Nuchal ligament reconstructions in diplodocid sauropods support horizontal neck feeding postures

Abstract

Historically, sauropods have been largely perceived as having vertical, ‘S’-curved necks which were hypothesised to allow them to feed from the canopy of trees. Within the past two decades, this popular perception has been questioned, resulting in a debate over neck posture. The osteological differences between sauropods with horizontal neck posture (Diplodocus), and less horizontally inclined necks (Brachiosaurus) suggest differing life and feeding styles. One differing vertebral feature between these polarised bauplans is the bifurcated neural spine. Regardless of the spine condition, sauropods with and without bifurcated spines have been reconstructed exhibiting the same neck posture. Corroborating histology and morphology in extant taxa highlights the presence of modified vertebral ligaments associated with bifurcated spines. Using these extant taxa to better understand the biomechanics of bifurcated spines, this study proposes alternative soft tissue reconstructions. Previous depictions had the bifurcation trough entirely open or harbouring pneumatic diverticula or muscles; conversely this study proposes that the apices of the bifurcated spines were the anchoring points for a split nuchal ligament, and that the trough of bifurcation was predominantly filled with interspinal ligaments. Ligaments provide energy-efficient elastic rebound, and a paired ligament in the cervical series would aid in prolonged, lateral movement in a horizontal plane (i.e. feeding).

Keywords:

• Sauropod
• diplodocid
• bifurcation
• nuchal ligament

Acknowledgements

I would like to dedicate this manuscript to my esteemed colleague and friend Kent Stevens. I fully support Kent’s alternative cervical reconstructions, but his work should be recognised for the far greater significance as the paradigm shift in sauropod neck posture (a topic which had remained rather static for 122 years). As in all paradigm shifts, what we once knew will be later modified and corrected, and Kent’s work should not be treated as definitively right or wrong; instead it should be viewed as a novel topic that engages the scientific community and rallies us into further investigation: and this is exactly what he has achieved. So I thank you Kent for your contribution to the progression of our field. Additionally, I would like to thank J. Horner, K. Padian, H. Woodward, J. Scannella, D. Fowler, K. Nordén, E. Freedman-Fowler, and two anonymous reviewers who generously reviewed earlier drafts of this manuscript.