Materials and Methods
A set of doubled haploid (DH) lines was derived by anther culture from the spring barley cross Blenheim x E224/3 at the Scottish Crop Research Institute (SCRI). Blenheim was a popular malting barley cultivar in North West Europe but suffered from water sensitivity and, in the field, susceptibility to mildew and leaf rust. E224/3 was an SCRI breeding line with good yield and disease resistance. The DH lines were grown in replicated trial along with their parents each year at SCRI from 1989 to 1992 and a range of agronomic, yield, grain quality, germination, disease and malting characters were scored on them. In addition the DH lines were scored for 144 genetic markers, mostly RFLPs and RAPDs but also some SSRs, an STS, some biochemical markers, a morphological marker ( the denso dwarfing gene) and Rhynchosporium resistance. From these markers, the genetic map in Figure 1 was produced. This map has been used to search for QTLs, controlling the overall means of characters scored on the DH lines, by the regression technique of Haley and Knott (1992). The most significant QTL was fixed and the search repeated for significant QTLs up to a maximum 6. We adopted this approach since QTLs of large effect, particularly those at the denso dwarfing gene locus, often masked smaller QTLs in a single scan. The results for 4 of the characters; PY, plot yield (t/ha), TGW, thousand grain weight (g), N nitrogen content of the dry grain (%) and HWE, hot water extract of a malted sample of grain (L°/kg) are presented.

Results
Two QTLs were found for yield with increasing alleles being derived from Blenheim in both cases (Figure 2a). One was located at the denso dwarfing gene locus and increased yield by approximately 0.5t/ha and the other was located in the region of Rubisco activase on chromosome 6 but had a much smaller effect. Four QTLs were revealed for TGW with three increasing alleles being derived from Blenheim (Figure 2b). Two QTLs appeared to be linked in repulsion on a segement of chromosome 1 and the others were located on chromosomes 3 and 7. The QTL on chromosome 3 may be loosely linked to

Figure 1. Chromosome maps for the Blenheim x E244/3. Letters indicate separate linked segments of chromosomes. The gaps between segments represent no significant linkage for LOD=3.0

Figure 2. Location of QTLs in Blenheim x E224/3. Size and direction of bar is the deviation of the Blenheim allele from the mean of the cross.

the denso dwarfing gene. Six QTLs were found for N but major effects were confined to regions on chromosomes 3 and 5 (Figure 2c). Three increasing QTLs were derived from Blenheim and 2 QTLs appeared to be linked in repulsion in the region of the Hor1 and Hor2 on chromosome 5. The QTL from Blenheim on chromosome 3 producing a large decrease in N was located in the region of the denso dwarfing gene. One, producing a smaller decrease in N, was on chromosome 7, in the region of the water-soluble protein loci. A major effect on HWE was found in the same region with evidence of a repulsion linkage between 2 QTLs (Figure 2d). Two further QTLs for HWE were located on the largest mapped segment of chromosome 3 in this study, one in the vicinity of Rhy.

Discussion
These results suggest that the denso dwarfing gene has a large effect upon any characters associated with it. In contrast to most previous studies, the dwarf allele was associated with a yield increase but previous studies, apart from Gymer (1991) who also found the gene to be associated with a yield increase, have examined alternative sources of the dwarfing gene. Blenheim carries the Diamant source of the gene found in Triumph and its derivatives. Its higher yield would seem to derive from increases either in tillering or grain number as the other yield component (TGW) has no association with the dwarfing gene other than a very loose linkage which would not explain the effect upon PY. Given that the Diamant source of the dwarfing gene increases yield and has no adverse effects upon quality, it is not surprising that so many of the North West European barley cultivars now carry the gene. The effect of the gene upon N would appear to be a dilution effect due to its increase in PY, as no effect on nitrogen yield can be found at the locus. Large effects upon N were also found in the region of Hor1 and Hor2 on chromosome 5 but, as there was no effect upon PY in this region, these effects must reflect differences in nitrogen uptake and/or deposition.

Comparison of the map of Blenheim x E224/3 with those of Steptoe x Morex (Kleinhofs et al. 1994) and Harrington x TR306 (Kasha et al. 1994) is difficult due to few common markers and the region around denso is not covered by the skeletal map for Harrington x TR306 (Tinker et al. 1994). However some rough comparisons can be made and of the 17 QTLs detected for the 4 characters in the present study, 10 would appear to be located in similar regions to QTLs detected in the North American germplasm, mainly in Steptoe x Morex (Hayes et al. 1993, Han and Ullrich 1994). Both yield QTLs are in similar regions to QTLs from Steptoe x Morex as are 2 of the QTLs for TGW. Three of the QTLs for N are also in similar regions to some of those found for Steptoe x Morex, although there was no evidence of QTLs around the hordein loci in either Steptoe x Morex or Harrington x TR306 (Tinker et al. 1994). The QTL on chromosome 2 for HWE was found in both sets of North American data and there was also evidence of a QTL in Steptoe x Morex extending into the region around the water-soluble protein loci.

In conclusion, whilst there is evidence of some similarity in the results from North American and North-West European germplasm, there is also considerable diversity between the two pools of germplasm. This emphasises the need to conduct mapping investigations in the target envionment with germplasm adapted to that environment. and the desirability of using common markers in different populations to establish the transportability of QTLs across different genotypes

References
Gymer PT (1991) Inheritance and linkage of some quality-related characters in barley. In Munck L (ed) Barley Genetics VI (I). Munksgaard Int Publ, Copenhagen, pp 254-257.

Haley CS, Knott SA (1992) A simple regression method for mapping quantitative trait loci in line crosses using flanking markers Heredity 69 : 315-324.

Han F, Ullrich SE (1994) Mapping of quantitative trait loci associated with malting quality in barley. Barley Genetics Newsletter 23: 84-97.

Hayes PM, Liu BH, Knapp SJ, Chen F, Jones B, Blake T, Franckowiak J, Rasmusson D, Sorrells M, Ullrich SE, Wesenberg D, Kleinhofs A (1993b) Quantitative trait locus effects and environmental interaction in a sample of North-American barley germ plasm Theor Appl Genet 87: 392-401.

Kasha KJ, Kleinhofs A and the North American Barley Genome Mapping Project (1994) Mapping of the barley cross Harrington x TR306. Barley Genetics Newsletter 23: 65-69.

Kleinhofs A, Kilian A, Kudra D and the North American Barley Genome Mapping Project (1994) The NABGMP Steptoe x Morex mapping progress report. Barley Genetics Newsletter 23: 79-83.

Thomas WTB, Powell W, Chalmers KJ, Barua UM, Swanston JS, Waugh R, Forster BP, Ellis RP, Jack P, Lea V (1993). Quantitative trait loci for yield and quality components in spring barley. Aspects of Appl Biol 36: 125:134.

Tinker NA, Mather DE and the North American Barley Genome Mapping Project (1994) Main effects of quantitative trait loci in Harrington/TR306 two-row barley. Barley Genetics Newsletter 23: 72-78.