Abstract. The National Small Grains Collection (NSGC) maintains more than 24,000 accessions of cultivated barley. Passport and descriptor data are maintained on the Germplasm Resources Information Network (GRIN). Descriptor data for reactions to disease and insect pests, agronomic, and quality traits were analyzed. Key countries and regions of the world where particular traits are found were identified based on the accession improvement status. This information can be utilized by the curator and the requesting scientist in identifying appropriate germplasm for research and breeding.

key words: genebanks, disease resistance, insect resistance, descriptors, cultivars, landraces

Introduction. The National Small Grains Collection (NSGC), a component of the U.S. National Plant Germplasm System, maintains 24,842 accessions of cultivated barley from all parts of the world. Passport and descriptor data are maintained on the Germplasm Resources Information Network (GRIN). This large amount of data presents both opportunities and problems for the curator and for the scientist seeking germplasm for breeding and research. What strategies can be used to select germplasm that is diverse and best answers the scientist's needs? An analysis of searches for disease and insect resistance, agronomic, and quality traits in the barley collection is presented.

Materials and Methods. Descriptor data were extracted from the GRIN database (http://www.ars-grin.gov/npgs). Accessions showing a resistant reaction to any isolate or biotype were counted, ignoring isolate or biotype differences. Analyses were based on the accession improvement status and country of origin. In GRIN the relevant improvement status categories are: CULTIVAR/BREEDING (developed) and LANDRACE (collected).

Descriptors. Russian Wheat Aphid (Diuraphis noxia), seedling test, 24,451 accessions with data; Stripe Rust (Puccinia striiformis f.sp. hordei), natural infection in field tests at Cochabamba, Bolivia, 24,202 accessions with data; Net Blotch (Pyrenophora teres), natural infection in field tests at Langdon, North Dakota, 14,101 accessions with data; and Spot Blotch (Cochliobolus sativus), natural infection in field tests at Fargo, North Dakota, 18,250 accessions with data; Hull Cover, laboratory test, 15,395 accessions with data; Spike Row Number, laboratory test, 19,492 accessions with data; Growth Habit, field test, 24,459 accessions with data; ß-Glucan, laboratory test, 7,667 accessions with data.

Results:

Stripe Rust: Resistance is found in landraces mainly from Ethiopia. Resistance in developed materials is mainly from Europe and U.S. Net Blotch: Resistance is found in landraces from Ethiopia, China, and many other areas. Resistance in developed materials is mainly from North America and Europe. Spot Blotch: Resistance is found in landraces from Ethiopia, China, Turkey, and others. Resistance in developed materials is from North America and Europe. Russian Wheat Aphid: Resistance is found in landraces from West Asia, mainly Afghanistan and Iran. Resistance in developed materials is primarily from the U.S. Hull Cover: Hulless trait is found in landraces from Peru, Ethiopia, Himalayas, China, and Japan. Hulless trait is found in developed materials wherever active breeding programs exist. ß-Glucan, Spike Row Number, Growth Habit: These traits are widely dispersed and no clear geographic trends are apparent.

Discussion. CULTIVAR and BREEDING accessions will usually originate where active breeding programs have existed. LANDRACE accessions originate in areas of origin of the crop or where the accessions have been in long-term cultivation. A choice must be made between utilizing LANDRACE accessions where new, unexploited sources for a trait will most often be found vs developed materials (CULTIVAR and BREEDING) where the trait will be in a more adapted genetic background. Searches can often be narrowed further by taking into account race and biotype information, particular adaptive environments of interest, and descriptor data for other agronomic traits. The data presented describe the situation in the NSGC, but the general principle should apply to other genebank

Acknowledgments. Appreciation is expressed to the following scientists who were instrumental in collection of the descriptor data: J. Webster and S. Kindler (Russian Wheat Aphid data), W. Brown, V. Velasco, and J. Hill (Barley Stripe Rust data), B. Steffenson and V. Pederson (Net and Spot Blotch data), D. Peterson (ß-Glucan data).