Methods of Analysing Diversity Within and Between Taxa

New technologies are rapidly developing and increasingly provide results towards the status and evolution of populations. Heterogeneity and heterozygosity have characteristic functions inside the genetic structure of populations. Genetic erosion is a specific problem, also inside genebanks: collected samples may be lost during maintenance in genebanks, and the allelic composition of populations may change. In the last 20 years, landraces have gained new interest as sources for extended variation (Zeven 1998). They are usually characterized by complex morphological diversity. For such variation, diagnostic infraspecific classifications have been used (e.g. Percival 1921; Mansfeld 1951; Dorofeev et al. 1979), and they proved useful for characterizing and handling landraces. For example, Dorofeev et al. (1979) (Table 3.2) recognises 27 species with 17 subspecies, 32 convarieties and 1,055 botanical varieties (Knüpffer et al. 2013). If infraspecific forms are not named and

Fig. 3.1 Genepools in Hordeeae (formerly Triticeae) (After Bothmer et al. 1992)

described systematically, their diversity is at risk of being lost. Modern cultivars usually show only few morphologically discernible variants, since breeders selected only a fragment from the previously existing diversity, and, therefore, they do not see the need for traditional classification systems using botanical varieties. Scholz (2008), for example, observed that in T. aestivum only a single botanical variety, var. lutescens (Alef.) Mansf., is still present in modern cultivars, with very few exceptions.

Methods of Evaluation

Molecular markers in the form of DNA segments, even if they do not always represent functional genes, are used to identify genetic differences on a fairly simple level without reference to ecological adaptation. Traditionally many other evaluations are carried out in the breeding process. Genebanks should increase or newly

Table 3.2 Classification of Triticum according to Dorofeev et al. (1979), with minor changes. Authors of scientific names omitted

establish the classical evaluation programmes. Screenings for disease resistance or reaction to abiotic stresses have been carried out in Gatersleben for long time (e.g. Nover 1962 and other publications listed by Hammer et al. 1994; Börner et al. 2006). Pre-breeding (also called germplasm enhancement) will gain importance. It is necessary to bridge the gap between geneticists (aiming at excellent research and high-ranking publications), breeders (aiming at developing new cultivars), and genebanks (aiming at conserving the existing diversity). None of them has the capacity to do pre-breeding alone. Only a combination of efforts developed by all three players can help overcoming this situation.

Storage and Reproduction in Genebanks

Plant genetic resources are usually preserved in genebanks effectively and costefficiently under long-term conditions, although the mutation rate may increase during storage, leading to genetic changes (Stubbe 1937). However, strategic concepts are needed for reproduction. This seemingly simple procedure is full of problems and needs higher scientific and technical inputs. For example, genebanks as a rule cannot provide sufficient seed for immediate use of accessions in experiments on larger plots. Perhaps this problem is closely related to pre-breeding.

Outlook

As is the case with all major methodological and technological changes, it is dangerous to neglect the repertoire of methods formerly used and to over-emphasize modern technologies. Landraces of crops are a challenge for maintaining in genebanks and on farm (Maxted et al. 2008). In genebanks, initially diverse landraces may lose rare alleles, due to reproduction and storage conditions, but hundreds and thousands of landraces cannot be efficiently maintained alone on-farm in their regions of origin; costs and logistics requirements are prohibitively high. The historical background and evolutionary history of landraces can be investigated in a first step by using traditional methods. Landraces show the structures for which the traditional methods have been developed. The work with PGR is conservative because we have the task to conserve them. The subsequent examination with the help of molecular methods can resolve specific questions in a satisfactory and meaningful fashion.

Acknowledgments We thank Roland von Bothmer, Alnarp, Sweden for his permission to use the diagram in Fig. 3.1.