580 Mr. Tennant’s Account of an easier Mode 
to protect it from the air, was exposed to a strong heat, wiiilst 
the upper part was kept cool, and upon opening the barrel, 
when it had become quite cold, it was found that potassium 
had sublimed into the upper part. The potassium, however, 
so produced, though it burnt on contact with water, had not 
the purely metallic aspect of that formed in the common 
mode. It was of a more dusky appearance, resembling a mix- 
ture of some black powder with potassium. As it seemed 
probable that some of the potash had risen along with the 
])Otassium, I repeated the experiment with attention, to heat 
the barrel to a greater length, so as to force the potassium to 
rise further from the ingredients below, but the potassium 
formed with this precaution had nearly the same dusky ap- 
pearance as before. 
After trying different means for obviating this imperfection, 
I found the following to be quite effectual. Into the upper 
part of the barrel a narrower piece, nearly fitting it, was in- 
serted, open only by a perforation at the lower end to admit 
the^ vapour of the potassium to pass into it, and upon dis- 
tilling potash and iron turnings with this addition, the potas- 
sium rose into the narrow tube, quite pure, with its usual 
brightness. 
The most convenient dimensions of the apparatus, are for 
the external barrel to be about a foot and a half long, and the 
internal one about seven or eight inches. The latter should 
not be wholly inserted in the former, but about an inch of it 
left out for the greater ease in withdrawing it. The width is 
in general determined by that of a common gun barrel, but 
may be increased to a certain degree. I have had the thick 
part of a gun barrel so much enlarged by hammering it thinner. 
