484 On some new analytical Researches 



fort and after the process, the volatile alkali employed having 

 been very dry, I found that it bad increased more than two 

 grains ; the rapidity with which the product acts upon mois- 

 ture, prevented me from determining the point with great 

 minuteness ; but I doubt not, that the weight of the olive* 

 coloured substance and or' the hydrogen disengaged precisely 

 equals the weight of the potassium, and ammonia consumed. 



MM. Gay Lussac and Thenard* are said to have procured 

 from the fusible substance, by the application of a strong 

 heat, two fifths of the quantity of ammonia that had disap- 

 peared in their first process, and a quantity of hydrogen and 

 nitrogen in the proportions in which they exist in ammonia, 

 equal to one fifth more. 



My results have been very different, and the reasons will, 

 I trust, be immediately obvious. 



When the retort containing the fusible substance is ex- 

 hausted, filled with hydrogen and exhausted a second time, 

 and heat gradually applied, the substance soon fuses, effer- 

 vesces, and, as the heat increases, gives off a considerable 

 quantity of elastic fluid, and becomes at length, when the 

 temperature approaches nearly to dull redness, a dark gray- 

 solid, which, by a continuance of this degree of heat, does 

 not undergo any alteration. 



In an experiment, in which eight grains of potassium had 

 absorbed sixteen cubical inches of well dried ammonia in a 

 glass retort, the fusible substance gave off twelve cubical 

 inches and half of gas, by being heated nearly to redness, 

 and this gas analysed, was found to consist of three quarters 

 of a cubical inch of ammonia, and the remainder of elastic 

 fluids, which, when mixed with oxygen gas in the proportion 

 of 6 j to fi, and acted upon by the electric spark, diminished 

 to 5\. The temperature of the atmosphere, in this process, 

 was 57° Fahrenheit, and the pressure equalled that of 30*1 

 inches of mercury. 



In a similar experiment, in which the platina tray contain- 

 ing the fusible substance was heated in a polished iron tube, 



* No notice is taken of the apparatus used by MM. Gay Lussac and The- 

 nard in the Moniteur; but, from the tenour of the details, it seems that they 

 must have operated in jjlas.3 vesseL in the wav hi retur'ore adopted over mercury, 



filled 



