DETERMINATION OF ALKALIES IN MINERALS. 207 



the method about to be mentioned was contrived. It recom- 

 mends itself both on account of its simplicity and certainty 

 of operation. 



24. Having some time back noticed the decomposing effect 

 produced by heating sal ammoniac with nitric acid, the nature 

 of the decomposition was investigated to see how far it could 

 be made use of to decompose entirely the sal ammoniac. The 

 result of the investigation was that the sal ammoniac could 

 be conrpletely decomposed at a low temperature into gaseous 

 products, and it was immediately adopted in my analytical 

 process with the greatest satisfaction, both as to accuracy of 

 results as well as economy of labor.* 



^Formation of almost pure Protoxide of Nitrogen by the action of Nitric 

 Acid on Sal Ammoniac. — The experiments made with the nitric acid heated 

 with sal ammoniac to test the character of the decomposition have resulted 

 in the discovery of a new method for procuring protoxide of nitrogen with 

 the aid of a very low temperature. Among the experiments the following 

 were quantitative. Two grammes of sal ammoniac were placed in a glass 

 flask, and half an ounce of nitric acid poured upon it; the flask was connected 

 with a small wash-bottle containing a little water, and from this latter a tube 

 passed into a pneumatic trough filled with hot water; heat was applied to 

 the flask, and before the temperature reached 140° Fah. a gas began to be 

 given off, and at 160° it came off rapidly, and continued to do so after the 

 lamp was withdrawn. A small amount of red fumes appeared in the flask 

 that were condensed in the wash-bottle. The gas that passed over was col- 

 lected in a receiver, and measured one thousand and eight cubic centimetres. 

 The gas smelt of chlorine. The flame of a candle burnt with an increased 

 brilliancy when introduced in it. The candle was reignited when extin- 

 guished if a burning coal remained on the end of the wick. ISTo red fumes 

 were formed when it came in contact with the air, and the gas was absorb- 

 able by cold water. The properties were those of protoxide of nitrogen. 

 In another experiment the gases were collected at different stages of the 

 process, in vials over hot distilled water, and a solution of caustic potash 

 introduced and shaken up for some time. This latter was subsequently ana- 

 lyzed for the chlorine it absorbed, and in three different portions, collected 

 at the beginning, middle, and end of the process; the proportions of the 

 chlorine to the whole bulk of the gas were one fifty-seventh, one twenty- 

 ninth, and one sixteenth. The amount of protoxide of nitrogen due to the 

 ammonia in two grammes of sal ammoniac and its equivalent of nitric acid 

 is eight hundred and eighty-seven cubic centimetres. The gas freed from 

 chlorine, on being shaken up with cold water for some time, was found to 

 be almost entirely absorbed by the water. What remained was a mixture 

 of nitrogen and a little air. Some nitrous or hyponitrous acid forms during 



