PURPOSE: The purpose of this paper was to compare the ability of fluorescence-aided caries excavation (FACE) to remove infected dentin in primary teeth with that of conventional methods. METHODS: Sixty-six extracted carious primary teeth were divided according to lesion size into 3 groups of 22 teeth. Caries excavation was carried out with a slow-speed handpiece and round burs for all groups. In the first group, caries was excavated conventionally using visual tactile criteria. In the second group, a caries detector dye was used to detect carious dentin. In the FACE group, cavities were excited with violet light (370-420 nm) and observed through a 530 nm highpass filter. Orange-red fluorescing areas were removed. Undecalcified thin slices were prepared, stained with Giemsa, and examined for presence of infected dentin using light microscopy. Four samples were lost during processing. RESULTS: Histology showed infected dentin in significantly less FACE samples (3 of 22) compared to conventional excavation (9 of 20; P=.03), but not significantly less compared to caries detector (5 of 20; P=.35). CONCLUSIONS: Within the limitations of this in vitro study, it can be concluded that fluorescence-aided caries excavation is more effective than conventional excavation in removal of infected primary dentin.