The so-called "breaks" in the voice occur at points where one register passes into another; and it should be the aim of proper instruction in voice-culture to eliminate the breaks. They are due to the change in adjustment which each register calls for. The best method of "blending the registers"--of smoothing out the breaks--is to bring a higher register several tones down into the one below and thus bridge over the passage from one adjustment to another. To do this consciously would defeat its aim. It must be done in spontaneous response to the mental conception of the tone or phrase to be emitted. It must become second nature with the singer, a physiological adjustment in answer to a psychical concept--a detail, in fact one of the most important details, in that true physiology of voice-production which also takes psychical conditions into consideration. CHAPTER IX THE STROKE OF THE GLOTTIS The _coup de glotte_, translated as "stroke of the glottis," refers to the manner in which a note should be attacked. This matter of attack already has been covered by inference many times in the course of this book. For, as the effectiveness of vocal attack depends upon the way in which the air-column strikes the vocal cords, it follows that the advice constantly given and in accordance with which the entire vocal tract of the singer should adjust itself as if by second nature to the tone that is to be produced, each time places the cords in the correct position to receive the stroke of the outgoing air. It does away with all danger of the "audible stroke" which occurs most frequently on the very open vowel-sounds, when the air reaches the glottis too late and is obliged to force its way through, the result being a disagreeable click; and it also obviates the defect from the opposite cause, when the air passes through the glottis too soon and results in an aspirated sound, an H before vowels, the voice, for example, emitting "Hi" for "I". Mackenzie remarks on these points that the great object to be aimed at is that no air should be wasted or expended improvidently; that just the amount required for the particular effect in view must be used. Too strong a current tends to raise the pitch, a result which can be prevented only by extra tension of the vocal cords, which, of course, entails unnecessary strain. Again, the air may be sent up with such velocity that some of it leaks through before the glottis has time to intercept it; or with such violence as to force the lips of the chink a little too far apart. In either case so much motive power is thrown away and to that extent the brilliancy and fullness of the tone are lost. The _coup de glotte_, or exact correspondence between the arrival of the air at the larynx and the adjustment of the cords to receive it, is a point that cannot be too strongly insisted on.