Many existing studies of vocal fold geometry are based on anatomical measurements made on histologically fixed laryngeal tissues using formalin. However, the validity of these geometric data is questionable because of the potentially significant tissue deformation associated with formalin fixation, particularly tissue shrinkage. Previous experiments have shown that valid geometric data may be obtained on vocal fold tissue samples quickly frozen with liquid nitrogen. Based on this finding, the present study attempted to quantify the geometric deformation of formalin-fixed vocal fold tissues with respect to quick-frozen tissues. Six freshly harvested canine larynges were quickly frozen with liquid nitrogen and sectioned at the mid-membranous coronal plane. Each larynx was thawed and divided into halves along the midsagittal plane, one of which was not fixed; the other was fixed with formalin, from which histological tissue sections were also prepared. Measurements of vocal fold geometry were made on digital images of mid-membranous coronal sections of the tissue samples, based on linear dimensions of vocal fold depth and thickness defined functionally for biomechanical applications. Significant shrinkage of the vocal fold (particularly the vocal fold body) and considerable distortion of the vocal fold contour (particularly at the free edge) were observed for the formalin-fixed samples and the histological sections in comparison with the unfixed samples. Results of the present study suggested that significant geometric artifacts are induced by conventional histological fixation of laryngeal tissues using formalin. These artifacts should be carefully considered for interpreting any vocal fold geometric data obtained through formalin fixation.