High-resolution melting analysis (HRM) is a novel, highly sensitive method in qPCR technology, which facilitates the detection of minor mutations, such as microsatellites (SSRs), insertions/deletions (indels) and single nucleotide polymorphisms (SNPs) down to class 4 (A/T transversion). Due to complex PCR products of codominant markers the interpretation of HRM curves offers some challenges. Explicit assignment of melting profiles to specific homozygous and heterozygous genotypes is only possible via alignment with direct sequencing or through analysis and comparison with mixed samples. In the present thesis a strategy for a systematic workflow and evaluation of HRM runs implementing such mixed samples was developed. This protocol, exclusively relying on PCR and HRM, is useful for development and analysis of codominant markers and was tested elaborately in a population study. Nineteen accessions of garden sage (Salvia officinalis L.) from the Genbank Gatersleben (Germany) were analyzed with nine newly developed markers and the results were used to calculate population genetic parameters. Expressed sequence tags(ESTs) from Salvia fruticosa were scanned for single nucleotide polymorphisms(SNPs) and microsatellites crossamplifying in other Salvia species. The population data of these markers (PIC ranging from 0.12 to 0.61) show complex overlappings and similarities of the 19 accessions at a variable broadness within the populations. Calculated genetic distances coincide with phytochemical data of the same sample set, especially in the clear separation of a new chemotype which occurs in two Romanian accessions. Detailed results, such as the Nei's genetic distances (ranging from 0.03 to 0.75) permit inference of ancestry and relationship of the various accessions. It can be concluded that HRM is easy to handle and a fast, clean and inexpensive method for development of codominant markers and genotyping.