Among epigenetic aberrations, DNA methylation itself features a diverse presence. Recently, 5-hydroxymethylcytosine has been identified as a constituent of mammalian DNA and described as the sixth base of the genome. The loss of 5-hmC has been highlighted as a hallmark of melanoma by a single, remarkable study, whereas interesting clues as to the role of 5hydroxymethylcytosine are still emerging. In contrast to 5hmC, the importance of 5-methylcytosine in cancer cells is much more firmly established. Aberrant promoter DNA hypermethylation or localised methylation preferably occurs in CpG dinucleotide-dense regions, resulting in the down-regulation of the corresponding gene. It has recently become apparent that malignant melanomas feature hypermethylation, and currently more than 80 genes mainly in promoter regions are hypermethylated at a single-gene level. Taking a global view of the available data, the number of primary tumour samples involved in the studies and the frequency of positive results do not allow determining whether the hypermethylated genes described are appropriate for diagnosis or can be considered candidate therapeutic targets. Moreover, most of the data provided are derived from cell lines and estimated methylation values indirectly consisting of three steps: measuring mRNA or protein expression in cell lines, treating samples with a specific drug that acts against the process of methylation and measuring gene expression again. Nonetheless, powerful arguments have been presented in the literature that support direct experiments being less ambiguous; furthermore, most of the groups conducting direct measurements have applied candidate gene approaches. In addition to the rapid progress that has been made in studying promoter hypermethylation at the single-gene level, only two groups have attempted to conduct array-based experiments to identify the methylation pattern of thousands of gene promoters. Regrettably, one group has focused only on comparing the methylation level of primary invasive melanomas with benign melanocytes and has clearly identified a group of genes in a statistically powerful interpretation that can be used to discriminate naevi from melanomas based on their methylation signature. Another group has examined the short-term cultures of homogeneous stage III specimens. As no data are currently available regarding the methylation markers of diverse melanomas with different clinical behaviours, we performed a systematic comparison of localised methylation patterns among 42 primary melanomas using the Illumina Golden Gate Cancer Panel Bead Assay. We found differentially methylated CpGs altogether among melanoma subgroups and the majority of CpG sites were hypermethylated in melanomas that represent more favourable prognoses including a non-ulcerated tumour surface, superficial spreading histological subtype, nonmetastatic subgroup and smaller tumour thickness.