344 New analytical Researches 



In tblsexpennicnt the heat wasapjilied much more slowly 

 than in any oF those in which the iron tube was used, and 

 even at the end of the operation the temperature was Ultle 

 more than that of cherry red. 



In the upper part of the stop-cock there was found a 

 minute quantity of gray powder, which gave ammonia by 

 the operation of moisture. 



In no case, in which I have used the copper tube in like 

 processes of slow distillation, has there been less than four 

 grains of potassium revived ; and the proportion of nitro- 

 gen to the hydrogen in the gas evolved has been uniformly 

 much greater than in processes of rapid distillation in the 

 tubes of iron ; but the whole quantity of elastic matter pro- 

 cured considerably less. 



Copper has a much stronger affinity for potassium* than 

 iron. It occurred to me as probable, that this atlraction^ 

 by preveutins; the potassium from rising in vapour at it^ 

 usual temperature, and likewise by the general tendency of 

 such com'uination to s;ive greater density, might occasion a, 

 diminution of its action upon the nitrogen in the nascent 

 state. Amniuaia has a strong attraction for the oxide of 

 copper, and it consequently is not unlikely that the fusible 

 substance may combine with metallic copper, and that this 

 compound may not be entirely destroyed in the distillation. 

 And assuming this, it may be conceived that the loss of 

 hydrogen partly depends upon some combinalion of the 

 basis of ammonia wiih copper. 



I had a tube, of the capacity of 2t cubical inches, made 

 of wrought platina, cemented by means of fine gold solder. 

 The fusible substance was obtained (as usual from six grains 

 of potassium) in a tray of platina, where it was brought in 

 contact with a large surface of platina wire ; the distillatiort 

 was slowly conducted ; but before the temperature of the 

 tube had approaclied to that of ignition, it dissolved and 

 gave way at the points where it was soldered, and a violent 

 conibusiion took place. Only seven cubical inches of gas 

 were collected; but of this, allowing for the hydrogen that 

 filled the uibe, nearly three-fifths were nitrogen. 



I am making preparations for performing the experiment 

 in a bored tube made from a single piece of platina, and 

 likewise in tubes made' of other metals, and I hope to be 



• Cooper heated in pot;i!:bium spcec'ily dissolves, and diminishes its fusi- 

 bility; but poiassium retjuiros :i white heat to enable it to combine with 

 iron. In another experiment, in which I distilled the fusible substance in 

 an iron tray, contained in the copper li:be, a considerable quantity of cop- 

 per, that had been dissolved, was fouifl in the state of powder deposited 

 upon the uay, or loose in the bottom of the tube. 



ablcj 



