At treeline, selection by harsh environmental conditions sets an upward limit to arboreal vegetation. Increasing temperatures and the decline of traditional animal raising have favoured an upward shift of treeline in the last decades. These circumstances create a unique opportunity to study the balance of the main forces (selection and gene flow) that drive tree migration. We conducted a parentage analysis sampling and genotyping with five microsatellite markers in all Norway spruce individuals (342 juveniles and 23 adults) found in a recently colonized treeline area (Paneveggio forest, Eastern Alps, Italy). Our goal was to evaluate local reproductive success versus gene flow from the outside. We were able to identify both parents among local adults for only 11.1% of the juveniles. In the gamete pool we sampled, two-thirds were not produced locally. Effective seed dispersal distance distribution was characterized by a peak far from the seed source (mean 344.66 m ± 191.02 s.d.). Reproductive success was skewed, with six local adults that generated almost two-thirds (62.4%) of juveniles with local parents. Our findings indicate that, although a few local adults seem to play an important role in the colonization process at treeline, large levels of gene flow from outside were maintained, suggesting that the potential advantages of local adults (such as local adaptation, proximity to the colonization area, phenological synchrony) did not prevent a large gamete immigration.

Understanding the genetic structure and diversity of edge populations can shed light on the role of peripheral populations and their relevance for conservation strategies. In this study, three fragmented and isolated Apennine populations of Scots pine (Pinus sylvestris L.) belonging to the rear edge of the species' distribution were analyzed using both nuclear (nSSR) and chloroplast (cpSSR) microsatellites and were compared with an Alpine population belonging to the species' main range. Although small population size and considerable isolation have probably reduced the genetic variability of Apennine populations, these fragmented populations maintain a high level of within-population genetic diversity. A signiWcant amongpopulation diVerentiation was found using both nSSR (F ST = 0.08) and cpSSR markers ( = 0.14). Analysis of molecular variance (AMOVA) on the nSSRs attributed all variabilities to the among Apennine populations component supporting the theoretical predictions regarding fragmentation eVects on genetic structure. On the other hand, AMOVA on the cpSSRs attributed all variances to the between-region component and no diVerentiation was found within region, among the Apennine populations. This result suggests the importance of pollen gene Xow in homogenizing populations on this geographical scale. Our results conWrm the genetic distinctiveness of Apennine populations and their possible derivation from diVerent glacial refugia than those of the Alps. Considering their peculiarity and the high level of intrapopulation genetic diversity that they still retain, fragmented Apennine populations should be considered of high priority for conservation.

The genetic linkage map of European beech ( Fagus sylvatica L.) that we report here is the first to our knowledge. Based on a total of 312 markers (28 RAPDs, 274 AFLPs, 10 SSRs) scored in 143 individuals from a F(1) full-sib family. Two maps (one for each parent) were constructed according to a "two-way pseudo-testcross" mapping strategy. In the male map 119 markers could be clustered in 11 major groups (971 cM), while in the female map 132 markers were distributed in 12 major linkage groups (844 cM). In addition, four and one minor linkage groups (doublets and triplets) were obtained for the male and female map respectively. The two maps cover about 82% and 78% of the genome. Based on the position of 15 AFLP and 2 SSR loci segregating in both parents, seven homologous linkage groups could be identified. In the same pedigree we investigated the association with genetic markers of several quantitative traits: leaf area, leaf number and shape in 2 different years, specific leaf area, leaf carbon-isotope discrimination and tree height. A composite interval-mapping approach was used to estimate the number of QTLs, the amount of variation explained by each of them, and their position on the genetic linkage maps. Eight QTLs associated with leaf traits were found that explained between 15% and 35% of the trait variation, five on the female map and three on the male map.

No abstract