332 Mr. Knox's experiments and observations 
if the mass be dried on the sand bath at a heat below red- 
ness, when it is afterwards diffused in water, the salts will be 
dissolved, and the iron and silica will be obtained on the 
filter. 
With respect to the alkalies, the fusion with nitrate of ba- 
rytes is a tedious process, and liable to loss from the frequent 
filtrations, and also from the necessity of evaporating off the 
sulphate of ammonia. 
That of Sir Humphry Davy, by boracic acid, is much neater ; 
but, probably from want of analytical skill, I found it so 
difficult to get rid of the boracic acid, that I was not able to 
rely upon my result. 
The new method, by lead, has its embarrassments also, 
particularly from the fusibility of the glass which acts upon 
platina, and penetrates clay crucibles. There is also a good 
deal of difficulty in getting entirely rid of the lead. 
The process by acids, where the nature of the stone will 
admit of it, is obviously the best ; and I found that the nitric 
acid was to be preferred, as, by its peroxidizing the iron, and 
destroying the bitumen, those two troublesome substances 
were disposed of. 
Bitumen. 
It has appeared, from the foregoing analysis, that this sub- 
stance may be obtained both in the usual mode of analysing, 
and by distillation of the stone, but quite pure only in the 
latter mode. It seems to consist of two inflammable sub- 
stances, the one much more volatile than the other, but both 
inseparable from the stone, except at a heat approaching, if 
not entirely amounting to, whiteness. I imagine that it is in 
combination with the iron, as it seemed in general to accom- 
