Data from - A new tale of lost tails: correlates of tail breakage in the worm lizard Amphisbaena vermicularis

Predator-prey interactions are important evolutionary drivers of defensive behaviours, but they are usually difficult to record. This lack of data on natural history and ecological interactions of species can be overcome through museum specimens, at least for some reptiles. When facing aggressive interactions, reptile species may exhibit the defensive behaviour of autotomy by losing the tail, which is also known as ‘urotomy’. The inspection of preserved specimens for scars of tail breakage can reveal possible ecological and biological correlates of urotomy. Herein we investigated how the probability of urotomy in the worm lizard Amphisbaena vermicularis is affected by sex, body size, temperature, and precipitation. We found higher chances of urotomy for specimens with larger body size and from localities with warmer temperatures or lower precipitation. There was no difference in urotomy frequency between sexes. Older specimens likely faced – and survived – more predation attempts through their lifetime than smaller ones. Specimens from warmer regions might be more active both below- and aboveground, increasing the odds to encounter predators, and hence urotomy. Probability of urotomy decreased with increased precipitation. Possibly, in places with heavier rainfall worm lizards come more frequently to the surface when galleries are filled with rainwater, remaining more exposed to efficient predators, which could result in less survival rates and fewer tailless specimens. This interesting defensive behaviour is widespread in squamates, but yet little understood among amphisbaenians. The novel data presented here improve our understanding on the correlates of tail breakage and help us to interpret more tales of lost tails.
Methods
All analyses in this study were performed in the software R v. 3.5.3. The dataset is thus composed of two files:

1. R_script.R

2. raw_data.csv

The R-script is commented by the authors and cointains all code used to manipulate and analyse data tables, and create the figures in the manuscript.