The west Mexican cotton rat (Sigmodon mascotensis) is found from Nayarit and Zacatecas and south to Chiapas in most types of tropical vegetation, but is ecologically associated to grass, which provides food and cover.  It has been considered as a potential pest, but it is usually found in very low densities.  As it survives in disturbed areas, the species is not protected under Mexican or international laws.  The ecology of S. mascotensis has never been studied thoroughly.  The aim of this study was to obtain information about S. mascotensis’ demography based on the analysis of the population parameters and reproductive condition, and to investigate its habitat use based on 14 vegetation-structure variables.  In January 2003 to 2005, we conducted mark-and-recapture studies at Playa de Oro, Colima, Mexico.  Each year, five 100-station live-trap grids (each station with one ground and one arboreal trap; 10 by 10 configuration) were set for 8 nights in areas with dense vegetation; in addition, for each station we measured 14 vegetation-structure variables.  The results were analyzed using stepwise logistic regression (with Systat 12) and a nonparametric multiplicative regression (NPMR; with HyperNiche 1.32).  

Results/Conclusions

We trapped 95 individuals of S. mascotensis, 81 (88%) of which were adults with a sex ratio M:F of 1.00:0.76.  Most were captured at ground level (86.2%).  S. mascotensis was found on 10 of the 15 grids, with densities estimated using MARK of 0.84-24.73/ha.  The trap stations at which S. mascotensis were captured had higher percentages of ground cover formed by woody plants, forbs, and grasses; lower percentages of litter and dead wood; a lower and more open canopy; and greater distances to the nearest trees.  The resulting logistic-regression equation included 4 of the 14 variables; cotton rats preferred sites with a more open canopy, dense low vegetation, little litter, and sparse high vegetation.  NPMR does not result in an equation, but can provide a three-dimensional landscape of response to the variables.  It showed that the likelihood of occurrence of the species decreased as the canopy closed and litter increased, while likelihood of occurrence increased slightly as the nearest trees were farther away, and sharply as the percent of woody plants increased in the ground cover.