An actualistic and phylogenetic approach to identifying and interpreting crocodylian bite marks

Bite marks provide direct evidence of trophic interactions, feeding behavior, and inter- or intraspecific conflict in the fossil record. However, their utility as a source of taphonomic and paleoecologic data requires differentiation from traces left by other processes. Since the 19th century, taphonomists have often relied on actualistic observations of modern bite marks and feeding behaviors in order to identify diagnostic traces and patterns. A recent increase in interest in taphonomic research has resulted in a large body of work describing patterns of bite marks from many different clades. Most research has been focused on mammalian taxa, but a smaller number of non-mammalian groups, including crocodylians, have also drawn interest.

Crocodylians are taphonomic agents who consume and modify bones, often depositing them in the active depositional systems in which they live. However, actualistic observations of crocodylian bite marks have been limited to forensic case studies and surveys of two taxa: Crocodylus niloticus and Crocodylus porosus. Both surveys utilized captive animals, which often exhibit atypical morphologies that may bias ensuing bite mark datasets. In order to address this issue, a 2D morphometric analysis of Alligator mississippiensis crania from captive and wild specimens was performed. A principal component analysis and a canonical variates analysis revealed some statistically significant differences between the two groups, while crossvalidation had mixed resuts. An ANCOVA test of the covariance of centroid size and origin against shape (principal component scores) revealed that the effects of ontogeny introduced a stronger signal than captivity. This implies that while using captive crocodylians in fine scale analyses should be avoided, they are suitable for gross scale research, such as bite mark analyses.

To explore crocodylian bite mark patterns in greater depth, a large scale survey of traces left by A. mississippiensis was performed. Bite mark types on samples taken from individual feedings were discussed in light of vital statistics and collection protocols. Bite mark types on samples taken from group feedings were classified by location and orientation on bone and type. The results were compared to pre-existing crocodylian datasets with regards to potentially diagnostic traits: bisected marks, hook scores, and a lack of furrows. Bisected marks were found in rates similar to those seen in C. niloticus, and rates of hook scoring and bone breakage were higher. These traces were present in higher rates than those reported in C. porosus. Furrows were identified, but rare.

Finally, a survey of bite marks from 21 of the 23 generally recognized species of extant crocodylians was performed to better characterize marks found across Crocodylia and to test methods for synthesizing taphonomic datasets. Bite marks were identified, and specimens were then coded for presence or absence of mark subscores. Attempts to find statistical correlation between mark types, animal vital statistics, and sample collection protocol were unsuccessful. Mapping bite mark character states on a eusuchian phylogeny successfully predicted the presence of bisected marks in previously published, extinct taxa. Predictions for clades that may have created multiple subscores, striated marks, and extensive crushing were also generated. Inclusion of fossil bite marks which have been positively associated with extinct species allow this method to be projected beyond the crown group. The results of this study indicate that phylogenies can and should be explored further for use as predictive tools in a taphonomic framework.