Objectives: The presence of sexual dimorphism in the birth canals of anthropoid primates is well-documented, and birth canal dimorphism tends to be especially robust among species that give birth to relatively large neonates. However, it is less clear whether birth canal dimorphism is accompanied by dimorphism in parts of the pelvis not directly under selection for birth, particularly including bi-iliac breadth, biactetabular breadth, lengths of the ischium and ilium, and 3D shape. This study investigates the patterns of dimorphism among anthropoid primates in those parts of the pelvis which do not directly contribute to the bony birth canal, here termed the non-obstetric pelvis.

Methods: 3D landmark data were collected on the bony pelves of 899 anthropoid primates. Specifically, landmark data were collected on parts of the pelvis not thought to be directly involved in selection for parturition, including portions of the posterior and superior ilium, acetabulum, and lateral ischium. principal components analysis and Euclidean distance matrix analysis were used to ascertain sexual dimorphism in pelvic sizes and shapes within each species.

Results: Results show that dimorphism in non-obstetric pelvic size and shape exists across anthropoids, just as is seen in the birth canal. However, the magnitude of dimorphism in non-obstetric pelvic shape tends to be greater among anthropoid species that give birth to relatively large neonates compared to those birthing smaller neonates relative to maternal pelvic size.

Conclusions: Though all anthropoids included in the study show some degree of sexual dimorphism in non-obstetric pelvic size and/or shape, species which give birth to large neonates relative to maternal pelvic size have the highest levels of dimorphism in pelvic shape. Moreover, the magnitude of dimorphism in certain parts of the non-obstetric pelvis mirrors patterns seen in the birth canal. The results of this study are promising for ascertaining pelvic dimorphism and relative neonate size in fossil primates, particularly in fragmentary remains which do not preserve a complete bony birth canal.

The 3D coordinates in this dataset were collected using a Microscribe MX 3D Digitizer. The dataset included here are the raw landmark data for each specimen. Skeletal pelves were articulated using putty and rubber bands, and the following landmarks were collected:  

The following link is a figure depicting the location of each landmark collection location (only the left side is shown for bilaterally collected landmarks):

https://docs.google.com/document/d/1CJvDDgeio2lVgV32lewatqEBzdR6SIrGD01tIdu1_-4/edit?usp=sharing

The dataset included here has 921 specimens, as some sample sizes collected were too small for sufficient power to include in the published AJPA article. However, these extra specimens are included in the dataset for completeness.