Immortelle (Helichrysum italicum /Roth/ G. Don) is a perennial plant species of the genus Helichrysum and the family Asteraceae. It is spread in the Mediterranean area inhabiting the sea coast as well as the mountains up to 2,200 meters above sea level. In Croatia, it grows along the Adriatic coast, on islands, and rocky pastures of the Dalmatian hinterland. H. italicum has been used in folk medicine for centuries due to many medicinal properties. Many studies have confirmed its anti-inflammatory, antioxidant, antimicrobial and antiviral activity. With the widespread use in the cosmetic pharmaceutical and food industries, the demand for H. italicum essential oil has increased in recent years. Immortelle makes 95% of the total wild harvested medicinal and aromatic plants in Croatia. Due to the overexploitation of immortelle wild populations, their collection was limited, which led to the development of plantation cultivation and new scientific research. This doctoral thesis aims to determine the genetic diversity and population structure of immortelle, which has not been researched for this area and also to determine the biochemical diversity. Eighteen wild populations along the Adriatic coast and islands were sampled, representing the entire natural area of its distribution in Croatia. Accessions were planted in the experimental field to avoid the influence of different environmental factors. DNA was isolated from young leaves. Genetic diversity was assessed using the AFLP molecular marker technique used to detect a large number of polymorphic markers and to examine genetic diversity and structure in species for which there is no prior knowledge of the genome. Using four combinations of primers, 693 polymorphic AFLP markers were obtained. The total phenotypic diversity of all analyzed populations calculated by the Shannon information index was Ht = 0.414. Analysis of molecular variance (AMOVA) revealed that most of the genetic diversity was attributed to differences among individuals within populations (93.08%), which is expected due to the outcrossing nature of the species. Wright’s index of genetic differentiation was significant but low (FST = 0.036), indicating extensive gene flow between populations. The unrooted Fitch–Margoliash tree based on Nei’s genetic distance matrix showed a separation of 18 H. italicum populations into two groups (supported by 96% bootstrap values) according to geographical origin. The first group included the northern Adriatic populations, and the second the populations of the central and southern Adriatic. The Bayesian model-based clustering using Structure showed the highest ΔK value for K = 2 and the second highest ΔK value for K = 3. At K = 2, the representative populations of ancestral gene pool A were mainly those from the northern Adriatic, while the populations from the central and southern Adriatic were representative of a separate gene pool B. At K = 3, a newly formed gene pool C was identified. Populations in which Q < 75% were observed as a hybrid because they were formed by mixing the two original populations. By using BAPS, two distinct ancestral clusters were obtained: A: with populations from the north and B: with the rest of the populations from the central and southern Adriatic which showed consistency with the results of the Structure at K = 2. These results are also in concordance with the Fitch–Margoliash tree that confirmed the existence of two different genetic groups. Adaptive genetic variation in H. Italicum natural populations was examined using landscape genetic methods Mcheza and BayeScan. Mcheza revealed 4.71% loci possibly under selection, among which 2.24% of loci exhibited higher FST values and 2.47% of loci exhibited lower FST values than the majority of loci. BayeScan detected the lower proportion of loci under divergent selection (2.02%) than Mcheza. The spatial analysis method Samβada detected 50 individual markers associated with bioclimatic variables, among them five were identified by both Mcheza and BayeScan as being under directional selection. A high correlation was found between 19 bioclimatic variables. Out of the 171 pairs examined, a strong positive correlation (r > 0.70) was found in 26 cases, while a strong negative correlation (r < -0.70) was found in five cases. The principal components analysis (PCA), based on a correlation matrix, showed that the first four major components together explained 93.82% of the variance. The chemical composition of essential oils for all populations was determined by gas chromatography and mass spectrometry. A total of 90 compounds were obtained of which 84 compounds were identified. Fifty compounds were isolated in 18 populations. Seventy-two compounds were present in a concentration of less than 5%. The maximum proportion of 18 essential oil compounds was higher than 5% in at least one sample from 18 populations. Populations differed significantly in nine essential oil compounds: limonene, linalool, nerol, neryl acetate, trans caryophyllene, neryl propanoate, α-curcumene, β-selinene, and δ-selinene that were taken for chemotype identification. Multivariate analyzes of chemical diversity enabled the identification of three chemotypes. Based on stepwise discriminant analysis, the most informative compounds for discriminating the three H. italicum chemotypes were: nerol, neryl propanoate, trans-caryophyllene, and β-selinene. With these compounds, after a cross-validation procedure, it was possible to correctly classify 100% of the populations into presumed chemotypes. Correlations between geographical, genetic, and biochemical distances were calculated. A significant correlation between biochemical and genetic distance was found and it was shown that 18.3% of biochemical differentiation between populations can be explained by genetic distance. Due to the increased demand for high-quality immortelle essential oil, new data about this plant species are crucial. The results of the research are a valuable contribution to the development of future breeding programs and commercial varieties of H. italicum that do not currently exist. The obtained results will also contribute to the development of conservation strategies of these valuable genetic resources.