TRANSACTIONS OF THE SECTIONS. 51 
In this Table V is the volume of the gas, T its temperature in Centigrade 
degrees, B the height of the barometer in millimetres, and C the height of the 
mercury in the tube in which the observations were made. From these data, and 
the results of the previous action of the caustic potash and pyrogallic acid, it fol- 
lows that the composition of the gas was :— 
Marsh-gas (C H,).....+-.. nchioda ga see bans eed O 
Carbonic acid ..... ae WEA he 8) GOAL Ede ib dope a4 
Oxypen) 3 Mipsis. ts coms ohana oot aonedcon 1-06 
Nitroreha traits oslnat. 062 aw Resayhin opts on ghar (O 
The density of the gas (air=1) was found to be 0661, which corresponds nearly 
to the foregoing composition. The gas was inodorous, and contained no compound 
of carbon and hydrogen except marsh-gas. 
From this analysis it is evident that the gas formed in this subterranean sheet 
of water isin all respects the same as that which is produced in stagnant pools 
containing leaves and other vegetable matters, 
On an Aspirator. By Dr. AnpRuws, F.2.S. 
On the Joint Action of Carbonic Acid and Cyanogen on Oxide of Iron and 
on Metallic Iron. By I, Lowratan Bett, F.RS., FCN. 
In the operation of smelting iron the reducing agent for all practical purposes is 
carbonic oxide. The power this substance possesses of depriving an ore of iron of its 
oxygen is greatly weakened by the presence of the resulting carbonic acid. From 
my own experiments, it would appear that when one third of the carbon in the gases 
of a blast-furnace is raised to its highest state of oxidation, further action is so re- 
tarded as virtually to place a limit on the economy of fuel in the process in question. 
Besides this reducing property exercised by carbonic oxide, there is a second one 
which, like the former, takes place in the upper and cooler parts of the furnace, viz. 
the splitting up of itself in considerable quantities into carbon and carbonic acid by 
contact with iron. This rearrangement of elements is accompanied by an eyolu- 
tion of heat, and at the same time returns a quantity of carbon to the operation, 
Here also, when carbonic acid exceeds certain limits, this reaction ceases. 
In certain smelting-works in Austria, where charcoal is employed, white pig is 
Ge uest with a smaller quantity of fuel than I would have supposed possible, 
eeping the law just mentioned in view. This difference might be supposed due to 
some peculiarity in the ore itself, which was of the spathose variety. It was found, 
however, when coke was substituted for charcoal in a properly constructed furnace, 
haying a height of 60 feet, the fuel required rose from 14 to nearly 24 ewt. to the 
ton of metal, which is something more than that required in this country. 
In a furnace in the county of Durham, where coke was exclusively used, I 
found cyanogen combined with sodium and potassium exist in more considerable 
quantities than had been hitherto suspected. Supposing it possible that these 
cyanides might be more abundant in charcoal-furnaces, where the fuel is richer in 
alkaline substances than is the case with mineral fuel, I had the following experi- 
ments performed in the laboratory of the Clarence Iron Works by Mr. Rocholl, to 
determine the effect relatively of carbonic acid in restraining the reducing and 
carbon-depositing powers of carbonic oxide and cyanogen. 
Carbonic acid and cyanogen, carefully prepared and dried, were introduced in 
measured quantities into a mercurial gas-holder. The specimens of oxide of iron 
were exposed in porcelain tubes to a current of the mixed gases, all air having been 
previously expelled. Heat was applied by means of a Hofmann’s gas-furnace, 
and the temperature of the interior of the tube was ascertained by means of the 
electric pyrometer of Dr. Siemens, which during the experiment did not indicate 
a greater fluctuation than 25°C, 
I formerly ascertained that when equal volumes of carbonic oxide and carbonic 
acid were passed over peroxide of iron at a red heat, it was impossible to remove 
oxygen balow that required to form protoxide of the metal; and in like manner, 
