This work provides the experimental confirmation of a generalized theory for thermal diffusivity measurements of solid samples using the ‘‘mirage’’ technique presented by the authors in a previous paper [A. Salazar, A. Sanchez-Lavega, and J. Fernandez, J. Appl. Phys. 65, 4150 (1989)]. Moreover the influence of different effects on such determinations has been analyzed, and the limits of validity of the method have been established. The most important influence relies on the finite nature of two parameters: the height of the probe beam above the sample surface h and the radius of the exciting beam a. Such effects in all solids are discussed in detail according to their bulk thermo-optical properties and for two ways of measuring: ‘‘bouncing’’ (probe beam sent to the sample surface at a small angle) and ‘‘skimming’’ (probe beam parallel and grazing the surface). The first way of measuring gives reliable results while the second one must be managed with care. In any case, significant shifts in the thermal diffusivity measurement for nonzero values of h and a are specially observed when the thermal diffusivity of the sample is much lower than that of the deflecting medium.