Are there any crystallographic effects we can see in the reflection of visible light from metal surfaces? I started thinking about this in a discussion in comments.
One can start by thinking of the reflection of visible light by most metals as similar to the reflection of radio waves in that it's an interaction between an electromagnetic wave and an infinite half-plane of dense electron plasma. So long as the frequency is somewhat below the plasma frequency, this is a good starting point.  Once in a while there are strong, obvious atomic effects such as gold metal being gold in color but many/most crystalline, amorphous and molten metals have high specular reflectivity and little obvious spectral variation, i.e. most of them look more or less "silvery". 
Are there any crystallographic effects in the specular reflection from smooth metal surfaces that are visibly detectable either by looking at a reflection, or doing a simple experiment without an ellipsometer or other special equipment? Perhaps a simple polarizer film or plastic diffraction grating or something around a school science lab or home?
A visible "crystallographic effects" might be for example a difference in appearance or easy-to-observe optical behavior between 


*

*two polished faces of a metal crystal having different crystallographic orientations,

*two polished faces of two metal samples of different crystallographic grain size, or one being totally amorphous

*a polished face of a metal crystal and the surface of the metal liquid at nearly the same temperature.


or it could be something visible that "happens" at a certain angle or under a certain condition or color of illumination for one sample but not another.
 A: Maybe it partially depends on what is meant by "smooth".  A coin, viewed in nondiffuse light, often displays a speckle pattern that shifts as the coin is tilted or as the point of view is shifted.  Perhaps that pattern relates to the microcrystalline structure of the coin's metal.  This paper relates to the topic. 
A: Sticking with looking for metals with lower symmetry than usual (i.e. not fcc or bcc), one comes to metals such as Antimony and Bismuth. Pulling up an early paper (Optical Properties of Antimony and Bismuth Crystals, A.P. Lenham et al., J. Optical Soc. Americal 55(9) 1072-1074 (1965)), they measured the optical properties parallel to and perpendicular to the basal planes. 
In the figures, the difference in (n,k) in the visible looks pretty small, compared with the infrared. However, early on in the paper they note 'When viewed under a polarizing microscope, the anisotropy was pronounced.'
A question to ask, of course, is about the surface quality and the possibility of, say, oxidation. As noted in the paper,

In studies of the optical parameters of metals, great attention must be given to surface preparation. The absolute values of the optical constants are often found to be sensitive to the method of surface preparation, but the shapes of the dispersion curves and especially the energies of the absorption maxima and minima, and the sense of the anisotropy, are preserved for various sensible surface preparations.

Which, of course, gives no definitive description of what 'sensible' means. But, the bismuth was both hand polished (diamond grit) and electropolished, while the antimony was only hand polished. For thin oxides on the surface, there should be little influence on the relative optical properties, as noted above, because they will be optically thin. 
Cadmium also shows reasonable anisotropy. 
Continuing on, the authors have a variety of other papers on other materials. One you may or may not consider a metal, graphite, is covered in The Observatory (1966), showing a larger anisotropy there than for the more traditional metals. 
Again, the key is to look for non-cubic metals. In addition the 'poorer' a metal likely the better. As a counterpoint, the noble metals (Cu, Ag, Au) are all fcc and all have nearly-free-electron like Fermi surfaces, so exhibit a very high degree of symmetry. 
