92 Experiments with Potassium and Sodium. 
5. On charcoal, which is not mentioned by Serullas, sodium never 
fails to inflame, with brilliant scintillations. This is the mode which 
I adopt with most confidence, for firing sodium in contact with cold 
water. It confirms the truth of the reason given by Serullas, why 
sodium will not inflame.under the same circumstances as potassium 3 
namely, the superior temperature which the latter acquires, during its 
combination with the oxygen of the water: hence the necessity of 
placing the former on a bad conductor, in order to avoid the too ra- 
pid abstraction of caloric, which prevents a sufficient elevation of 
temperature for manifesting the phenomenon of combustion. 
6. It is commonly stated, that in the decomposition of water by 
sodium, pure hydrogen is evolved. This isa mistake. A portion 
of the metal, as in the case of potassium, combines with the hydro-— 
gen, as may be shown by the following experiment.—Take a globule 
of sodium, wrap it up in a small piece of paper, and introduce it un- 
der a small receiver provided with a stop-cock and jet, filled with 
water and standing over the pneumatic trough. The decomposition 
of the water will. be effected as usual, and sodiwretted hydrogen 
will be collected, which (on opening the jet attached to the receiver) 
being inflamed, burns with a characteristic bright yellow.flame. Po- 
tassiuretted hydrogen, obtained under the same circumstances, burns 
with a rose-colored flame fringed with blue. The potassium in sev- 
eral repetitions of this experiment always emitted light; the sodium 
did not. 
7. A globule of potassium placed on a bath of mercury gradually 
. amalgamates with the latter, without any rotary motion, if the atmos- 
phere be dry ; but when breathed upon, it immediately acquires, as ob- 
served by Serullas, a very rapid revolving motion, which continues for a 
long time. The surface of the mercury becomes tarnished, apparently 
by the accumulation of minute particles of the amalgam formed, which, 
at intervals are seen to emerge from beneath the surface of the mercu- 
ry, and at some distance from the large globule. The surface of the li- 
quid metal, within a circle of half an inch to an inch in diameter, re- 
tains its brilliancy. The minute particles of amalgam, which I sup- 
pose to be the cause of the tarnish, seem to be repelled by the large 
globule of potassium, and, occasionally, as new accessions are made 
to them, they become singularly agitated, exhibiting somewhat of the 
appearance observed when a drop of vinegar, or of an acid, comes 
in contact with a drop of water. 
