To explain the use of the fourth tube, I must enter into some discussions. Suppose the vessel LMNO, Pl. VIII. Fig. 1. full of water, and the jar A partly filled with gas, and partly with water; it is evident that the weights in the bason P may be so adjusted, as to occasion an exact equilibrium between the weight of the bason and of the jar, so that the external air shall not tend to enter into the jar, nor the gas to escape from it; and in this case the water will stand exactly at the same level both within and without the jar. On the contrary, if the weight in the bason P be diminished, the jar will then press downwards from its own gravity, and the water will stand lower within the jar than it does without; in this case, the included air or gas will suffer a degree of compression above that experienced by the external air, exactly proportioned to the weight of a column of water, equal to the difference of the external and internal surfaces of the water. From these reflections, Mr Meusnier contrived a method of determining the exact degree of pressure to which the gas contained in the jar is at any time exposed. For this purpose, he employs a double glass syphon 19, 20, 21, 22, 23, firmly cemented at 19 and 23. The extremity 19 of this syphon communicates freely with the water in the external vessel of the machine, and the extremity 23 communicates with the fourth tube at the bottom of the cylindrical vessel, and consequently, by means of the perpendicular tube st, Pl. IX. Fig. 4. with the air contained in the jar. He likewise cements, at 16, Pl. VIII. Fig. 1. another glass tube 16, 17, 18, which communicates at 16 with the water in the exterior vessel LMNO, and, at its upper end 18, is open to the external air. By these several contrivances, it is evident that the water must stand in the tube 16, 17, 18, at the same level with that in the cistern LMNO; and, on the contrary, that, in the branch 19, 20, 21, it must stand higher or lower, according as the air in the jar is subjected to a greater or lesser pressure than the external air. To ascertain these differences, a brass scale divided into inches and lines is fixed between these two tubes. It is readily conceived that, as air, and all other elastic fluids, must increase in weight by compression, it is necessary to know their degree of condensation to be enabled to calculate their quantities, and to convert the measure of their volumes into correspondent weights; and this object is intended to be fulfilled by the contrivance now described.