The Canary Island pine, Pinus canariensis, is an endemic tree that forms one of the main forest ecosystems within the archipelago, and whose distribution has been reduced in the last five centuries by clear cutting for the extraction of timber and tar. It was in the XXth century that exploitation declined and reforestation programs were brought forward for the restoration of an ecosystem that harbours a number of endangered endemic species of plants and animals. In addition to reforestation efforts, an understanding of population genetic processes is also necessary for the successful conservation management of the Canarian pine forest, particularly in light of gathering evidence for local adaptation. In this thesis historical and contemporary gene flow within P. canariensis was studied with nuclear and chloroplast microsatellite markers. High immigration rates (0.68–0.75) were estimated as expected for an outcrossing wind-pollinated tree. Nevertheless, significant population differentiation (FST=0.019, RST=0.044) was detectable for sites separated by only a few kilometres. Within the context of reforestation programs the high levels of gene flow detected would appear to have a positive effect, by facilitating the immigration of local alleles from natural stands into potentially genetically depauperate first generation gene pools of reforested stands. Historical population growth was revealed with chloroplast microsatellites for most populations of P. canariensis. Population expansions for the pine parasite weevil Brachyderes rugatus were also detected, broadly coinciding with the population expansions within the Canary Island pine forests. Given the estimated times of expansion, these population demographic increases would seem likely related to the process of colonisation of newly emerged islands or local patches after volcanic disturbance. Detection and dating of these expansions from chloroplast microsatellites was, to some degree, negatively affected by homoplasy (i.e. parallel and back mutations). Coalescent simulations of the evolution of chloroplast microsatellites were applied to study the effects of homoplasy in the statistical analysis of population structuring. Measures of genetic diversity based on number of haplotypes and measures based on genetic distances were differently affected. Genetic distances were underestimated but were proportional to the actual value. These effects help to explain the lower performance of statistical analyses for the detection and dating of population expansions. Further research on the effects of homoplasy in the analysis of population differentiation using chloroplast microsatellites is essential.