New paper out by Dr Karl Bates and Dr Peter Falkingham:

The importance of muscle architecture in biomechanical reconstructions of extinct animals: a case study using Tyrannosaurus rex

In living animals, we measure their morphology (e.g. muscle size, muscle fibre and tendon lengths) directly, either through dissection or medical imaging techniques. We can also of course measure their biomechanical performance by observing them in the wild and/or in certain cases by measuring them in a controlled laboratory setting. This is central to what we do in the Evolutionary Morphology & Biomechanics Group at Liverpool. However, as a group we are also interested in how and why the morphology and biomechanics we see in living animals came to evolve. Understanding morphology and biomechanics in an evolutionary context requires that we also study fossils. However, in long extinct animals like dinosaurs measuring their anatomy and biomechanics directly just isn’t possible; the long process of fossilisation has (mostly) removed all soft tissue leaving us with just the lifeless bony remains of such animals. To get around this limitation we, like many other groups, we use computer modelling approaches to first reconstruct soft tissue anatomy, and subsequently to reverse engineer how extinct animals moved and what they were capable of.

In a new paper published in the Journal of Anatomy this week EMB’er Karl Bates and colleague Peter Falkingham (Liverpool John Moores University) set out to explore how accurate and consistent computer models of extinct animals currently are. This is an important question because biomechanical assessments of extinct animals play a pivotal role in understanding how these animals lived, interacted with each other, and why they evolved. If our estimates of the biomechanical capabilities (e.g. running speeds, bite forces) of extinct animals are poor then this potentially means that our ability to reconstruct past behaviours, ecology and selective evolutionary pressures may also be limited.

To do this, Karl and Peter used a well-studied ‘exemplar’ taxon: Tyrannosaurus rex. T. rex has arguably been studied more intensively than any other extinct animal (maybe Australopithecus would give it a run for its money in this respect) for a number of reasons, but mostly because of its large body size, complete skeletons and of course it’s celebratory status and popular appeal! However, what made T. rex interesting in this case was that recent computer modelling studies had applied extremely similar methods to estimate how hard T. rex could bite but arrived at very different quantitative predictions. In their new paper, Karl and Peter argue that by understanding why these studies arrived at different predictions, despite using the same methods, will tell us something important about how accurately and consistently we are using biomechanical modelling approaches to study extinct animals. So what did they find?


Figure 1: A computational biomechanical model of T. rex used to reconstruct jaw-closing muscles and estimate bite force (modified from Bates & Falkingham 2018).
In a nutshell, Karl and Peter argue that studies of bite force in T. rex have differed so considerably in their predictions because of different assumptions made about the length and arrangement of muscle fibres in the jaw-closing muscles of T. rex. The length and arrangement of fibres plays a really important role in how much force muscles can produce and subsequently on how hard an animal can bite. In this case, differences in the way previous studies had estimated fibre lengths and arrangements in the jaw muscles of T. rex were so great that their predictions of bite force varied by more than 50%! In attempt to improve this situation Karl and Peter analysed existing data from over 1000 muscles measured in living animals in an attempt to better predict muscle properties in T. rex. Unfortunately while this helped in some respects, this new analysis really highlights that more data is needed (particularly on jaw muscles from living animals). Karl and Peter went on to use their new fibre length data to remodel in T. rex and unfortunately showed that muscle fibre length and arrangement remains the biggest single source of uncertainty in bite force estimates. As always then – more research is needed.


Link to paper: