Reliance on the accuracy of the process cannot rest upon the supposition that the cyanide required for decoloration is proportional to the copper present, for varying quantities of ammonia salts, ammonia and water, and differences of temperature have an important effect. The results are concordant and exact only when the cyanide is standardised under the same conditions as it is used. It is best to have the assay solution and that used for standardising as nearly as possible alike, and to titrate the two solutions side by side. This demands an approximate knowledge of the quantity of copper contained in the ore and a separation of the bulk of the impurities. For the titration there is required a standard solution of potassium cyanide made by dissolving 42 grams of the salt, known to dealers as Potassium Cyanide (Gold), in water and diluting to one litre: 100 c.c. of this will be about equivalent to one gram of copper. For poor ores the solution may conveniently be made half this strength. The solution of the ore and the separation of the copper as sulphide are effected in the same ways as have been already described for electrolysis. Similarly, too, the sulphide is attacked with 15 c.c. of nitric acid and the assay boiled down to 10 c.c. Add 20 c.c. of water and warm, filter into a pint flask, wash well with water, and dilute to about 150 c.c.; add 30 c.c. of dilute ammonia, and cool. Prepare a standard by dissolving a quantity of electrotype copper (judged to be about the same as that contained in the assay) in 20 c.c. of water and 10 c.c. of nitric acid, boil off the nitrous fumes, and dilute to 150 c.c.: add 30 c.c. of dilute ammonia and cool. Fill a burette with the standard cyanide solution. The burette with syphon arrangement, figured on page 52, is used. A number of titrations can be carried on at the same time provided the quantity of copper present in each is about the same. This is regulated in weighing up the ore. The flasks must of course be marked, and should be arranged in series on a bench in front of a good light and at such a height that the liquid can be looked through without stooping. Supposing about 50 c.c. of cyanide will be required, 30 c.c. should be run into each, and each addition be recorded as soon as made; then run 15 c.c. into each. The solutions will now probably show marked differences of tint: add 1 c.c. of cyanide to the lighter ones and more to the darker, so as to bring the colours to about the same depth of tint. They should all be of nearly equal tint just before finishing. At the end add half a c.c. at a time until the colours are completely discharged. A piece of damp filter paper held between the light and the flask assists in judging the colour when nearly finished. Overdone assays show a straw yellow colour which deepens on standing. The following will illustrate the notes recorded of five such assays and one standard:--