The level of β-glucan which is a major soluble dietary fiber found in the grain endosperm cell wall was highly variable among 25 barley genotypes grown at four locations including Suwon, Naju, Jinju, and Jeju. Statistically significant genotypic effects were observed for β-glucan content at each or across growing sites (P<0.001). On average, 'Chalssalbori' showed the lowest percentage β-glucan (4.04%) among genotypes in the grain, whereas 'Yonezawa Mochi' was highest in percentage β-glucan (6.46%) compared to other genotypes. The significant difference between genotypes was approximately 1-2% across environments. The effects of location or interaction between locations and genotypes were not significant on the variation of β-glucan contents. High β-glucan content seemed to be greatly associated with such grain traits as waxiness and presence of husk except for 'Chalssalbori'. The waxy genotypes had a mean of 5.37% and values ranging from 5.28 to 5.47%, but normal genotypes had a mean of 4.78% and values ranging from 4.69 to 4.88% over environments. Hulless barley genotypes were also higher than hulled barley genotypes for the average β-glucan content in both individual and over all environments. The difference between the hulled and hulless gene pools was on average of 0.37% with ranges from 0.19% to 0.56% at four environments. β-glucan content measured from a mapping population of F5 -derived 107 lines derived from the cross between 'Yonezawa Mochi' and 'Neulssalbori' was not significantly associated with other agronomic traits except for 1,000-kernel weight at the '01 Suwon environment. Not too much information on the relationship of β-glucan content to agronomic traits was available.

Molecular-based genetic diversity for a set of 141 accessions of Korean barley cultivars and 24 accessions of foreign exotic cultivars were analyzed using random amplified polymorphic DNAs (RAPDs). Different level of genetic variability was observed with 30 random decamer primers in the Korean barley varieties and breeding lines which were preliminarily classified by morphological (hulled & hulless barley) and end-use (malting barley) and/or by the released periods. A total of 74 RAPD bands were scored, and the number of bands per primer varied from 1 to 7 with an average of 2.74. The hulled barley pool had one more marker genotype per primer than the hulless barley pool. The polymorphic information content (PIC) values based on the band pattern frequencies among genotypes varied depending on genetic pools where mean PICs of hulled, hulless and malting barleys were 0.62, 0.57, and 0.43, respectively. Certain genomic loci amplified by opR04, opF01, opB05, and opC13 were highly polymorphic with PIC>0.8. Patterns and temporal trends of genetic diversity assessed over the period from 1970s to 1990s had a tendency to increase, and in particular, this upward slant was quite clear and significant for the hulless barley pool. In the cluster analysis using genetic similarity matrix calculated from RAPD profiles, two major groups and several small subgroups were classified. Major grouping of materials was not affected by the presence of the husk but by their genetic background and the spike-row type. The validity of information on the genetic diversity and relationships between genotypes will have been reviewed to predict their yield potential.

Allelic diversity of 44 microsatellite marker loci originated from the coding regions of specific genes or the non-coding regions of barley genome was analyzed for 19 barley genotypes. Multi-allelic variation was observed at the most of marker loci except for HVM13, HVM15, HVM22, and HVM64. The number of different alleles ranged from 2 to 12 with a mean of 4.0 alleles per micro-satellite. Twenty-one alleles derived from 10 marker loci are specific for certain genotypes. The level of polymorphism (Polymorphic Information Content, PIC) based on the band pattern frequencies among genotypes was relatively high at the several loci such as HVM3, HVM5, HVM14, HVM36, HVM62 and HVM67. In the cluster analysis using genetic similarity matrix calculated from microsatellite-derived DNA profiles, two major groups were classified and the spike-row type was a major factor for clustering. Correlation between genetic similarity matrices based on microsatellite markers and pedigree data was highly significant (r=0.57** ), but these two parameters were moderately associated each other. On the other hand, RAPD-based genetic similarity matrix was more highly associated with microsatellite-based genetic similarity (r=0.63** ) than coefficient of parentage.

Small seed size is one of the major traits of soybean cultivars for sprouts with regard to high sprout yield. This study was conducted to identify quantitative trait loci (QTL) for seed size and weight in a set of F 6 seeds of 89 lines derived from a cross between 'Pureunkong', a soybean cultivar developed for sprouts and 'Jinpumkong 2', a soybean cultivar with no beany taste in seed due to the lack of lipoxygenases. The genetic map of 25 linkage groups with a total of 98 markers including RFLP, RAPD, SSR and classical markers was constructed from this F/sbu 5/-derived population and was used for QTL analysis. 'Pureunkong' was significantly smaller (P＜0.01) than 'Jinpumkong 2' in seed size and seed weight. Genetic variation was detected and transgressive segregation was common in the population for these traits. Seven DNA markers including opT14-1600 in LG A2, opF02-400 in LG B2, Satt100, opC09-700, opG04-730 and opQll-650 in LG C2, and opY07-1100 ＆ 1000 in LG(unknown) were significantly associated and accounted for 4.7 to 10.9% and 5.1 to 10.1 % of the phenotypic variation in seed size and seed weight, respectively. 'Pureunkong' alleles increased seed size and seed weight at the all four significant marker loci on the LG C2. These marker loci in LG C2 were closely linked and were presumed to be a single QTL. Overall, at least three independent QTLs from 3 linkage groups (A2, B2, and C2) were putatively involved in the control of seed size and seed weight.

The objective of this study was to develop a linkage map of soybean under the genetic background of Korean soybean. A set of 89 F/sub 5/ lines was developed from a cross between 'Pureunkong', which was released for soy-bean sprout, and 'Jinpumkong 2', which had no beany taste in seed due to lack of lipoxygenase 1, 2, and 3. A linkage map was constructed for this population with a set of 113 genetic markers including 7 restriction fragment length polymorphism (RFLP) markers, 79 randomly amplified polymorphic DNA (RAPD) markers, 24 simple sequence repeat(SSR) markers, and 3 morphological markers. The map defined approximately 807.4 cM of the soybean genome comprising 25 linkage groups with 98 polymorphic markers. Fifteen markers remained unlinked. Seventeen linkage groups identified here could be assigned to the respective 13 linkage groups in the USDA soybean genetic map. RFLP and SSR markers segregated at only single genetic loci. Fourteen of the 25 linkage groups contained at least one SSR marker locus. Map positions of most of the SSR loci and their linkages with RFLP markers were consistent with previous reports of the USDA soybean linkage groups. For RAPD, banding patterns of 13 decamer primers showed independent segregations at two or more marker loci for each primer. Only the segregation at op Y07 locus was expressed with codominant manner among all RAPD loci. As the soybean genetic map in our study is more updated, molecular approaches of agronomically important genes would be useful to improve Korean soybean improvement.