2005ApJ...619..558T

We reexamine the photometric and spectroscopic evidence available for the star OGLE-TR-56, recently discovered to harbor a giant planet presenting transits and orbiting with a period of 1.21 days. We investigate the possibility that the observational signatures reported might be the result of ``blending'' with the light of an eclipsing binary along the same line of sight. Using techniques developed earlier we perform fits to the light curve under a variety of blend configurations, subject to all observational constraints, and we infer further properties of these possible blends. We then carry out realistic end-to-end simulations based on those properties in order to quantify the spectral line asymmetries and radial velocity variations expected from such scenarios. We confront these calculations with the observations. The results from these models are clearly inconsistent with the measured radial velocity and bisector span variations, ruling out blends and confirming the planetary nature of the companion. The example of OGLE-TR-56 serves to illustrate the sort of tests that can and should be performed on transiting planet candidates to eliminate the possibility of ``false positives'' and, in particular, line-of-sight contamination from eclipsing binaries.