greggii, P. maximinoi, P. oocarpa and P. tecunumanii

are at the stage where second and third-generation field trials have been established ( Camcore Annual Report, 2012). In Europe, national research institutions operated 15–20 separate breeding programmes often on the same species until 1990 (Pâques, 2013). This changed dramatically in the 1990s when budgets of many research institutes were cut and the interest of policymakers in tree breeding decreased. As a result, tree breeding programmes in Europe were forced to change their operating practices and to seek greater synergies through increased international collaboration and coordination, sharing responsibilities and targeting fewer tree species. During UMI-77 clinical trial the past 20 years, a number of projects, and especially the TreeBreedex project (2006–2010), have supported the transformation of European tree breeding into a collaborative effort, carried out by a network of national institutions sharing their research facilities, breeding material and field tests (Pâques, 2013). This new modus operandi now resembles the way tree breeding has been carried out elsewhere for decades. During the past

decade or so, genetic analysis of forest tree populations with molecular markers has strengthened R&D efforts and has increased the transfer of DNA samples. Range-wide genetic surveys were initiated for temperate tree species (e.g., Petit et al., 2002 and Magri et al., 2006) and they are now increasingly also conducted for tropical species (e.g., Jamnadass et al., 2009 and Kadu et al., 2011). These studies have Natural Product Library in vitro provided useful information on the geographic structure of genetic diversity, knowledge of importance for the management of natural tree populations and for the formulation of conservation strategies. Site-specific studies with molecular markers have also been essential

to better understand ecological and genetic Terminal deoxynucleotidyl transferase processes within tree populations (e.g., Lee et al., 2006), and the impacts of forest fragmentation and logging on them (e.g., Rymer et al., 2013 and Wickneswari et al., 2014). Genomic developments and new markers, such as those based on single nucleotide polymorphisms (SNPs), also offer possibilities to survey adaptive diversity within tree populations (Neale and Kremer, 2011). With the advent of new, ‘next generation’ sequencing technologies, genetic markers for almost any tree species can now be developed at low cost (van der Merwe et al., 2014 and Russell et al., 2014). Tree seed crops often have high year-to-year variation, causing remarkable fluctuations in seed availability. This makes it desirable to maintain seed storage capacity and maximise seed harvest during mast years. However, many tree species (e.g., around 70% in humid tropical forests; Sacandé et al., 2004) produce recalcitrant or intermediate seed which lack dormancy and which are sensitive to both desiccation and low temperature (see Pritchard et al.