Our objective was to evaluate whether or not recent mortality data for the region of Tuscany confirm the hypothesis that an epidemic in the incidence of melanoma is an apparent phenomenon reflecting an overdiagnosis of indolent cases. We considered 1755 melanoma deaths in Tuscany in the period 1987-2003, and 2644 incidence cases of melanoma diagnosed in 1985-2003 in a subset of the same population. We calculated annual mortality and incidence trends using the National Cancer Institute's Joinpoint Program (version 2.6). We observed an increasing mortality from melanoma from 1987 to 2003 in both sexes, but mainly in women (estimated annual percentage changes=2.25; P<0.05). We also observed a statistically significant rise in melanoma incidence in both sexes, mainly of thin lesions. Furthermore, we observed an increase in thick lesions, especially in females (estimated annual percentage changes=2.9; P<0.05), and for lesions without Breslow definition. In conclusion, the rise in melanoma mortality and incidence, especially of thick lesions, suggests that the observed growth in melanoma incidence is not wholly apparent.