It is practically helpful to the voice-producer and the teacher to think of the resonance-chambers and the ear as bearing a close relationship to the movements essential to tone-production. The sensations from these parts are of importance above all others in voice-production. They are the chief guides, and the attention may to advantage be concentrated on them. No doubt the question of registers for the speaker must be considered, but this can be done to greater advantage in a later chapter. SUMMARY. All good definitions of a register must recognize two things: change of quality in the voice, and change of mechanism in the vocal apparatus. A break is not a register, but occurs because of the existence of registers. The abrupt transition, or break, is to be avoided by covering, or modification of the upper tones of the lower (at least) register. For an adequate scientific examination of the question of registers, many qualifications are required in the investigator; and the student, when not an investigator, should endeavor to weigh the evidence presented so as to choose with caution from among conflicting opinions. He should be suspicious of those who scout the value of scientific study of this or any other subject, and also of those who claim that experience is of no importance in settling such a question. Though several well-qualified persons who have written on the subject differ in some respects, they are in agreement as to many of the more important points. They are practically all convinced that there is commonly a change of register for all voices, at or near one point in the scale (F), and that if this be practically disregarded, dangerous straining may result. Conclusions drawn from trained singers, alone, may be misleading. All classes of persons should be examined with the laryngoscope, if correct and far-reaching generalizations are to be safely made. The precision and rigidity of physics and mathematics cannot be introduced with safety into a subject of this character; otherwise the division and limits of registers will be fixed with a narrowness of margin that does not comport with Nature's methods. In all questions of register, the method of breathing--_i.e._, the nature of the application of the expiratory blast--must be duly considered. With male voices, the subject is usually considered much simpler than in the case of female voices. Men sing mostly in the chest register; basses and barytones wholly so, with the rarest exceptions. Tenors are taught to do so. Whether there might not be a subdivision of this register made to advantage in training, the author leaves as an open question; but about straining, in the case of tenors and all others, and as to the importance of recognizing three registers for female voices, there is in his mind no question. The fact that some may not be able to produce head tones does not justify carrying up the chest register to any appreciable extent, even by altos.