Also, just to further emphasize that reflection densitometry results from photographic papers can be influenced by many variables, here is a quote from Proposed American Standard Method of Processing for Sensitometry of Photographic Paper—Z38.2.3*
Nothing wrong with using reflection densitometry and there is usually no difficulty with mid zones and slopes (as PE pointed out) but if you are using reflection densitometry in your darkroom to include a reliable D-max in the curves, you need to make sure you account for all the variables to get useful results.

The sample should be illuminated at an angle of 45 degrees to the surface and the density
measured normal to the surface. Samples having a pronounced surface characteristic should have the densities read first in one position perpendicular to the viewing axis. Then the specimen should be rotated through an angle of 90. degrees about the viewing axis and the density measured again. The two values thus obtained should be averaged and recorded as the density for that measurement.