Roads inhibit the movement of many terrestrial species, including mammals and reptiles, by increasing mortality or through avoidance by wildlife (Trombulak and Frissell 2000). In many cases, evidence for use of habitat is depressed near roads with effects that increase with traffic volume (Foreman 2000; Nafus et al. 2013). For terrestrial vertebrates these effects can be mitigated for through the installation of road-side fencing (Clevenger et al. 2001). While there is some evidence that road fencing can “reclaim” habitat for surrounding wildlife, there may be other deleterious effects. One such negative effect is transitioning roads into an impermeable barrier, and thus enhancing fragmentation of the habitat (Jaeger and Fahrig 2004). Fragmented wildlife populations are more vulnerable to localized extinctions through mechanisms such as demographic or environmental stochasticity and inbreeding depression (Higgins and Lynch 2001; Trombulak and Frissell 2000). Installation of wildlife over- and underpasses can mitigate for these effects and maintain population connectivity. The majority of research on roads and mitigation strategies has focused on large mammals (but see Fahrig and Rytwinski 2009) or forested environments. Very little, however, is known about creating effective underpasses for smaller terrestrial vertebrates, especially in desert ecosystems. Biological and behavioral processes can be fundamentally different depending on the habitat in which a species evolved. In contrast to species that inhabit forest habitat (Gillies and St. Clair 2008), many desert species appear to prefer open canopy areas, such as washes, as movement corridors (Baxter 1988). Consequently, understanding the behavioral processes and landscape variables that affect underpass use in a wide-variety of ecosystems is a critical necessity. We propose to assess the use of pre-existing underpasses with distinctly different qualities for small terrestrial vertebrates, including the federally-protected Mojave desert tortoise. Our first objective will be to evaluate the relative use of underpasses with different dimensions (width, height, length) and characteristics (temperature, humidity, light) using trail cameras to record species use and dataloggers for environmental data. Our second objective will be to determine if placement is predictive of relative use by terrestrial vertebrates including the Mojave desert tortoise, a variety of other reptiles, rodents, lagomorphs, skunks, foxes, bobcats and coyotes. Placement will account for distances between culverts as well as the structure of the habitat, in the form of substrate, canopy cover, and the presence or absence of a wash, at both culvert entrances. The outcome of this study will be increased understanding of best practice culvert designs for maintaining wildlife population connectivity in a greater diversity of ecosystems.