Well, Finite Element Analysis actually, but it amounts to the same thing. This is where engineering and the study of human evolution overlap: the realm of physical anthropology.
The large majority of what follows is based on reference .
FEA: Finite Element Analysis.
FEA is widely used in engineering to model what happens to buildings and other structures when forces are applied to them by things like wind, earthquakes and their own weight, as well as in orthopaedics in the development of artificial joints and in modelling bloodflow. All this is thanks to the exponential increase in computing power in the second half of the 20th century.
It is only over the last 15 years that the technique has really caught on in physical anthropology although until very recently it has mainly been used to analyse the bones in the skull.
Loads of Force, Stress and Strain.
It sounds like what you go through to get an engineering degree, but these things also apply to bones as they are found in your skeleton, supporting your weight, helping move you around and chewing your food.
A load applied to a bone at a certain point exerts a force on the bone at that point.
That force causes stresses in the bone.
Stresses transfer the force to other regions of the bone.
That displaced force strains the structure of the bone.
That strain causes the bone to deform slightly.
Finding the Finite Elements
As you will probably be aware, the surface of a bone is smooth and continuous. Essentially, then, it is made up of infinitely many infinitely small individual points, in the same way that a line can be thought of as consisting of infinitely many infinitely small dots. Good.
Now, computers are bad at dealing with this sort of infinity because doing the maths takes an awfully long time. An infinitely long time, really.
So the first trick is to create a digital representation of the bone in question using either computer-aided design software, or a laser scanner as I mentioned in my last post. FEA software then breaks up the infinitely smooth surface into merely a very large number of points (nodes) which it joins with lines. These lines can be thought of as forming the edges of tetrahedral or cuboidal structures.
It is these structures that are the finite elements.
What you are left with is a finite element mesh which represents the entire structure:
Each of the finite elements is then given the physical properties (such as density and elasticity) of the bone it’s representing, while realistic anchor points and constraints on mobility that the real structure would experience are added to the whole model. For example, a jaw can only open so wide and the joint has very little side to side movement. These sorts of things are important in order for the simulation of stress and strain to be accurate. together with these properties, a simulated realistic load applied to the finite element mesh model of the bone completes the boundary conditions.
Under these boundary conditions the nodes move slightly from their original positions, and it is these displacements that are analysed to give a measure of the strain experienced. From this, the direction and size of the stresses can be calculated and represented using false-colour images:
Notice how the areas of high stress (orange) are regions of low strain and deformation (greeny-yellow with a bit of orange) and vice-versa. This is because, as mentioned above, stresses transfer energy to other areas of the bone causing strain and deformation in these regions.
Finding Exact Accuracy
Once these images have been generated, they are compared with data gathered from the real structure. If they are significantly different the nodes and the finite elements are modified until the simulation matches with real life. From here, experimenters can subject the simulation to unusual forces safe in the knowledge that their model will faithfully reproduce the behaviour of its solid counterpart.
But how do we know how the real bone would act, I hear you ask.
This is actually where a lot of the error comes from in FEA. For the bones of animals that are alive today such as the long bones of cows and humans they can be scanned by ultrasound while real mechanical loads are applied to them. Differences in the acoustic properties which result from deformation of the bone show up as changes in the ultrasound readings. However, these readings differ between species.
Fossilised bone presents a different problem altogether as it has completely different properties from fresh bone. This is because fossilisation entirely replaces the protein-rich bone matrix with minerals, essentially giving it the same properties as stone.
The difficulty then comes in knowing what sort of modern bones to compare it to. Do you compare it to bones which are in the same position in modern animals, or those that perform the same function? For example, the femur of a seal is in the same position as its four-legged landlubber ancestor which it shared with weasels and bears but maybe it has more in common functionally with the femur of the more distantly related whales and dolphins.
Or do you compare it to bones of modern animals which walk grow in the same way; should fast-growing bones be compared no matter where they come from as they are likely to grow in the same manner? (reference ).
One More Problem
For fossilised bones we still don’t know what’s inside them which might affect their physical properties. If it’s solid bone, then great. If it’s hollow as in birds and their ancestors, theropod dinosaurs (which includes T. Rex [ref. 3]), then that changes things. So does the presence of a medullary cavity full of bone marrow. For modern bones we can perform medical-style CT scans which tell us the density of the bone all the way through (Fig. 4). (Find out how bones are actually made here).
But that doesn’t help with fossil bones because they’re essentially stone all the way through.
So what do you compare a fossilised bone with? (Hmm, there’s a pattern here…)
Basically, do all of those things and you should have a pretty good idea of how the original handled the the pressure.
Finding Evolutionary Applications
All this strain and deformation is useful both in developmental and evolutionary terms. These forces can be detected by the body and acted upon by the process of bone remodelling, a slow but constant process.
Osteoblasts and osteoclasts work together to ensure that bones are in tip-top condition by ensuring constant turnover of calcium minerals as well as collagen and other proteins which make up the bone matrix. An extreme example of this can be seen in professional bodybuilders and weightlifters whose bones have muscle attachments far larger than those of ordinary people. The same can be said of Neanderthal man compared to modern humans (reference ).
See if you can see any differences in the size of the joints in these pictures. The modern human skeleton is the pale one.
Over an evolutionary timescale, individuals whose bones are less well-shaped or arranged are less likely to pass on their sub-optimal genes as they will have to expend more energy moving about in day-to-day life at the expense of reproductive success, meaning that everyday stressing of bones will, in the long run, result in changes in bone shape in a population.
Overall then, the shape of a bone and how it dissipates forces is determined by its function, which means that forces exerted by experimentally loading a single bone, and measuring how the bone responds, can be used to figure out the role it would have played in an intact individual. So analysis of the shape of a femur could tell us whether a hominid was bipedal or not.
Plus, as mentioned right at the start, most of the research has been done on the skull. In terms of human evolution the structure of the skull can shed light on the diet of the species. The prolonged chewing required to extract nutrients from tough foods such as grasses, roots and nuts requires a different jaw motion from that required to tear meat and will produce different patterns of loading on the bones of the skull. These loading patterns can be simulated through FEA to determine what the individual would have eaten. This is turn tells us something of the environment in which the individual lived and helps to fill in our evolutionary story.
 Rayfield, E. J. (2007). Finite Element Analysis and Understanding the Biomechanics and Evolution of Living and Fossil Organisms. Annual Review of Earth and Planetary Sciences, Vol. 35: 541-576. (Not freely available).
 Puymerail, L., et al. (2012). A Neanderthal partial femoral diaphysis from the “grotte de la Tour”, La Chaise-de-Vouthon (Charente, France): Outer morphology and endostructural organization. Comptes Rendus Palevol, 11(8), 581–593. (Not freely available).
 Padian, K., Hutchinson, J., & Holtz, T. (1999). Phylogenetic definitions and nomenclature of the major taxonomic categories of the carnivorous Dinosauria (Theropoda). Journal of Vertebrate Paleontology, 19, 69–80.
 SHOWCASE – The annual University of Leeds postgraduate research conference, University of Leeds, UK. www.leeds.ac.uk. Accessed 11:51, 13/09/2013.
 Sawyer, G.J. and Maley, B. (2005), Neanderthal reconstructed. The Anatomical Record. Vol. 283B: 23–31. Freely available! And some very cool pictures – a MUST SEE.