the next question is how do we visualize those data in a way that’s accessible and meaningful and allows us to get the information we want from them, if you think of every single joint pose as having an excursion in each of the three degrees of freedom? For example, if I’m holding my hip joint at any pose my leg is some amount forward and backward, some amount in and out to the side, and twisted to some extent. That gives me three numbers about that, describes that specific joint pose, and I can plot that joint pose as a 3D point in space. And if I do that for every single pose, that’s possible for example with the ligaments intact in the quail hip, I end up with this 3D point cloud of all the poses that are biologically possible at my joint. So, what I’m doing in this range of motion mapping method as I’m then fitting a nice little 3D polygon, just the 3D shape, to shrink-wrap that point cloud and represent it with one shape in 3D space. And I’m saying if I put that 3D shape in a space that represents all possible joint poses I can do the same thing again with for example bones only from the quail hip make another 3D shape and then I can say how this range of motion envelopes these 3D shapes of joint poses overlap or not overlap or interact in the space of all possible joint poses.

Doug Leigh: Ryan and I were interested in getting under the hood of the technology used in “ROM” mapping, and learned from Armita that it involves at least three main systems: x-ray, digital video recording, and three-dimensional reconstruction of the x-rayed specimen, as she describes next.

Armita Manafzadeh: When I dissect their hip joint capsules what I am left with is the pelvis and the femur, so that’s the two bones that make up the hip joint with the joint capsule attaching them. And then what I did, is to implant tiny little metallic beads into both the pelvis and the femur so the two bones at hand. And I tied a really long stick to the thigh bone of a bird so truly just a three foot long dowel rod from a hardware store, and then I mounted the bird in the middle of these two x-ray machines. So, essentially what we’re doing is taking x-ray videos from two different views of this bird. And I’m standing outside of the x-ray field as far as I can get with my three foot stick and I’m standing there waving around this thigh bone with my stick trying to move the hip joint through all its poses while wearing a lead jacket Alette apron, a lead neck collar, eye protection, and try to stay as far away from the x-rays as possible but I need to get this hip joint through all its poses. So, it’s a day it’s a lot of work. Once I do all my bone waving I end up with two x-ray videos of my bones moving through all the different hip poses that are possible and because I’ve implanted those little metal beads into the bones, I can then track the positions of those beats in each of the two videos, and then I can reconstruct in three dimensions from those two views what the 3D positions of those beads are. Because they’re radiopaque I can see them on the x-ray and then I can use the positions of those beads in 3D to place the bones. This is a technique that was developed at Brown called x-ray reconstruction of moving morphology and it was developed by my advisor, Steve Gates II, and my committee member, Brainerd. By creating the 3D motions of these bones and animating the 3D motions of these bones I’ve essentially reconstructed my quail hip on my computer screen in animation software similar to what animators for movies use. And then I can use the animations of these bones moving around on my screen to measure the rotations that are happening at the joints.

Rya Watkins: With its joint capsule intact, Armita found that quails’ hip bones could potentially assume a bat-like posture when flying, but only in the absence of hip ligaments. With the bone movement naturally restricted by ligaments, however, such a posture wasn’t viable, as she discusses next.

Armita Manafzadeh: Even if a ligament were attached the same way a pterosaur, and even if we can infer that based on how ligaments seem to be oriented relative to bones, a counter-argument that someone could raise would be well even if that ligament was attached in the same way it could be stretchier. right? It could allow your leg to swing out farther to the side. But what we’re finding is that in the quail that we studied the ligament would have had to stretch 63% beyond its maximum experimental length to get into a bat-like hip pose. So, although there’s some variation possible in ligament material properties and we can even see that in living animals, there is some degree of variation in the few that have been studied. That’s a really extreme amount of variation field to stretch 63% farther, I mean maybe if our results had shown that it was a 10% stretch beyond maximum experimental legs or 20% I think that maybe that amount of variation would be within the realm of reason, but 63% is a lot and nothing in the existing ligament literature suggests to us that that amount of variation does exist among living animals. So, there would have to be really extraordinary evidence to suggest that pterosaurs were that different. The other things to consider are that as your ligaments start to allow more mobility you’re doing that at the cost of stability, so the purpose of ligaments, one of their purposes seems to be that they make joints more stable by limiting motion you’re trying to prevent, extra motion that would cause your joint to disarticulate to dislocate. So, if a ligament were more stretchy to allow a bat-like hippos and that hip would be less stable and the most important thing to remember on top of all of this is that in addition to ligaments, there are other soft tissues that also restrain range of motion further. So, there are still muscles on top of all of that will further constrain range of motion in part because they have a size and shape, they have mass that gets in the way, but also because they cross joints and limit mobility. In addition to that, studies have shown that in a wide variety of the behaviors that animals use in life on a day-to-day basis they’re using a very small part of the range of motion that you would measure from a dead animal. What this really says to us is that if with just the ligaments intact in a dead quail it already can’t get into the bat-like pose that seemed possible based on bones alone, then imagine how much further constrained it would be with muscles and skin and how much of that would be even further constrained in life. So, there’s really a lot to get past to be able to argue litera sores would have been so different than they could have done this.

Doug Leigh: So if pterosaurs didn’t fly like bats, we wondered: just how did they hold their legs during flight anyhow? And should we expect museums to be changing their pterosaur exhibits anytime soon?

Armita Manafzadeh: When I walk into museums now, most of the pterosaurs and most museums are about are mounted with a bat-like hippos. So, I feel like it ruins museums for me. I just can’t look at them anymore but I think it’s important that we don’t fall into a trap of saying that pterosaurs flew like baths or flew like birds. Obviously, that’s the easiest way to simplify it, especially when we’re trying to dilute down what we’re saying. When we say that they didn’t fly like bats what we are referring to is this hip pose. So, if they couldn’t swing out their legs to the side, then our reconstructions of what a pterosaur wing membrane and a pterosaur airfoil. So, the surface they’re using, the fly look like couldn’t have actually been the way that they’ve been drawn and been used in studies of pterosaur flight. What the study doesn’t do much to the dismay of many people but that’s not what it’s meant to – and say how pterosaurs did hold their legs during flight so this is a constraint study right we’re saying what they could not do not what they did do what this does is it leaves open a couple of possibilities which are studied this current study can’t speak to so did they crouch their legs underneath their bodies like birds ligaments wouldn’t have prevented that that’s possible did they hold their legs straight back as far as the study goes ligaments wouldn’t have prevented that that’s a viable hip pose – on the basis of ligaments so I think the real implications of our study are just that the typical hip pose that’s been used in studies of pterosaur flight that’s been taken for granted and assumed to be possible now seems very very very unlikely so what I think this means is that people who are interested in the evolution of pterosaur flight and pterosaur flight mechanics need to start incorporating some alternate hip poses that wouldn’t have been prevented by ligamentous constraints into their reconstructions and into their models when they study pterosaur flight and I think only then can we start to figure out whether this will have a bearing on what we understand about how pterosaurs flew.

Ryan Watkins: Next, Doug and I were also interested in learning what implications Armita’s methods and findings might have on the future study of pterosaurs by other paleontologists.

Armita Manafzadeh: Our results that 95 percent of the poses that seem possible from bones alone were actually prevented by ligamentous constraints goes beyond just the specific study system and quail and pterosaurs and what this could mean and it really tells us that we need to exercise caution as paleontologists trying to reconstruct function and extinct animals clearly bones tell us a lot about what’s possible and they can give us a really way in reconstructing extinct animals but in order to really understand the details of how an extinct animal moved we need to understand the soft tissues that would have been present and the only way we can really do that or the best way that we can do that is to look at the living relatives of those animals so only by familiarizing ourselves with how living animals move and how their soft tissues affect how their joints work can we then even hope to begin to understand how joints worked in extinct animals where we have limited evidence to understand their morphology so in this case although perhaps our flashiest conclusion and the one that’s gotten the most press is that this felt like pose would have been impossible for pterosaurs I think it’s important to remember that yeah 95% of the poses that seemed possible from bones alone were actually prevented by ligaments and even more would be prevented if the rest of the soft tissues were intact so I think it causes us to be it causes us to have to be more humble as paleontologists and realize what a small subset of the information we’re left with and that we need to be working within a phylogenetic framework and we need to be looking at the relatives of these animals that are still alive today.

Doug Leigh: Some paleontologists have taken exception with Armita’s conclusions regarding pterosaurs’ range of motion. So we asked her what it’s like to be critiqued by more senior researchers.

Armita Manafzadeh: I guess I knew when I took on this project as an undergraduate that I was taking on a project that potentially had really controversial implications. Because if it seemed like ligaments really would be preventing this bat-like hippos. If what we thought to be the case was true and that would be going against the prevailing opinion for about two centuries and that’s a kind of stressful thing to engage in. And when I first presented this work for the first time at a conference in our field, I did start to get quite a bit of backlash from people who are very invested in these animals and spend much of their career studying them. I think the only thing I can really say to that, as the my goal, is to adhere to the principles of comparative biology and just the basic foundations of our field. So I’m not treating pterosaurs as special, I’m not giving them any kind of special status, I’m really just looking at structures and how they’ve evolved and how we can infer which extinct animals would have had them and how they limit mobility. I’m really just working from basic principles and in this case that happened to have implications for pterosaurs. So although the way this turned out is that our results are counter to what many people think to be true about pterosaurs, I really hope that the people who disagree with our findings are able to engage with our methods and our evidence and if they have critiques I’m perfectly happy to be wrong. I just want their critiques to be about our evidence and our methods so that they can really engage with our attempts to solve this problem in this way.

Ryan Watkins: Especially since she’s also taught a course in the design of robots, we were interested in learning what applications Armita thinks that the methods she used in her study might also have outside of the field of paleontology.

Armita Manafzadeh: I think the methods that I develop here can definitely be applied to orthopedics and in understanding joint injury and rehabilitation. Because there’s a lot of interest on the orthopedic side of things and how joint injury limits range of motion and how range of motion changes during healing. So that would be a really great application of range of motion maps just progressive mobilities from the same individual mapped over time. The ligament simulation method I used is actually modified from early attempts at ligament simulation from orthopedist again, so there’s a lot of crossover there. So, that’s a very natural connection in terms of joints whether they’re in humans or other animals. Joints are things that people care about. I think that ultimately these types of research will have implications for robotics or potential grounds for collaboration just because we’re asking a lot of the same questions. How does a joint move? How does that affect where an animal can place its entire limb? How does that affect? How the animal ends up walking, and those are questions that roboticists have to ask to in determining how the degrees of freedom at the different joints of their robots should work, and how their robots limbs should be placed. So, I think there are a lot of opportunities for collaboration and will be exciting moving forward to continue to pursue some of those.

Doug Leigh: To close out our conversation, we asked Armita a speculative question: namely, whether there’s anything in science that she might believe is true, even if it’s not possible to prove it.

Armita Manafzadeh: At least the way I operate in science there’s really one fundamental belief that we all have to have as scientists and I hope that everything follows from there and can’t really be classified as belief. But, I think we all have to believe that the methods we are using are just empirical observation and hypothesis testing and looking at the natural world and studying it in that way. I think we have to believe that that’s a viable way to look at the world. We have to believe that these fundamental things that we’re doing can reveal those secrets of the natural world and I hope that nothing we do from there is a belief. I hope that we continue to operate on that philosophy and really trust our data from there, but I don’t know if that’s something we can ever prove. You know I think that’s just that’s our philosophy as scientists, that’s our belief. We have to think that the methods we are using can answer the questions that we’re asking and I really hope that’s true and I really hope that’s the only thing that we’re choosing to believe without testing it.

Ryan Watkins: That was Armita Manafzadeh , discussing her article “ROM mapping of ligamentous constraints on avian hip mobility: implications for extinct ornithodirans, which she published with Kevin Padian in the May 27, 2018 issue of Proceedings of the Royal Society B. You’ll find a link to their paper on parsingscience.org along with bonus content and other material that she discussed during the episode.

Doug Leigh: We’re releasing this episode of Parsing Science from the Podcast Movement conference in Philadelphia. If you’re also attending and would like to meet up, send us a message in the Podcast Movement app: just search for Parsing Science and you’ll find Ryan and me. If you’re not attending, check out our posts at parsingscience.org/pm18 for updates on what we’re up to here this week.

Ryan Watkins: Next time on Parsing Science, we’ll be joined by Neil Lewis, Jr. from the University of Michigan. He’ll talk with us about his research into what differentiates college students who experience difficulty in school as a sign of the importance of succeeding academically from those who believe that difficulty means that completing their schoolwork is impossible.

Neil Lewis, Jr.: If you’re in a situation where say you’re working two or three jobs but still not moving up very much it’s hard to agree that difficulty is important and is sort of worthwhile.

Ryan Watkins: We hope that you will join us again.