The results of the calculations as a whole correspond to the results of field investigations of shear of rock pillars and confirmed the picture of failure, examined in detail in [3], obtained on the basis of their analysis. I. 2. . 4. 5. LITERATURE CITED O. Zienkiewicz, Finite-Element Method in Engineering [Russian translation], Mir, Moscow (1975). M. G. Zertsalov, V. A. Ivanov, L. S. Stepanova, and V. V. Tolstikov, "Investigation of the failure of a plate-foundation system under shear conditions with the use of the finite-element method," in: Application of Numerical Methods to Problems of Geomechanics [in Russian], MISI, Moscow (1986). Yu. A. Fishman, "Establishment of standard and design characteristics of the shear strength of rock masses," Tr. Gidroproekta, No. 68 (1980). C. Fairhurst, "On the validity of the 'Brazilian' test for brittle materials," Intern.On=site observations by geodetic methods became widespread at Soviet hydro developments in 1950 and, despite the labor intensity, as before they are among the necessary and most essential on-site observations in the overall complex.Geodetic measurements have expanded and deepened our notions about the settlement and displacement of dams and their foundations and have made it possible to reveal and evaluate a number of causes of deformations which were not known or were not taken into account in calculations, since their effect on the overall result was considered negligibly small. They include, for example, the convergence or divergence of the banks at the site during filling of the reservoir (on the Zeuzer arch dam with a height of 160 m convergence of the banks led to a large deflection of the crown section toward the upper pool and to the creation of an emergency situation, which was revealed only thanks to geodetic measurements). For a long time the rise of the foundation and banks during filling of a reservoir recorded by leveling at a number of hydro developments (Toktogul, Inguri, etc.) did not find an explanation. Only in recent years was it proved on the basis of calculations [12,13] that the phenomenon is due primarily to the effect of buoyant hydrostatic forces upon a change in the reservoir level.By now considerable experience has been gained in observations by geodetic methods and a vast volume of practical data has been obtained. However, due attention is still not being devoted to its study and generalization. During an analysis the difference between the measured and calculated deformations is usually estimated [2]; with the use of statistical methods one tries to reveal the relation between the total displacements and change in the reservoir level and seasonal fluctuations of the temperature of the concrete or even the air [5].But from geodetic measurements it is possible to obtain more diverse information about the causes and character of deformation processes. In particular, from the total deformations it is possible to isolate components caused by different factors. An analysis of these components enables a more reliable determination o...