melanoma (OM) is the most common primary intraocular malignancy, with an annual incidence of 6 per million in the United Kingdom and in many other countries. 1 Environmental factors known to increase the risk of cutaneous malignant melanoma (CMM), such as UV radiation, appear to be less important in OM, and little is known about other risk factors. 2 Systematic studies on familial factors are also lacking. The coexistence of OM and CMM in some patients suggests a predisposition to both types. 3 Mutations in the p16 (CDKN2A) gene have been implicated in a proportion of these cases, but no strong evidence has come forth. 1 An association between OM and breast and/or ovarian cancer has also been observed, and results from breast-cancer families have implicated BRCA2 as the predisposing gene. 4 BRCA2 sequence variants were observed in families where OMs and breast/ovarian cancers were diagnosed. Among the 62 cases, 7 germline BRCA2 alterations were observed and 3 of these were thought to be associated with the disease. Only 1 of these 3 families had a history of any of these cancers. It was suggested that, in addition to BRCA2, other genes predisposing to OM and breast cancer may be found. 4 Because of the lack of population-based data on familial OM, we examine here these neoplasms using the nationwide Swedish Family-Cancer Database, including all Swedes and "offspring" born since 1932 with their parents who were recorded at birth. The parental population can assume any age in the database, but the offspring population ranges between 0 and 66 years. The database was updated in 2001 to include over 10 million individuals and over 1 million malignancies notified to the Swedish Cancer Registry between 1958 and 1998. 5 In addition to familial relationships and neoplasms, the database contains information on residential area and socioeconomic status. Familial risks were calculated for offspring separately for families where only a parent or only a sibling was affected with OM or with any other malignancy; risk from parents or offspring may suggest a different genetic mode or other etiological mechanisms. 5,6 Standardized incidence ratios (SIRs) were calculated by dividing the observed numbers of cases by the expected ones and expressed as person-years at risk based on age (5-year age groups)-, period (5-year periods)-, area of residence (2 categories)-, socioeconomic status (4 categories)and sex-specific incidence rates for all persons in the database. 7 Confidence intervals (95% CIs) were calculated assuming a Poisson distribution. 7 The annual number of cases has ranged from 50 to 90, and the total numbers of affected cases, offspring, fathers and mothers were 408, 996 and 877, respectively, in the Family-Grant sponsor: Swedish Cancer Society.