Back from burn out: are experimentally charred grapevine pips too distorted to be characterized using morphometrics?

Abstract

This article examines the impact of charring on the possibility to characterize grape pips, at compartment (wild versus domesticated) and cultivar level, using morphometrics. Two morphometric methods have been used, one based on linear measurements (traditional morphometrics) and one on elliptic Fourier transforms (EFT; morphogeometrics). Charring experiments were performed using a laboratory muffle furnace and various charring conditions. Despite the strong impact of heating, results showed that wild and domesticated Vitis seeds can be reliably discriminated using both traditional morphometrics and morphogeometrics, even when charring has been done at high temperature (450 °C). The characterization of charred pips at cultivar level using EFT is very powerful when the seeds are charred at 250 °C, but the risk of misclassification is, as expected, higher at 450 °C. Results suggest that the characterization at cultivar level should only be attempted with large assemblages of well preserved archaeological pips, and only after a first classification at compartment level. Our approach was applied on a case study of two assemblages of waterlogged and well preserved charred pips from the archaeological site of Lattara. The results are consistent both between the two morphometric methods and between waterlogged and charred remains.