INSTYTUT DENDROLOGII

Diversity of mountain pine Pinus uncinata Ram. ex DC. (Pinaceae) in the distribution range

Pinus uncinata Ram. ex DC. (mountain pine, known also as hooked pine) was described as a species in 1805, by Louis Ramond de Carboniére, in Lamarck and De Candolle's Flore Française. The epithet 'uncinata' was caused by the hooked apophyses of the cones. P. uncinata is found in the Pyrenees and Western Alps, Massif Central, Jura, Vosges, and in isolated stands in the Sierra de Gúdar and Sierra de Cebollera. It forms forests at altitudes of 1400-2700 m above sea level.

P. uncinata has been treated by various authors as a species or a subspecies or even as a variety of P. mugo Turra (= P. montana). Two hypotheses on the origin of P. uncinata have been discussed. The first one suggests that the taxon is a result of an ancient hybridization between P. mugo (dwarf mountain pine) and P. sylvestris (Scots pine). In the other hypothesis, P. uncinata originated as a result of isolation from P. mugo.

The main goals of my research were: (1) to investigate the variation of needles and genetic structure of populations of this species; (2) to estimate genetic distances between the populations; and (3) to test how strong the connection between P. uncinata populations and populations of P. mugo, P. sylvestris and P. × rotundata is. Populations of P. uncinata from the centre of its range (Pyrenees), from the edges of its range (Massif Central, Alps) and its isolated stands, were compared with populations of P. mugo, P. sylvestris, and P. × rotundata on the basis of anatomical, morphological and genetic data. Material of P. uncinata was collected outside the typical P. mugo range, to exclude the potential influence of the latter.

The biometric analysis included 11 natural populations of P. uncinata in the central and Eastern Pyrenees, Massif Central, Western Alps, Sierra de Cebollera, and Sierra de Gúdar, 2 populations of P. mugo from the Alps, 2 populations of P. sylvestris from the Sierra de Baza and Sudetes, and one population of P. × rotundata. For the biometric analysis of 20 morphological and anatomical needle characteristics, 10 previous-year short-shoots were sampled from the sunny side of the crown of 30 individuals in each population. This analysis was performed to describe the variation of the populations and taxa compared. Data obtained were subject to multivariate statistical analysis.

The genetic analysis was based on seeds that came from 7 natural localities of P. uncinata, 2 of P. mugo (Karnishe Alps, Salzburgian Alps), and 2 of P. sylvestris (Sierra de Guadarrama, Eastern Alps). Individual trees were compared by using 7 macrogametophytes on 8 enzyme systems encoded by 15 loci (Table 2). The following enzymes were studied: malate dehydrogenase (Mdh, E.C. 1.1.1.37), 6-phosphogluconate dehydrogenase (6Pgd, E.C.1.1.1.44), glutamate dehydrogenase (Gdh, E.C.1.4.1.2), sorbitol dehydrogenase (Sdh, E.C.1.1.1.21), glutamate oxaloacetate transaminase (Got, E.C.2.6.1.1), alcohol dehydrogenase (Adh, E.C. 1.1.1.1), isocitrate dehydrogenase (Idh, E.C.1.1.1.42), and shikimate dehydrogenase (Shdh, E.C.1.1.1.25). Starch-gel electrophoresis was used to separate the isoenzymes.

The presented results suggest that the most stable traits include: number of stomata on both sides of the needle (traits 4 and 5), needle dimensions and shape (traits 7, 8 and 14), and dimensions and shape of epidermal cells (traits 10, 17 and 18). The most variable were: number of resin canals (trait 6), distance between vascular bundles (trait 9), and Marcet's coefficient (trait 12). Apparently P. uncinata differs from the other taxa mostly in needle length (trait 1) and shape (trait 14), shape of epidermal cells (trait 18), and distance between vascular bundles (trait 9). The type of sclerenchyma cells is specific too. Most frequently, the cells between vascular bundles are not fibre-like, and their walls are thin or of medium thickness (mixed type BC or CD). Around resin canals, fibre-like cells with thick walls and small lumen (type a or b) are found in P. uncinata.

Results of the present work show that P. uncinata is much more closely related to P. mugo than to P. sylvestris. Nevertheless, P. uncinata remains distinct from both taxa in most of the examined characteristics. The species differs significantly from P. mugo in needle length (trait 1) and distance between vascular bundles (trait 9), while from P. sylvestris in number of resin canals (trait 6), thickness of epidermal cells (trait 11), and distance between vascular bundles. Width of epidermal cells (trait 10) was similar in all taxa studied, but the shape of epidermal cells (trait 15) differed between the taxa. All taxa also differ in type of sclerenchyma cells around resin canals (trait 20) and between vascular bundles (trait 19). Similar conclusions were drawn on the basis of analyses of cones, but they indicated that P. mugo is more closely related to P. sylvestris than to P. uncinata.
P. × rotundata differs significantly from P. uncinata in the first 18 traits, while the type of sclerenchyma cells was the most different. This seems to confirm the opinion of various authors that P. × rotundata is a natural hybrid between P. uncinata and P. mugo.

Out of the 15 loci analysed in 7 populations of P. uncinata, 12 loci were polymorphic and 3 loci were completely monomorphic (Idh, Mdh-1, Shdh-2). Allozyme frequencies at some loci varied markedly between populations, but the most common alleles were the same in all 7 populations. The low values of F-statistics and small genetic distance confirm that there is relatively little genetic differentiation in P. uncinata.

The mean number of alleles per locus ranged from 2.5 in P. sylvestris to 3.1 in P. uncinata. Average expected heterozygosity in 3 compared taxa varied from 0.32 in P. sylvestris to 0.34 in P. uncinata. Gene flow was the highest in populations of P. mugo and reached Nm = 27.195, while in the other 2 taxa it ranged from 7.20 to 8.19. On the basis of the frequency of alleles, 3 genetically different groups of populations can be distinguished. The first group consisted of the 7 populations of P. uncinata, the second group was formed by P. mugo, and the last group includes populations of P. sylvestris. The highest genetic distance values were observed between P. mugo and P. sylvestris (D = 0.270). However, the mean genetic distance between P. mugo and P. uncinata (D = 0.131) was only slightly smaller than between P. uncinata and P. sylvestris (D = 0.155). The mean genetic distance within P. uncinata was always much smaller than between taxa.

In conclusion, the studied taxa form 3 distinct groups, which are internally quite homogeneous. My results seem to confirm the hypothesis that P. uncinata is a result of an ancient hybridization between P. sylvestris and P. mugo, and to confirm the species status, as it is clearly separate from P. mugo.