To prepare photographic quality prints, an optimization of surface structure of the print is very important. Conventionally, the structure has been evaluated in terms of the geometric profile obtained with a surface roughness-measuring device. However, the geometric profile does not directly reflect the appearance of the print surface. Therefore, by simulating the unconscious visual operation, a microscopic surface analyzing method has been devised: a print surface was illuminated with a light flux transmitting through a slit, and the distribution of surface reflected light was captured by an area sensor placed in the specular direction. Evaluation of various print surfaces by this method proved that the surface can be characterized in terms of the following two indices; the average width and the gravity variation of the reflected image. This optical analysis correlates well with the subjective texture of print surface.