350 On PyrophoruSy and the making of Potassium. [May, 



serves perhaps to be known, though it did not yield me quite so 

 large a quantity as the apparatus of Gay-Lussac and Thenard, 

 which, however, does not yield so good potassium, and comes 

 much more expensive. 1 have made a good deal of potassium, and 

 found, 1. That the chief secret to obtain it in considerable quantity 

 is to use a very strong tire, in order to drive every particle of it 

 over. 2. To prevent the barrel from melting in sucii a high degree 

 of heat a good tube is wanted. 3. The banel, notwithstaneling 

 these precautions, will generally serve only once. 4. For this 

 reason all the contrivances, where there are different parts combined 

 bv grinding them together, or by means of screws, cpme too ex- 

 pensive, as they arc lost after having been used once. I used to 

 make my potassium in old gun-barrels, which were bent (like Gay- 

 Lussac's and Thenard's) in two places, and had no other piece added 

 to them. The middle part of this barrel is placed almost horizon- 

 tally (only a little elevated towards the open end) through a cylin- 

 drical furnace, and contains the iron turnings ; the thicker end, 

 which projects on one side from the furnace, is bent upwards, and 

 contains the potash. It can easily be tightened by screwing in the 

 breach, and putting a little clay into the touchhole. To the open 

 end, which projects on the other side from the furnace, and is bent 

 downwards, at first I adapted, as the French chemists did, a 

 receiver made of iron, from which a tube descended into a vessel 

 with quicksilver. I likewise, according to their prescription, 

 screwed into the upper part, containing the potash, an iron tube, 

 the end of which was placed under mercury also; but I found this 

 to be a very troublesome and expensive apparatus, as I could not 

 use it a second time ; nor did 1 always observe tlie gas coming out 

 through the quicksilver, as it would find its way out somewhere else, 

 either where the receiver was adapted, or where the pipes were 

 screwed in. Finding, therefore, these pipes quite useless, I closed 

 the upper end of the barrel, after the introduction of the potash, 

 entirely by screwing in the breech and putting some clay into the 

 touchhole, and instead of adapting to the lower end any receiver, I 

 put this end, open as it is, into a small iron bason, containing a 

 little vegetable oil previously well heated, in order to drive ofi' the 

 watery particles. (Petroleum could not be used for this purpose, as 

 it would be set on fire by the gas, and even by the heat of the furnace.) 

 When the formation of potassium was going on, one bubble after 

 the other of hydrogen gas holding potassium in solution rose 

 through the oil, and coming in contact with the air took fire, like 

 phosphoric hydrogen gas, but with a more considerable explosion. 

 The purest part of the potassium run down in drops into the oil, in 

 which it rose to the surface, from whence I took it with a spoon, 

 and put it into petroleum. The potassium so obtained did not 

 contain any iron, but was of a very bright lustre, much like silver, 

 whereas that collected in the barrel is generally alloyed with iron, 

 and of a bluish colour, more like lead. Living near an iron- 



