How the practical method of the old masters came to be forgotten is perhaps the most mysterious feature of this puzzling system. There has been a lineal succession of teachers of singing, from the earlier decades of the eighteenth century down to the present. Even to-day it is almost unheard of that any one should presume to call himself a teacher of singing without having studied with at least one recognized master. Each master of the old school imparted his knowledge and his practical method to his pupils. Those of his pupils who in their turn became teachers passed the method on to their students, and so on, in many unbroken successions. Yet, for some mysterious reason, the substance of the old method was lost in transmission. What little is now known about the old method is derived from two sources, the written record and tradition. To write books in explanation of their system of instruction does not seem to have occurred to the earliest exponents of the art of Voice Culture. The first published work on the subject was that of Pietro Francesco Tosi, _Osservazione sopra il Canto figurato_, brought out in Bologna in 1723. This was translated into English by M. Galliard, and published in London in 1742; a German translation by J. F. Agricola was issued in 1757. The present work will call for several citations from Tosi, all taken from the English edition. Only one other prominent teacher of the old school, G. B. Mancini, has left an apparently complete record of his method. His _Riflessioni pratiche sul Canto figurato_ was published in Milan in 1776. Mancini's book has never been translated into English. Reference will therefore be made to the third Italian edition, brought out in Milan, 1777. Tosi and Mancini undoubtedly intended to give complete accounts of the methods of instruction in singing in vogue in their day. But modern vocal theorists generally believe that the most important materials of instruction were for some reason not mentioned. Three registers are mentioned by Tosi, while Mancini speaks of only two. Both touch on the necessity of equalizing the registers, but give no specific directions for this purpose. About all these early writers have left us, in the opinion of most modern students of their works, is the outline of an elaborate system of vocal ornaments and embellishments. On the side of tradition a slightly more coherent set of rules has come down to us from the old masters. These are generally known as the "traditional precepts." Just when the precepts were first formulated it is impossible to say. Tosi and Mancini do not mention them. Perhaps they were held by the old masters as a sort of esoteric mystery; this idea is occasionally put forward. At any rate, by the time the traditional precepts were given to the world in published works on the voice, their valuable meaning had been completely lost. Gathered from all available sources, the traditional precepts are as follows: "Sing on the breath."