Abstract : The first part will deal with the description of all actual ground movements and their varied aspects in order to obtain a classification , based on criteria from observation, enabling to distinguish several types of movements to be studied with several processes . This study leads to the definition and the description of landslide and is illustrated by the landslide in MIONTAUD (Isère) . Our next concern is with the stability analysis of slopes as we first review all the factors ruling this stability, and the present state of knowledge about these factors and the means they can be assessed with, then the methods that enable this stability analysis . He have looked for a method easy to use owing to its mathematical basis as well as the materials it requires , so that it can be useful to a greater number of people . A paragraph records all the factors ruling the stability of slopes mechanical properties and boundary conditions . Then we study the two ways that helps to the stability analysis of slopes : description, first, in which observation and investigation of all datas given by each case of stability or unstability (that is, solved by nature) are needed ; this first way can lead either to comparative study of problems, thus appealing to the experience of the observer, or to a statistical treatment, should the number of observations be large enough, which enables to define the relative influence of the different parameters : a method is suggested , illustrated by the examples of the slides in EN BRUNET (Savoie) ; this descriptive way, which can mainly be used by the naturalist, is of noteworthy help to the engineer, faced with great natural phenomena ; the theoretical study, on the opposite, is to give the geologist a simple and efficient tool to estimate the stability. After recalling all the different methods of reckoning, we have chosen FELLENIUS ' s one and BISHOP's simplified one, that can be used by manual calculating ; to assess the validity of these two methods , we have applied them to some cases (stable and unstable slopes) well discribed in previous literature ; which has enabled us to say that the FELLENIUS's method, with a corrective coefficient, gives satisfying results to define the stability limit of a slope ; out of this most simple method, we have produced a probable safety factor (FP), equal to FELLENIUS's factor (FF) plus 0.12; and the limits of accuracy of this factor FP are stated in concordance with the accuracy which can be expected from different field and laboratory investigations and tests. An example of reverse usage of this method is set forth (slide in SAINT JEAN DU BRUEL - Aveyron) which consists in looking for the values of the mechanical parameters of a soil on the moment of its failure (FP = 1.0) so that the evolution of its stability can be studied, according to the various changes in the soil masses conditions. The last part of this work is devoted ta the study of the influence of water on the stability, because very often changes of hydraulic conditions in a soil mass lead to its failure ; first we review the successive evolution of each parameter ruling the stability according to the changes of the hydraulic conditions, in order to discuss the relative importance of their influence on stability ; a short paragraph is then dealing with the behaviour of water in the ground according to its nature and geological structure, in order to point out which hydraulic problems are given birth to by different types of grounds ; at last we wanted to figure out the influence of the position of a ground water table inside a slope, on the stability of the latter ; the variation of the probable safety factor of the slope has been figured out according ta its height and its gradient, to the position of the ground water table, to the density, the cohesion and the internal friction of the sypposedly homogeneous soil; which leads to a design charts book which is inserted at the end of the book.