Recent and historical biomarkers assess chronic or subchronic exposure to fluoride. The most studied recent biomarkers are nails and hair. Both can be non-invasively obtained, although collection of nails is more accepted by the subjects. External contamination may be a problem for both biomarkers and still needs to be better evaluated. Nails have been more extensively studied. Although the available knowledge does not allow their use as predictors of dental fluorosis by individual subjects, since reference values of fluoride have not yet been established, they have a strong potential for use in epidemiological surveys. Toenails should be preferred instead of fingernails, and variables that are known to affect nail fluoride concentrations - such as age, gender and geographical area - should be considered. The main historical biomarkers that could indicate total fluoride body burden are bone and dentin. Of these, bone is more studied, but its fluoride concentrations vary according to the type of bone and subjects' age and gender. They are also influenced by genetic background, renal function and remodeling rate, variables that complicate the establishment of a normal range of fluoride levels in bone that could indicate 'desirable' exposure to fluoride. The main issue when attempting to use bone as biomarker of fluoride exposure is the difficulty and invasiveness of sample collection. In this aspect, collection of dentin, especially from 3rd molars that are commonly extracted, is advantageous. However, mean values also span a wide range and reference concentrations have not been published yet.