TL;DR
This study introduces a novel geometric morphometric approach to accurately distinguish between sheep and goat remains, specifically focusing on third lower molars and mandibles. This method offers a more objective and quantitative alternative to traditional visual identification, crucial for understanding ancient herding practices.
What Problem Does This Paper Address?
Sheep and goats, while distinct species (Ovis aries and Capra hircus), are often herded together and exhibit significant morphological similarities, making it challenging to differentiate their skeletal remains in archaeological contexts. This ambiguity hinders our ability to accurately reconstruct past agricultural and pastoral economies. While discrete morphological criteria exist for some bones and teeth, and quantitative methods have been applied to post-cranial elements, the identification of isolated third lower molars and mandibles has remained a persistent challenge. This paper aims to bridge this gap by developing and validating a geometric morphometric protocol for precise species identification from these commonly found skeletal elements.
Methods
The researchers employed 2D geometric morphometrics, a technique that quantifies shape by analyzing the positions of homologous landmarks and sliding semi-landmarks on an object. For this study, they used two main approaches:
- Landmark-only analysis: This simpler method uses a defined set of specific points on the teeth and mandibles.
- Complete protocol: This more comprehensive approach incorporates both landmarks and sliding semi-landmarks, which are points that can move along curves to better capture subtle shape variations.
After data acquisition, Generalized Procrustes Analysis (GPA) was used to align and standardize the shapes, removing variations due to position and scale. Principal Component Analysis (PCA) was then employed to visualize shape variation, and Linear Discriminant Analysis (LDA) with cross-validation was used to assess classification accuracy. The study also investigated the influence of size, form (size and shape combined), allometry (size-shape relationships), age, and sex on species identification.
Datasets
The study utilized two primary datasets:
- Modern reference collection: This dataset comprised 102 mandibles and 133 third lower molars from 143 modern specimens. These specimens represented 6 breeds of sheep (Ovis aries) and 5 breeds of goats (Capra hircus).
- Archaeological specimens: This dataset consisted of 32 third lower molars from the Middle Ages, recovered from the site of Missignac-Saint Gilles le Vieux, Aimargues, France.
Key Insights and Ideas
- Shape is King (for identification): While size plays a role, the shape of both the third lower molar and the mandible proved to be the most powerful discriminator between sheep and goats. This highlights how subtle morphological differences, captured by geometric morphometrics, can be highly informative.
- Isolated teeth are valuable: The study demonstrates that even isolated third lower molars, often overlooked due to identification difficulties, can be reliably identified using this geometric morphometric approach. This significantly expands the utility of dental remains in zooarchaeological studies.
- Age and Sex are Secondary Factors: Despite known sexual dimorphism and age-related changes in these animals, these factors had a surprisingly limited impact on the species identification accuracy when using the developed protocols. This suggests the core species-specific shape differences are robust.
- Archaeological vs. Modern Discrepancies: Applying the method to archaeological samples revealed that geometric morphometric identifications didn't always perfectly align with traditional visual identifications. This points to the need for caution when interpreting archaeological data and suggests potential biases in older identification methods.
Key Results
- High Classification Accuracy: The geometric morphometric approach achieved impressive classification accuracies:
- Third lower molar shape: 93.3%
- Mandible shape: 95.2%
- Mandible size: 84.0%
- Third lower molar size: 62.7% (less discriminatory than shape)
- Shape Outperforms Size: Shape-based analyses consistently yielded higher classification accuracies than size-based analyses for both teeth and mandibles.
- Archaeological Sample Analysis: The 32 Middle Ages third lower molars from Aimargues were identified as a mixed assemblage, predominantly sheep (62.5%) with a notable proportion of goat (37.5%). This differed from previous visual identifications, highlighting the potential of the new method.
- Archaeological Specimens are Smaller: The identified archaeological specimens were generally smaller than their modern counterparts, regardless of species.
Limitations and Future Work
The authors acknowledge several limitations and suggest future research directions:
- Modern Breed Representation: While 13 modern breeds were used, they may not fully capture the diversity of ancient sheep and goat populations. Future studies could benefit from a broader range of breeds and historical samples.
- Archaeological Identification Discrepancies: The partial congruence between geometric morphometric and visual identifications of archaeological specimens warrants further investigation. The authors suggest that future research should incorporate molecular identification methods (e.g., ancient DNA, palaeoproteomics) to validate the geometric morphometric results and assess their reliability across different time periods and geographic regions.
- Protocol Refinement: While the "complete protocol" (including sliding semi-landmarks) generally offered higher accuracy, the simpler "landmark-only" protocol also provided good results and is faster to implement. Further research could explore optimizing the balance between accuracy and efficiency.
- Broader Application: The developed protocol shows promise for identifying isolated teeth, which are abundant in archaeological contexts. Future work should focus on applying and validating this method across a wider range of archaeological sites and time periods to build a comprehensive understanding of past caprine husbandry.
Summary updated