Thermodynamic analysis of the conditions under which the environment has a physicochemical effect on the mechanical properties of solids (the Rebinder effect) gives grounds for the view that fracture in solids may be facilitated by irreversible chemical reactions taking place on their surfaces [1, 2] as well as by the action of related (similar in molecular nature) liquid media. This permits examination from a common viewpoint of an extensive series of processes in which the environment affects the mechanical properties of solids: from purely physical adsorption to purely chemical reaction, for example, during the stresscorrosion cracking of metals and alloys. It is an essential condition for these processes that the energy of the solid-environmen t reaction should be commensurate with the surface energy [3], similar to the situation occurring when a solid is in contact with a related environment [4]. The purpose of our work was the experimental study of the effect of a chemically active environment on the mechanical properties of a solid material under conditions as favorable as possible for fracture to be facilitated as a result of surface reactions. Forced localized fracture (scratching with a diamond point) makes it possible to compare the results obtained with the facilitation of metal machining in the presence of metallic fusions previously studied in [5], a classic example of a reduction in strength resulting from a crop in surface energy; the system studied (brass solutions of ammonia and its derivatives) provides an opportunity for comparing the facilitation of fracture discovered with the patterns of stress corrosion. The investigation was carried out with a specially designed pendulum-type scratch sclerometer. The linear dimensions of the scratch (its width for the most part) were recorded at a constant level of dissipated energy (identical initial amplitude of pendulum swing), as was the speed of pendulum oscillation damping, which is proportional to the coefficient of friction (the linear reduction in amplitude with time characteristic of dry friction was established by preliminary experiments). The relative percentage increase in scratch width was taken as the measure of the effect of the environment on strength. During the experiments, the concentration of ammonia and of bivalent copper in the solution was varied over a wide range. The test agent was prepared by mixing solutions of ammonia and of copper salt of identical percentage concentrations in specified proportions. The liquid obtained was applied by pipette to the carefully polished surface of a brass plate on which a diamond cutting tool connected to the pendulum had made a scratch in about 100 consecutive passes; the cutting tool was pressed against the plate by a constant force. The test results showed that solutions of cuprammonium complexes in specified concentrations could greatly facilitate the fracture of alpha brass (Fig. 1). In these circumstances an increase in the coefficient of friction can be detected in this concentration region. Pure ammonia does not facilitate fracture; what is more, at high ammonia concentrations there is a considerable drop in friction and a reduction in the depth of the scratch as a result of tool slippage. It was necessary to study the connection between the data obtained by us and the composition of the solutions used in the experiments, because contradictory opinions are expressed in the literature on stress corrosion of alpha brass in cupramm0nium solutions as to precisely which ammonia complex is responsible for the fracture.