The wrybill (Anarhynchus frontalis
) is an endemic plover that breeds only in braided rivers east of the main divide in the South Island of New Zealand. It is threatened by a range of factors, including loss and degradation of habitat, flooding, and predation. We monitored wrybills in 2 sites in the Tekapo River and 2 in the Tasman River in the Mackenzie Basin, South Canterbury, during 3 breeding seasons (1997/98–1999/2000). We aimed to compare survival and productivity between areas with and without trapping (mammalian predator control) to determine whether predator control was associated with higher survival and/or breeding success of wrybills. In the Tekapo River, results were similar between trapped and un-trapped areas, suggesting that control had little effect. In the Tasman River, there were large differences between the two sites and trapping appeared to be beneficial; in the upper river (un-trapped), productivity and survival were very low and in the lower (trapped) site they were high. Over the whole study, 67.3% of nests hatched, and depredation was the largest cause of nest failure. Fledging success (the proportion of chicks hatched that fledged) averaged 35.4%. Losses at the chick stage were higher than at the egg stage, and there was only a weak correlation between nesting success and overall breeding success; we therefore caution against the use of nesting success as a proxy for overall breeding success. Productivity averaged 0.49 chicks fledged per pair over the whole study; when the very low values from the upper Tasman site were excluded, productivity averaged 0.61. Survival of adult male wrybills was lower than survival of females in all four study sites. Measurement of adult survival is important in determining the full effect of predator control (and in determining population trends) but is often overlooked. At the time of our study, wrybill populations in 3 of our 4 study sites appeared not to be self-sustaining and, in the absence of immigration, were in decline. A number of factors, including depredation by mammals, can affect breeding success. Trapping may be beneficial, but temporal and geographic differences in predator densities, as well as variability in other threats (such as flooding and levels of avian predation) mean that predicting when and where mammalian predator control may benefit wrybills is currently difficult.