Although benign and atypical moles are considered key melanoma risk factors, previous studies of their influence were small and/or institution-based. We conducted a population-based case-control study in the state of New Hampshire. Individuals of ages 20-69 with an incident diagnosis of first primary cutaneous melanoma were ascertained through the New Hampshire State Cancer Registry. Controls were identified through New Hampshire driver's license lists and frequency-matched by age and gender to cases. We interviewed 423 eligible cases and 678 eligible controls. Host characteristics, including mole counts, were evaluated using logistic regression analyses. Our results showed that pigmentary factors, including eye color (OR = 1.57 for blue eyes compared to brown), hair color (OR = 1.85 for blonde/red hair color compared to brown/black), freckles before age 15 (OR = 2.39 for freckles present compared to absent) and sun sensitivity (OR = 2.25 for peeling sunburn followed by no tan or a light tan and 2.42 for sunburn followed by tan compared to tanning immediately), were related to melanoma risk; these associations held after adjustment for sun-related factors and for moles. In analyses confined to skin examination participants, the covariate-adjusted effects of benign and atypical moles were moderately strong. Compared to 0-4 benign moles, risk increased steadily for 5-14 moles (OR = 1.71), 15-24 moles (OR = 3.55) and 25 moles (OR = 4.33). Risk also increased with the number of atypical moles; compared to none, the ORs for having 1, 2-3, or 4 atypical moles were 2.08, 1.84 and 3.80, respectively. Although risk was highest for those with multiple benign and atypical moles, the interaction was not of statistical significance. Our findings, arising from the first population-and incidence-based study to evaluate atypical moles in relation to melanoma risk, confirm the importance of host susceptibility, represented by pigmentary factors and the tendency to develop benign or atypical moles, in the etiology of this disease. ' 2005 Wiley-Liss, Inc.Key words: melanoma; moles; atypical moles; pigmentary characteristics; sun sensitivity Numerous epidemiologic studies of melanoma risk have shown a modest to moderate influence of pigmentary factors.1-4 Associations with sunburn or sun exposure are also widely reported, 3,[5][6][7] although findings are inconsistent.8 Over the last 2 decades, in response to clinical reports of multiple benign and atypical moles occurring in nonfamilial melanoma patients, 9 epidemiologic studies have sought to quantify the relationship between these lesions and melanoma risk. Research results have demonstrated strong associations between melanoma risk and benign mole counts, whether self-counted 1 or assessed by examiners. 3,10-20 Although several studies have also shown strong associations between atypical moles, the putative melanoma precursors and melanoma risk, these efforts were wholly or in part institution-based and/or involved small numbers of cases. [11][12][13][14][15][16][19][20][21][22...