Voyager disk-resolved images of Triton in the violet (0.41 μm) and green (0.56 μm) wavelengths have been analyzed to derive the photometric characteristics of terrains on Triton. Similar conclusions are found using two distinct but related definitions of photometric units, one based on color ratio and albedo properties (A. S. McEwen, 1990, Geophys. Res. Lett. 17, 1765-1768), the other on albedo and brightness ratios at different phase angles (cf. P. Lee et al.; 1992, Icarus 99, 82-97). A significant diversity of photometric behavior, much broader than that discovered so far on any other icy satellite, occurs among Triton's terrains. Remarkably, differences in photometric behavior do not correlate well with geologic terrain boundaries defined on the basis of surface morphology. This suggests that in most cases photometric properties on Triton are controlled by thin deposits superposed on underlying geologic units.Single scattering albedus are 0.98 or higher and asymmetry factors range from -0.35 to -0.45 for most units. The most distinct scattering behavior is exhibited by the reddish northern units already identified as the Anomalously Scattering Region (ASR) by Lee et al., which scatters light almost isotropically with g = -0.04. In part due to the effects of Triton's clouds and haze, it is difficult to constrain the value of θ̄ , Hapke's macroscopic roughness parameter, precisely for Triton or to map differences in θ̄ among the different photometric terrains. However, our study shows that Triton must be relatively smooth, with θ̄ less than 15-20°, and suggests that a value of 14° is appropriate. The differences in photometric characteristics lead to significantly different phase angle behavior for the various terrains. For example, a terrain (e.g., the ASR) that appears dark relative to another at low phase angles will reverse its contrast (become relatively brighter) at larger phase angles. The photometric parameters have been used to calculate hemispherical albedos for the units and to infer likely surface temperatures. Based on these results, we determine that all but the most southerly regions (≃south of the equator) of the reddish northern terrains are likely to have been covered with deposits of nitrogen frost at the time of the Voyager flyby, in agreement with the suggestion from the photometry that these units are overlain by a thin veneer of material.