| Autor: |
Strikić, Frane, Klepo, Tatjana, Rošin, Jakša, Radunić, Mira, Perica, Slavko, Šatović, Zlatko |
| Jazyk: |
angličtina |
| Rok vydání: |
2012 |
| Předmět: |
|
| Popis: |
Olive (Olea europaea L.) cultivation in Croatia, like in other Mediterranean countries, is very ancient. The estimated number of cultivars considered as autochthonous is more than 40, represent valuable germplasm resources. The aim of this study was to evaluate the efficiency of morphological traits in identification and classification of Croatian olive cultivars. Morphological variability of a set of 31 most common cultivars was assessed using UPOV descriptors for olive. A total of 32 qualitative morphological traits related to tree, leaf, inflorescence, fruit and endocarp features were analyzed. Cluster analysis based on dissimilarity matrix was performed using the UPGMA method. Shannon information index and discriminating power were calculated for each trait. A total of 88 states have been observed. Cultivars were clustered into 4 groups. The most similar cultivars were ‘Buža’ and ‘Kosmača’ differing in six out of 32 traits while the most distant pairs of cultivars were ‘Dužica’/‘Slivnjača’ and ‘Murgulja’/‘Drobnica’ differning in 22 out of 32 traits. The most discriminating trait was endocarp weight followed by tree vigor and longitudinal curvature of the blade, while the least discriminating one was length of the internodes. Shannon's information index was highly correlated with the number of states as well as with the discriminating power. Although the traits exhibiting 'rare' states are generally not suitable for efficient cultivar classification, those cultivar-specific characteristics could be used as readily observable diagnostic markers for a cultivar in question. According to our results, morphological traits provide an inexpensive and reliable method for routine screening of a large number of trees, in order to monitor and manage olive germplasm. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
