Market policies have profound implications for consumers as well as for the management of resources. One of the major concerns in fish trading is species mislabelling: the commercial name used does not correspond to the product, most often because the product is in fact a cheaper or a more easily available species. Substitution rates depend heavily on species, some often being sold mislabelled while others rarely or never mislabelled. Rates also vary largely depending on countries. In this study, we analyse the first market-wide dataset collected for France, the largest sea food market in Europe, for fish species substitution. We sequenced and analysed 371 samples bearing 55 commercial species names, collected in fishmonger shops, supermarkets and restaurants; the largest dataset assembled to date in an European country. Sampling included fish fillets, both fresh and frozen, and prepared meals. We found a total of 14 cases of mislabelling in five species: bluefin tuna, cod, yellowfin tuna, sole and seabream, setting the overall substitution rate at 3.7% CI [2.2-6.4], one of the lowest observed for comparable surveys with large sampling. We detected no case of species mislabelling among the frozen fillets or in industrially prepared meals, and all the substitutions were observed in products sold in fishmongers shops or restaurants. The rate of mislabelling does not differ between species, except for bluefin tuna. Despite a very small sample size (n = 6), the rate observed for this species (83.3% CI [36-99]) stands in sharp contrast with the low substitution rate observed for the other substituted species. In agreement with studies from other countries, this work shows that fish mislabelling can vary greatly within a country depending on the species. It further suggests that more efforts should be directed to the control of high value species like bluefin tuna.