Habitat fragmentation can alter the trophic structure of communities and environmental conditions, thus driving changes in biodiversity and ecosystem functions. Quantifying niches of generalist predators can reveal how fragmentation alters ecosystems. In a habitat fragmentation experiment, we used stable isotopes of a generalist predator skink to test predictions from spatial theory on trophic structure and to quantify abiotic changes associated with fragmentation among continuous forest, fragments, and matrix habitats. We predicted that in fragments and the matrix, isotopic niches would shift due to decreases in skink trophic positions (δ15N) from reductions in trophic structure of arthropod food webs and abiotic changes over time (δ13C) relative to continuous forest. Contrary to theoretical predictions, we did not find evidence of reductions in trophic structure with fragmentation. In fact, skink δ15N values were higher in the matrix and fragments than continuous forest, likely due to changes in distributions of a detritivorous prey species. In addition, δ13C values in the matrix decreased over years since fragmentation due to abiotic changes associated with matrix tree maturation. We show how isotopic niches are influenced by fragmentation via shifts in biotic and abiotic processes. The potential for either or both spatial and abiotic effects of fragmentation present a challenge for theory to better predict ecological changes in fragmented landscapes.