The widespread natural occurrence of two-pyroxene assemblages and the temperature effect produced by the component redistribution between the constituent minerals have long been used to obtain temperature information, and quite a number of two-pyroxene geothermometers have been derived over the past decade or so. At first there was considerable optimism about applying these thermometers to determine the physico-chemical conditions of natural mineral formation. However, as new experimental data (especially on Fe–Mg minerals) were accumulated and more thorough investigations on particular geological objects became possible, it was found that various thermometers differed considerably and did not fit to the actual geological situation. Rightly, two-pyroxene thermometry was criticised and there were even some quite pessimistic conclusions (Bohlen & Essene, 1979, Lindsley, 1983, Saxena, 1983 and oth.) concerning its potentials. Indeed, a paradoxical situation has developed where the very abundance of temperature values yielded by the many types of two-pyroxene thermometers has resulted in their devaluation. Moreover, this leads to errors in the petrogenetic interpretation, especially when different complexes are compared on the basis of results obtained with different thermometers (depending on the authors’ preferences). Clearly, a situation has arisen where it has become necessary to evaluate critically all the thermometers available, choosing only those which yield more reliable and consistent results, relegating the rest to history. Only then will it become possible to achieve correct petrogenetic comparisons of the data obtained by different authors on different geological objects.