The use of local seed sources for revegetation is accepted practice to reduce the potential that propagules will be poorly adapted to site conditions. However, data are often lacking to determine the distance within which seed sources represent local genotypes. Short-term reciprocal transplant studies represent a class of tools to detect local adaptation of target species. We conducted a reciprocal transplant of Nassella pulchra between two central California locations to test for adaptation to local environmental conditions over a 3-year period. Experimental plots at one location were split between grazed and ungrazed sites to evaluate the potential influence of livestock grazing on the detection or magnitude of local adaptation. During each year of the study, evidence of a home-site advantage depended on the location, traits studied, and population. At the end of the 3-year study period, however, we detected consistent evidence of a home-site advantage for seedling biomass among grazed sites at one location and ungrazed plots at the other location. In effect, local adaptation was only apparent in the final year of the study. Short-term reciprocal transplant studies are an effective tool to guide the selection of seed sources most likely to germinate and to become established at a restoration site, but such studies cannot rule out local adaptation, which may not be immediately detectable.