Translocation is an increasingly popular conservation management activity worldwide, but the success of translocation is often not measured or reported. A population of the endangered Macquarie perch was imperilled by the damming in 1977 of the Queanbeyan River, near Canberra in south-eastern Australia. In November 1980, 66 adult Macquarie perch (309–389-mm total length) individuals were collected from the newlyformed reservoir, and translocated approximately 4 km upstream into the Queanbeyan River past a waterfall (which prevented access to spawning habitat). Five years of post-translocation monitoring at the release sites resulted in the capture of only a single individual in late 1981. Consequently, monitoring ceased because the translocation was assumed to have failed. However, subsequent angler reports and a preliminary survey in 1991 confirmed that some translocated fish had survived, and a small recruiting population had established. More intensive follow-up surveys and subsequent monitoring from 1996 to 2006 demonstrated an established population with consistent recruitment until 2001. However, after 2001, there was no evidence of recruitment and the population is now undetectable, with the prolonged ‘millennium drought’ (1997–2010) being the most plausible cause. The present study demonstrates the potentially ephemeral nature of assessments of success and failure, and the importance of targeted long-term monitoring programs.