The weight of the furnace, ABC, is first accurately determined, then introduce the tube RS, of known weight, by the chimney, GH, till its lower end S, rests upon the grate, d e, which it occupies entirely; in the next place, fill the furnace with charcoal, and weigh the whole again, to know the exact quantity of charcoal submitted to experiment. The furnace is now put in its place, the tube, l m n, is screwed to that which communicates with the gazometer, and the tube, o p, to that which communicates with the bottles of alkaline solution. Every thing being in readiness, the stop-cock of the gazometer is opened, a small piece of burning charcoal is thrown into the tube, RS, which is instantly withdrawn, and the tube, o p, is screwed to the chimney, GH. The little piece of charcoal falls upon the grate, and in this manner gets below the whole charcoal, and is kept on fire by the stream of air from the gazometer. To be certain that the combustion is begun, and goes on properly, the tube, q r s, is fixed to the furnace, having a piece of glass cemented to its upper extremity, s, through which we can see if the charcoal be on fire. I neglected to observe above, that the furnace, and its appendages, are plunged in water in the cistern, TVXY, Fig. 11. Pl. XII. to which ice may be added to moderate the heat, if necessary; though the heat is by no means very considerable, as there is no air but what comes from the gazometer, and no more of the charcoal burns at one time than what is immediately over the grate. As one piece of charcoal is consumed another falls down into its place, in consequence of the declivity of the sides of the furnace; this gets into the stream of air from the grate, d e, and is burnt; and so on, successively, till the whole charcoal is consumed. The air which has served the purpose of the combustion passes through the mass of charcoal, and is forced by the pressure of the gazometer to escape through the tube, o p, and to pass through the bottles of alkaline solution. This experiment furnishes all the necessary data for a complete analysis of atmospheric air and of charcoal. We know the weight of charcoal consumed; the gazometer gives us the measure of the air employed; the quantity and quality of gas remaining after combustion may be determined, as it is received, either in another gazometer, or in jars, in a pneumato-chemical apparatus; the weight of ashes remaining in the ash-hole is readily ascertained; and, finally, the additional weight acquired by the bottles of alkaline solution gives the exact quantity of carbonic acid formed during the process. By this experiment we may likewise determine, with sufficient accuracy, the proportions in which charcoal and oxygen enter into the composition of carbonic acid.