Worldwide germplasm collections contain about 7.4 million accessions of plant genetic resources for food and agriculture. One of the 10 largest ex situ genebanks of our globe is located at the Leibniz Institute of Plant Genetics and Crop Plant Research in Gatersleben, Germany. Molecular tools have been used for various gene bank management practices including characterization and utilization of the germplasm. The results on genetic integrity of longterm-stored gene bank accessions of wheat (self-pollinating) and rye (open-pollinating) cereal crops revealed a high degree of identity for wheat. In contrast, the out-pollinating accessions of rye exhibited shifts in allele frequencies. The genetic diversity of wheat and barley germplasm collected at intervals of 40 to 50 years in comparable geographical regions showed qualitative rather than a quantitative change in diversity. The inter- and intraspecific variation of seed longevity was analysed and differences were detected. Genetic studies in barley, wheat and oilseed rape revealed numerous QTL, indicating the complex and quantitative nature of seed longevity. Some of the loci identified were in genomic regions that co-localize with genes determining agronomic traits such as spike architecture or biotic and abiotic stress response. Finally, a genome-wide association mapping analysis of a core collection of wheat for flowering time was performed using diversity array technology (DArT) markers. Maker trait associations were detected in genomic regions where major genes or QTL have been described earlier. In addition, new loci were also detected, providing opportunities to monitor genetic variation for crop improvement.

The deterioration in the quality of ex situ conserved seed over time reflects a combination of both physical andchemical changes. Intraspecific variation for longevity is, at least in part, under genetic control. Here, the grain of 183bread wheat accessions maintained under low-temperature storage at the IPK-Gatersleben genebank over somedecades have been tested for their viability, along with that of fresh grain subjected to two standard artificial ageingprocedures. A phenotype–genotype association analysis, conducted to reveal the genetic basis of the observedvariation between accessions, implicated many regions of the genome, underling the genetic complexity of the trait.Some, but not all, of these regions were associated with variation for both natural and experimental ageing, implyingsome non-congruency obtains between these two forms of testing for longevity. The genes underlying longevityappear to be independent of known genes determining dormancy and pre-harvest sprouting.