i^*6 The American Geologist. August, i89& 
scopic crystals seen on the surface of the pyrite must be gold. 
This method of treatment is the same as that which plays 
such an important role in the cyanide process as practiced in 
the Witwatersrand. The chemical examination thus leads to 
the same conclusion as the microscopical ; the gold is depos- 
ited on the pyrite and is of later origin. Since pyrite precip- 
itates gold quantitatively from almost all its solutions it is 
natural to reason as follows regarding the auriferous con- 
glomerate: Beds of pebbles composed of quartz and small 
pyrites crystals were involved in percolating siliceous waters 
carrying gold in solution. The gold was precipitated by the 
p,yrite and deposited upon it. The occurrence of gold in the 
absence of pyrite. which might have precipitated it, may be 
explained by assuming the presence of a small amount of 
easily soluble ferrous sulphate, which has also the power of 
precipitating gold from its solutions. f 
The source of the auriferous solutions remains to be ex- 
plained. It is well known that the advent of acid eruptives 
is alw^ays accompanied by silica-bearing waters which have in 
many cases extensively silicified the rocks through which they 
passed. That such solutions can carry metals is proven by 
the numerous contact veins and ore deposits with which we 
are familiar. And that they sometimes carry gold was shown 
above by reference to the gold content of geyser deposits. 
Now the schist complex of the Swazi formation as well as 
the Cape formation is cut by numerous dykes M'hich are some- 
times 30 meters in thickness. A number of these dykes have 
been studied microscopically by Dr. Koch. He distinguished 
among them quartz hornblende dioryte, quartz diabase, quartz 
proterobase (three dykes), quartz enstatite or bronzite ensta- 
tite, olivine diabase, olivine norite and tholeite. Faults and 
overthrusts are not uncommon in the whole gold jiroducing^ 
area. 
It is certainly not by chance that rocks of the dioryte group 
occur in close relation with gold districts in regions far re- 
*It .should not be forgotten that gokl occurs in quartz unassociated 
with pyrite. In that case we can only explain its presence by dilution 
of the original solution and precipitation of the gold. In most cases, 
however, pyrite was originally present, as is shown by the hexahedral 
cavities or by the brown stain on the quartz from which the pyrite has 
been removed by decomposition and solution. 
