42 Mr. Davy’s Lecture on some new analytical Researches 
I soon substituted for it the metal obtained by the action of 
ignited iron upon potash, in the happy method discovered by 
M. M. Gay Lussac and Thenard, finding that it gave the same 
results, and could be obtained of an uniform quality,* and in 
infinitely larger quantities, and with much less labour and 
expense. 
When ammonia is brought in contact with about twice its 
weight of potassium at common temperatures, the metal loses 
its lustre and becomes w T hite, there is a slight diminution in 
the volume of the gas ; but no other effects are produced. 
The white crust examined proves to be potash, and the ammo- 
nia is found to contain a small quantity of hydrogene, usually 
not more than equal in volume to the metal. On heating the 
potassium in the gas, by means of a spirit lamp applied to the 
bottom of the retort, the colour of the crust is seen to change 
from white to a bright azure, and this gradually passes through 
shades of bright blue and green into dark olive. The crust and 
the metal then fuse together ; there is a considerable effer- 
vescence, and the crust passing off to the sides, suffers the 
brilliant surface of the potassium to appear. When the potas- 
sium is cooled in this state it is again covered with the white 
crust. By heating a second time, it swells considerably, be- 
comes porous, and appears crystallized, and of a beautiful 
* When the potash used for procuring potassium in this operation was very pure, 
and the iron turnings likewise very pure and clean, and the whole apparatus free 
from any foreign matters, the metal produced differed very little, in its properties, 
from that obtained by the Voltaic battery. Its lustre, ductility, and inflammabi- 
lity were similar. Its point of fusion and specific gravity were, however, a little higher 
it requiring nearly 130° of Fahrenheit to render it perfectly fluid, and being to 
water as 7960 to icooo, at 6o° Fahrenheit. This I am inclined to attribute to its 
containing a minute proportion of iron. 
