THE RAND GOLD MINING INDUSTRY. 121 



to about iiO/ per ton for a plant crushing, say, 3,000 tons per 

 working day.* In the older plants much of the advance made 

 during recent years in economy and efficiency has been secured 

 by application of the foregoing principles, which has frequently 

 involved elimination of existing devices and appliances. Among 

 these are the entire elimination of stamp-mill amalgamated 

 plates, f the substitution of a few large steel cone diaphragm 

 classifiers for existing nests of numerous small pyramidal 

 wooden spitzkasten4 and the continuous collection of sand by 

 vacuum sand filter tables, § thus avoiding the cost of sand coUec- 

 ing vats for drainage and storage only. At the present time the 

 process of application and of full utilisation of the knowledge 

 already gained is more pronounced than any apparent impend- 

 ing new developments, though minor advances are continually in 

 progress. 



In considering the local conditions of gold extraction, to the 

 prime factors of continuity of large scale operations, the cheap 

 cost of coal for power, a healthy climate and accessible locality, 

 must be added the simple nature of the banket ore, and its 

 amenability to amalgamation and cyanide treatment. Consist- 

 ing, as the ore essentially does, mainly of silica with some com- 

 bined silicates and about 3 per cent, of pyrite, its constituents 

 offer few difficulties in treatment, and the chemistry of the 

 processes involved has been worked in its essentials. || The 

 possible greater compactness due to greater compression at in- 

 creased depth merely involves somewhat finer crushing, though 

 in no case has the proposal, based on experience on other fields, 

 to all-slime the ore in place of crushing to fine sand and slime 

 been found either necessary or economically desirable. On cer- 

 tain mines on the Eastern Rand portion of the gold appears 

 almost uniformly diffused in an extremely fine state of division 

 throughout the siliceous matrix, and in such case very fine, 

 though leachable sand, is desirable. Similarly, the various 

 costly attempts in the past to concentrate out the bulk of the 

 gold into a rich pyritic production of small weight and to discard 

 a valueless tailing has proved futile, since the gold and pyrite 

 are not proportionately distributed in the ore, and hydraulic 

 classification for the re-crushing of the tailing pulp ensures the 

 proportionately finer reduction of the specifically heavier pyritic 

 particles, which their value warrants. Attempts, accompanied 

 by considerable expenditure, for rapid continuous treatment of 

 the ore as one product, or of slime, have likewise failed. This 

 was mainly because a considerable time is required to dissolve 

 the gold particles in banket, and to separate the gold-bearing 



* Chairman's speecli at Simmer Deep Meeting, xSth March, 19T0; 

 also "Rand Metallurgical Practice," 2, 291, 2>2>7- 



\ Joiirn. Chem. Met. and Miii. Soc. of S.A. 11 (1911), 4i4- 



t "Rand Metallurgical Practice," 1, 99. 



^Journ. Chciii. Met. and Min. Soc. of S.A.. 10 (1909), 43. 



II W. Bettel : "The Cyanide Process on the Rand," in 21st Anniversary 

 No. of S.A. Mining Journal, (1912), 274. 



