A major science goal of future, large-aperture, optical space telescopes is to directly image and spectroscopically analyze reflected light from potentially habitable exoplanets. To accomplish this, the optical system must suppress diffracted light from the star to reveal point sources approximately ten orders of magnitude fainter than the host star at small angular separation. Coronagraphs with microdot apodizers achieve the theoretical performance needed to image Earth-like planets with a range of possible telescope designs, including those with obscured and segmented pupils. A test microdot apodizer with various bulk patterns (step functions, gradients, and sinusoids) and 4 different dot sizes (3 µm, 5 µm, 7 µm, and 10 µm) made of small chrome squares on anti-reflective glass was characterized with microscopy, optical laser interferometry, as well as transmission and reflectance measurements at wavelengths λ=600 nm and λ=800 nm. Microscopy revealed the microdots were fabricated to high precision. Results from laser interferometry showed that the phase shifts observed in reflection vary with the local microdot fill factor. This effect is not explained purely by interference between reflected fields from the chrome and glass portions. Transmission measurements showed that microdot fill factor and transmission were linearly related for dot sizes ≥5 µm. However, anomalously high transmittance was measured when the dot size is <5× the wavelength (i.e. ∼3 µm) and the fill factor is approximately 50%, where the microdot pattern becomes periodic. The transmission excess is not as prominent in the case of larger dot sizes suggesting that it is likely to be caused by the interaction between the incident field and electronic resonances in the surface of the metallic microdots. We used our empirical models of the microdot apodizers to optimize a second generation of reflective apodizer designs, specifically for demonstrating end-to-end instrumentation for planet characterization at Caltech's High Contrast Spectroscopy Testbed for Segmented Telescopes (HCST), and confirmed that the amplitude and phase of the reflected beam closely matches the ideal wavefront.