Comparative sacral morphology and the reconstructed tail lengths of five extinct primates: Proconsul heseloni, Epipliopithecus vindobonensis, Archaeolemur edwardsi, Megaladapis grandidieri, and Palaeopropithecus kelyus

Abstract

This study evaluated the relationship between the morphology of the sacrum—the sole bony link between the tail or coccyx and the rest of the body—and tail length (including presence/absence) and function using a comparative sample of extant mammals spanning six orders (Primates, Carnivora, Rodentia, Diprotodontia, Pilosa, Scandentia; N = 472). Phylogenetically-informed regression methods were used to assess how tail length varied with respect to 11 external and internal (i.e., trabecular) bony sacral variables with known or suspected biomechanical significance across all mammals, only primates, and only non-primates. Sacral variables were also evaluated for primates assigned to tail categories (‘tailless,’ ‘nonprehensile short-tailed,’ ‘nonprehensile long-tailed,’ and ‘prehensile-tailed’). Compared to primates with reduced tail lengths, primates with longer tails generally exhibited sacra having larger caudal neural openings than cranial neural openings, and last sacral vertebrae with more mediolaterally-expanded caudal articular surfaces than cranial articular surfaces, more laterally-expanded transverse processes, more dorsally-projecting spinous processes, and larger caudal articular surface areas. Observations were corroborated by the comparative sample, which showed that shorter-tailed (e.g., Lynx rufus [bobcat]) and longer-tailed (e.g., Acinonyx jubatus [cheetah]) non-primate mammals morphologically converge with shorter-tailed (e.g., Macaca nemestrina) and longer-tailed (e.g., Macaca fascicularis) primates, respectively. ‘Prehensile-tailed’ primates exhibited last sacral vertebrae with more laterally-expanded transverse processes and greater caudal articular surface areas than ‘nonprehensile long-tailed’ primates. Internal sacral variables performed poorly compared to external sacral variables in analyses of extant primates, and were thus deemed less useful for making inferences concerning tail length and function in extinct primates. The tails lengths of five extinct primates were reconstructed from the external sacral variables: Archaeolemur edwardsi had a ‘nonprehensile long tail,’ Megaladapis grandidieri, Palaeopropithecus kelyus, and Epipliopithecus vindobonensis probably had ‘nonprehensile short tails,’ and Proconsul heseloni was ‘tailless.’