Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Animal Science (Animal, Dairy, and Poultry Sciences)

First Advisor

John K. Cullen, Jr


Since its introduction in 1974, Johan Sundberg's model of the laryngeal system as the resonance source of the singer's formant (Fs) has gained wide acceptance. There have heretofore been no studies directly testing its validity in vivo. The purpose of this study was to undertake a direct test of that hypothesis, utilizing as subjects professional male singers trained in the western Classical tradition. The vocal behaviors of three trained singer-subjects were evaluated during modal and pulse register phonation via magnetic resonance imaging (M.R.I.), strobolaryngoscopy, and acoustical analysis. Dr. Sundberg's hypothesis rests upon two premises: (1) that the laryngeal system is acoustically isolated and therefore capable of independent resonation during artistic singing, and (2) that the laryngeal ventricle contains an air volume adequate to function as the volume element of the proposed two-tube resonating system (Sundberg, 1974). Results of the above analyses revealed that none of the subjects achieved the requisite 6:1 laryngopharynx:laryngeal outlet area ratio to support acoustic isolation and independent resonation of the laryngeal system. Further, subjects demonstrated robust and stable singer's formants in pulse register phonation concomitant to the occlusion of the laryngeal ventricular spaces as documented by M.R.I. Therefore, these data indicated that the subjects' behaviors do not fit the model of the laryngeal system as the resonance source of the singer's formant, and that the model is inadequate to account for the generation of the singer's formant in these three subjects. Further analysis of these data suggested that the singer's formant is resolvable into two component formants, termed Fs1 and Fs2. These formants are apparently analogous to F4 and F5 of speech, but are approximated by the singer to produce the desired high amplitude energy concentration. It was hypothesized that Fs1 arises from excitation of the fourth natural mode of the quarter wave resonance of the vocal tract by the optimized voice source of the trained singer. Application of this model to data obtained in this and previous studies reported in the literature predicted the frequency locus of Fs1 with an accuracy of 92-100%.