Nest abandonment prior to laying is a behaviour that may have evolved in response to predation risk in birds, but is poorly understood.  For example, the grey fantail (Rhipidura albiscapa) has been shown to build and abandon multiple nests through a breeding season without laying eggs. It has been suggested that this costly strategy has evolved to avoid even greater costs of losing eggs/chicks because this species experiences high nest predation.  We tested this hypothesis by placing artificial nests in the exact locations as natural nests from the previous breeding season (spanning a large altitudinal gradient) where the fate was known for each natural nest (abandoned, predated, or fledged).  Trials were conducted early and late in the breeding season to test for temporal patterns.  If nest abandonment does indeed reduce predation risk, then artificial nests placed at previously abandoned nest sites should have a greater risk of predation than nests placed at predated or fledged nest sites.  Because nest predation tends to be lower at high elevations, we expected nests at high elevation and nests with greater concealment would have higher survival.  We observed that artificial nest predation did not vary by previous nest fate, but there was an interaction between elevation and season: predation was similar initially, but increased at high and decreased at low elevation late in the breeding season. We suggest this change in predation may be due to partial migration of bird predators to high elevation to escape high late season temperatures at low elevation.