The optimal treatment of uveal melanoma would destroy the intraocular tumor, retain good vision, and avoid the development of metastatic disease. Unfortunately, inherent limitations make the development of such an optimal therapy unlikely in the near future. 1 One, approximately two thirds of choroidal melanomas examined in most ophthalmic oncologic practices are within 3 mm of the optic nerve or fovea. Even with a very focused type of treatment-delivery system, to avoid a marginal miss, usually such eyes have significant long-term visual damage. Two, whereas only 1% of patients present with simultaneously detectable metastases and a newly discovered uveal melanoma, the majority of patients who metastasize have already developed micrometastatic disease before the detection of their intraocular lesion. At present, there is no effective treatment of liver metastases from uveal melanomas. Three, in approximately 20% to 35% of patients, at the initial presentation, the tumor is too large for treatment with current eye salvage techniques, and these eyes are treated with primary enucleation.In this issue, several different radiation delivery systems are presented, including those that rely on brachytherapy (radioactive plaques) as well as teletherapy (protons, intensity-modulated conformal therapy, and cyberknife). Over 10,000 patients with uveal melanoma have been treated around the world with charged particles (helium ions and protons).There are 4 theoretical advantages of proton radiation. One, because of the inherent Bragg peak, the lateral and distal falloff with this form of treatment results in more focused radiation delivery than any other form of ionizing radiation. In most proton facilities, the lateral and distal falloff 41