Thus far in the present work, the usage of the terms empirical and scientific, accepted by vocal theorists generally, has been adopted. A distinction has been drawn between knowledge of the voice obtained through the study of the vocal mechanism and that obtained through observation of any other kind. The purpose will best be served by continuing this same usage. It must be apparent to the reader, from the analysis of modern methods, that no real Science of Voice has thus far been erected. This is due to the fact that the general principles of scientific investigation have not been applied to the study of the voice. Under the influence of the idea of mechanical vocal management the attention of all investigators has been turned exclusively to the mechanical features of tone-production. Meanwhile the empirical knowledge of the old masters seems to have been forgotten. As a matter of fact, as will now be seen, this empirical knowledge has never been lost. Every modern teacher of singing shares the empirical knowledge which formed the sole material of the old method. But this knowledge is not applied effectually in modern instruction for two reasons. First, modern teachers do not realize the importance of this knowledge; indeed, they are practically unaware of this valuable possession. Although in fact the basis of nearly all modern instruction in singing, empirical knowledge is always unconsciously used. Second, empirical knowledge is always applied in the prevailing mechanical spirit. The attempt is always made to translate the sub-conscious empirical understanding of the voice into rules for direct mechanical management. Under the influence of the mechanical idea the modern teacher's most valuable possession, empirical knowledge of the voice, becomes utterly unserviceable. Thus far, the whole result of this work has been destructive. The accepted Vocal Science has been shown to be erroneous in its conception and unsound in its conclusions. The work cannot halt here. Vocal Science must be reconstructed. This can be done only by following the general plan of all scientific investigation, beginning with the observation of all ascertainable facts bearing on the voice. How can any facts be observed about the voice other than by the study of the vocal mechanism? An answer to this question is at once suggested so soon as scientific principles are applied to the subject. Strictly speaking, the voice is a set of sounds, produced by the action of the vocal organs. The scientific method of inquiry is therefore to begin by observing these sounds. Sounds as such can be observed only by the sense of hearing. It follows then that the attentive listening to voices is the first step to be taken.