Dynamic Duos: Why Examining More Bones is (Mechanically) Better Than One
Eadweard Muybridge's pioneering photographs of a human male running. (https://thebioscope.net/2010/09/22/the-running-man/)
A common question raised in research by morphologists and functional anatomists, whether experimentally observing the movement of an organism in the lab or developing a model based on fragmentary fossils, is: “How do we better understand the movement of this creature?” From the work of the earliest naturalists, descriptions of the shape and size of bones were key aspects of this research endeavor. Further, because of the fragmentary nature of the fossils of many ancient organisms, scientists learned to draw conclusions from isolated or a few skeletal elements. These models were used to describe how organisms moved and, through comparative anatomy, to extrapolate their ecology.
Perhaps because of decades of using this approach, we have grown accustomed to examining certain single bones over others, even when we have completely intact skeletons to consider. In the limbs, primates have two parallel bones both in the forearm (the ulna and radius) and in the anatomical leg (the tibia and fibula). For most osteological studies, there is a strong bias toward measuring the radius and the tibia over the ulna and the fibula, respectively; a cursory scan of the literature suggests that this bias is more prevalent in studies of the leg than in the forearm. There are sound functional and anatomical reasons for focusing on one particular skeletal element over another. The radius has more direct articulation with the wrist joint than the ulna. The tibia is much larger than the fibula, and transmits greater loads as a result. It is therefore no surprise that, for instance, most studies considering locomotion in hominin legs tend to exclusively measure the tibia. Yet, limiting examination to one element in each of these sets may be ignoring subtle but important contributions that better explain variation in the morphology of both bones.
Why You Should Keep Your Notes from Grad School
Over thirteen years ago (!), when I was a naïve doctoral student (as opposed to the naïve professor I am today), this habit of examining one element without its partner piqued my interest. After all, though minor, the fibula has a role in the mechanics of the lower limb. Nine muscles attach to it in humans, it helps stabilize the lower limb when traversing uneven ground, and it mitigates some of the forces imposed on the tibia by the femur and by sideways movement (Goh et al., 1992; Thomas et al., 1995; Thambyah and Pereira, 2006; Carlson, 2014). I had a hunch that the shape of the tibia might, in part, reflect variation in the shape and position of the fibula. So I performed a small pilot analysis comparing the biomechanical properties of the tibia with its paired fibula on donated skeletons at Johns Hopkins, and concluded that there were signs that the shape of the former was influenced by variation in the shape of the latter together with their relative positions within the leg. Distracted by other research questions, I shelved the project for over a decade, and a growing body of literature since has suggested that we should not be too hasty in disregarding the importance of that long, narrow bone when trying to understand its larger companion (Funk et al., 2004; Thambyah and Pereira, 2006; Marchi, 2007; Rantalainen et al., 2010; Marchi and Shaw, 2011; Carlson, 2014; Sparacello et al., 2014).
CT scan image taken from data collected by Dr. Colin Shaw and used in the forthcoming publication, showing the two measures of relative position between the tibia and fibula.
I returned to this study recently with the help of colleagues: OVAL doctoral candidate Alice Gooding, Dr. Adam Sylvester, and Dr. Colin Shaw. The results we obtained are reported in a forthcoming paper in the American Journal of Physical Anthropology. This paper emphasizes the importance of examining paired skeletal elements simultaneously in biomechanical studies. My coauthors and I followed up on previous mechanical studies of the fibula that showed it has a significant function in transmitting loads during bipedal locomotion, to the point that the mechanical properties of the fibula covary with different types of habitual activities (Marchi and Shaw, 2011). In our study, we looked at the cross-sectional geometric (CSG) properties of the tibia and the fibula, measured from computed tomography (CT) scans taken from athletic and non-athletic students by Colin Shaw (Shaw and Stock, 2009). CSG properties reflect the dynamically changing resistance of bone to mechanical forces encountered in reaction to impact with the ground, joints, and soft tissues. We supported previous findings about the fibula, but also showed that the locations of the two bones relative to each other—how far posterior or lateral the fibula is to the tibia—affects the CSG properties of each bone. The CSG properties of the fibula covary with the distance between the centers of the two bones, and the shape and properties of the tibia covary with the angle between the bones. There are also differences in the relative positions of the bones in swimmers versus terrestrial athletes, though we cannot address whether there is a causal relationship between activity and anatomy.
The main take-away from this study is that we need to be cautious about drawing conclusions from unitary skeletal elements under the assumption that variation in those bones is driven by a limited set of factors. That is, variation in the shape of the tibia has been interpreted to reflect locomotion and activity imposed on that element alone, but we show that some of the variation is explained by the location of the fibula, which likely also covaries with variation in movement in ways different from the tibia. Undoubtedly, effects driven by the femur and by soft tissues likewise contribute to this variation as well. Furthermore, these considerations focus on shape variation that arises from mechanical effects. Genetic covariation with other traits, developmental variation, metabolic effects, and the non-mechanical interaction of hard and soft tissue are all also factors to be considered.
Two (or More) is Often Better Than One
A theme that emerges from this research and related posts by Kristen, Sam and Liz on this blog, which reflect a perspective that colleagues and I advocate (Savell et al., 2016), is that we atomize and Balkanize the skeleton to our detriment. We must develop a more whole-organism approach in our anatomical studies, whether they are developmental studies with model organisms, evolutionary quantitative genetic analyses, or functional anatomical research. As noted in previous posts here as well, deciding which traits to include is a challenge in this approach, given the many factors and interactions that take place in an organism. However, we should not feel like this incurs an enormity of irreducible complexity. Rather, we choose traits based on the research question, and sometimes listen to our hunches about including obvious and not-so-obvious characteristics. Starting with a model is essential. Examining isolated bones has its utility, and sometimes is unavoidable, but drawing broad conclusions from an isolated skeletal element alone without regard for the whole environment in which that element is situated leads to incomplete and potentially misleading conclusions. More importantly, we should remember that this endeavor is a science, and as such we may have to add or remove trait measurements as we learn more about an anatomical system. It is through this informed trial-and-error that we arrive at a more complete understanding of how organisms evolve and vary with respect to their genetic histories and their environments.
References