448 
Wed GOR FE 
[Marcu 11, 1897 
the uranium opposite the insulated polished aluminium, a devia- 
tion of —84 sc. divs. from the metallic zero was found in about 
half a minute. After that the electrometer reading remained 
steady at this point, which we may call the uranium rays-zero 
for the two metals separated by air which was traversed by 
uranium rays. If, instead of having the uranium opposite to 
the aluminium, with only air between them, the uranium was 
wrapped in a piece taken from the same aluminium sheet, and 
then placed opposite to the insulated polished aluminium disc, 
no deviation was produced. Thus in this case the rays-zero 
agreed with the metallic zero. 
With polished copper as the insulated metal, and the uranium 
separated only by air from this copper, there was a deviation of 
about +10 sc. divs. With the uranium wrapped in thin sheet 
aluminium and placed in position opposite the insulated copper 
disc, a deviation from the metallic zero of + 43 sc. divs. was 
produced in two minutes, and at the end of that time a steady 
state had not been reached. 
With oxidised copper as the insulated metal, opposed to the 
uranium with only air between them, a deviation from the 
metallic zero of about +25 sc. divs. was produced. 
When the uranium, instead of being placed at a distance of 
one centimetre from the insulated metal disc, was placed at a 
distance of two or three millimetres, the deviation from the 
metallic zero was the same. 
These experiments show that two polished metallic surfaces 
connected to the sheath and the insulated electrode of an elec- 
trometer, when the air between them is influenced by the 
uranium rays, give a deflection from the metallic zero, the same 
in direction, and of about the same amount, as when the two 
metals are connected by a drop of water. 
THE EXTRACTION OF GOLD BY CHEMICAL 
METHODS. 
FLXCLUDING mechanical, smelting, and amalgamation pro- 
cesses, the methods of extracting gold from its ores may 
conveniently be grouped together under the heading of wet or 
chemical methods. In these, the gold is dissolved by some suit- 
able solvent, and is then separated from the unaltered ore by 
washing, and recovered by precipitation. The processes owe 
their origin to the rapid advance in the science of chemistry 
which has been made during the present century, and, although 
they are now of vast importance, and give results which would 
astonish our grandfathers, it is, perhaps, somewhat surprising 
that chemistry has not done more for the gold-mining industry. 
At the present day, the wet methods produce little more than a 
tenth of the total output of gold, while mechanical improvements 
in the old processes, made during the last half-century, are 
probably answerable for four or five times as much, 
Gold exists in nature practically only in one form, the metallic 
state, and the differences in treatment of the ores are necessitated 
by the variations in the physical condition of the metal, and by 
changes in the other constituents of the rock. Where the 
particles of gold are large enough to be seen by unassisted 
vision, they can usually be collected by means of mercury, and, 
on the other hand, are not dissolved in a reasonable time by any 
of the solvents of gold yet applied in practice. In these cases, 
therefore, chemical methods are not advantageous. Neverthe- 
less, it usually happens that some, if not all, of the gold in an 
ore is in an extremely fine state of division. It has recently 
been shown by Edman that a great proportion of the gold in 
American ores consists of particles less than y,/55 inch in 
diameter, and that some of these are less than ys, inch. 
Sometimes, gold in an ore is not visible even under the micro- 
scope, though readily detected by chemical means. Metal in 
such a condition is far more readily dissolved by a mobile liquid 
than by a viscous one like mercury, which does not wet the 
grains of sand between which the gold is hidden. Moreover, 
mercury may be prevented from doing its work by the presence 
of substances on which it exerts chemical action, such as the 
sulphides of antimony, or arsenic, or which protect the gold 
from its action by coating the particles with insoluble films. 
From such causes as these, it has long been recognised that 
the treatment of gold ores by mercury is very imperfect in a 
great many cases. The method is, speaking generally, unsatis- 
factory in extracting gold contained in pyrites or other sulphides, 
and it is in the treatment of these substances that the chlorina- 
tion process, now nearly fifty years old, has its main value. 
NO. 1428, VOL. 55] 
Chlorine is a somewhat slow solvent for gold, but the time 
occupied by it in dissolving the fine flakes existing in pyrites is 
not excessive. Unfortunately, chlorine has a strongly prefer- 
ential action on sulphides, and, to avoid the enormous waste of 
the gas which is entailed in the oxidation of a small percentage 
of these substances, it is necessary to precede chlorination by 
careful and complete roasting. Even in the rare cases, such as 
that of the Mount Morgan ore, in which the use of chlorine on 
completely oxidised ores is found to be desirable, the pre- 
liminary roasting is not omitted; as the peicolation of liquids 
through the roasted mass is far easier than through the raw ore. 
After roasting, there is little difficulty in the process. Oxides 
of the metals, except the alkaline earths, are very slowly 
attacked by chlorine ; and when the alkaline earths are present 
salt is added in the roasting furnace. Here one of the sources 
of loss in the process is encountered, chloride of gold being 
formed and volatilised at all temperatures above 200°, when 
common salt is mixed with the ore. In long-bedded furnaces, 
however, this loss is reduced to a minimum ; chloride of gold is 
prevented from formation by the presence of large quantities of 
unoxidised pyrites, and when formed, in the oxidised product in 
the hottest part of the furnace, it is in great part decomposed 
and re-absorbed during its passage over the bed of comparatively 
cool ore, which has just been charged into the furnace. 
It was formerly the universal practice to apply the chlorine to 
the slightly-damped roasted ore in the form of gas, and this 
method has never been entirely abandoned. Subsequently, 
after Dr. Mears had discovered that compressed chlorine was 
more rapid in its action than the same agent under ordinary 
atmospheric pressure, strong aqueous solutions were used, the 
ore being agitated with the solvent in revolving barrels. ’ This 
practice is still adhered to in several works in the United States. 
Elsewhere, however, it has been completely set aside. For 
example, at Mount Morgan, in Queensland, the largest chlori- 
nation mill in the world, stationary vats have been reverted to, 
aqueous solutions of chlorine being, however, still used. At 
this mill about 1,500 tons of ore are treated every week at a 
cost of about 18s. per ton, or little more than one-sixth of the 
value of the yield in gold. 
After the ore has been treated with chlorine for a period 
varying in different mills from an hour to one or two days, the 
liquid is filtered off and the gold precipitated by ferrous sulphate, 
sulphuretted hydrogen, or charcoal. As regards the relative 
advantages of these methods, it may be noted that charcoal only 
acts well with boiling solutions, and that sulphuretted hydrogen 
is now recommended by its advocates even when copper is 
present in the ore, Rothwell having recently pointed out that in 
acid solutions there is partial precipitation, all the gold being 
removed from solution before the copper begins to come down. 
The chlorination process, though perhaps unrivalled in the 
percentage of extraction which can usually be attained, labours 
under two serious disadvantages. Roasting the ore is often so 
expensive as.to be impracticable, and the silver is, in any case, 
all lost. Both of these disadvantages are avoided by the 
use of the cyanide process. This was introduced by MacArthur 
and the Forrests after prolonged researches, having for their 
object the discovery of some chemical process which would not 
requirea preliminary roasting of the ore. 
The action of cyanide solutions on the precious metals had 
long been known. Elsner had stated, in 1846, that the presence 
of air was necessary for the dissolution of gold or silver by 
potassium cyanide, and, subsequently, it was suggested that the 
action was represented by the equation 
4Au + 8KCy + O, + 2H,O = 4K AuCy, + 4KHO. 
This equation has recently been established by Maclaurin 
(Jour. Chem. Soc. vol. \xiii. (1893) p. 7243 vol. Ixvii. (1895) 
p- 199), who also showed that the dissolution of gold and silver 
becomes slower in proportion as free oxygen is more and more 
carefully excluded from the system. Thus, when a plate of gold 
was treated with a solution containing I per cent. of cyanide of 
potassium ina stoppered bottle filled with oxygen, the loss of 
weight was 0'24 gramme in 96 hours; in a shallow vessel 
exposed to the air, the loss was 0°00835 gramme in 24 hours, 
and in a flask, freed from air as completely as possible, the loss 
was only 0’0002 gramme in the same time. In addition, Mac- 
laurin prepared the curves of solubility of gold and silver in 
cyanide solutions, and showed that the maximum rate of dis- 
solution of both metals is reached at 0°25 per cent. of KCy, and 
