This chapter focusses on the probability of a caries lesion detected during a clinical examination being active (progressing) or arrested. Visual and tactile methods to assess primary coronal lesions and primary root lesions are considered. The evidence level is rated as low (R(w)), as there are few studies with proper validation. The major problem is lack of an accepted clinical gold standard. Evidence from high-quality basic research and epidemiological, clinical and intervention studies is therefore discussed. High-quality basic research has mapped the patho-anatomical changes occurring in response to cariogenic plaque as well as lesion arrest. Based on this understanding, different clinical scoring systems have been developed to assess the severity/depth and activity of lesions. A recent system has been devised by the International Caries Detection and Assessment System Committee. The literature suggests that there is a fair agreement between visual/tactile external scripts of caries and the severity/depth of the lesion. The reproducibility of the different systems is, in general, substantial. No single clinical predictor is able to reliably assess activity. However, a combination of predictors increases the accuracy of lesion activity prediction for both primary coronal and root lesions. Three surrogate methods have been used for evaluating lesion activity (construct validity); all have disadvantages. If construct validity is accepted as a 'gold standard', it is possible to assess the activity of primary coronal and root lesions reliably and accurately at one examination by using the combined information obtained from a range of indicators--such as visual appearance, location of the lesion, tactile sensation during probing and gingival health.