TRANSACTIONS OF THE SECTIONS. 
49 
Proof spirit dissolves from it a very minute quantity of a sub¬ 
stance which seems to be a gum resin. It is entirely decomposed 
by a red heat, in close vessels, and also by concentrated and boiling 
sulphuric acid, which reduces it to charcoal, and a substance ap¬ 
parently analogous to artificial tannin. 
The bistre, or colouring-matter, obtained from the turf is not af¬ 
fected by carbonic acid, nor by sulphuretted hydrogen, nor by proto¬ 
chloride of tin: strong nitric acid will not change its colour, although 
by long standing it is decomposed by it. Chlorine bleaches it slowly ; 
caustic alkalies redissolve it. It is scarcely bleached at all by the 
sun’s rays, nor does it when properly washed and dried show any 
tendency to deliquesce; it is therefore an excellent colour for pa¬ 
per-staining and other such purposes, as few common agents will in¬ 
jure it, and it can be readily removed from surfaces by an alkali. 
The proportions of useful products above given can only be con¬ 
sidered as approximations, having been deduced from experiments 
on a small scale; they would probably be much increased, and the 
relative expense of preparing the material reduced, if the process 
were carried on with greater quantities. 
On some singular Phcenomena of Flame from Coal-Gas . 
By R. Mallet. 
If an Argand gas-burner be lighted, and a conical tube of a cer¬ 
tain diameter be inserted concentrically within it, with its extremity 
entering a certain distance, within the burner, and, while the gas is in¬ 
flamed, a current of air be propelled through the conical tube in the 
same direction with the streams of gas, under certain conditions, the 
whole of the gas-flame will retract or be. drawn back between the 
internal surface of the burner and the external surface of the conical 
tube, and nothing whatever will pass forward but a stream of 
strongly heated carbonic acid and aqueous vapour. This very sin¬ 
gular phenomenon of the passage in opposite directions of two cur¬ 
rents in such close contact does not appear to be affected by the size 
of the burner, provided a certain proportion be preserved between 
it and the conical air-tube. The experiments were made with two 
burners chiefly, one of which was three quarters of an inch internal 
diameter and one inch and a half deep, measured along its axis, 
and the other seven sixteenths of an inch internal diameter, and one 
inch and three eighths deep. 
With these it was found that the retraction of the flame was pro¬ 
duced most perfectly in the case of the large burner by a tube of 
five sixteenths of an inch diameter, but yet took place to a certain 
extent until the diameter of the tube was reduced to one eighth of 
an inch, and in the case of the smaller burner it was most perfectly 
produced by an air-tube of three sixteenths of an inch diameter; 
