84 I. Introduction: evolving concerns over loss of crop diversity 85 II. Defining and measuring crop genetic erosion 89 III. Evidence for, and drivers of, changes in crop diversity over time 92 IV. Steps needed to advance knowledge about crop genetic erosion 99 V. Conclusion: mitigating, stemming and reversing losses of crop diversity 102 Acknowledgements 103 References 104 Appendix A1 111

Ensuring the availability of the broadest possible germplasm base for agriculture in the face of increasingly uncertain and variable patterns of biotic and abiotic change is fundamental for the world's future food supply. While ex situ conservation plays a major role in the conservation and availability of crop germplasm, it may be insufficient to ensure this. In situ conservation aims to maintain target species and the collective genotypes they represent under evolution. A major rationale for this view is based on the likelihood that continued exposure to changing selective forces will generate and favor new genetic variation and an increased likelihood that rare alleles that may be of value to future agriculture are maintained. However, the evidence that underpins this key rationale remains fragmented and has not been examined systematically, thereby decreasing the perceived value and support for in situ conservation for agriculture and food systems and limiting the conservation options available. This study reviews evidence regarding the likelihood and rate of evolutionary change in both biotic and abiotic traits for crops and their wild relatives, placing these processes in a realistic context in which smallholder farming operates and crop wild relatives continue to exist. It identifies areas of research that would contribute to a deeper understanding of these processes as the basis for making them more useful for future crop adaptation.

Our aim was to assess the level of use of conserved plant genetic resources (PGR) in crop genetic research. To do this, we analyzed in detail the reports from four internationally recognized journals published in 1996. These journals were Crop Science, Euphytica, Plant Breeding, and Theoretical and Applied Genetics Our results indicate that about 23% of the articles published in these journals reported work conducted with material that originated in an ex situ PGR collection or was collected directly in the field. We also analyzed our results to determine the range of species and research topics involved, the sources of materials used, and the users of PGR for research. Of the material used, 80% was accessed from ex situ germplasm collections housed in genebanks in 27 countries and seven centers of the CGIAR. The rest of the studies utilized material gathered directly in the field. Almost all work was produced at national research centers and universities; very little work from private industry was published in the journals covered. Just under 20% of the institutions represented by authors were located in developing countries. Work published involved 112 species and included assessments of genetic diversity among accessions (42% of the studies), studies of the inheritance of biotic stress resistance (29%), taxonomic and phylogenetic analyses (16%), as well as cytologic, molecular and conservation research. Our research demonstrates a significant use of conserved material in research.

Informed collecting, conservation, monitoring and utilization of genetic diversity requires knowledge of the distribution and structure of the variation occurring in a species. Hordeum vulgare subsp. spontaneum (K. Koch) Thell., a primary wild relative of barley, is an important source of genetic diversity for barley improvement and co-occurs with the domesticate within the center of origin. We studied the current distribution of genetic diversity and population structure in H. vulgare subsp. spontaneum in Jordan and investigated whether it is correlated with either spatial or climatic variation inferred from publically available climate layers commonly used in conservation and ecogeographical studies. The genetic structure of 32 populations collected in 2012 was analyzed with 37 SSRs. Three distinct genetic clusters were identified. Populations were characterized by admixture and high allelic richness, and genetic diversity was concentrated in the northern part of the study area. Genetic structure, spatial location and climate were not correlated. This may point out a limitation in using large scale climatic data layers to predict genetic diversity, especially as it is applied to regional genetic resources collections in H. vulgare subsp. spontaneum.

Successful conservation strategies require that taxa are prioritized because resources for planning and implementation are always limited. In this study, we created a partial checklist of crop wild relatives (CWR) that occur in the Southern African Development Community (SADC) region and identified the taxa of highest priority for regional conservation planning based on their importance for food and economic security. We found that the region contains over 1900 wild relatives of species cultivated for food, beverages, ornamental, forage/fodder, forestry, medicinal, environmental and other uses. Prioritization of these species was based on two criteria: (i) the value of the related crop for human food and economic security in the region and/or globally, and (ii) the potential or known value of the wild relatives of those crops for crop improvement. The region contains 745 CWR species related to 64 human food and beverage crops that are of high socioeconomic importance and 100 of these are of immediate priority for conservation action. The results of this study show that the SADC region contains a wealth of CWR diversity that is not only of value for food and economic security within the region but also globally. Furthermore, this study represents the first step in developing a CWR conservation and sustainable use strategy for the region, where its implementation would contribute to food security and well-being.