Disturbance is one of the central concepts explaining how diversity arises and is perpetuated in ecological time. A good model system for testing hypotheses related to disturbance is the longleaf pine ecosystem in the southeastern U.S. because in this ecosystem frequent, low-severity fires acts as a disturbance that maintains a unique vegetation structure and high species richness. Vegetation structure influences animal distributions; in fire-dependent ecosystems many animals rely on open-structured, fire-maintained vegetation but shrubs and trees encroach into fire-dependent ecosystems where fire has been excluded. Prescribed burning and mechanical removal are commonly used as restoration tools to control encroachment. To better assess and compare the restoration potential of these tools, a more thorough understanding of how they change vegetation structure and habitat suitability for animals is necessary. The southeastern U.S. is a diversity hot-spot for amphibians, many of which require ephemeral wetlands embedded in longleaf pine uplands for the aquatic phase of their life cycle. Amphibian diversity has been declining in recent decades and habitat loss/degradation has been cited as one of the leading causes. Although often overlooked in studies of fire ecology, the ephemeral wetlands required by many amphibians are also fire-dependent habitats that have been negatively impacted by lack of fire. To understand how disturbance interacts with wetland vegetation and aquatic-phase amphibians, three disturbance treatments meant to mimic the effects of natural disturbance on vegetation structure were applied randomly to 28 dry ephemeral iii wetlands in the Lower Coastal Plain of South Carolina, U.S. The treatments consisted of early growing-season prescribed fire, mechanical vegetation removal (a proposed fire surrogate), and a combination of mechanical removal plus fire; some sites were left untreated for reference. Vegetation structure was quantified and amphibian assemblages were monitored before and after treatments. In addition, one species of amphibian was used in a tadpole survival experiment to examine differences in performance among treatments. Other factors that could be affected by treatments and in turn influence amphibians were measured, including water chemistry, wetland depth, quantity and quality of epilithon, and leaf litter composition. Amphibian survival was lowest, and species depauperateness highest in untreated wetlands. Depauperateness of species whose range was restricted to the range of longleaf pine was lowest in sites that had mechanical treatment plus fire. The mechanical plus fire treatment created the most open vegetation structure with lowest leaf litter accumulation, especially of hardwood litter, conditions correlated with high amphibian survival and diversity. When data from this study was combined with data from a previous study of similar nearby wetlands, a pattern emerged in which one suite of species was absent from recently burned sites, while an entirely different suite of species was absent from long-unburned sites. This evidence suggests that disturbance is related to a shift in amphibian assemblage possibly due to changes in vegetation structure and perhaps wetland ecology in general, from an algal-based system maintained by frequent fire to a detrital-based system that develops in the absence of fire