The geologist and the engineer are both faced with specific difficulties when they attempt to forecast the behaviour of a large rock mass. All the difficulties come from the fact that a rock mass is always crossed by a number of joints, or cracks, or fractures, or shear faults. In addition, these discontinuities are often open due to stress relief near the free surface of the mass. Knowledge of the mechanics of a discontinuous medium is still new and so rock testing cannot be considered to have any rigorous theoretical basis and, despite the considerable increase in the number and diversity of tests made over recent years, progress has been rather uncoordinated. Some trends, however, are noticeable in the approach used by practitioners, mainly dam designers. In the past the engineer or the geologist faced with the decision of building or not building a dam on a given rock site was asked to guess its future behaviour. There was a temptation to display a gift of divination. More and more engineers however refuse to gamble or to make believe that they are especially gifted and instead they use rock mechanics. In other words, the time has come when not only experts are consulted but the rock itself.
2. Geological description
Validity of tests. The rock mass takes the form of an assemblage of blocks with a very low void ratio (Fig. 1). The surfaces of separation are called major discontinuities to distinguish them from fine