From myths to metrics: a preliminary new look at measuring coiled, preserved snakes

Abstract

Accurate length measurements of preserved snakes are fundamental for morphological, ecological, and taxonomic research. However, specimens stored in three-dimensional coils present a significant methodological challenge, and a lack of measurement standardization will likely lead to unquantified errors that compromise data comparability. Based on our dissatisfaction with the state of the science, we conducted a brief investigation into the accuracy and consistency associated with snout–vent length (SVL) measurements on two coiled, preserved Amazon False Coralsnakes (Oxyrhopus rhombifer). We evaluated the impact of measurement surface (dorsal vs. ventral), specimen size, implement type, and measurer experience. Five measurers with varying levels of scientific expertise took repeated SVL measurements using four different implements (three braided polyester strings and one flexible wire). Using "Objective Length" (OL), the mean of all measurements, as a reference, our results reveal a previously overlooked and substantial systematic error: dorsal SVL measurements were consistently 20–25% longer than ventral measurements on the same coiled specimen. Implement choice also affected accuracy, with one of the braided polyester strings performing best (mean error: 0.2% below OL) while the wire consistently underestimated length (2.2% below OL). Measurer experience did not predict accuracy but was associated with higher consistency in some cases. There was no overall effect of specimen size on measurement error, but individual measurers performed worse on the smaller snake. We conclude that the true length of a coiled, preserved snake is unknowable, let alone with high precision, but that the associated errors can be characterized and minimized. While we acknowledge that our limited study cannot be a definitive work on how to optimize this methodology, it is undeniable that a problem exists. We therefore advise caution when using and making measurements, and we strongly recommend that future studies explicitly report whether measurements are taken dorsally or ventrally and which with type of implement, as these factors appear to be those most critical for data comparison and reproducibility.