The experiments have not been sufficiently numerous to determine with precision which of these candles affords the most light at a given expence, but the few experiments which have been made seem to indicate, that the quantity of light is nearly as the quantity of combustible matter consumed, and thus a candle which is used in the manner pointed out gives more light than a candle of the same dimension set perpendicularly and snuffed, because one part of a candle that is snuffed, is thrown away, and another part flies off in the form of smoke. And this is not the only inconvenience that attends the using candles in this manner, and which the other method is free from, for the light which it gives is of a bad quality, on account of its being variable and undulating. From the time that a candle is snuffed till it wants snuffing again, its strength of light scarcely continues the same for a single minute. And that variation which frequently takes place in the height of the flame, is a matter of still more serious consequence. The flame of a long candle placed vertically when it is snuffed burns steadily, is about two inches high, but it very frequently rises to the height of four inches or upwards; drops down again in a moment, till it is less than three inches, and then rises again. In this manner the flame continues in motion for some time before it returns to its original dimensions. But it does not continue long in a quiescent state before it begins a new series of undulations. In this manner the candle burns till the top of the wick is seen near the apex of the flame, carrying off clouds of smoke. In this state of things the eye becomes uneasy for want of light, and the snuffers are applied to remove the inconvenience. Mr. WALKER further observes, that it is these sudden changes, and not the nature of candle-light itself, that do so much injury to the eye of the student and artist; and that that injury may be easily prevented, by laying aside the snuffers, and in the place of one large candle, let two small ones be used in the manner stated. The following observations on this subject are copied from the Monthly Magazine, 1805, p. 206. “It is scarcely necessary to observe, that the combustion of candles proceeds the quicker in proportion as the inclination is greater. From the experiments which I have made, I should consider an angle of forty degrees with the perpendicular as the maximum of inclination, beyond which several considerable inconveniencies would occur; and I should take 25 degrees as the minimum of inclination, less than which does not sufficiently expose the point of the wick to the action of the air.