We report the results of chemical sourcing of obsidian artifacts from Tres Zapotes using X-ray fluorescence analysis. This is the first obsidian sourcing study for this major Olmec and Epi-Olmec center in which samples are drawn from secure archaeological proveniences specifically assigned to Early, Middle, Late Formative, and Protoclassic periods. We employed a stratified random sampling strategy to select 180 obsidian artifacts from excavated assemblages, supplementing the random sample with another 24 specimens drawn from rare visual categories. Consequently, we are able to characterize changes in the relative importance of different obsidian sources in the political economy of Tres Zapotes across the critical transition from Olmec to Epi-Olmec society with greater confidence than has been possible for the Gulf lowlands while extending our observations to the full sample of 5,713 visually characterized obsidian artifacts—2,695 of which come from the well-dated Formative contexts examined in this article. Our study confirms the absence of obsidian from Otumba and from Guatemalan sources in the excavated Olmec assemblage in favor of sources from eastern Puebla and Veracruz, supporting a model of overlapping autonomous networks for obsidian procurement at Gulf Olmec sites. Presence of the Guatemalan San Martín Jilotepeque source in Epi-Olmec contexts may relate to the reestablishment of trans-Isthmian contacts, while increasing prevalence of Zaragoza-Oyameles obsidian from eastern Puebla marks the beginning of a long-term trend. Although more even representation of obsidian sources in Epi-Olmec contexts is consistent with the hypothesized transition from an exclusionary Olmec political economy toward a more “corporate” system associated with power sharing among factional leaders at Tres Zapotes, neither Olmec nor Epi-Olmec elites monopolized a particular obsidian source or technology.