Background: The clinical differentiation between lentigo senilis/initial seborrheic keratosis and lentigo maligna on the face can be difficult. Objective: Our purpose was to determine whether dermatoscopy (eg, skin surface microscopy at 10× magnification) can reliably differentiate between these entities. Methods: Dermatoscopic slides of 87 consecutive patients presenting 37 malignant and 50 benign pigmented skin lesions on the face were analyzed with the use of 27 dermatoscopic criteria. Results: Univariate analysis selected two criteria specific for lentigo maligna: asymmetric pigmented follicular openings and dark (brown or black) rhomboidal structures. Location-specific importance in relation to facial location was attributed to the color “slate-gray,” especially in combination with structures such as dots, globules, streaks, and homogeneous areas. Multivariate analysis (logistic regression model) revealed the 4 most important features to be asymmetric pigmented follicular openings, dark rhomboidal structures, slate-gray globules, and slate-gray dots with a sensitivity of 89% and a specificity of 96%. Conclusion: Three conclusions can be drawn from our study: With a set of 4 dermatoscopic features, early lentigo maligna can be detected with high accuracy; dermatoscopic features on the face differ from criteria used in other locations; and our progression growth model for lentigo maligna delineates the different steps of malignant growth in lentigo maligna. (J Am Acad Dermatol 2000;42:25-32.)