Tunas are the focus of significant fisheries in the Pacific Ocean, where landings of four species – skipjack tuna (Katsuwonus pelamis), yellowfin tuna (Thunnus albacares), bigeye tuna (Thunnus obesus) and albacore tuna (Thunnus alalunga) – constitute approximately 70 % of the global tuna catch. Stock assessments for skipjack, yellowfin and bigeye tunas in the Pacific Ocean currently assume eastern and western stocks. For albacore tuna, separate North Pacific Ocean and South Pacific Ocean stocks are currently assumed. In each case, these geographic definitions reﬂect the historical development of fisheries management across the Pacific rather than biological considerations. There is widespread agreement that uncertainties surrounding the stock structures of these four tuna species could have important impacts on the population dynamics models used to assess their status and inform management options. Knowledge of stock structure is also essential for improved modelling of the effects of climate change on tuna distribution and abundance and associated implications for fisheries. This paper reviews current knowledge and understanding of the stock structures of skipjack, yellowfin, bigeye and South Pacific albacore tunas in the Pacific Ocean, by exploring available literature relating to their biology, movement and spatial dynamics. As a guide for future research in this area, we identify the main uncertainties in defining the stock structure of these four tunas in the Pacific, including i) spawning dynamics; ii) the degree of spawning area fidelity and localised residency; iii) the provenance of individuals in, and proportional contributions of self-replenishing populations to, fishery catches within the Pacific Ocean; iv) linkages with adjacent ‘stocks’; v) the effects of climate change on stock structure and proportional contributions of self-replenishing populations to fisheries; and vi) the implications of improved knowledge of tuna stock structure for stock assessment and climate change model assumptions and fisheries management. We also briefly propose some approaches that future studies could use to address these uncertainties.