Methods

Capture-Mark-Recapture was conducted on square or rectangular grids of parallel trap lines spaced at a set distance apart. In the study by Hubert (1982), as well as during the 1984-1986 period (Granjon 1987), the grids comprised 21 lines (A-U) of 21 traps (1-21), with an inter-traps and inter-lines distance of 10 m, thus covering a 200m x 200m (4 ha) area. In the 1998-2001 period (Bâ 2002), a smaller grid was sampled, consisting of four lines (A-D) of 10 traps (1-10), with an inter-traps distance of 10 m and an inter-lines distance of 20 m, yielding a grid area of 0.54 ha (60m x 90m). Between 2007 and 2012, the grid comprised 15 lines (A-O) of 19 traps (1-19), with inter-traps and inter-lines distances of 10 m, hence a grid area of 2.52 ha (140m x 180m). Traps were placed at the best available spot (e.g., under a bush, along a fallen tree, or in front of an apparently active burrow) within ca. 1 m distance of the grid node. In all　instances, trapping sessions lasted 6 days, including 5 nights during which M. erythroleucus captures occurred. The traps used were wire-meshed, single-capture live-traps (Manufrance© before 1998, then locally-made), (re)baited every afternoon with peanut butter, and checked every morning. At first capture, each rodent was individually marked by toe and/or ear clipping, and the following data were recorded: location on grid, weight (to the nearest gram, using a Pesola© brand scale), sex; position of testes for the males (scrotal or abdominal); for the females, vagina perforate or not, nipples small or large (and possibly lactating), pregnant or not (as assessed visually or via palpation). The remarks “young adults” and “juveniles” have been mentioned at one occasion, in April 1986, for individuals that were not weighed. These field observations were based on size and general appearance of the individuals concerned and are only indicative of a general age class, in the absence of any quantitative data. On recapture during the same trapping period, only the number and location were routinely recorded.