It has been known for many years that p-doped and n-doped regions in semiconductors reveal different contrasts (p-doped normally brighter than n-doped) when imaged in a scanning electron microscope (SEM). This effect could be very useful to the semiconductor industry to determine dopant concentrations at nanoscale dimensions if it could be understood and made quantifiable. Highly doped n+ and p+ samples were studied in a SEM with two different oxide thicknesses. The samples were initially studied without applying any initial treatment and then the oxide was removed by dipping in diluted HF. The samples were studied as a function of primary electron beam energy. It was found that at low primary beam energy (1–2 keV), the p-type material was brighter than the n-type for both oxide thicknesses. However, at higher primary beam energy, the contrast reversed for the sample with the thicker oxide above a primary beam energy of 2 keV. The role of the oxide in these contrast variations is explored and the consequences for the various theories are examined. A comparison is made between the results found here and published results from other techniques which also involve the emission of electrons from surfaces such as photoemission and field emission. It is concluded that oxygen plays a significant role in the dopant contrast (DC) mechanism and that the results are inconsistent with all current DC theories.