406 On the Definite Proportions in which the [Dec. 



containing 33 cubic inches. The carbonic acid gas was absorbed 

 by caustic potash in a small glass exactly weighed, and represented 

 of its natural size in fig. 5. The increase of weight of this small 

 glass indicated the quantity of carbonic acid. 



The caustic potash employed in these experiments ought not to 

 be heated red-hot beforehand, otherwise it will contain too little 

 water for the formation of carbonate. It will soon become covered 

 with a hard impenetrable crust, which will put an end to the 

 absorption. We obtain it in a proper state when we evaporate it till 

 it becomes hard on cooling. We cannot employ quick-lime, which 

 absorbs scarcely any thing, while hydrate of lime lets go its water 

 in proportion as it unites with the acid, and this water condenses in 

 the glass jar. To determine exactly the weight of the small glass 

 containing the potash, I cork it well, and weigh it cork and all. 

 Then 1 take out the cork, and cover the mouth of the glass with a 

 thin piece of glove leather, which I fasten round its neck by means 

 of a silk thread. It is then immediately introduced into the mer- 

 curial jar. That I may be able to regulate the movements of this 

 little glass, I tie to its bottom a small iron wire, by means of which 

 I could withdraw it out of the jar, if that were necessary. The 

 glass jar which I employed had a stop-cock at its upper extremity, 

 to which I could attach a small air-pump. By this means I filled 

 the jar with mercury. 



When the decomposition was finished, I removed the tube, D, 

 fig. 6, from the mercurial trough, and wiped away the mercury 

 that might be attached to it. I then cut the neck of the tube 

 remaining in the fire at B by means of a file. Lastly, I separated 

 the receiver from the tube filled with muriate of lime. I now re- 

 moved the silk threads at A and C, and weighed the receiver 

 together with the piece of glass cut from the great tube. I then 

 removed this piece of glass, dried it thoroughly, and weighed it by 

 itself. By subtracting this weight from the increase of weight in 

 the receiver I learned the quantity of water contained in the 

 receiver. I then weighed the tube with the muriate of lime, which 

 never gained in weight more than O'l grain, and often a great deal 

 less. ( In this way I was sure of determining the whole of the water 

 formed during the combustion ; and if in some cases the experi- 

 ments have given inaccurate results, we are neither to ascribe this 

 to the apparatus nor the method, but to the substances introduced, 

 which were not absolutely dry. The only circumstance connected 

 with the method which can occasion any inaccuracy with respect to 

 the quantity of hydrogen is neglecting to pulverize the mass in a hot 

 mortar. Error may result also if the hands of the operator be in a 

 state of perspiration, or if he allows his breath to play upon the 

 mass which he is pulverizing. I have always taken care to keep the 

 mortar as far off as possible, and to wear gloves during the whole 

 arrangement of the experiment. We shall see by the analysis of 

 pxalic acid how delicate this analytical method is, and how exactly 

 \t points out the most minute quantities of hydrogen. In that 





