THE RUSSELL PROCESS IN AUSTRALIA. 313 
heated by an open fireplace—a method which cannot be recom- 
mended. The dried precipitates were shipped to Swansea for sale. 
The sulphide of sodium used for precipation was made exactly 
as recommended by Russell. 
The mill, as a whole, worked well, very little handling being 
required. No difficulty was encountered in carrying out the process, 
and the workmen soon learned their duties. Since the mill was 
closed, the writer has visited the Marsac Mill, at Park City, and 
he was pleased to find that, in most respects, the Rivertree Mill 
compared favourably with that home of the Russell process, and, 
in some points, was decidedly its superior. 
The mill was worked at intervals during 1893 by the mining 
company which built it. The extraction from the roasted ores 
was reported to have been good, but the losses during roasting 
were said to be very heavy. At the latter end of 1894 the mill 
came into the hands of the writer, who tried to work it as a 
customs mill, by buying ore from the miners on a fixed scale of 
charges. An etfort was made to obtain accurate statistics of the 
work so done, and also to gain some further insight into the 
chemistry of the process. With this in view, Mr. E. 8. Simpson, 
B.E., a graduate of the Sydney University Mining School, under- 
took a series of analyses of the ore treated and the various 
products. 
Unfortunately, the failure of the ore supply compelled the 
closing of the mill, and curtailed Mr. Simpson’s work, the results 
of which, however, so far as they go, will be given by that gentle- 
man in the second part of this paper, which deals with the chemical 
side of the process. The present writer will confine himself to the 
practical and metallurgical results of the operations of the mill, as 
a customs mill, during the year 1895. The ore was bought upon 
assay, the sample being taken after the ore had passed through 
the rock-breaker, by the automatic sampling device before men- 
tioned. For medium and low grade ores this method was found 
satisfactory ; for high grade ores it gave results which were found 
untrustworthy. Automatic sampling of high grade ores has been 
abandoned at customs mills elsewhere, owing to similar experiences. 
The ore treated was of higher grade, and much less siliceous than 
the ore supplied for the experiments in 1891, and was not by 
any means an ideal ore for the process. Its composition is given 
in Mr. Simpson’s paper. The high percentage of lead necessitated 
a very careful roasting, in order to prevent sticking, and doubtless 
added to the loss by volatilization, as the analyses show a great loss 
of lead in roasting. The sulphur in the ores existed in combination 
with the arsenic and zinc, and a portion of the lead and iron. The 
salt used was 12 per cent. of the weight of the ore. 
Owing to the variety of the ores bought the chloridising varied 
daily, and the results were obtained only after the ore had lain 
