Józef Michal Chominski's postulate of investigating the 'real sound' of a composition is the most essential aspect of the theory of musical sonology; yet the existing analytical literature regards it primarily as a metaphor. In Chominski's approach, the 'auditive shape of sound' provides the true basis for investigation, whilst its music notation records the projection of the composer's creative intention. However, the sonoristic re-definition of the functions of particular elements of music is carried out exclusively on the basis of the score. Its results are thus not fully satisfactory, because of the gap which exists between the etymology of the concept of 'real sound' (a physical-acoustic fact) and the nature of the source - the written text of a work which is being analysed (a symbolic-sign code). If the theory of sonology is to develop, it seems necessary to relocate the point of gravity, from an interpretation of the sound phenomena encoded graphically in the score to the issues of their acoustic shape and their psychologically conditioned perception. As a condition of adopting such an approach, one has to enter the area of empirical musicology and to identify the 'real sound' using the methods and tools appropriate to it. The model of a sonoristic analysis of a musical composition presented in the first part of the article is based on three kinds of sources: the score (the music notation of a composition), a recording (the acoustic text of a music composition) and a sonogram (a mathematical-information text), which is a form of making the real sound of a composition, taken from its phonographic recording, 'permanent' by means of a visual time-frequency representation. In the later part of the article the authoress presents an analytical study of 'Threnody to the victims of Hiroshima' by Krzysztof Penderecki. This work is one of the most representative examples of the avant-garde style of the 1960s 'Polish school of composition', which the composer himself strongly associated with the idea of music sonoristics/sonology. An analysis of those aspects of sonoristic regulation of 'Threnody' which are given above allows one to observe that the score notation and the spectral image of the work - what might be regarded as a visualisation of auditive perception - provide a very convergent picture, confirming the composer's belief that the 'graphic logic' is reflected in the 'sound logic'. Moreover, the construction of 'Threnody' has certain features of symmetry already apparent on the diagram of the shape and amplitude of the sound wave, which are confirmed on the basis of theoretical-musical criteria associated with the distribution of states of density and dilution of sound and the movement of sound events.