The Arboretum of the Croatian Academy of Sciences and Arts with its historic gardens is considered to be one of the most valuable works of landscape architecture in Croatia. The Arboretum was established on the former land estate of the Dubrovnik patrician Gučetić-Gozze in 1948 and has been maintained for more than 500 years. The first written record dates back to 1494 which also reveals the year of its foundation where a contract was drawn up in which Ivan Marinov Gučetić ordered from the Korčula's stone carvers Bartul and Fran Karlić various carved parts for his summer villa in Trsteno. From the very founding of the estate at the end of the 15th century to today, depending on the purpose and wishes of the owners, hundreds of plant species have been planted. The introduction of many plants over the years reveals the image of the constantly evolving historical gardens of the Arboretum. The plant collections of the Trsteno gardens have evolved through historical stages and therefore in the present time chart the entire evolutionary course. From the modest, renaissance choices, including a few baroque novelties, to the introduction of the richness and diversity of new species from all continents during the 19th and 20th centuries, the garden began to look like a genuine arboretum. The collection of cultivated plants, most of which can be found in the historic gardens accompanying architectural and garden components, is of particular significance. It should therefore be immediately revitalized, with special focus being paid to the plants from the Renaissance period, which needs to be protected from further decay. Particular plants have different requirements for nutrients and especially soil reaction plays an important role for its uptake. Inadequate selection of soil and/or plants, regarding the soil pH (soil reaction), eventualy results in poor appearance and decay of plants. The reason for this is the lack of some biogenic elements that become unavailable to plants at certain soil reactions, which causes various physiological disorders that manifest themselves in visual symptoms: deformation of leaves, deformation and colour changes of flowers, various chlorosis and plant necrosis, eventual decay of the whole plants. The aim of this research is to determine the plant species whose origin dates back to the Renaissance period and which make an expressive design element in the Dubrovnik Renaissance garden. Five plant species were identified and selected: grape vine (Vitis vinifera L.), bitter orange (Citrus aurantium L.), olive tree (Olea europaea L.), boxwood (Buxus sempervirens L.) and bay tree (Laurus nobilis L.). The grape vine has been selected because of its indispensable qualities as a climber in the Dubrovnik Renaissance garden. The bitter orange was selected because it has traditionally been grown in Dubrovnik, the olives because of its economic importance in the history of this garden. Boxwood and bay tree were selected as important forming element of garden: boxwood because it is very important plant in development of the garden, especially in the baroque period, and the bay tree because it exists in the garden from its inception. After the selection of the plant species, data collection was undertaken in the field, which included various elements: a record of visual observations in the Arboretum, an assessment of the health status of the plant species according to the intensity of physiological damage – chlorosis (grades 0-5 or 0-100 % of chlorosis) and the field measurements of soil reaction using soil reaction model (pHH2O) (class A-very acidic to E-very alkaline). The grape vine has been graded according to visual evaluation with 3, 4 and 5, bitter orange with 4, 5 and 5, olive tree 2, 3 and 4, boxwood 2, 2 and 3, bay tree with 2, 3 and 3. All the soil under the selected plants showed a neutral to slightly alkaline reaction. Based on the field research, micro-locations were selected for the purpose of this research, where average samples of soil and plant material were taken. Soil was sampled in autumn 2010 at two depths (0-30 and 30-60 cm) in the width of the crown of three plants of all 5 plant species to determine soil reaction and soil nutrients status. For each plant species, a series of plant material sampling was undertaken from plants showing acute or latent chlorosis symptoms. The sampling has been carried out on three occasions (autumn 2010, spring 2011 and autumn 2011). Statistical analysis was used to estimate the correlation between intensity of chlorosis and soil reaction (pHH2O) with biogenic elements by Pearson’s correlation coefficient (n=9) for each researched plant species. A canonical discriminant analysis was also carried out for each researched plant species in order to estimate the multivariate distance between the chlorosis grades assigned by the visual field observation based on the results of the chemical analysis of the biogenic elements (N, P, K, Mg and Ca). The results of studying the rich literary material showed that all 5 selected plant species were planted in the Renaissance period in the Trsteno garden. The results of chemical analyses have shown that the nutrition status of grape vine, bitter orange and olive tree is not optimal. The numerous imbalances, which cause various physiological disorders, were noted. In the case of boxwood and bay tree the values of most elements are within optimal values, and the good health status of boxwood and bay tree also confirms low grades according to the intensity of physiological damage – chlorosis on the field. In grape vine there is an imbalance between the ions on alkaline soil or the lack of potassium, and the excess of calcium and magnesium. The correlation between soil reaction, biogenic elements and chlorosis also indicated the problem of nutrient imbalance overall. The correlation between pHH2O and chlorosis was also noted in the strong positive correlation (r=0,78, p<0,0136) and that chlorosis is increased by increasing the amount of phosphorus (r=0,81, p<0,0086), while phosphorus increases by increasing the amount of dry matter in the leaves of grape vine (r=0,78, p<0,0135). There is also a negative correlation between K and Ca and K and Mg and a positive correlation between Ca and Mg. The canonical discriminant analysis found a significant difference between grades 3 and 4, 4 and 5 and 3 and 5, and phosphorus had the greatest impact on this difference. In the case of bitter orange, the comparing of the correlation between biogenic elements, pHH2O and chlorosis showed that the pHH2O is in positive correlation with the intensity of chlorosis (r=0,96, p<0,0001) and that the higher dry matter content increases the Ca in the leaves and the higher Ca content decreases N. The antagonistic relationship was confirmed by the negative strong correlation of K and Mg (r=-0,90, p<0,0011). An insufficient level of zinc and manganese was also confirmed. This could be related to the pHH2O of the analysed soil which is 8.11. The canonical discriminant analysis revealed a significant difference between grades 4 and 5 according to the results of the chemical analysis of leaf. In the case of the olive tree, negative medium strong correlations between dry matter and potassium (r=-0,70, p<0,0341) and between potassium and the intensity of chlorosis (r=-0,71, p<0,0338) were identified. As the optimal amount of potassium and calcium was determined and the lack of magnesium, it is possible that intensive chlorosis appears due to the antagonism of those ions. No significant difference was found in the canonical discriminant analysis between the grades 2, 3 and 4. In the case of boxwood there was determined a medium strong correlation between dry matter and magnesium (r=0,80, p<0,0099) and a negative medium strong correlation between dry matter and zinc (r=-0,75, p<0,0188), which could lead to poorer enzymatic activity in the plant and less resistance to stress, especially at high and low temperatures. The canonical discriminant analysis showed no statistically significant difference between grades according to visual evaluation (2 and 3) compared to the results of the chemical analysis of the biogenic elements N, P, K, Mg and Ca. In the case of the bay tree a strong negative correlation was found between dry matter and nitrogen (r=-0,92, p<0,0004), dry matter and phosphorus (r=-0,83, p<0,0055), dry matter and potassium (r=-0,84, p<0,0045), dry matter and copper (r=-0,95, p<0,0001), and a medium strong negative correlation between dry matter and zinc (r=-0,78, p<0,0129), suggesting possible imbalance of nutrients in the future. The canonical discriminant analysis showed that there is no statistically significant difference between grades according to visual evaluation (2 and 3) compared to the results of the chemical analysis of the biogenic elements N, P, K, Mg and Ca. The lack of phosphorus in the analysed plant material of boxwood and bay tree is directly related to the lack of available phosphorus in the analysed soil as a consequence of rinsing due to the mechanical composition of the soil under these two plant species (3238% of sand). In all five selected plant species, it was confirmed that soil reaction is one of the most critical factors in selecting plants in planning and preserving the identity of the landscape. Based on the results, certain measures of revitalization according to the methods of the restoration of the historical gardens have been proposed. These include the conservation and restoration methods for grape vine, conservation, restoration and reconstruction of the bitter orange, conservation and restoration for the olive tree, conservation, restoration and partial reconstruction for the boxwood and bay tree conservation method. The proposed fertilization design based on nutrients status and plant nutrients requirements for each researched plant species involves the application of acidic organic and mineral fertilizers into soil and foliar application with a mixture of macro and micro elements on several occasions during vegetation. The outcomes of this research have demonstrated that a multidisciplinary scientific approach in the area of optimizing fertilization in historic gardens is vital and that the soil reaction is considered to be an important research parameter, which could then be utilized in the revitalization and maintenance of other historical gardens. This is also relevant in choosing plants in designing and preserving the identity of the landscape. The results also indicate that in general, and especially for historically valuable gardens, it is important to monitor the condition and plant needs in terms of their constant care and fertilization to develop and maintain designed landscape as a whole. In addition to this, an additional contribution to this dissertation is the established list of plants from the Renaissance period with archaic, Croatian and Latin names.