6B M. Berzelius 07i Compou7Kh 
the boiling temperature, it lost the water of crystallisation, be^ 
came opaque and of a milk-white colour, but did not change its 
form. At a more elevated temperature it softened into blisters, 
which gave out carbonic acid gas. By roasting, it was half 
melted, and acquired a strong alkaline taste. 
This salt was analysed in tlie following manner. It was intro* 
duced into a bulb of glass, formed at the end of a barometric 
tube. Tlie bulb bad been v/eighed when empty, and its weight 
when charged, served to determine tlie weight of the salt which 
it contained. The tube was then softened, at the distance of 
an inch from the bulb, and drawn into a very slender shank, 
wdiich, being bent at the same time, formed a species of cornute. 
It was next attached, by means of a tube of elastic gum, to 
a SKjall recipient made like an enameller’s lamp. This re- 
cipient ended in a shank equally slender, and was connected 
by an elastic tube, with a somewhat larger recipient, which w^as 
filled with dry muriate of lime. These two recipients, with their 
attached tubes, ri^ere weighed. The salt was then heated in 
the bull) by the hame of a spirit-of-wine lamp ; and the heat 
was continued till the mass had been red, for a quarter of an 
hour. By this process, ail the water, and a part of the carbo- 
nic acid were expelled from the salt. The water was, of course, 
retained by the apparatus, wliile the acid escaped into the air. 
To determine the quantity of tliis water with all possible ac- 
curacy, the neck of the small cornute was cut by a dia- 
mond ; the carbonic acid which remained in the apparatus, 
■was extracted at the end of the recipient, and the tv/o reci- 
pients with their segment of the tube were weighed. To cut 
the tube is a necessary precaution, because the small drop of 
Avater that always adheres to the end of it, which enters the 
recipient, would otherwise escape our notice. The segment of 
tlie tube was then detached, dried and Aveighed ancAv. The 
difference between its Aveight and the v/eight last obserA^ed, 
gave Avhat the two recipients had gained by the addition of the 
Avater: 100 parts of the salt had yielded 31.24 parts of water. 
Of the same salt, 100 other parts, Avhen Aveighed, after being 
lieated to incandescence, in a crucible of platina, left 42.62 
parts of half-melted salt. Cold Avater, when added to this re- 
siduum, dissoked carbonate of potash from it, and left caustic 
