In order to assess the economic value of a geothermal field, it is important to know the temperature distribution and in particular the temperature in the main part of the reservoir. A description of the development of a chemical geothermometer based on the concentrations of the four main cationic constituents of a ground water, namely Na +, K +, Ca 2+, and Mg 2+. Chemical composition and underground temperature data for a number of well water samples were assembled. The list includes samples from low-, medium- and high-temperature reservoirs. Assumptions were adopted which permitted the formulation of a simple linear relationship between the reciprocal of the absolute temperature, and the concentration of cation species in the ground water, for any hypothetical rock-solution cation exchange reaction. A systematic, computerised procedure was devised, which analysed a large number of possibilities for the rock-solution cation exchange reaction, and for partitioning of the data into two, three and four subgroups based on the composition parameters √ [ Ca 2+] [ Na +] and √ [ Mg 2+] [ Na +] . The problem of the mixing of geothermal and cold shallow waters was considered from a prospecting standpoint. A dilution-intensitive geothermometric expression, to be applied to samples showing ‘preliminary signs’ of a geothermal origin, was developed and integrated within the geothermometer. The resulting geothermometer is able to predict the reservoir temperature, based on the composition of relatively dilute hot-spring waters. Moreover, the new geothermometer shows a fit to the well water data, superior to that of published geothermometers based on cation composition of natural waters, over the entire temperature range.