The objective of this study was to assess the abrasive effects of toothpastes and acidic F gels on sound and demineralized enamel. Pairs of enamel specimens were cut from bovine incisors, embedded in epoxy resin and polished. An artificial subsurface lesion of 80–90 µm depth was created in one specimen from each pair. The samples were covered with adhesive tape, thereby exposing the enamel for abrasivity testing. All samples were divided into six groups of 15 and brushed with a slurry (1:3) of F gel or toothpaste and human saliva. Brushing with water (control) or with slurry was carried out (16,000 strokes) using a medium toothbrush (load 275 g) mounted in a brushing machine. Abrasion was evaluated using laser profilometry, and was about 50% less on sound than on demineralized enamel (p < 0.001). In the latter, brushing with water (0.09 ± 0.03 µm) or with fluoride-free gel (0.08 ± 0.03 µm) resulted in negligible wear. With a medium-abrasive paste (1.76 ± 0.85 µm) and an acidic F gel (2.48 ± 0.72 µm), brushing abrasion was significantly greater (p < 0.001) than with a low-abrasive paste (0.84 ± 0.38 µm). The greatest wear (16.6 ± 10.8 µm) was observed with high-abrasive paste (p < 0.001), and here transversal microradiography revealed a complete loss of the pseudointact surface after brushing. In vitro formed caries-like lesions can be abraded (by toothbrushing) more easily than sound enamel; hence, initial white spot lesions should preferably be brushed with oral hygiene products of low abrasivity.