Respecting the quantity of oxygen in the red oxide, Lavoisier finds 9 oxygen to 100 lead, Thomson 13.6, and Berzelius 11.55. This last is partly from experience and partly from a supposed analogy, that the successive oxides of the same metal contain oxygen as 1, 1½ and 2 respectively; and having found (I believe) correctly, that the brown oxide contains just twice as much oxygen as the yellow, this ingenious and generally accurate author adopts the theoretic inference in this instance at least prematurely, and concludes the red oxide is the mean between the yellow and the brown. But we must appeal to experience. It has already been stated that when red lead is exposed to heat, oxygen gas is given out, and it may be added, a small trace of water; and yellow oxide remains. This experiment requires considerable skill. If too great a heat is used, a part of the lead is reduced or revived as it is termed; if too little heat, then a part of the red lead remains unaltered. In performing this experiment I use a small clean iron spoon to hold the red lead, and cover it by another iron spoon; the whole is then held by a pair of tongs in a red fire till the spoon exhibits a uniform moderate red, and some time after. It is then withdrawn and cooled, and the oxide weighed. The average loss of weight is nearly 2 grains per cent. If only 1 grain or less, a considerable portion of red oxide remains mixed with the yellow; if 3 or more grains, then the margin of the oxide exhibits particles of lead amounting to ⅒, less or more, of the original weight; this can be easily separated from the oxide if necessary, but it is apt to adhere to the iron; when red oxide remains, it is so mixed with the yellow as not easily to be separated, but its quantity may be determined by nitric acid, which dissolves the yellow, and ⅘ of the red, leaving a residuum of brown oxide, from which the quantity of red is inferred. Now if the loss of weight of 100 red oxide be only 2 grains, and a part of that be water, it is impossible that 115.55 should lose 3.85 grains of oxygen, according to Berzelius. Another experiment, equally decisive of the question, is to determine the quantity of oxygenous gas to be obtained by heat or acids from a given weight of red lead. In one experiment made with great care, 500 grains of red oxide gave 6 grains of oxygenous gas by sulphuric acid; in another, 200 yielded 2½ grains. In order to vary the mode of determining the quantity of oxygen, into 210 measures of test green sulphate of iron solution, (1.156) = 16.8 green oxide, put 160 grains of minium; to this was added dilute muriatic acid more than sufficient for the minium: The oxymuriatic acid from the oxygen of the minium was instantly seized by the oxide of the iron, the whole of which was found by precipitation to be changed from green to red and an excess of oxymuriatic acid appeared. Now 16.8 oxide would require 1.86 oxygen to become red, which it must have acquired