Regeneration of germplasm samples: Wild versus cultivated plant species

The growth in size and species diversity of germplasm collections held around the world has led to several managerial problems. For dealing with the increasing size of collections, the ''core collection'' offers one strategy. However, the problems of managing increasing species diversity are only just emerging, particularly in the germplasm collections of wild species. Wild collections encompass those of forest trees, forage plants, medicinal and industrial plants, the wild relatives of cultivated plants, and plant species designated as endangered. Regenerating such collections is a particular problem. In all cases, the accessions of wild species are generally more costly to regenerate than those of cultivated species. This is because of their life history, breeding system, genetic structure, ecology, and lack of domestication. Indeed it is worth asking why bother to regenerate wild accessions at all, and instead rely on resampling conserved sources in situ, or on frozen DNA libraries, or both. Modern empirical studies of the population genetics of wild populations focus on host-pathogen coevolution, spatial subpopulation structure and dynamics, restricted sampling strategies, breeding system variation, and colonizing history and molecular divergence and phylogeny. Each of these topics has important lessons for the optimum prioritizing of accessions, or the best methods of their regeneration. Optimum regeneration of wild species requires a clear definition of objectives and priorities among accessions, monitoring of mating system and genetic structure, maintenance of accession purity and associated passport data, and biologically realistic and flexible guidelines for sample size.