"The expression “unsaturated soil” is used in geotechnical engineering to identify a class of soils whose pores are partially filled with water and partially filled with air. Unsaturated soils occur naturally above the water table as a superficial ground layer, whose thickness depends on the balance of precipitation and evapo-transpiration and, hence, on the prevailing local climatic conditions. Unsaturated soils also occur in manmade structures that are built by compacted earth such as embankments, gravity dams, barriers for underground nuclear waste repositories and flood defences. The engineering properties of unsaturated soils are therefore crucial to many civil engineering applications as well as to geohazards engineering (e.g. slope instabilities and landslides). The response of unsaturated soils to loading and environmental actions is crucially dependent on the mechanical anisotropy of this material. Anisotropy in unsaturated soils may be caused by two concurring factors, i.e. the nature of the soil fabric and the deviatoric component of the inter-granular stress generated by capillary pore water. This project aims at investigating anisotropy in unsaturated soils based on the appreciation of the microscopic interactions between solid grains, pore water and pore air. The research is divided in four main tasks: a) performance of laboratory tests on unsaturated soil samples compacted under different conditions to achieve different degree of initial cross-anisotropy, b) interpretation of test results in terms of theoretical models based on the mechanics of granular materials and continuum mechanics, c) microscopy analysis of soil samples subjected to cycles of wetting and drying and 4) study of boundary value problems where the anisotropy of unsaturated soils is particularly relevant. One of the main project deliverables will be the formulation of a constitutive model capable of representing the anisotropic mechanical behaviour of unsaturated soils."