Annual recruitment of the NewZealand longfin eel (Anguilla dieffenbachii) hasdecreased by 75 per cent since significant levels of commercial fishing began in the early 1970s. This motivates application of a multiple-cohort bioeconomic model to a New Zealand longfin eel fishery to investigate its optimal management and ascertain the suitability of existing regulatory policy. The use of historical harvest to calculate total allowable catch is asserted to be unsustainable based on recovery dynamics. In addition, individual transferable quota systems are argued to be fundamentally flawed for the protection of longfin fisheries because of high-grading, low-surplus production and a current lack of effective stock-assessment procedures. Area closure and the spatial definition of harvest rights are attractive alternatives given the territoriality of longfins and high larval spillover. The importance of unfished reserves is reinforcedwhen significant uncertainties regarding population strength, harvest intensity and growth dynamics are considered. Restriction of exploitation to older cohorts in fished areas is demonstrated to maximise economic yield.