In Mediterranean-type environments, grain weight of barley (Hordeum vulgare L.) is influenced by the presence of terminal abiotic stresses such as drought or elevated temperature. The relative importance of these climatic constraints acting together prior to and after anthesis, as well as the specificity of genotypic responses, have not been clearly documented. Field experiments were carried out in 12 rainfed environments of northern Spain involving five six-rowed and five two-rowed cultivars. A large variability in individual grain weight (IGW) among environments was recorded (34.8-59.3mg). An analysis of variance for IGW revealed genotype by environment (GxE) interaction. Environments were characterized by climatic covariables (temperature, rainfall and potential evapotranspiration) measured during three consecutive periods of plant development: jointing, heading and grain filling. Genotypes were described by covariables related to differences in morphological (ear-type), development (anthesis date) and physiological (carbon isotope discrimination, response to chemical desiccation) traits. Possible factors underlying GxE were examined by additive main effects and multiplicative interaction (AMMI) and factorial regression models. Genotypic and environmental descriptors were used as concomitant variables at the levels of the genotypic and environmental factor to partition GxE. The factorial regression model confirmed that the incidences of drought and of high temperatures during grain filling were responsible for the differential genotypic responses found in IGW: some cultivars were affected by both constraints, others by a single one, and some proved to be insensitive. These responses could be partially ascribed to the contrast between two- and six-rowed barleys: six-rowed types were more prone to be negatively influenced by terminal abiotic stresses than two-rowed cultivars. For breeding purposes there are opportunities to exploit the apparent genetic variability for tolerance to drought and/or thermic stress.