During the nineteenth century physiologists and clinicians developed several graphical recording systems for the mechanical registration of heart sounds. However, none of these replaced traditional methods of auscultation. The paper describes criticism of the aural sense as one of the driving forces behind the development of phonocardiography and analyses its variants from a technological and clinical perspective. Against the background of the physiological “method of curves,” the parameters that prevented the implementation of phonocardiography against overwhelming odds are highlighted. Contemporaries denied specific evidence beyond auscultation. Many clinicians also feared that the art of auscultation was being undermined by the new, reproducible mechanical methods. The paper argues that phonocardiography was on the one hand regarded as impractical in clinical settings; on the other hand—and even more important—implicit practices, tacit knowledge and cultural models fostered skepticism against the new method. The argument of “self-evidence”—often connected to medical images, curves, graphs or tables—was not valid for the visualization of cardiac sounds in the opinion of the promoters of acoustic heart sound registration and its individual interpretation. Rather, the acts of subjective hearing and objectively reporting what was heard seemed “self-evident” for pathophysiological characteristics and the development of a diagnosis. Therefore, auscultation and phonocardiography coexisted with different emphases. While auscultation remained the method of choice for a bedside diagnosis, phonocardiography played its role in differential diagnostics or research settings.