This project drew the previous work on mapping rice root traits and field-droughtreaction QTLs to a conclusion. The project addressed the specific issue of the effect of the environment on QTL expression in the field. Prior to this project most root studies in this population had been carried out in artificial, non-field conditions. In particular this project tackled the influence of the soil in field situations. The field site at WARDA where previous QTL mapping for drought reaction traits had been carried out was one of the sites compared for soil physical properties. Other fields at WARDA and IRRI were analysed and root trait analysis was done in fields with contrasting soil penetration resistance. Roots studies were for the parents (Azucena and Bala), the mapping population (recombinant inbred lines, RILs) and for nearisogenic lines (NILs) derived from marker-assisted selection to introgress Azucena root QTLs into both Kalinga III and Bala. The results were complex, underlying the complex nature of QTLs for roots and drought reaction. There was some agreement for QTLs detected in the field and previously detected QTLs. A QTL on chromosome 2 for root density in the field matched root penetration QTLs detected using a wax layer in pots. Detection of QTLs in the field was found to be more difficult than in artificial conditions because of the very large environmental effect. This was exemplified in the lack of significance in detection of improved root distribution in NILs containing Azucena root growth QTLs. The QTL x environment interaction was investigated and is discussed in the resulting publications, which also include the QTL maps for the traits studied. The molecular map and trait data is freely available for researchers on the internet and this information can now be used by rice and other cereals breeders wishing to study or select root and drought related traits. The molecular linkage map was improved by the addition of 19 microsatellite (SSR) markers. These linked split linkage groups completing 9 of the 12 chromosomes to the target of mapped markers at every 20 cM or more. Only one chromosome (chromosome 12) is now left as two separate linkage groups, and chromosome 8 still lacks markers at the top end. The additional markers enabled more precise QTL analysis of existing and new trait data to be completed with this map. Some of these SSR markers are now being used in marker-assisted selection in rice breeding projects.