The genetic basis for some stability statistics used in plant breeding was investigated in this study by quantitative trait loci (QTLs) analysis. The mapping population comprised 180 doubled haploid lines from a cross between Australian (VB9524) and North American (ND11231-12) two-rowed barley lines. Using a 211-point genetic linkage map, we identified six QTLs segregating for heading date in the population. These mapped to the approximate position of the major earliness genes (Ppd-H1 and Ppd-H2), and several genes for earliness in a narrow sense (earliness per se). By aligning the QTL positions with those for calculated stability statistics, we used the co-location of QTL positions to infer genetic basis. The stability statistics considered were Wricke's [Wricke, G., 1962. Über eine Methode zur Erfassung der oekologischen Streubreite in Feldversuchen. Zeitschr f Pflanzenz 47, 92-96] Ecovalence (Wi), Eberhart and Russell's [Eberhart, S.A., Russell, W.A., 1966. Stability parameters for comparing varieties. Crop Sci. 6, 36-40] regression coefficient (SLOPE) and deviation from regression (MSDEV), Lin and Binns' [Lin, C.S., Binns, M.R., 1988. A superiority measure of cultivar performance for cultivar × location data. Can. J. Plant Sci. 68, 193-198] cultivar superiority measure (Pi) and the first two principal component axis scores (IPCA1 and IPCA2) from the additive main effect and multiplicative interaction (AMMI) model. Based on a genome-wide LOD threshold of 3.0, we found little evidence for genetic control of MSDEV. However, Wi was considered to be under genetic control, due to a significant linkage (peak LOD = 4.85) to a locus on chromosome 2H where the Ppd-H1 gene is located. For SLOPE and IPCA1, strong evidence of genetic control was indicated by linkage to three unlinked loci. Two of these occurred at regions on chromosomes 2HS and 5HL, close to the approximate locations of Ppd-H1 and eps5HL, and where heading date QTLs showed significant (P < 0.001) QTL × E interactions. For Pi, the genome-wide LOD profile matched that for heading date QTLs, with peak LOD scores at all regions where the QTLs were identified. The pattern of co-localisation of QTLs for stability parameters with those for heading date suggests that, in this population, G × E interactions might be largely controlled by allelic sensitivity.