“…In the Sierra Nanchititla Natural Reserve (SNNR), central Mexico, 10 years of trophic niche studies have determined that the cougar (P. concolor) is an generalist predator, consuming 21 different mammal species, of which the armadillo (Dasypus novemcinctus) is the most highly predated (54% occurrence in cougar scats), followed by coati (Nasua narica, 16% occurrence) and white-tailed deer (Odocoileus virginianus, 8% occurrence), with other species contributing less than 5% (Monroy-Vilchis et al 2009a;Gómez-Ortiz et al 2011). Here, we assess the relative influence of top-down and bottom-up controls on these mammals in the endangered pineoak woodland food web, based on three a priori biological hypotheses: (1) Given that the cougar is a strict carnivore, dependent on prey availability, is dominant over resources in the environment, and shows territorial behavior (Sunquist and Sunquist 2002), we predicted that the per capita changes in cougar abundance (pca) would be positively affected by increasing relative abundance of their main prey species (interspecific interactions) and negatively by individuals of the same species (intraspecific interactions/self-regulation); (2) since focal prey species are herbivorous/insectivorous, depending on vegetation directly or indirectly (Valenzuela 1998;Aguilera-Reyes et al 2013), we predicted that seasonal normalized difference vegetation index (NDVI), as an approximation of primary productivity in SNNR, will affect the pca of prey species, resulting in bottom-up control of intermediate consumers. This index has been found to provide a strong vegetation signal and is a good indicator of plant biomass (Oindo and Skidmore 2002); and (3) the pca of the three prey species, D. novemcinctus, N. narica, and O. virginianus, will be affected by the relative abundance of cougar, indicating top-down control.…”