In 1994, the United States National Marine Fisheries Service (NMFS) implemented a series of time/area closures for the Gulf of Maine sink gillnet fishery to reduce the bycatch of harbour porpoises (Phocoena phocoena). The present study evaluates the effectiveness of the Mid-Coast closure area, implemented during November 1994. Rates of porpoise bycatches are analysed prior to, during and after the closure. In addition, individual vessels are tracked and the spatial distribution of fishing effort examined to determine how fishermen responded to the closure. The highest bycatch rate occurred in September in the Mid-Coast region, well before the closure. During November, fishermen concentrated much of their effort adjacent to the closed area in unrestricted waters, where bycatch occurred. The Mid-Coast closure was not in place for a long enough period, nor was it large enough, to be effective in reducing bycatch rates of harbour porpoises. The failure of the Mid-Coast closure is attributed to temporal and spatial variation in patterns of bycatch rates, and to the displacement of fishing effort and porpoise bycatch outside the closed area.