Previous studies have shown that congenital as well as acquired melanocytic naevi indicate an increased risk for developing melanoma in individual patients. Most studies stress the importance of melanocytic naevi as melanoma markers, while in some studies they are considered to be direct melanoma precursors. The latter opinion is favoured by the common co-occurrence of melanoma and naevus within one biopsy. The present study examines the question of whether the co-occurrence of melanoma and naevus is a random event or whether melanomas significantly co-localize with pre-existent naevi, which would suggest a precursor role for these naevi. Seven hundred biopsies of primary melanoma were examined for the presence of congenital or acquired naevi according to standard histological criteria. A naevus was found in 143 of the 700 biopsies (20.4%), of which 90 were acquired (12.9%) and 53 were congenital (7.6%). Within each biopsy the exact location of the melanoma and the naevus was determined using an ocular micrometer at a final magnification of 20 x. From these data the frequency of finding a naevus with increasing distance from the melanoma margin was calculated. The frequency of finding a naevus decreased from 2.6% immediately at the melanoma border to 0.3% at a distance of 4.5 mm and 0% at a distance of 5.0 mm. This decrease was statistically highly significant (P<0.001). Similar results were obtained when congenital and acquired naevi were evaluated separately. These data strongly indicate that melanoma and naevi are non-randomly distributed and that both congenital and acquired naevi may be precursors of melanoma.