In this study we investigate the appearance of combination tones in violins. Most authors in recent times have emphasised that combination tones occur inside the ear exclusively (intra-aural). This assumption will be subjected to scrutiny based on evidence found in an empirical study in which combination tones were measured outside the ear (extra-aural).

Measurements were performed in which a violinist played two tones of a particular musical interval simultaneously. This was recorded and subsequently analysed using a Fourier Transformation. In addition to the partial tones of the primary interval, the resulting spectrum showed frequencies corresponding to combination tones. Similar measurements on the viola and violoncello also revealed the existence of extra-aural combination tones. Such frequencies may influence the timbre of simultaneous intervals played on string instruments. In another experiment the violin was excited using an electrodynamic mini-shaker with the aim of localising the origin of extra-aural combination tones. A newly devised tone matrix was used as a theoretical approach which computes all potential combination tones that may occur between any pair of partial tones. The detailed analysis of musical intervals by both the frequency spectrum and the tone matrix shows characteristic mirror and point symmetries in the partial tone structure. The discussion focuses mainly on the audibility of extra-aural combination tones and on ‘the combination tone 1’. This research opens up new perspectives and questions relevant for interpreters, composers, violin makers and violin acousticians.

Absolute pitch is a unique feature of the auditory memory which makes it possible for its possessors to recognize the musical name (chroma) of a tone. Six musicians with absolute pitch, selected from a group of 250 music students as best scoring in musical pitch-naming tests, identified the chroma of residue pitch produced by harmonic complex tones with several lower partials removed (residual sounds). The data show that the percentage of correct chroma recognitions decreases as the lowest physically existent harmonic in the spectrum is moved higher. According to our underlying hypothesis the percentage of correct chroma recognitions corresponds to the pitch strength of the investigated tones. The present results are compared with pitch strength values derived in an experiment reported by Houtsma and Smurzynski (1990) for tones same as those used in this study but investigated with the use of a different method which consisted in identification of musical intervals between two successive tones. For sounds comprising only harmonics of very high order the new method yields a very low pitch recognition level of about 20% while identification of musical intervals remains stable at a level of about 60%.