This work provides descriptive epidemiological data of malignant mucosal and uveal melanomas and adnexal skin carcinomas in Europe as defined as in the RARECARE project. We analysed 8669 incident cases registered in the period 1995-2002 by 76 population-based cancer registries (CRs), and followed up for vital status to 31st December 2003. Age-standardised incidence to the European standard population was obtained restricting the analysis to 8416 cancer cases collected by 64 not specialised CRs or with information available only for some anatomical sites. Period survival rates at 2000-2002 were estimated on 45 CRs data. Twenty-two CRs which covered the period 1988-2002 were analysed to obtain the 15-year prevalence (1st January 2003 as reference date). Complete prevalence was calculated by using the completeness index method which estimates surviving cases diagnosed prior to 1988 ('unobserved' prevalence). The expected number of new cases per year and of prevalent cases in Europe was then obtained multiplying the crude incidence and complete prevalence rates to the European population at 2008. We estimated 5204 new cases per year (10.5 per million) to occur in Europe, of which 48.7% were melanomas of uvea, 24.8% melanomas of mucosa and 26.5% adnexal carcinomas of the skin. Five-year relative survival was 40.6% and 68.9% for mucosal and uveal melanomas, respectively. Adnexal skin carcinomas showed a good prognosis with a survival of 87.7% 5 years after diagnosis. Northern Europe, United Kingdom (UK) and Ireland showed the highest 5-year survival rate for uveal melanomas (72.6% and 73.4%), while Southern Europe showed the lowest rate (63.7%). More than 50,000 persons with a past diagnosis of one of these rare cancers were estimated to be alive at 2008 in Europe, most of them (58.8%, n=29,676) being patients with uveal melanoma. Due to the good prognosis and high incidence of uveal melanomas, these malignancies are highly represented among the long-term survivors of the studied rare cancer types. Therefore, maximising quality of life is particularly important in treatment of uveal melanoma. As regards mucosal melanomas, the centralisation of treatment to a select number of specialist centres as well as the establishment of expert pathology panels should be promoted. The geographical differences in incidence and survival should be further investigated analysing the centre of treatment, the stage at diagnosis and the treatment.