All gamma cameras used in nuclear medicine have finite energy resolution. Therefore, they cannot perfectly discriminate scattered from unscattered radiation. Different scatter correction methods have been developed to reduce this effect, but it is better to reject scatter events in data acquisition instead of doing scatter correction afterwards. Most conventional ways of rejecting scatter events use energy windows (EWs). An alternative introduced by Milster et al. is the likelihood window (LW). We use this method in our modular gamma scintillation camera that has a 2×2 array of photomultipliers (PMTs) coupled to a NaI(Tl) crystal. We have developed yet another scatter rejection scheme, the Bayesian window (BW). In contrast to the LW and EW, the BW takes into account the simulated or actual scatter spectrum in making a decision to accept or reject a particular photon. The purpose of this paper is to compare the effectiveness of scatter rejection by BW, LW, and EW.