Mackenzie has defined a register as a series of tones of like quality producible by a particular adjustment of the vocal cords. Mills defines register as a series of tones of a characteristic clang, timbre, color or quality due to the employment of a special mechanism of the larynx in a particular manner. Both definitions practically mean the same thing. What I object to in them is their use of the word "quality," and Mackenzie's limitation of the adjustment to the vocal cords and Mills' to the larynx. The adjustment takes place throughout the entire vocal tract. Indeed, one of the claims I make for this book is, that it does not limit the voice-producing factor to the vibrations of the vocal cords, but while recognizing the importance of these, also considers the importance of the rest of the vocal tract in relation to them. Other writers hold that voice is produced solely by the vibrations of the vocal cords, and that the rest of the vocal tract is concerned merely with determining the timbre of the voice. But I do not limit the function of the vocal tract below and above the cords simply to voice quality. To produce a given tone requires not only vibration of the cords but an adjustment along the entire tract and especially a change in the size and shape of the cup space. If one wished to be exasperatingly accurate one might say that each adjustment constituted a register, and that in every voice there were as many registers as there are tones. But surveying the progress of the voice up the vocal scale, and as a whole, it is found that up to a certain point the general character of adjustment within the vocal tract is the same, that beyond that point there is a change to another adjustment of a general character, and further beyond still another--in other words, that there are three registers. Some writers recognize only two physical changes in the mechanism of the vocal tract and consequently only two registers instead of three. They dispense entirely with the middle register because the general change there in the adjustment within the vocal tract is not, in their opinion, sufficient to determine a new register. In point of fact, however, while the lower vocal range calls the vocal cords into vibration along their entire length, and while for the highest range only a portion of the edges of the vocal cords vibrate, the adjustment for the medium tones shows a gradual change from the first condition to the third. It is a bridge by which the voice crosses in safety from the lowest to the highest register--a register of transition, but a register withal.