bed itself, but is also the initial background on which dynamic loads due to seismic effects are superposed. The mechanism responsible for interaction between the body of the dam and the rock bed ultimately reflects the seismic safety of the structure and requires increased attention, especially in the stage when the entity is brought on line. Of paramount significance here are experimental observations of the mechanical processes that take place both in the body of the dam, and in the rocks of the bed [1]. It is important to compare computed estimates obtained in design with actual data on the one hand, for example, to adapt the three-dimensional finite-element model of the dam to actual conditions. On the other hand, empirical information on the stress-state of specific segments will contribute to the database of the system used for geomechanical monitoring of processes occurring in the zone of the shore abutments under cyclically varying loads caused by the regime under which the reservoir is filled and drawn down. A program for complex experimental investigation of the stress-strain state of the rock bed of the high dam for the Sayano-Shushenskoe hydroelectric power plant* calls for, among other things, stationary observations by several geophysical (ultrasonic profiling, seismic inspection) and geomechanical methods, including the hydraulic rupture of rock in a hole. In our study, we discuss the results of two series of experiments conducted by the method of hydraulic rupture of holes in a section of the left-bank abutment Of the dam when the reservoir had been filled close to, respectively, the maximum 540-m (October 1996) and minimum 500-m (May 1997) elevations.t The procedure used for the experimental studies, the makeup of the equipment, the recording apparatus, and the method employed for data interpretation are described in [2]. The selection of this method was not random and was predetermined by its correspondence to a series of requirements set forth for long-term stationary observations: firstly, by the repeatability of measurements, i.e., the possibility of taking measurements at the same point of the mass as external conditions vary; and, secondly, by the serviceability under rather complex conditions where the water content of the rocks is high and the natural jointing of the mass is increased; these conditions are critical for the majority of familiar methods based on strain measurement. Attention as also focused on the scale of the method (the possibility of conducting measurements at a distance to tens of meters with a measurement base of the order of 0.5 m) and a quality rather rare for mechanical methods: direct measurement of a parameter possessing the dimensions of stress; this precludes additional determination of some deformation constants of the rocks. Of course, the question concerning the sensitivity and resolution of the shaft variant of equipment, which is intended for operation in deep underground workings remains unanswered as applies to different rock and hydrogeological conditions. True, preliminary estimates of the expected stress level of the order of 10 MPa in sections of the shore arch supports, which had made it possible to calculate a positive result, and also the data derived from a trial series of experiments with the