Juniperus thurifera is an important component of woodland communities of dry sites within the West Mediterranean region and is characterised by a strongly disjunctive geographic range. Two subspecies were recognised, subsp. thurifera in Europe and subsp. africana in Africa. The aim of the study was the comparison of phenetic diversity to the pattern of AFLP geographic differentiation of the species described in the literature. The examination of phenetic diversity was based on the biometrical analysis of 17 populations using 12 morphological characters of cone and seed. The differences among populations were analysed using Student's t test, analysis of discrimination, UPGMA agglomeration and hierarchical analysis of variance. The majority of morphological characters differentiated at a statistically significant level between populations and between J. thurifera subsp. thurifera and subsp. africana. Three groups of populations were detected using multivariate statistical analyses. The first, well separated, is subsp. africana, while the following two concern subsp. thurifera.The morphological differentiation of populations appeared similar to that described on the AFLP. The Gibraltar Straight appeared to be the most important barrier.

Juniperus excelsa and J. thurifera are considered to originate from the same ancestor. As a result of occurring in geographically isolated refuges -J. excelsa in SE Europe and SW Asia and J. thurifera in SW Europe and NW Africa -their divergence could have taken place at the end of the Tertiary, with the climate cooling. Juniperus foetidissima occurs in SE Europe and SW Asia in territories shared with J. excelsa and is similar to and sometimes misidentified with it. The occurrence of the latter two species over the same area suggests they should be more similar to each other than to the geographically distant J. thurifera. Four populations of J. excelsa ssp. excelsa, four of J. thurifera and two of Juniperus foetidissima were examined on the basis of features of 10 cones and 10 shoots of 18-36 specimens each. Results of discrimination analysis, Mahalanobis distances and cluster analysis showed great similarity to J. excelsa and J. thurifera, while J. foetidissima was more distant. The results support the hypothesis of a common ancestor of J. excelsa and J. thurifera, but suggest another origin of J. foetidissima.

No abstract

Abstract• Introduction, Material and Methods The genetic structure and diversity of ten natural populations of Juniperus phoenicea L. from the western part of the species range have been studied using random amplified polymorphic DNA (RAPD) markers.• Results and discussion Among 10 analyzed primers only 3 reproduced consistently across successful PCR reactions and gave 45 loci. The percentage of polymorphic loci (P) and Nei's heterozygosity (H e ) have average values of 64.9% and 0.177. The average expected heterozygosity of particular populations positively correlate with latitude and negatively with altitude (C=0.556, P=0.025; C=−0.494, P=0.047, respectively). The proportion of genetic variation contributed by the differences between populations was low (G ST = 0.056). The gene flow (N m ) has an average value of 4.2, and was higher in subsp. turbinata (7.3) than in subsp. phoenicea (4.1). Significant proportion of the variation (Φ ST =0.106) was attributable to differences among populations, as revealed in analysis of molecular variance analysis of pair-wise RAPD distances. No evidence for isolation by distance was detected in Mantel test on genetic (Φ ST ) and geographic distances. European populations differed at a higher level from the African, subsp. phoenicea from turbinata (3.97% and 3.14% of total variance, respectively). The significant level of differences between European and African populations can result from (1) the earlier divergence and considerably low level of gene flow between them, or (2) a different mutation rate within population of different continent.• Conclusion The results suggest rather local forest economy with J. phoenicea, without seed exchange on large distance.

No abstract