Self-etching primers are thought to offer significant advantages over total-etch adhesive systems. The hypothesis tested in this study was that there was no difference in etching characteristics between a self-etching primer and a phosphoric acid solution at the same pH. Etching was assessed using atomic force microscopy (AFM) evaluation of site-specific changes in the height of the peritubular and intertubular dentin as a function of exposure time. Human dentin disks (n = 6/group), prepared with an acid-resistant glass reference layer, were etched with a self-etching primer and with 0.0134 M phosphoric acid (both pH approximately egual to 1.94). Depth changes relative to the reference layer were measured with the AFM after each etching interval, at 15 different locations, each in the peritubular and intertubular dentin. The total demineralization depth was measured in a scanning electron microscope. Peritubular dentin etching rate was linear while it could be measured (up to 15 s) and was greater for the self-etching primer (p < 0.0001). Intertubular dentin displayed a similar demineralization pattern with both acids, ultimately reaching a plateau in the majority of specimens. The self-etching primer attained a plateau after less recession than phosphoric acid (p < 0.0001). Dentin demineralization appears to be affected by other factors in addition to the pH of the etchant solutions.