We have conducted a large-scale experimental manipulation of vertebrate assemblages in semi-arid Chile for almost 28 years. Established to assess the relative influence of predation, interspecific competition, and herbivory on the structure and composition of a dryland biota, we selectively exclude vertebrate predators (mammals, raptors), the ecologically dominant degu (Octodon degus), both, or neither (control) from replicate 0.56 ha exclosures; in 2001 we converted degu exclusions to all-small-mammal exclusions. Whereas this research has documented only modest and transitory top-down regulation by predation and limited evidence for negative interspecific competition among small mammals, it has revealed subtle to strong influences of small mammals on ephemeral and shrub vegetation, as well as on soil mycorrhizae and bacteria. In contrast to the limited role of biotic influences on small mammals here, episodic high-rainfall/ENSO periods and resulting vegetative growth release small mammals from bottom-up regulation, allowing for demographic expansion. Since about 2002, both mean annual rainfall and interannual variation in rainfall have decreased. In response, degus have increased numerically at the expense of other small mammals, and now regularly comprise 60-85% of small mammal biomass, with cascading effects on plants. Degus are important prey for culpeo foxes (Lycalopex culpaeus), and analysis of culpeo diets over 23 years has documented the first Type IV functional response of a vertebrate to changes in prey numbers. This research underscores the value of long-term research in understanding the broader impact of transient or modest influences on the ecology and behavior of dryland mammals.