How to weigh a fossil mammal? South American notoungulates as a case study for estimating body mass in extinct clades

Body mass (BM) is a fundamental variable for many paleobiological investigations that is challenging to accurately infer for species that lack living representatives and/or close morphological analogs. This study explores this issue using notoungulates, a diverse group of extinct South American herbivorous mammals with an extensive fossil record. We use a new dataset of 1,900+ extant mammal species (from similar to 80,000 specimens) to estimate notoungulate BM based on head-body length and a published dataset of 400+ species (similar to 2,100 specimens) to estimate BM based on occipital condyle width. Condylobasal length, stylopod diameter and circumference, and neck length data are used to explore factors that can confound BM predictions. We estimate the following BM ranges for 10 osteologically well-characterized species and calculate similar ranges for 30 others known from less complete remains: Toxodontia: Thomashuxleya externa (80-120 kg), Homalodotherium cunninghami (250-350 kg), Scarrittia canquelensis (450-550 kg), Adinotherium ovinum (75-90 kg), Nesodon imbricatus (350-400 kg), and Toxodon platensis 1,000-1,200 kg); Typotheria: Interatherium robustum (1.9-2.0 kg), Miocochilius anomopodus (9-14 kg), Protypotherium australe (3.5-4.0 kg), and Pachyrukhos moyani (1.2-1.6 kg). We suggest that species such as these can be used as calibration points when inferring BM of species known from more limited remains. Discrepancies between our estimates and previously-published studies are primarily due to the distinctive craniodental morphology of notoungulates and the robust limb bones of toxodontians. There is significant, non-random error correlated with body habitus (i.e., being relatively robust or gracile) in many variables traditionally used to estimate BM, including femur circumference, and new methods are needed to compensate for this.

Automated summary

This study uses a new dataset to estimate the body mass of extinct South American herbivorous mammals, and finds discrepancies with previous studies, which are primarily due to the unique craniodental morphology of these animals. Therefore, new methods are needed to account for these differences in body mass inference.